Top News SEOs & Audience Growth Experts to Follow in 2026

John Shehata has put together a handy directory of the people working on news SEO and audience growth around the world. A useful way to freshen up your reading list, although your LinkedIn feed may never recover. Most of what I learn about news SEO comes from people sharing what they’ve tested, and this is a quick way to find more of them.

Dan Petrovic: Is Apple building a search engine?

I’m a big fan of Dan Petrovic’s blogs and research into LLMs. He recently found that Applebot was the busiest crawler of his own website clocking in 73,232 of 190,662 requests - roughly twice as many as Baiduspider and well ahead of every AI crawler. I think the takeaway here is simply check your log files regularly, because there’s a lot of movement happening right now.

Dan Hinckley: Grade Every Page the Way Google’s Quality Raters Would

Elsewhere, Dan Hinckley has done something equally funky - and that’s funk in the good sense. He used an AI model to assess 1,600 pages against Google’s Quality Rater Guidelines in under two minutes, for less than $1. He plotted the pages by topic, with each point sized by its Search Console clicks, to highlight weaker pages that still attract traffic and others worth reviewing for removal. A really interesting way of approaching a potential content audit.

Sean B. on reviewing an SEO report written by Claude

This deserves a high five as it’s something I see ALOT. Sean reviewed a recent SEO report a client had produced with Claude. It looked pretty convincing, but its conclusions were completely wrong as it lacked any expert context. AI driven audits without an expert human in the loop can go badly wrong.

Myriam Jessier: Multimodal GEO

Myriam has done a wonderful job here on how to help AI find and interpret your images, video and audio. Her practical tips include keeping product shots and text overlays on screen for at least three seconds, allowing for models that sample video at around one frame per second. She also suggests using Google’s Vision API to check whether it identifies what you intended to show in your photos which is a great tip.

Why it matters: As AI search draws on images, video and audio, the way you shoot, script and label your work can affect how easily it is found and understood. Get multimodal!

Everyone wants to be your personal assistant - and media should notice

Florent Daudens explores what happens when personal AI assistants choose our news and read our newsletters for us.

Why it matters: If AI assistants are going to decide what reaches the reader, publishers will need to fully understand how these agents work.

AI Search KPIs: Measuring Success in the New SEO Era

In his lightning talk at Google’s Search Central Live Deep Dive, Eduard Blacquière argues that we focus too much on whether AI selects our content and too little on the steps that come before it. He proposes measures for five stages: whether AI knows you exist, associates you with a topic, finds you in the searches it runs to build an answer, retrieves your pages and, finally, selects your content.

Why it matters: These measures can help distinguish problems with recognition, crawling and content. Top job, Eduard!

The key to prompt tracking

Jan Ehrlinspiel at Peec AI explains why an AI visibility score can move even when nothing has changed. Drawing on an analysis of 120 million chats, he compares each prompt to a ‘biased coin’: some outcomes are more likely, but individual answers still vary. In his example, a daily score based on 50 runs can shift by up to 12.4 percentage points through chance alone. His advice is to track a wider range of prompts and assess changes over a month rather than a day.

AI Initiatives in News Media

This is a really useful tool for AI-tool enthusiasts and news junkies alike. Sergei Yakupov’s database now covers more than 1,300 AI initiatives from over 900 newsrooms in 112 countries. Filter by use, from content production to monetisation, to find real examples when someone asks what other publishers are doing.

Dan Hinckley: New SEO Tool, Quality Rater Jev

Dan Hinckley introduces Herringbone’s free Quality Rater Jev tool, which he says can assess a page against Google’s Search Quality Rater Guidelines in under a second. It examines the page itself, without checking the site’s wider reputation. Use it to identify pages that deserve a closer look.

Why it matters: Applying a 180-page set of guidelines to every article takes time. A quick first pass could help you prioritise the pages that need a fuller review.

Here’s the tool itself. Quality Rater Jev

Gemini 4 Argon - One Step Closer to ‘Model as an Employee’ Paradigm

Dan Petrovic explores Gemini 4 Argon and the idea of a ‘model as an employee’: AI that can carry out long, complex tasks with less supervision.

Why it matters: These advances are useful to clock. The actual model might not be accessible yet, but you should be aware of what it’s capable of before it drops.

Reddit is killing RSS feeds and ending public API access because of AI bots

ICYMI: Reddit is switching off RSS on 13 November and closing its public API in March 2027, blaming scraping by AI bots. Developers of approved apps have until 12 January 2027 to register. Sarah Perez at TechCrunch points out that Reddit’s ‘other revenue’, which includes its AI licensing deals, grew 24% year on year to $43 million in the second quarter.

Why it matters: Reddit’s move highlights a question facing publishers too: how much access should AI companies get, and on what terms?

Google Lighthouse Adds Audit For AI Agent Resource Discovery

Lighthouse 13.5 has just added a check for Agentic Resource Discovery, a proposed standard that helps AI agents find a website’s tools and services. The check looks for an Agentmap entry in robots.txt, an ai-catalog link or a file at /.well-known/ai-catalog.json.

Until next time,

Steve

…but first a reminder of how whacky working with AI can sometimes be. Yes, we all know bots hallucinate but this one took me completely off guard.

I was trying to find some info about an old rave I used to go to years ago (as you do) and Google’s AI Mode gave me quite a decent response as there’s never been a lot of detail about that specific event.

However, something in the response felt a little familiar so I asked AI Mode where it had sourced the info from. It referenced a specific DJ interview:

But as I read the answer, I realised that the copy was not from NARR Radio but actually from ME.

Google had pulled in the transcript from an Instagram video post I’d uploaded earlier this year, so the citation and quoted source were completely made up.

This was so odd, at one point I thought I was hallucinating.

Its apology was equally suspect:

What the hell Google AI??? ¯\_(ツ)_/¯

Just another reminder that, as we absorb more AI into our lives - freely handing over API details, passcodes and personal information - the system is, at best, flaky.

Anyway, here’s the Couscous track I wrote and let’s crack on.

SEMRush AI Visibility Index Report 2026.
Of course, it’s a promo for a suite of Semrush tools but I thought it did a pretty good job of outlining the basics when it comes to AI Visibility with some smart recommendations - including just how different each of the bots are in what they cite and mention.

AI visibility is a topic-level game: A study of 50,000 brands in ChatGPT [Study]
Some standout insights here from , reminding us to delve deeper into topic level analysis. “Broad SEO signals do not consistently predict topic ownership. Domain-level metrics like Authority Score and organic traffic correlate with ownership only about half the time. The competitive picture only becomes clear at the topic level.”

I also like his point: “Grow your topical authority. Topical presence comes from three things working together: broad content coverage across the questions inside a category, relevant mentions in the sources ChatGPT draws from (Reddit, review sites, industry publications, expert commentary), and strong entity signals (structured data, consistent brand naming, third-party citations that tie your brand to the topic).”

And Speaking of entities there was a great video on this very subject by Kim Harris which you can watch here.

Kim also had some excellent tips on ‘writing for the next question’ so go and check it out. Likewise, also has a great summary of the steps you need to consider when it comes to entity search.

Louisa Frahm also have a nice post about optimising for conversational search which also includes this notion of ‘the follow-up question’.

Bot Management In Search 2026
has some excellent advice here on bot management - a really strong overview of everything you should consider before typing that disallow rule. He also has several neat tools via his website so go and check them out.

Did I mention cool seo tools? I also launched a few myself this week. If nothing else you will love the Google Core Update predictor. Then again, if you have an actual life, you will move swiftly on.

International SEO for Publishers
always chooses the key topics being discussed in SEO-land and this one is no exception.

Google announces web multimodal Search performance reporting in Search Console
Publishers can now start to see which stories and topics are being discovered through visual search, and how that behaviour differs from traditional Google searches.

The Generative AI Report Sucks - Let’s Make It Brilliant
Harry delivers another useful method of connecting data to make powerful commercial decisions, this time via the Generative AI Report available in Google Search Console.

50 things you should know about AI
I found this fabulous podcast this week and I now my tiny brain feels levelled up.

The Godfather of User Needs x A Stern Look
Dmitry Shishkin and Selma Stern have a great chat here about short-termism in newsrooms that limits innovation. Thankfully our AI overlords will soon elevate us all from this messy business of being human.

@robertsaiz0I'm starting to think this movie was a documentary 👀😂... #memes #theterminator #ai #foryoupage #funny

Tiktok failed to load.

Enable 3rd party cookies or use another browser

Final Thought: I am coming round to the idea that using the term ‘commodity content’ is becoming redundant, at least in the news industry. I think the term has become too pejorative, a polite way of saying ‘don’t do crap content.’ This avoids the detail that a lot of newsroom output is necessarily functional i.e a lot of users just want accurate breaking news served without any frills.

So, I think we, as an industry have to be careful when we use this term ‘commodity’, because it belittles the good work of journalists working every minute to get those stories out.

That still doesn’t change the fact that all publishers are facing an existential moment if if they don’t offer something distinctive or try to put their personalities front and centre - they are going to struggle. But let’s be kinder when we are trying to problem-solve here.

See you next week.

Now, if you’re an SEO/GEO specialist, you might reasonably ask why I’ve revisited some very familiar AI visibility fundamentals in this research. But from conversations with senior leaders in the news industry, I think there’s real value in revisiting the essential mechanics because these can get buried really quickly when everyone is discussing, for instance, the multiple layers of AI retrieval.

And, as a news junkie, I find this whole space fascinating. 🤓

Methodology:

My research uses the same dataset as before: 201 questions tested across ChatGPT, Copilot, Gemini and Google AI Mode between 2 and 6 September 2026. I wrote the questions to cover the UK, the US and continental Europe, and four themes: trust, identity, subscription value and subject expertise. Of these, 105 are longer, analyst-style questions and 96 are short queries of the kind people actually type, such as “is the guardian reliable”. Each question was run five times on each platform, because the same question can get a different answer each time. That should have produced 4,020 answers, but 17 runs failed to return an answer, leaving 4,003 answers and 20,016 citations. Google AI Mode records its sources as Google redirect links, so I followed each one to the page it points to. After excluding 11 Google ad and shopping links, that left 20,005 citations that I could trace to a website. Phew!

Some standout findings were:

• AI platforms rely heavily on third-party sources: 62% of all citations pointed to third-party websites rather than publisher-owned sites.

• Being mentioned does not mean being cited: when an answer named a publisher, 78% of the time it did not cite that publisher’s own website. This varied considerably by platform, from 52% for ChatGPT to 94% for Copilot and 92% for Gemini.

• Third-party sources frequently appear first: Copilot and Gemini placed a third-party source first in around 84% of their answers, compared with 70% for Google AI Mode and 44% for ChatGPT.

How these figures were calculated. For the 78% figure, each tracked publisher named in an answer counts once, so an answer naming both the Guardian and the BBC gives two pairings. A pairing counts as cited only if that answer also cites the publisher’s own website, including its corporate and parent-company sites. Answers with no citations count as not cited, and only the 98 publishers I track are included. “First” means the first source in the list each platform returned with its answer, in the order it returned them, which does not always correspond to where a source appears on screen. That measure covers only answers with at least one traceable citation, and a source counts as third party when none of the 98 tracked publishers owns it.

AI frequently relies on pages published by others about a news brand, sometimes alongside and often instead of the publisher's own pages. That makes a publisher's off-site reputation a critical part of the evidence AI uses to describe it.

Trust is shaped by what others say about you

Trust questions relied most heavily on third parties, which supplied 77% of citations. Bias and reliability raters such as Media Bias/Fact Check, AllSides and Ad Fontes Media accounted for 25.4%, more than all publisher-owned pages combined (22.6%). Two thirds of trust answers (66%) did not cite a publisher’s own website.

Asked “is the guardian reliable”, Gemini cited Media Bias/Fact Check first in all five answers, usually more than once. Its “High” score for factual reporting did most of the work. The Guardian’s own website appeared in only one answer.

Copilot did the same with the New York Times. Across five answers to “can you trust the new york times”, every cited page was third-party. Sources included Ad Fontes Media, Media Bias/Fact Check and an AI fact-checking site. The New York Times itself was absent.

Short questions relied even more on external evidence. Third parties supplied 72% of citations for short questions, compared with 55% for longer ones. The share from bias raters rose from 2% to 12.1% of citations, while the share from encyclopaedias rose from 1.4% to 8.9%. Asked “is npr biased”, all four assistants cited AllSides’s page on NPR.

Corrections logs and standards pages remain trust assets, but raters read those pages too and their verdicts are often what AI quotes. Publishers should know what each rater says about them, check that the assessment is current and challenge errors directly.

AI can give weak sources too much weight

Some third-party sources are authoritative. ChatGPT drew heavily on audience research: the Reuters Institute’s Digital News Report was cited 567 times, including 381 by ChatGPT. Its UK country page alone was cited 79 times. YouGov’s trust surveys, Pew Research Center and Ofcom also appeared regularly.

Others were much weaker. Directories, listicles and content farms made up 10.1% of all citations, more than Wikipedia and other encyclopaedias combined (4.7%). A newsdata.io blog post listing “unbiased news sources” was the joint most-cited third-party page, alongside the Reuters Institute UK page. It appeared 79 times, including 24 times when Gemini was asked “which us news source is least biased”.

Factually.co was cited 354 times, all but once by Copilot. Created in November 2024 by a single developer, it uses AI to reach its conclusions. Media Bias/Fact Check rates it only “Mostly Factual” because its reliance on AI introduces the potential for error. Several of the pages Copilot cited summarise Media Bias/Fact Check and Ad Fontes ratings. In those cases, Copilot is quoting an AI-generated summary of somebody else’s judgement.

Asked “which uk newspaper is most accurate”, Copilot cited just three websites across five answers: factually.co, BriefMyNews and a blog on allattiamo.eu. It cited no newspaper, regulator or audience research.

For publishers, the implication is clear: monitor influential third-party pages, because AI may present information from a low-quality content farm just as confidently as research from a respected organisation such as the Reuters Institute.

When retrieval is blocked, third parties may fill the gap

This raises a question about blocking AI crawlers, particularly for questions of trust and credibility.

The BBC, New York Times, NPR and The Economist, which block some or all of the main AI crawlers, were named in 13.5% of answers on average. Publishers with open access, including the Guardian, The Times, ProPublica and Reuters, were named in 12.6%. But third parties supplied 59% of the pages cited about the blocked publishers, compared with 24% for the open group. Here, “main AI crawlers” refers to GPTBot, OAI-SearchBot, ChatGPT-User, Google-Extended, PerplexityBot and CCBot, based on each publisher’s robots.txt file when I checked on 2 September 2026. Three publishers with mixed rules, the Financial Times, Le Monde and the Washington Post, were excluded from both groups.

When ChatGPT was asked “can you trust the new york times”, one answer cited 21 pages. Its first three sources were Ad Fontes Media, Media Bias/Fact Check and AllSides. The answer reached the newspaper’s own ethics guidelines through copies elsewhere: two on web archive sites, a 2003 version on Poynter, a Scribd upload and several more on document-sharing and journalism-ethics sites. The answer also cited the ethics page of an unrelated website called East New York Times. Almost none of the evidence came from the New York Times itself.

The BBC showed the same pattern. For “is the bbc trustworthy”, ChatGPT cited Ofcom in all five answers, alongside the BBC’s annual report on gov.uk and apparent copies of its editorial guidelines on two unfamiliar domains. The sources shaped the answer’s framing: “Ofcom has found BBC breaches”.

Only four blocked publishers were compared with seven open ones, and the two groups differ in other ways too. So this shows an association, not proof that blocking causes the gap. Blocking training crawlers can be a sound commercial decision. Blocking crawlers that retrieve pages for live answers is different. It can mean your standards are quoted from archives, mirrors and decades-old copies you do not control. Those decisions should be made separately.

Wikipedia often tells AI who you are

Encyclopaedias supplied 11.6% of citations for identity questions, more than for any other theme. Gemini and Google AI Mode relied on them most. Of the study’s 760 Wikipedia citations, 341 came from Gemini and 299 from Google AI Mode.

Asked “what is the i paper”, Gemini cited Wikipedia first in all five answers and used it for the paper’s launch history and ownership by DMG Media.

Ownership matters because it changes. Asked “who owns the telegraph”, ChatGPT cited Axel Springer’s press release in all five answers, alongside coverage from RTÉ, Axios and The National. Telegraph Media Group’s website appeared once. Across the study, 447 citations about a publisher came from another publisher’s website.

Wikipedia often acts as a publisher’s second About page. Check that yours is accurate and current, especially on ownership, funding and editorial leadership. Because Wikipedia discourages organisations from editing their own articles, suggest sourced corrections on the article’s talk page. When ownership changes, publish a clear statement on your own site too.

Ambiguous names create a more basic problem. For “what is nrc known for” and “how much does nrc cost”, the assistants returned 187 citations across 40 answers. Every citation referred to another NRC, such as the US Nuclear Regulatory Commission, the Norwegian Refugee Council or Nike Run Club. None referred to the Dutch newspaper. The Atlantic and Semafor had smaller versions of the same problem: 22 of 52 citations for “what is the atlantic known for” concerned the Atlantic Ocean, while 15 of 105 for “what is semafor” concerned semaphores or unrelated uses of the name.

Readers tell AI whether your subscription is worth it

Readers’ voices appeared mainly when the question was whether a publisher was worth paying for. Reddit and other forums supplied 6.5% of citations for those questions, app stores and review sites 3.6%, and discount, reseller and price-guide sites 9.4%. Review sites barely appeared on other themes.

Asked “is the atlantic worth subscribing to”, Gemini cited Reddit in four of five answers, including a thread titled “The Atlantic’s Deceptive Subscription Practices”. It also cited App Store reviews and a discount and review site, alongside The Atlantic’s subscription comparison page. The answer warned that introductory rates can rise sharply at renewal.

Reddit was cited 436 times across the study, mostly by Google AI Mode (249) and Gemini (162). Copilot did not cite it once.

For AI, renewal pricing, cancellation and customer service are part of a publisher’s value proposition. A subscription page can explain the benefits clearly while the answer still leads with a complaint thread.

Mentions and citations are not the same thing (obvs)

Although this report focuses primarily on citations, I also analysed mentions.

Across the study, the assistants named tracked publishers 15,104 times. At least one tracked publisher appeared in 94% of answers, with roughly four brands named per answer. Only 22% of those mentions came with a citation to that publisher’s own site.

The gap depends on the assistant. The share of mentions accompanied by a citation to the publisher’s own site was 48% for ChatGPT, 22% for Google AI Mode, 8% for Gemini and 6% for Copilot.

The Guardian was named in 25.7% of answers, and 37% of those mentions included a citation to its own site. The New York Times was named nearly as often, but only 11% of its mentions included a citation to its own site. ProPublica was named in 6.7% of answers, with half of those mentions citing its own site. The Athletic was named in 6.1% of answers, but only 1.6% of those mentions included a citation to its own site, the widest gap in the study.

The gap is widest where trust is at stake. On trust questions only 13% of mentions came with a citation to the publisher’s own site, against about 25% on every other theme. That is also where raters and encyclopaedias are most prominent, making it especially important for publishers’ trust-related pages to be specific, current and reachable by the crawlers that fetch pages for live answers.

The same pattern appears in the crawler comparison. For publishers that block the main AI crawlers, 14.5% of mentions were accompanied by a citation to their own site, compared with 28.6% for publishers with open access, roughly twice the rate.

It is worth knowing which of your pages are used when you are cited. Half of all publisher-owned citations were articles or section pages, 20% were About or corporate pages, 12% were subscription or pricing pages and 10% were home pages. Standards and ethics pages accounted for 3%, and corrections pages for less than 1%, while raters were often quoted instead.

Treat share of voice as the starting point, not the final measure of AI visibility. A mention shows that AI knows you exist. A citation shows that it chose your page as evidence and gives readers a link to follow.

Your own pages still matter

Publishers’ own pages still matter. Across the 98 publishers tracked, being named in AI answers correlated more strongly with citations to their own sites (0.81) than with third-party pages about them (0.68). These are Spearman rank correlations. For each publisher, I compared the number of answers that named it with the number of citations to its own websites, and separately with the number of citations to other websites whose web address or page title includes its name. Brand size affects both figures, so this does not prove that either type of citation causes a publisher to be named.

AI builds its picture from both. The results suggest that when first-party evidence is missing, inaccessible or weak, third parties are more likely to fill the gap. Raters and encyclopaedias also draw on information publishers provide, so strong first-party evidence can help shape the off-site picture.

AI visibility is increasingly reputation management through retrieval

Publishers cannot shape how AI describes them simply by optimising their own websites. AI systems assemble answers from a much wider information ecosystem: publisher pages, Wikipedia, research organisations, bias and reliability raters, other publishers, Reddit, app-store reviews, directories and sometimes low-quality sites.

The research suggests three distinct layers that publishers now need to manage: what they say about themselves, what authoritative third parties say about them, and what customers and readers say about them.

The balance between those layers changes with the question. Trust pulls heavily on raters and research. Identity pulls towards Wikipedia. Subscription value brings in Reddit, reviews and price sites. That means AI visibility is not simply an on-site optimisation challenge. It is increasingly reputation management through retrieval.

Audit the sources that shape your AI reputation

Publishers should audit the sources AI uses to describe them.

What do Media Bias/Fact Check, AllSides and Ad Fontes Media say about you? Is your Wikipedia entry current? How are you represented in research from the Reuters Institute, YouGov or Pew Research Center? What do Reddit and app-store reviews say about subscriptions, cancellation and customer service? And which AI crawlers are you deliberately allowing to retrieve your pages?

Test the short questions readers actually ask about you too. Some surprisingly weak listicles and AI fact-checking pages are influencing the answers.

My previous study asked whether all of your own pages tell the same story. This one raises the harder question: does the rest of the web tell the same story?

That makes AI visibility a wider organisational responsibility. Bias ratings, Wikipedia entries, research, reviews and forum complaints involve communications, audience, customer service and commercial teams as much as SEO or the newsroom.

AI rarely forms its view of a publisher from the publisher alone. Trust is shaped by raters and research. Identity is shaped by Wikipedia and other publishers. Value is shaped partly by readers. If your own evidence cannot be retrieved, archives and mirrors may speak for you.

You cannot control every source, but you can know what it says, correct errors where possible and give the rest of the web accurate evidence to work from.

I work with publishers, media businesses and content-led brands on SEO, GEO and AI visibility, from citation and source audits to retrieval strategy, content structure and team training.

If you’d like to discuss a project, get in touch.

To read the previous piece, visit: https://www.publishingstrategies.co.uk/p/which-publisher-pages-does-ai-cite

Stepping onstage, I couldn’t help feeling, well, existentially challenged. After all, this is the week Anthropic announced there was a 10% chance of AI killing humanity within the decade. Was that a PR blunder or genius on their part? ’Twas a topic passionately discussed at the event over copious amounts of red wine.

Speaking of wine, the award for best soundbite has to go to George Monaghan from the New Statesman, who, describing the advantages of expertly targeted newsletters over randomly assorted homepages, said ‘A homepage is like a wine list without a sommelier.’

I get you George, I have the same problem every time I visit Wetherspoons.

Reflecting somewhat bleary-eyed this morning, I think you could say D2C was the main topic of the day. The panic over zero-click seems to have subsided, and news publishers are now firmly focused on engaging more directly with their audiences, whether by launching new studios to embrace (or perhaps challenge?) the creator economy, sending habit-forming newsletters, or using AI to free up journalists’ time so they can focus on original storytelling.

At least, that’s the goal

As usual, I’ll finish with my interesting picks of the day. Here goes:

• Jon Slade, FT CEO since July 2025, talked about its journey from page views to fans, with Recency, Frequency and Volume all part of the mix, alongside Recognised Users. These are users who have given the FT their email address, opted into marketing, etc. In other words, they are sending potential loyalty signals.

• They also look at ‘Quality Views’, i.e. how far down an article a reader travels and whether that results in a share or comment.

• It was also interesting to note that the FT reaches the most under-35s in the UK, which surprised many in the room. Jon pointed to their ‘future audiences initiative’, which is already helping them towards their ultimate goal of achieving 850,000 recognised users under 35 by 2030.

• He mentioned their ‘geeky finance’ Alphaville Substack as one of the ways they engage with younger audiences.

• He also mentioned FT Edit, which was initially a standalone app until it was folded into their main app due to issues around discoverability. Interestingly, they promote FT Edit as a downgrade offering to users who are thinking of cancelling their subscription.

• Elsewhere, we learned that 70% of The Times of London’s audience reads via its app. Its Golden Metric is ‘Daily Return Rate’, which Anna Sbuttoni has spoken of before. It is daily active users divided by monthly users, expressed as a percentage.

• The Guardian is using AI to help with predictive models designed to keep its audience engaged. The Times of London is using AI to help with moderation in the form of OpenWeb.

• To help signal WHY you should subscribe, The Guardian often leads its messaging with key investigations or campaigns it has recently run.

• Press Gazette has launched a newsroom tool to stop AI fakes before publication

• On an excellent panel on how newsrooms can engage with the creator economy, I thought Graham Page (dig the SEO LinkedIn URL) had an excellent point: if you’re trying something new, what are you going to do less of? Otherwise, you’re going to overwhelm a team. We don’t hear much about what publishers are stopping. I’d like to hear more of this.

• Some of the video panellists were sceptical of Amol Rajan’s new venture. Can slow news really be delivered in two-to-three-minute videos? We shall see.

• Andrew Palmer, The Economist, emphasised that trust and connection were key to news creator output. You can’t have one without the other.

• Carly Steven said zero-click might be here, but the Daily Mail is perfectly positioned to support brands struggling to regain lost visibility.

• Tom Jackson or ‘Tagging Tom’ as he’d been called when trying to introduce more metadata at News UK, emphasised that metadata was still as crucial as before in an age of AI. News UK is using Netacea to track AI bot behaviour across its platforms. They currently block 6% of traffic from bots.

• Carly celebrated the new AI Overview control tool that is now available in Google Search Console, and hoped we would get more granular control soon. But she also emphasised that opting in or out of overviews was tricky when Google is constantly experimenting and occasionally adding Top Stories to AI Overview and AI Mode results. What’s it gonna be, Google?

• Politico is currently looking at how it can bridge the gap between its large free subscription base and its paid products. As Katie Palisoul explained, part of the work is deciding which free registered users should be sent on that conversion journey.

• Mads Holmen pointed out that everyone might be talking about AI-generated micropayments for publishers, but who is actually set up for this right now, or even investigating it as a possibility? Good point.

• A good session with Reach’s David Higgerson, who put the case forward for free local news coverage. If 90% of people in the UK don’t currently pay for news, how can we realistically expect them to? Hence its advertising model, alongside a healthy subscription model with 65,000 paying subscribers.

• He mentioned The Express was the first to go with its premium tier earlier this year, and more will follow. I think this is a really interesting space because, in order to be successful, newsrooms must clearly signal the difference between free and premium paid content. If it’s just news stories behind a paywall, I think users are reluctant to sign up. David also mentioned that they use the Synapse tool to help teams verify information.

• Anna Jeys (Reach) said that she might look at newsletter metrics like CTR, open rates, conversions, frequency, etc., but ultimately her team is asking how their initiatives are creating habits in their audience.

• The Logic’s David Skok said that although newsletters are a key focus, they should be part of a holistic content strategy that drives users across your portfolio. He also mentioned a common challenge for publishers: how long does it take for a free subscriber to become a paid subscriber? He said that, for The Logic, people either pay in the first 48 hours or it takes around seven months. So, again, do people understand your value proposition? In The Logic’s case, it’s ‘We are going to make you smarter in your life’, and ‘Why haven’t you signed up?’ rather than ‘Please sign up.’
  
  Congrats to and the team for putting together another great conference.

Following on from my previous article on audience growth, I’ve been reading a lot of claims recently about the importance of having a decent About page as part of your EEAT strategy to help bots (and people) understand your value proposition as a publisher: you know, things like “hey, we’re a trusted news provider” or “our subscription is worth paying for” or “we’re particularly strong on this subject”.

This all makes complete sense, but I’ve also seen a few comments suggesting that About pages and, heck, even a strong author bio are all you need to get the visibility you’re after in AI chatbots and to foghorn your value proposition to a fleeting audience scrolling from one news story to the next.

So, I wanted to do my own specific research: which pages do AI systems actually cite when they are judging a news brand, and what sort of evidence do those pages provide?

What I found was that the same publisher can be judged using very different evidence depending on a specific question. Trust pulls in one set of pages, subscription value another, specialist authority another, and identity another.

You might be thinking, yeah but we know AI can draw on multiple sources when forming an answer. But I thought it would be useful to point to the actual pages that are getting visibility so you can look at them yourselves.

Oh, and I admit now I didn’t look at third-party citations, which is a critical part of GEO. I might look at that in my next article, but for now, let’s hold hands and leap into this almighty rabbit hole together.

So what did I do? (Methodology)

Using an open-source AI visibility tool (Elmo + Bright Data), I put 201 questions to four AI search systems: ChatGPT, Copilot, Gemini and Google AI Mode. Each question was asked five times per model.

The final dataset contained 4,003 answers and 20,016 citations, collected between 2 and 6 September 2026. Questions covered the UK, the US and selected European countries, across four themes.

The prompt set had two parts: 105 longer comparative questions of the kind a strategist might ask, and 96 shorter queries designed to resemble the sort of thing a reader might actually type into a chatbot.

For every citation I recorded the URL, its position in the model’s source list and the question it was answering.

Citations were attributed to a publisher by matching the cited domain against a tracked list of 98 news brands, including subdomains, so, for instance, support.theguardian.com counts as The Guardian.

Each cited URL was then classified into one kind of page by rule, using the host and path, with specific categories tested before general ones.

There is an important limitation here. The classifier reads URLs, not the actual page content. So a standards document filed under an /about/ path may be counted as an About page even though, in reality, it is an ethics or standards page.

I thought this might be a bit iffy, so I checked that limitation in two ways.

First, I used page titles to sense-check the URL classification. The results barely changed. The biggest shift was ethics and standards, which rose from 2.6% to 4.5%.

Second, I tested a sample of 452 cited URLs against the actual page content. Of the 195 pages I could classify confidently, the URL rule was right 77.1% of the time. It performed particularly well for editorial articles, subscription and pricing pages, and ethics and standards pages, all at around 88% agreement.

The check also helped explain the large unclassified group. Of the unclassified pages I could assess, 79% were ordinary editorial articles. So much of that 41.5% unclassified share on specialist questions is likely to be good ol’ journalism that the URL rule simply failed to recognise.

So I trust the broader pattern that different questions surface different kinds of publisher evidence more than I trust every individual percentage by page type.

For each AI answer, I then looked at every publisher it cited and grouped together all the distinct pages cited from that publisher.

So if one answer cited four Guardian pages and two Reuters pages, that counted as one Guardian case and one Reuters case. This lets us see the mix of evidence AI surfaces for an individual publisher, rather than simply counting URLs across the whole answer.

Got it?

Ok, then. Here’s a selection of the prompts I used:

And some examples of the shorter queries:

(Hmm... hold on. Did I mention that all my prompts were in English? I know, I know this will skew the results and is something I will address in a future study!)

So, what did I find out?

What my analysis showed was that the same publisher is judged with different evidence depending on the question. The About page is part of that picture, not the whole of it. In 76.0% of cases, no About page appeared in the cited evidence at all.

So that means:

1. The evidence changes depending on what the model has been asked.

2. Trust questions draw heavily on journalism, alongside standards, ethics and accountability material.

3. Value-for-money questions bring pricing and subscription pages into the picture.

4. Questions about a publisher’s identity are much more likely to cite the About page.

That is why I think AI visibility for publishers is better understood as an ‘evidence architecture’ problem than an About-page problem.

What this is telling us is that the picture is not uniform.

In about half of the cases, AI cited just one page from the publisher’s site. But when it cited two or more, those pages were different types of content in around two thirds of cases, for example a news article alongside a subscription page or an ethics and standards page.

The type of evidence also changes depending on what the model is being asked about.

The About-page pattern shows this particularly clearly:

The remainder are cases where an About page was cited alongside other publisher pages.

What AI cites depends on the question

Different questions about the same publisher produce very different citation patterns.

Each publisher counts once here, regardless of its size.

Minor page categories are not shown, so totals may not add to 100%.

This is the central pattern in the research. Read across the About row.

It is worth a quarter of the cited evidence when the question is who you are, and almost nothing when the question is whether you are worth paying for.

The ethics row tells a different story.

Ethics and standards pages account for nearly one in ten trust citations and are effectively absent everywhere else.

These are not general-purpose AI credentials.

The important point is not that one page type wins. It is that different questions pull in different kinds of evidence.

For publishers, this means that the question is not simply which page to optimise, but whether the evidence needed for each claim exists, is discoverable and is consistent across the site.

Five random examples

The percentages are easier to understand when you look at individual answers.

ProPublica, asked to prove real-world impact

Question:

Which US news brands best demonstrate the real-world impact of their journalism? Identify examples of the types of outcomes or impact associated with them and explain how that strengthens their value proposition.

ChatGPT cited 14 distinct ProPublica pages across four kinds.

The general About page came fifth, behind three separate impact-related write-ups.

The point is that ProPublica is not relying on a page that simply claims it creates impact. It has an ecosystem of material that documents the impact.

Among the sources were:

• its investigations impact page

• an impact report

• its annual report

• further pieces documenting the outcomes of its journalism

• its general About page

That is what I mean by ‘evidence architecture’.

The Guardian, asked why anyone should pay monthly

Question:

Which UK news publisher has the strongest overall answer to the question “Why should I give you my money every month?” Explain the proposition, evidence and reader benefit.

The answer cited eight Guardian pages across four kinds: governance, pricing, mission, financial results, a Scott Trust statement and even a recruitment page.

The model did not simply cite a marketing statement saying “support independent journalism”.

The cited evidence stretched across:

• who owns the organisation

• what readers actually pay

• what the organisation stands for

• its financial position

• its approach to reader revenues

For a publisher trying to communicate why it deserves money, that is a much wider job than writing better checkout copy.

Mediapart, same question

Ask the same value question about European publishers and Mediapart produces seven cited pages.

These include:

• an activity report

• a subscription portal

• a help-centre pricing article

• the About page

• editorials about its independence

Mediapart’s value proposition is spread across several different kinds of page.

Reuters, asked about accuracy and reliability

Question:

Which US news brands have the strongest reputation for accuracy and reliability? Explain how each organisation communicates and demonstrates those qualities rather than simply claiming them.

Reuters produced one of the clearest examples.

Three Reuters pages were cited, with no general About page among them:

• a standards page

• an article explaining visual verification

• a piece involving editor-in-chief Alessandra Galloni discussing Reuters journalism

For publishers, that distinction matters.

Claiming that you are trustworthy is not the same thing as having visible evidence of how your journalism earns trust.

The Texas Tribune, asked to prove it deserves funding

Question:

Across US news publishers, which brands best communicate and prove the answer to three questions: “Why you?”, “Why should I trust you?” and “Why should I pay you?” Rank the strongest examples and explain your reasoning.

ChatGPT cited five Texas Tribune pages across four kinds, and no two of them were doing the same job:

the membership page

the donation campaign page

the ethics policy

the About page

the homepage

Only one of those is the kind of page a newsroom would think of as its own.

The Tribune is a nonprofit, and asked why a reader should trust and fund a publisher, the model assembled the funding route, the accountability policy and the identity statement together.

Two of the five sit on separate subdomains, support.texastribune.org and give.texastribune.org, which in most organisations belong to membership or development teams rather than to editorial.

So the answer to “why should I pay you?” was built largely out of pages the newsroom does not own.

Does the pattern change when the prompts get shorter?

Yes, it does. Short questions produce shallower and somewhat more ‘promotional’ citation patterns.

And the page mix changes too:

Minor page categories are not shown, so totals may not add to 100%.

Three things change when the question gets shorter.

The model cites far less journalism, 32% against 42%.

It cites more pricing pages and homepages, the commercial and navigational material publishers control most directly.

And it stops at one page nearly six times in ten.

The more promotional pattern comes mainly from pricing pages and homepages, not About pages. On short questions, About pages are cited less often than on longer ones.

That has a slightly awkward implication for publishers.

The deeper the question, the more actual reporting appears in the cited evidence. That gives a good newsroom more opportunity to demonstrate what makes its journalism distinctive.

Shorter questions are often closer to how readers actually ask, and the model appears more likely to settle on a handful of commercial and navigational pages.

Those pages may matter more for real reader queries than publishers assume.

The shorter-query sample covers 912 cases across 64 publishers, compared with 1,760 cases across 80 publishers for the longer questions.

Axios, asked a question a real person might type

The query was simply:

what is axios known for

Four words.

In its fullest answer ChatGPT cited seven Axios pages across four kinds. The largest single source of evidence was not the newsroom. It was the help centre. Among them:

what Axios is and how it works

what topics Axios covers

the Axios Bill of Rights

the newsletters page

the Axios About page

the Axios homepage

That was not a one-off. Across all six runs of that question, help-centre pages were cited seven times and the About page twice. The evidence was spread across four different Axios subdomains: www, help, pages and about.

A reader asks what a publisher is known for, and the model answers largely out of customer-support articles written to explain a product, on a subdomain most newsrooms never think about.

Build evidence for the claims you want AI to make

The actions below are driven by the data insights above: different claims need different evidence.

The temptation with AI visibility is to ask:

“Which page should we optimise?”

A better question is:

“What evidence would an AI need in order to support the things we want audiences to believe about us, and where does that evidence live?”

That is what I mean by evidence architecture: making the evidence for trust, value, authority and identity easy to find across the parts of the site where it naturally belongs.

If you want to be seen as trustworthy, the evidence will look different from the evidence required to demonstrate value for money.

If you want to be seen as the specialist in football, personal finance or travel, the evidence is different again.

Achieving this requires a content governance model that spans editorial, audience, product, subscriptions, customer service, corporate affairs, legal and marketing.

If you want to be called trustworthy

1. Don’t rely on one vague “editorial values” page.

If you have distinct information about editorial standards, corrections, complaints, fact checking or editorial independence, make sure each important piece of evidence can be found and cited in its own right.

Ethics and standards pages represented 9.8% of trust citations and almost nothing for other questions.

That does not prove that splitting pages causes more AI visibility. It does show that specific accountability ‘assets’ regularly appear in trust-related evidence.

Examples in the dataset included Reuters’ fact-checking material, Euractiv newsroom policies, El País’s ethical code, Guardian complaints and corrections and NRC’s principles.

2. Treat corrections as evidence, not housekeeping.

The models cited three different dated Guardian corrections pages, including very recent ones.

A static sentence saying “we correct our mistakes” is a claim.

A visible, continuously maintained record of corrections is evidence that the process actually exists.

There is a wider trust benefit here too, regardless of AI.

3. Explain how your journalism works.

Editorial articles made up 55.8% of trust citations, more than any other category.

Reuters provides a particularly good example with its piece explaining visual verification.

You do not necessarily need to create self-congratulatory corporate content.

But if your newsroom has distinctive verification processes, specialist expertise, reporting methods or editorial safeguards, explain them somewhere readers and machines can find them.

If you want to be seen as worth paying for

4. Treat the subscription page as evidence, not merely a checkout.

Subscribe and pricing pages were 33.7% of value citations, the single largest individual page category for this question.

That means a pricing page is doing more than converting somebody who has already decided to subscribe.

It may also be one of the main pieces of evidence an AI surfaces when somebody asks whether your product is worth paying for.

This matters particularly on short queries. When the question is four words rather than forty, pricing pages make up 18.8% of the cited evidence, compared with 9.9% for the longer questions.

Publishers should therefore ask whether these pages clearly explain:

• what the reader gets

• what is distinctive

• what is included in different tiers

• what the journalism enables or funds

• why the product is worth the price

A table of prices and a giant “subscribe now” button may not be enough.

5. Audit your help centre as part of AI visibility.

Help and FAQ pages made up 4.4% of value citations and almost nothing elsewhere.

That 4.4% may look small, but the organisational point is more interesting.

Some of the evidence shaping how your subscription product is represented may live inside customer-support software that nobody in editorial or SEO has looked at for years.

Check whether those pages are crawlable, accurate, current and consistent with the proposition being communicated elsewhere.

6. Make evidence of impact easy to cite.

Annual reports, impact reports and activity reports appeared repeatedly.

Mediapart’s activity report and impact material were particularly visible. ProPublica’s annual report was cited despite being a PDF.

If your journalism produces measurable outcomes, do not leave all of that evidence buried inside an annual PDF nobody reads.

Consider giving important impact material its own durable web presence.

If you want to be called the specialist

7. Accept that there may be no shortcut page.

Editorial articles accounted for 40.5% of citations in specialist questions.

A further 41.5% was unclassified, which is one of the main limitations of this analysis. I cannot say with confidence that those pages were section hubs, topic pages or anything else.

What I can say is that, in the validation sample, 79% of the unclassified pages I could fetch turned out to be ordinary editorial articles. That suggests the true share of journalism in specialist answers is likely to be higher than the 40.5% shown in the table.

So the broader finding is fairly clear: when AI is assessing a publisher’s subject expertise, journalism itself makes up a substantial share of the evidence it cites.

That should not be surprising.

If you want to be considered authoritative in personal finance, football, travel, politics or product reviews, you ultimately need to publish strong journalism in those areas.

Which means you can’t simply bolt a load of these ‘evidence architecture’ pages onto thousands of pieces of commodity content and hope for the best. The actual journalism still needs to rock.

8. Make specialist methodology visible where you have one.

There are some useful examples.

IndyBest has pages explaining how products are tested and who its experts are.

For review journalism, personal finance, health, science, investigations or any other field where methodology is part of the value proposition, explain it.

This does not guarantee citations.

It does, however, create clear evidence for why your specialist journalism deserves to be trusted.

If you want to be described correctly

9. Keep the About page, but don’t expect it to do everything.

Nothing above is an argument against About pages.

About and mission pages represented 24.9% of identity citations, and nearly a quarter of identity answers rested on an About page alone.

So your About page remains extremely important when the question is:

Who are you?

What do you stand for?

What makes you different?

The mistake is expecting the same page to prove that your subscription is good value, that your journalism is trustworthy and that you are the best source for personal finance.

One page cannot carry all of that.

Structural actions across all four areas

10. Make distinct pieces of evidence independently discoverable and citable.

The Guardian has separate pages for its general About information, journalism, history and organisation, and each appeared independently in the dataset.

I am not suggesting publishers should mechanically break a good page into four thin pages.

The useful point is simpler:

Do not bury several materially different propositions inside one generic corporate page if each deserves to stand on its own.

Your ownership is one thing.

Your standards are another.

Your impact is another.

Your subscription proposition is another.

Your methodology is another.

The goal is not more pages for the sake of more pages.

The goal is clearer evidence.

11. Keep important accountability pages on stable URLs.

Half of the dated citations were from 2025 or later, but the tail reached back as far as 2001.

Old evidence can continue to surface.

Mediapart’s 2019 material about its independence still appears. Le Monde’s 2010 ethics charter still appears.

That makes durable corporate, standards and accountability URLs potentially valuable long-term assets.

So avoid casually deleting or moving them.

If URLs genuinely need to change, handle the migration properly rather than assuming these pages have no audience value because humans rarely visit them directly.

12. Make crawler decisions deliberately.

Publishers that blocked the main AI crawlers were named at roughly the same rate as those that did not, 16.5% compared with 16.2%. But they were cited far less often in this dataset, averaging 61 citations compared with 200.

That does not mean crawler blocking caused the difference.

There are too many other factors involved, and this analysis was not designed to show that unblocking AI crawlers would lead to three times as many citations.

But it raises an important strategic question.

If you prevent AI systems from accessing your journalism but still expect them to describe your brand accurately, what evidence are you leaving available instead?

The BBC illustrates the problem.

Its main site was unavailable to the crawlers used here, so the evidence that surfaced included a downloads subdomain and recruitment pages.

That may be exactly what the organisation wants, but it should be a deliberate trade-off rather than an accidental one.

13. Audit the subdomains nobody thinks about.

Some of the most regularly cited evidence in this study comes from subdomains.

Examples include Guardian Support, Le Monde subscriptions, NRC’s code and Washington Post Help.

These properties may be owned by marketing, customer service, product or commercial teams rather than editorial.

To the model, however, they are still part of the evidence surrounding the brand.

That means AI visibility is not purely a newsroom SEO exercise.

The proposition needs to remain coherent across the whole publishing organisation.

What I think publishers should audit

If I were auditing a publisher on the back of this research, I would start with four questions.

Trust

Can a reader, or an AI system assembling sources, easily find evidence of:

• editorial standards

• corrections

• complaints processes

• fact checking

• expertise

• methodology

• editorial independence

Value

Can it find evidence of:

• what the subscription includes

• why it costs what it costs

• what subscriber funding enables

• journalistic impact

• distinct reader benefits

Specialist authority

Can it find:

• sustained high-quality journalism on the subject

• clear topic or section organisation

• specialist authors and expertise

• methodologies where relevant

• evidence showing why the journalism is credible

Identity

Can it clearly establish:

• who you are

• what you stand for

• who owns you

• how you are funded

• what makes you different

Then I would look across those pages and ask a fifth question:

Are we telling the same story everywhere?

That may be the most important organisational implication of all.

In AI visibility, when different pages contribute different parts of the evidence, consistency becomes an asset.

If your About page says independence is fundamental, your subscription page says subscriptions protect that independence, your standards page explains how editorial independence is maintained, and your journalism demonstrates it in practice, you have not repeated one claim four times.

You have produced four different kinds of evidence supporting the same proposition.

So what should you not bother doing?

Do not simply rewrite your About page in the hope that this alone will transform how AI systems describe your trustworthiness, subject authority or subscription value.

About pages accounted for 11.8% of the cited evidence around trust and just 4.6% around value.

And 93.1% of value cases contained no About-page citation at all.

Keep your About page good.

Just stop asking it to do everybody else’s job.

One caveat that applies to all thirteen recommendations

One important caveat: none of this proves causation.

The study records what models cite.

It does not establish that publishing an ethics page will make you more likely to be cited on trust questions.

It shows that ethics and standards pages appear disproportionately in the cited evidence associated with trust questions.

Those are different statements.

Likewise, I cannot prove from this dataset that breaking corporate information into separate URLs increases visibility, or that making a help centre crawlable changes an AI answer.

To establish cause, we would need proper before-and-after experiments.

So I would treat the recommendations above as things publishers should audit and test, not magic ranking factors.

Weaknesses of this analysis

With more time and budget, these are the areas I would push further.

This is an overview, not a complete answer, and there are some sizeable gaps.

• As I mentioned above, the page classification is not perfect. I grouped pages based on patterns in their URLs rather than reading every one by hand, so some inevitably ended up in the wrong bucket. That is also why there is a fairly large “unclassified” group, especially for specialist and shorter queries.

• A third of the cited evidence cannot be independently checked. Of 452 sampled URLs, 137 could not be fetched because they were blocked, paywalled or refused access. Help and FAQ pages were the worst, with two thirds unreachable. Anything I say about those categories rests on a smaller and more accessible subset of pages than the rest.

• Site structure can distort the classification. The Washington Post, for example, publishes “About Our Fact Checker” as a dated article within its politics section. Because the slug begins with “about”, my URL-based classifier counts it as an About page rather than as journalism. The Guardian’s editorial code sits under an /about/ path and is affected in the same way.
  

  Around a third of the URLs classified as About are dated articles of this kind. That means About content is somewhat overstated, while journalism is understated. Tightening the rule so that a dated, article-style URL remains classified as journalism, even when it sits under /about/, changes the results from 14.8% to 12.8% for About pages and from 38.7% to 42.4% for editorial articles.
  

  None of the conclusions in this article changes. In fact, the correction strengthens the overall argument. The figures quoted throughout are based on the original classification, so they should be treated as conservative estimates.

• A citation is not proof of use. We can see what the model listed as a source. We cannot see which sources actually shaped the sentence a reader received, whether every cited page was opened, or whether other pages were considered and dismissed.

• Position is not importance. The numbers shown are the order in the model’s source list. I have not established that position 0 mattered more than position 36.

• The study lasted five days, with data collected between 2 and 6 September 2026. There is no measure here of how much these patterns vary from week to week.

• I only included four models, with more time I would included Perplexity and Google AI Overview.

• European publishers were assessed in English, which inevitably favours those with strong English-language editions and English explanations of themselves.

• The shorter-query sample is smaller. There are 912 cases across 64 publishers, compared with 1,760 across 80 for the longer-question set. Thirty-one publishers clear the threshold for equal weighting in the shorter-query sample.

• Crawler access is a confound we did not control. Publishers that block the crawlers used here cannot contribute the same evidence mix. The BBC appears through a downloads subdomain and its careers site because its main domain was closed to the crawlers in this setup.

What I would change next time

• Extend the validation sample and hand-code it. The 452-URL check described above was automated and produced a confident classification for 195 pages. Hand-coding a larger sample, including paywalled pages that the crawler could not access, would make the 77.1% agreement figure more robust and give me greater confidence in the results for individual categories, particularly About and ownership pages.

• Fix the classifier where the validation says it is weak. Four-fifths of unclassified pages are editorial articles, so the rule that files long or dated URLs as journalism is too narrow. Rewriting it against the validation set would shrink the unclassified bucket materially.

• Separate cited from used. Distinguish between sources referenced directly in the answer and those merely listed in the source panel. That should give a better indication of which evidence is actually supporting the response.

• Run it repeatedly. Ask the same questions weekly for a quarter so every figure carries a range rather than a single value, and publisher changes can be observed over time.

• Run a real experiment. Identify publishers without a particular evidence asset, such as an ethics page or impact page, observe what happens if one is introduced and measure any subsequent change. Nothing here establishes causation, only patterns.

• Ask in the local language. Put French questions in French, German questions in German and so on. That would separate “legible to an English-language AI” from “strong in its home market”.

• Add the missing surfaces. Google AI Overview above all, followed by Perplexity.

• Weight by real question demand. Every theme currently counts equally. Real reader demand is unlikely to be evenly distributed across identity, trust, value and subject questions.

Of course, what really needs to happen here is to run a real experiment and track the results of the recommendations in my research.

And if you’d like someone to help with that, just please reach out. 😀

In summary

So there you have it.

Nerd tickle satisfied.

The useful conclusion, for me, is not that About pages are useless. It is that AI does not appear to form one fixed view of a publisher from one definitive page.

Instead, it assembles evidence around the claim it is being asked to judge.

• When the question is trust, the citations shift towards journalism, standards and accountability.

• When the question is value, subscription and pricing evidence becomes much more important.

• When the question is subject authority, journalism itself becomes central.

• When the question is identity, the About page finally earns its money.

That makes the challenge much bigger than optimising a couple of corporate pages.

It means understanding the evidence that supports the things you want audiences to believe about you, where that evidence lives, who owns it and whether it tells a consistent story.

And that is even before we get to third-party citations, which are a massive part of the picture and a subject for another day.

I was really responding here to the “quick fixers”.

There isn’t a quick fix.

AI visibility for publishers increasingly looks like the cumulative result of journalism, site architecture, corporate evidence, commercial information, accountability and consistency.

That is probably less exciting than adding some magic schema markup, but considerably more useful.

But I’m back.

Closes Asimov book, clears throat, presses play on Metallica’s ‘Enter Sandman’.

I’ve been talking to clients recently about how zero-click is forcing news publishers to reconsider something much more fundamental than search optimisation. It’s forcing them to ask: ‘Why should audiences choose us at all?’

As part of my audience strategy work, I keep coming back to three areas that can help brands gain visibility - although some of them may sound almost insultingly simple.

All I can say is go and look at some of the biggest news brands in the world and see if you can see evidence of what I’m about to write about. Even the biggest players can overlook some of these fundamentals.

So, with my tin hat firmly on in anticipation of the backlash, here they are.

1. Be Explicit: Make the value of your journalism visible.

Many publishers overlook this fundamental brand marketing rule: have a clear value proposition.

“Value proposition?” I hear you ask. “But we knew about that yonks ago!”

But ‘value proposition’ is a phrase I hear a lot but I don’t actually see a lot.

What many people don’t realise is that there are two facets to having a value proposition - the communicating bit and then the evidencing bit.

A brand can’t merely possess value; it has to make that value explicit, visible and provable.

This counts on so many levels, not just for readers but also for bots who are always looking for signs of credibility. The truth of the matter is that many news publishers could do a lot better in this space.

I think there are a few reasons why this happens:

• An internal assumption that everyone already understands the value of what you do
  

• A belief that delivering good content is enough to demonstrate that value
  

• A belief that this messaging belongs in the trade press or at conferences rather than on the product itself
  

• Brand messaging being tightly controlled at the top, creating a reluctance to ‘poke the bear’ by revisiting basic questions such as ‘Who are we?’ and ‘What do we do?’

I find this area particularly fascinating because if the value proposition is not at the forefront of everything you do as a publisher then you can’t blame the audience for wondering why they should care about your particular news coverage.

That question may never be voiced, but every ignored headline and absent click is, in some small way, an answer to it.

In no way am I suggesting that these brand values should be shoved in your face every time you read a news article. I mean, if I went into a supermarket and the store rep continually harked on about their commitment to youth employment every time I asked where I could find the grilled aubergines, I’d get pretty peeved.

Most people walk into a supermarket blissfully unaware of its brand values. But they would want to know these things if, like news, they had to pay a monthly subscription of £30 for the privilege of entering that store.

That’s why an explicit value proposition can help drive subscribers and retain them. Good examples of brands doing this are The Guardian and GB News. Both are very strong at communicating what they stand for and what they are aiming to deliver, even though their editorial propositions are very different.

Let’s illustrate this further.

Let’s say I had a news brand called ‘Clarity’ (I haven’t Googled that to check if one exists, so apologies if one actually does).

For a news brand, I would say that’s a pretty good name. It sort of does what it says on the tin, to use an old phrase.

But don’t expect users to connect that with the content they are reading unless you also illustrate the WHY and the IMPACT of what you are doing.

Be explicit and audiences are more likely to understand what you stand for, remember you and consider you when they next need news.

2. Be Clear: Make that value easy to consume.

Let’s stick with this word, ‘clarity’ for a moment.

Clarity is another point that is often spoken about, like ‘values’, but gets lost in the messy business of delivering the news, particularly on social media.

Of course, any editor worth his or her salt will talk about the importance of clarity in terms of making sense of the complex. It’s the essence of news reportage.

But again, we always have to look at the individual experience of the first-time user entering your world.

Here I want to look at one specific area to expose a bigger problem, and that’s Instagram carousels.

Instagram introduced carousels in 2017 as a way to tell more complete stories within a single post. For publishers, that makes them particularly useful: instead of passively registering a headline, the audience has to actively swipe through the story.

That makes it particularly useful for journalism where you want someone to engage with an idea rather than simply register a headline.

But the interesting thing is that you’d be surprised at how hard it can be, in many cases, to actually read the text in a carousel: either there’s too much text, the font is too small, or a busy background image makes you squint.

You could also add to that the wrong choice of topics.

Every time a publisher publishes, it should be an invitation to explore the product further.

In a zero-click world, that doesn’t necessarily mean generating an immediate click. It might mean earning a follow, a return visit, a branded search or simply enough recognition to be remembered next time.

This is not a dig at social media teams because achieving perfection in a busy news environment is impossible. But you can get close to it with a review of templates and a discussion around the ultimate aim of these posts.

Usually, that comes down to having a consistent approach across every touchpoint. Demonstrate the value through the way your news is delivered, in addition to the actual content.

In this case, let clarity be your focus.

3. Be Humble: Keep changing the proposition as audience behaviour changes.

This last point is more of a mindset than a practical example.

When I think of metaphors to describe news coverage, I sometimes think of a medieval joust: lots of banners waving, horns blowing and these big brands trying to knock each other out of the tournament to gain your attention.

But away from the bravado and noise, I think the most successful brands are those that really take the time to study their competition and admit that some of their competitors’ output might be better than theirs.

I think having that humble mindset allows innovation to flourish faster because it means admitting your weaknesses.

Or maybe it’s nothing to do with the competition. Maybe it’s to do with where the audience has moved to.

Maybe you have a legacy of presenting the news in a formal manner, but you are seeing audiences increasingly wanting something more accessible.

News is a product, and that product has to be in tune with an audience, no matter the gravity of the coverage.

Be humble enough to revise your position, your output and even your assumptions about what audiences need. Then let the evidence guide what you do next.

So there you have it. Three deceptively simple laws that keep emerging in my client work and training sessions.

Be explicit about why you matter. Be clear in the way you demonstrate that value. And be humble enough to change when audience behaviour tells you something isn’t working.

Zero-click may be changing the mechanics of audience growth, but the fundamental challenge is increasingly straightforward: give people a reason to choose you.

If you’d like to discuss any of this further, just get in touch.

Related Articles

• Competitor research starts with your own product

• Young Audiences Haven’t Abandoned News. They’ve Abandoned the Old Ways of Finding It.

• Austen, Eliot and the Messy Business of AI Visibility & Prompt Tracking

• “Stop renting your audience from other platforms”: What publishers can learn from Good Housekeeping, The Guardian and The Times

• Commodified News is Dead. Long Live Commodified News!

Before analysing rival websites, keyword gaps or search visibility, I always want to understand what the client believes it offers, how clearly that proposition is communicated and what audiences actually experience. The gap between those three things is often more revealing than anything found in a competitor report.

I learned this while working as an interim content strategist on a healthcare account. I had been asked to research the care home market and initially approached the task as an editor, which until then had been my main role. I read the leading providers’ websites, recorded what stood out and identified apparent gaps.

Then an SEO colleague showed me how to crawl the sites, export the data and analyse the results systematically. It changed how I approached the work. Editorial judgement was still essential, but it could now be tested and strengthened with evidence.

I’m writing about this because I believe competitor research is critical to the success of any product. It is also a lot of fun. I enjoy understanding why a brand has positioned itself in a certain way, where that positioning is working and why the organisation may not be firing on all cylinders.

Done well, competitor research gives you the evidence you need to test your proposition and map out the steps required to achieve your goals.

Here are some of the areas I always consider.

Is the proposition clear?

When I visit a website or app, I want to understand quickly what problem the organisation is trying to solve.

Many homepages do not function as effective marketing pages. The proposition may be buried deep within the site, reduced to a tagline that does not quite land or housed on a separate corporate website.

I understand that some problems are complex and cannot easily be summarised in a pithy sentence. But if you are the expert and cannot communicate your value in a way that is easily understood, why should an audience spend time trying to work it out?

What are others saying about the brand?

And when I say “others”, I do not necessarily mean humans.

What are Claude or ChatGPT saying about the brand? Where are they sourcing that information? What do review sites, Reddit users and Google results say?

The point is to understand whether the intended message is landing. Is it the right message? Does it reflect how the organisation is actually perceived? Do people care about the problem the brand says it is trying to solve?

Where are the search gaps?

A keyword gap analysis can be carried out quickly using tools such as Semrush or Ahrefs. It can help reveal how one website compares with another and where their areas of difference lie.

However, these insights should be treated as directional rather than absolute.

One of the worst things you can do is drop the results of a keyword gap report directly into a content plan. That is how organisations end up commissioning articles that may attract searches but are not properly aligned with their proposition, expertise or audience.

Optimisation for optimisation’s sake is rarely a good strategy.

What are answer engines surfacing?

Asking an answer engine a question such as “Is X a trusted brand?” can produce all sorts of revealing results.

You may discover that the shiny new corporate page containing your latest information is not the source being surfaced. Instead, the answer engine may be drawing from an old page that everyone thought had been archived years ago.

Reviewing answer-engine results can expose outdated information, inconsistent messaging and gaps between departments. It can also reveal just how siloed an organisation has become - something SEOs have been warning about for decades.

These results should not be treated as definitive measures of trust. They are diagnostic signals that help show which information is accessible, authoritative and being repeated elsewhere.

What is the client’s own audience experiencing?

One of the most useful things about competitor research is that it often highlights issues on the client’s website that the organisation may not be aware of.

Poor UX, slow-loading pages, advertising bloat and unexamined editorial workflows can all contribute to a frustrating reader experience.

This is why I take time to review the client’s website before examining its competitors. Sometimes the most urgent problems are closer to home.

Can real people complete basic tasks?

Competitor research does not always have to be expensive. Sometimes I simply ask people to visit a website and tell me what they think.

It is important to challenge your assumptions with data, but it is equally useful to test them with other people.

I used to begin content-strategy training sessions by asking participants to find the newsletter sign-up on the website belonging to the person sitting next to them. It was often an eye-opener. People quickly realised that basic calls to action could be surprisingly difficult to locate, even on their own platforms.

Has the role of video been considered?

I am not saying that video has to be part of every communications strategy. But if there is no video anywhere, that should be a deliberate decision rather than an oversight.

Competitor research can reveal whether audiences in a particular category use video to learn, compare services, evaluate products or assess credibility.

The video does not necessarily have to be about the organisation itself. It could simply be a useful resource that helps the audience understand a subject or solve a problem.

Which brings us neatly to value.

Is the value being made explicit?

Everyone talks about audience needs and value, but there is another important consideration.

In today’s attention economy, it is no longer enough to possess something you believe is valuable. You also need to communicate that value explicitly.

If your product saves people time or money, do not merely state the claim. Bring the benefit to life. What could someone do with the time they save? What frustration could they avoid? What opportunity could they pursue instead?

The same applies to publishing.

If you are a news brand, what effort went into breaking a story? What obstacles did the reporting team overcome? Why is a particular interview significant? What do its insights allow the reader to understand or do?

Never assume that the audience understands the full context or recognises the value automatically.

In Summary

Competitor research is most useful when it results in decisions, not another presentation that disappears into a shared drive.

Done properly, it can clarify a publisher’s proposition, expose weaknesses in the audience experience, identify realistic content opportunities and give teams a more confident basis for prioritising investment.

I help publishers and content-led organisations assess their positioning, content performance, search visibility, editorial workflows and audience experience.

To discuss a competitor review or content-strategy project, contact me at me@stevenwilsonbeales.com.

Related articles

• Why Your Blog Isn’t Driving Conversions

• Austen, Eliot and the Messy Business of AI Visibility & Prompt Tracking

• Commodified News is Dead. Long Live Commodified News!

• “Stop renting your audience from other platforms”: What publishers can learn from Good Housekeeping, The Guardian and The Times

• The Secrets Behind The Times’ SEO Success with Luke Halls

What I am writing about today is a really productive series of talks at Media Voices’ first-ever publishing summit, hosted by Esther Kezia Thorpe and Peter Houston. The event took place at The Huckster in Paddington and covered a wide variety of topics, namely publisher growth strategies in a difficult market.

Kicking things off was Liz Settle, Managing Director of Good Housekeeping, who introduced a number of significant changes when she joined 2023 - although she refused to go so far as changing the actual name of the brand. As Liz said: “All marketers know that you mess with brand at your peril.”

Liz commissioned an audience research (‘Gentelligence’) report that recognised the strength of their “intergenerational voices”, with a demographic that ranged from 20 years to 90 years. Faced with economic pressures, some publishers might typically focus on only a specific segment of their audience, but the report highlighted that, although the age range of its readers differed, they had more in common than not.

“The world is designed to keep people apart, but it’s our goal to bring them together.” I really liked that phrase.

I hadn’t heard of The Good Housekeeping Institute before, but it’s the backbone of their product reviews engine. I really liked the way they message the legacy of this team and the value it brings to the audience. If ever you needed an example of EEAT in action, this is it.

One of the advantages of being part of the Hearst family is that Liz and her team can take successful case studies from other brands in the portfolio and adapt accordingly. They recently ran a partnership with retailers over a weekend where GH subscribers could get various discounts online — something inspired by regular Elle shopping weekends across stores in the Netherlands, where customers get 20% off clothes at select stores.

For Liz, Good Housekeeping can’t win at the “media game” because there’s just so much competition out there, but it can win on client direct, tapping into “small slices of the big budgets”. She says: “Advertisers are sick of performance; they want to build something exciting, and publishers are perfectly positioned for this.” Advertisers want “tangible relationships” with readers, and GH can offer this not only through the physical product - the magazine - but through events as well.

She noted: “Trust in magazines is rising at the same point it’s dropping off a cliff for social. AI is driving people to our brands, not away.” I think a lot of newsrooms can learn from case studies like Good Housekeeping, so it was a great first discussion at the event.

Elsewhere, Liz Wynn, Chief Supporter Revenue Officer at The Guardian, had some excellent points about her role in driving conversions. Even though many people subscribe to The Guardian to support journalism, she says “emotional connection is not enough”. Direct connections need to be forged through daily habits and an effective CRM process.

Her team is particularly laser-focused on moving free subscribers into their full digital paid offering. This is when the “last mile” of the customer journey becomes critical, so it’s important to get packaging and messaging absolutely singing.

To help with this, they have introduced a “Supporter Editor” who acts as a bridge between editorial and the supporter team.

Liz noted that, rather than being a business that spreads its bets widely, they are really driven by the theory of marginal gains, really focusing on making small incremental changes across their touch points - although I did note they also have to fight for dev resource just like the rest of us!

Elsewhere, it was interesting to hear more about the “less is more” approach The Times and The Sunday Times are taking, courtesy of Deputy Head of Digital Anna Sbuttoni. This has already been covered in greater detail elsewhere, but this was a great session to find out a bit more about their drive to grow audience interest by producing less.

All good content audits have to start by establishing some kind of performance threshold, and for The Times, they looked at how many articles they were producing daily that delivered fewer than 2,000 subscriber reads. It turns out it was nearly a quarter of their overall output.

I thought it was interesting that the stories that hit the “red zone” weren’t just commodity news stories that you could read anywhere online. Stories may have underperformed due to poor headlines, overall presentation, or simply because they were published late. So topic choice itself might not be the issue.

This led to cutting 20% of their daily output down to approximately 160 articles in favour of articles that deliver more value. But what defines value exactly? Well, apart from the data, The Times also has access to a subscriber reader panel of over 12,000 people who can help sense-check their output.

Other changes have helped their newsroom grow. They have introduced a new live blog template, and timely comment pieces are now front and centre. Their TED analytics dashboard - Times Editorial Data - was also rejigged so subscriber activity was highlighted above casual users, for instance. These can be simple changes to an analytics dashboard, but they can completely pivot your editorial operation.

Malcolm Coles celebrated the power of spreadsheets when leading newsrooms through transformation. Creating a central document where the journalist writes not only their pitch, but also the headline and the specific platform they expect to drive the audience from, helps the newsroom become more data-driven. Interestingly, their threshold is 1,000 page views.

Sarah Ebner delivered a fantastic masterclass on driving audience loyalty ‘from reach to relationships.’ I really liked her idea of the ‘Golden Metric’ - something that really speaks to your overall business goal (other than revenue) that can be easily understood across departments. The point being, we all have many different metrics we are trying to hit so measuring actual success can get complex very quickly. So the Golden Metric reminds the business of what it is ultimately trying to achieve.

Other snippets:

I liked The Nerve’s approach to culture: “Culture is the Trojan horse to cover the most important news stories.” — Sarah Donaldson, Co-founder.

“You don’t need to be a big brand to be successful with newsletters. You just need to demonstrate utility and personality, and be designed for the inbox.” — Jake McGwire, Head of Product, beehiiv.

“What makes a newsletter different is that it has to come from the writer. This is your newsletter. The email is from the writer.” — Toby Moses, Head of Newsletters, The Guardian.

“People can read long texts, but you need short sentences where they can easily skip to something more relevant. Think about how you are organising the information. Condense what you are saying. Stop renting your audience from other platforms.” — Kate Day, Senior Executive Editor, Europe, POLITICO.

That’s it from me. I’m spending the rest of the week judging awards for Press Gazette. If there’s an event of story you’d like me to cover, or you’d like some help with your own content strategy, just get in contact via me@stevenwilsonbeales.com

Related articles

• Young Audiences Haven’t Abandoned News. They’ve Abandoned the Old Ways of Finding It.

• Commodified News is Dead. Long Live Commodified News!

• What Newsrooms can Learn from Duolingo – The ULTIMATE audience engagement and retention app

• The Secrets Behind The Times’ SEO Success with Luke Halls

• Competitor research starts with your own product

That was the response from one audience expert I spoke to this week after they read the latest Reuters Digital News Report.

But is it all doom and gloom?

Well, let’s take a closer look…

The report is always significant because it shapes how the news publishing industry talks about itself over the next 12 months.

But as a self-confessed cynic, I’m usually a little bit ‘meh’ when the report drops because it often reaffirms what we already know has been happening for years: direct traffic is declining, people are becoming more partisan, and there are now more ‘news avoiders’ than ever before.

I mean, there really was a distinct lack of gags in this year’s edition. I should know, I read all 178 pages and all I found were themes like this:

1. The Decline of Direct Access

Social media and video networks have now overtaken news brand websites and apps as the most widely used way to access news globally.

This is a major milestone in the “platformisation” of news, where the pathways between journalism and the public are becoming less direct, more fragmented and less intentional.

In other words, news organisations are losing control over the gateway to their own content. They are being forced to compete on third-party platforms where their branding is often diluted, their relationship with audiences is weaker, and their ability to build habit is much harder.

2. The Traffic Referral Crisis Continues

Publishers are also facing a “precipitous fall” in referral traffic from search engines, increasingly described as ‘Google Zero’.

That means:

Massive traffic drops: Organic search traffic to news sites is falling sharply, especially in major markets.

Future fear: Publishers expect search traffic to fall further as AI-generated overviews and answer engines satisfy user needs without requiring a click-through to the original source.

Google Discover fragility: Some major publishers have already reported significant drops in Google Discover traffic as platforms prioritise user-generated content and video over news links.

Note: We do need to be cautious when discussing Google Zero, or Zero Click. There are many ways publishers can still attract audience through this channel - but they do need to work harder to achieve results.

3. Trust and Interest Hit Record Lows (well, almost)

The report also highlights a darkening public mood around news itself.

Trust collapse: Global trust in news has fallen to its lowest recorded level since measurement began.

Waning interest: Since 2021, the proportion of people who say they are “extremely” or “very” interested in news has dropped significantly.

News avoidance: More people globally now say they sometimes or often actively avoid the news.

4. Stagnant Reader Revenue

Reader revenue also appears to be plateauing.

Stagnant pay basket: The percentage of people paying for online news remains stuck across many tracked countries.

Shrinking funnel: Because reach on owned news sites is declining, the top of the subscription funnel is shrinking too, making it much harder to acquire new paying users.

Uncertain value: In some major markets, payment for news has actually fallen.

5. The Rise of “AI Slop” and Fake News Concerns

Publishers are operating in an environment where concern about fake news remains high.

The proliferation of AI slop and deepfakes is escalating those anxieties. And yet the report also points to a paradox: audiences continue to use less trusted platforms because they are more convenient.

That means publishers are increasingly being forced to compete for attention in environments that are liked, but less trusted — and where the risk of misinformation is often highest.

So, you can kinda understand when someone says:

“We’re F@cked.”

But, I actually found there were quite a few positives in the report, particularly when it came to young audiences (specifically those under 35.)

Because the data does not suggest younger audiences have given up on news.It suggests they care more about format, ease of delivery, relevance and accessibility than any specific brand name.

Here’s why:

Platform-First Discovery

For 18–24s globally, social media, video networks and AI chatbots are now central ways they get news.

This is a very different behaviour from older audiences, who are more likely to have fixed habits around visiting specific websites, apps or broadcasters.

For younger audiences, the starting point is often the platform, not the publisher.

Incidental Consumption

Younger users are also less likely to have fixed news routines.

Their news consumption is often “incidental”, meaning they encounter news while on platforms for other reasons rather than intentionally seeking out a specific brand.

That matters.

Because it suggests the issue is not that young audiences do not care about news. It is that news often reaches them in different contexts, formats and moments than it does for older audiences.

Traditional vs Creator Trust

Generally, younger audiences find individual news creators and influencers more relatable and easier to understand than traditional news media.

However, on a global average, they still perceive established brands as more knowledgeable and impartial.

This is critical.

Young audiences may not always start with the institution, but that does not mean institutions have no value. It means the way that value is communicated is changing.

The “User” Exception

While the general population tends to trust creators less, young people who actually consume content from creators often view them as more trustworthy, authentic and knowledgeable than traditional outlets.

In some markets, 18–24s view creators as more trustworthy than mainstream media — a much starker contrast than we see in the wider population.

That should not be dismissed as young people being naive.

It should be understood as a signal that tone, format, relatability and perceived authenticity now play a much bigger role in how trust is built.

Institutional Fragility

Overall trust in news has fallen to its lowest recorded level, but trust in widely used individual news brands is holding up better than trust in “the news” as a general concept.

That distinction matters.

People may be sceptical about “the media” in the abstract, while still trusting specific brands, journalists, creators or formats they use regularly.

Enduring Support for Impartiality

Despite their preference for social platforms, younger audiences still largely support the ideal of impartial news.

Yes, 18–24s are slightly more likely than older groups to prefer news that shares their point of view. But a clear plurality still wants news that does not take a side.

Looking across social differences, young people and people with low levels of formal education are both groups that are less likely to say they prefer getting news from sources that don’t have a particular point of view. A slightly larger share of 18–24-year-old respondents say they prefer news from sources that share their point of view, or from sources that challenge their point of view.A significantly larger share of those with low levels of formal education answer ‘don’t know’. In both groups, however, a clear plurality still expresses a preference for impartiality. (Page 58)

So, in summary, younger audiences do care about information provenance.

Many are using AI chatbots specifically to verify news, understand context or find out more about a source. But they are also increasingly likely to trust personality-led, accessible formats over traditional institutional voices.

So, young audiences are still consuming news. They just don’t always get it from a homepage, an app or a brand they already know.

So the challenge is not to force them back to old habits. It is to make news work better in the places they already are.

That means clearer formats, better explainers, more human voices, and stronger signals about where information has come from.

There you go. A simple ray of light in an otherwise quite challenging report.

Related articles

• What does the Reuters Journalism and Technology Trends and Predictions 2026 report tell us about where newsrooms are heading?

• How Global South World Is Taking on Brainrot, Algorithms and News Avoidance

• What Newsrooms can Learn from Duolingo – The ULTIMATE audience engagement and retention app

• “Stop renting your audience from other platforms”: What publishers can learn from Good Housekeeping, The Guardian and The Times

• Best TikTok Publishing Times, Al Jazeera, Google Discover, Sun Video & William Gibson

Well, two problems, actually.

The first is this: trying to explain how AI actually works is not particularly easy, especially with a technology that’s evolving so quickly.

The second is what I’m going to call Austen’s Law of Predictability, which goes a bit like this:

“It is a truth universally acknowledged, that a single man or woman in possession of AI visibility data, must be in want of a LinkedIn post declaring they have cracked the code once and for all.”

And that’s because, when it comes to AI visibility, precision is tricky.

The biggest mistake I see businesses making right now is treating prompt-tracking tools like SEO ranking tools. In other words, they assume you can simply pop in the prompts you want to track and et voilà! - you can “track” your exact visibility in neat daily or weekly charts. If the tool tells you that you’re winning, then you must be winning.

The problem with this is what I’ll call Eliot’s Law which goes something like this:

‘“Humankind cannot bear very much reality. Faced with uncertainty, people will always choose a nice, shiny AI visibility dashboard.”

Now, I like a nice, shiny AI visibility dashboard as much as the next guy, but if you don’t have a strategy in place before generating your prompts, or understand how an LLM generates an answer through variance, then the value of that dashboard becomes redundant pretty quickly.

And that’s because:

• AI answers are probabilistic, not deterministic.
  

• MEANING. The same or similar prompt can produce different answers depending on the model, phrasing, location, timing, user context, retrieval sources, and even random variation in generation.
  

• WHICH MEANS A dashboard may tell you what happened for a sample of prompts at a particular moment. It does not tell you exactly what every user is seeing.

This is why I think when you’re reporting back on the main trends in your AI visibility report, you also need to flag the limitations behind the data. It should make clear whether any changes during the time period selected are statistically meaningful, directionally interesting, or just noise.

You should also caveat any presentations by explaining your prompt methodology: which prompts were tracked, why they were chosen, how often they were tested, and across which models or platforms.

Also, better reporting should always combine dashboard data with qualitative analysis: what answers say, which sources are cited, how brands are framed, and whether the responses are actually useful or accurate.

Sounds messy? Well, here are some tactics I’ve tried that have helped with reporting back on performance.

1. Start With the Business Objective

When starting out with AI prompt tracking it’s very easy to start generating a LOAD of prompts that COULD be related to your business but are not CORE to the business.

Many tools will offer the ability to AI-generate your prompts for you at a click of a button which is all very empowering, but do they reflect the CORE questions that your business needs to answer?

An example of that is using a prompt to compare yourself with a competitor. Insights like these can often be used to create a gap analysis to flag content that your competitor is doing and what you’re not.

But before you plunge into the task of writing 50 articles on topic A or B, stop. Is this content actually CORE to your business or are you just producing content for content’s sake? Is this the right competitor to track performance against? Is this even the right prompt?

You are ALWAYS in the stronger position if your prompts map to a specific and realistic business objective - which is why you need to establish those before you go anywhere near a prompt tracking tool.

Do you want to generate more sales? Are you looking to address negative brand reputation? Are you trying to promote a new product which is being misrepresented?

Nailing those objectives will really help you avoid generating hundreds of vaguely related prompts and being lost in the response data.

2. Make Every Prompt Actionable

If you’ve achieved step one, then step two should be a breeze. A prompt should not just be relevant. It should create a decision.

For example, if the answer changes, what would we actually do differently?

That question matters because when the connection between prompt and business goal is self-evident, stakeholder sign-off becomes much easier when you come to deliver your action plan.

So, have a strategy for your prompt generation and THEN you can get into things like entity consistency, schema, precision writing, what third parties are saying about you, topic clusters etc.

3. Map Prompts to the Buyer Journey

Your prompts need to reflect where the buyer is in relation to your brand which is typically:

• Awareness

• Consideration

• Decision

You need to look at the prompts that follow the audience all the way from discoverability through to final evaluation. I will often spend time here looking at the actual language that’s used by the potential customers - in surveys, review sites, subreddits etc. You really can’t overvalue this kind of research and it’s something you should be very careful about delegating to AI.

Once mapped, and depending on your weapon of choice, you can tag your prompts for easier reporting.

Very much connected with step 3 is to profile your actual consumers. However, as someone who has seen many beautiful persona slides in my life, I would always seek to ground this in actual data and not fanciful imaginings. Want to test this method quickly? Kevin Indig has a neat tool here.

Ok, so we are coming to the end of this ‘stack’ so let’s round off with a nice modest potential LinkedIn post title…

HERE’S THE ULTIMATE ANSWER TO ALL YOUR AI VISIBILITY WOES

Here goes. Want to excel at any type of visibility? Then get better at telling the story about your company. Who you are, what you do, the impact you are making, the audience you are serving. You’d be surprised at how poor we are about telling our own brand stories.

And that’s really it. All hail the brand marketing gods, you really were onto something.

Lastly, get someone with strong brand sensibilities who also understands the technical framework needed to make your brand legible to AI systems.

I may know a guy.

Related articles

• The Truth About AI Crawlers and Publisher Content

• Google’s New AI Dashboard – A Pyrrhic Victory for Publishers? PLUS World News Media Congress Highlights

• Google to Let Publishers Opt Out of AI Overviews: What It Means for Content Owners

• Why Your Blog Isn’t Driving Conversions

• Competitor research starts with your own product

You can always count on climate change to mess up the data. Thanks, greenhouse gases. Duh.

But there have been other things going on with Google that have made me wince the same way you might do when you hear your drunk Uncle at a party. I mean it started off ok. I quite liked their note on ‘Optimising your website for degenerative AI Features on Google Search.’ It’s all pretty standard stuff, but I think it’s important when a big org like Google comes out backing core optimisation principles and flagging certain practices that might get you in trouble further down the line.

(Eli Schwartz gave a great podcast interview talking further about this which I recommend for anyone who wants to go deeper.)

Then Google announced ‘New ways to find your favourite sources and original content in AI Search’ - and that’s when the wheels began to wobble. Sure, any tool that is going to lead to more publisher visibility is to be welcomed, but it all feels like begging for crumbs after the impact of AI Overviews.

And then we got the announcement of two new features in Google Search Console: one which will allow you to control whether your website appears in generative AI search features and the other which is a new performance report showing generative AI features across Search and Discover.

And that’s when the wheels came off. Again, these are all nice features to have but I think their practical use is limited. Firstly, the control is an ‘opt-out’ feature meaning a certain level of resource is needed to manage this and, well, I’m just not convinced about the value of opting out in the first place. The rationale here is that publishers wanted a way to protect their content from AI summaries so that users would click through to the actual content. But what we have here instead is a situation where NO ONE is going to opt out simply because someone else will only fill their place.

The only way this functionality could ever be effective is if the whole publishing community bonded together to opt out of certain topics. And even then, that sounds like a perfect opportunity for some tech bro to spin up a site à la Lovable. The misinformation risks are HUGE.

Then we get to the actual performance dashboard and…well…it’s all a bit ‘meh’ right? We get some impressions data but no clicks. WHY GOOGLE, WHY? ¯\_(ツ)_/¯ Doesn’t mean I won’t be using it though - it’s just that’s it’s all getting a bit ‘The Hand that Feeds’ around here.

Side Note: Did you know I read one of the best explainers on the the difference between Parametric Memory and Retrieval (RAG) this week from Stephen Burns at Common Crawl? Go ahead and read it.

Top Insights from World News Media Congress

I wasn’t there in person but I was certainly there in spirit as I popped open another bottle of Château Miraval and read Ezra Eeman’s keynote on AI & News Media. The message is clear - if you’re not looking at how AI is already changing audience expectations about how they want to consume their media, then you need to have a long hard look in the mirror. Essential reading - as was New York Times publisher A. G. Sulzberger’’s speech on why (and how) news publishers should fight AI platforms.

Released also this week was the FT Strategies and WAN-IRFA ‘collab’ in the form of their ‘Future Newsrooms’ study. There are so many great insights in this report that I could share, so I really recommend you read it. Here’s just a few of the many that caught my eye.

There’s probably no surprises in this one, but the report does build a powerful argument for the ‘structurally aligned newsroom’ - and that, I can tell you, isn’t about delivering three thousand updates on Love Island.

It’s always interesting to see what formats 448 senior publishing leaders are currently focusing on. And even more interesting to see that it’s still text. I’m not sure if that’s worth celebrating or crying over.

In summary, the report makes for grim reading in some areas: a lot of decisions made in newsrooms are not data-driven, staff lack strategic direction, and training in AI is pretty low. But I did like how the report highlighted how newsrooms can adapt to meet their audience in the places and formats that suit their audience whilst still delivering quality journalism.

And to that, I will raise my glass.

Come and join us at Glide Live in London on June 17th! All the details here.

Related articles

• Google to Let Publishers Opt Out of AI Overviews: What It Means for Content Owners

• The Truth About AI Crawlers and Publisher Content

• Austen, Eliot and the Messy Business of AI Visibility & Prompt Tracking

• Commodified News is Dead. Long Live Commodified News!

• What does the Reuters Journalism and Technology Trends and Predictions 2026 report tell us about where newsrooms are heading?

It’s no secret. With the current challenges to traffic, brands are looking at how they can squeeze more from every aspect of their content operation. Blogs have become a core focus because they’re not only a great way of educating audiences about your products and values, but—now with LLMs in the mix—they are also a great way of gaining visibility for key prompts.

However, many blogs fail to drive conversions because of a lack of strategic focus. This often leads to a failure to attract the ‘right’ kind of visitors such as casual readers with no intent to purchase. The issue is also compounded with content teams being ‘pressured’ to use AI-generated blog posts to rapidly achieve scale without precisely nailing those strategic goals. The results are often less than ideal.

So, what’s the answer? Well let’s take a look at some of the most common issues and how to tackle ‘em.

You’re writing for traffic, not intent

Traffic is useful, but only when it comes from the right audience.

An issue I see quite a lot is brands choosing to write about certain topics because they have a high search volume. It’s an easy mistake to make - more searches should mean more opportunity, right? But high-volume keywords are often broad, early-stage and commercially weak.

For example, a project management software company could publish an article on “How To Be More Productive.” That topic may attract plenty of readers, but many of them are not actively looking for software. Some may be students. Some may be freelancers. Some may simply want a few productivity tips.

Compare that with topics like “Best Project Management Software For Agencies,” “Trello Alternatives For Growing Teams,” or “How To Manage Client Projects Across Multiple Departments.” These searches are narrower, but they reveal far more about the reader’s problem and buying context.

A good content strategist will always ask: Who is searching? What are they trying to solve? How close are they to needing a product or service? What should they do next?

Your calls to action are weak

Many blog posts end with a vague call to action: “Contact us,” “Learn more,” or “Subscribe to our newsletter.”

These CTAs aren’t particularly wrong, but they are often disconnected from the actual message of the article.

If someone reads a post about fixing content decay, the next step should not be a generic company newsletter. A stronger CTA might offer a content decay audit, a checklist, a diagnostic call, or a guide to updating underperforming pages.

In other words, the CTA should match the reader’s intent.

Someone reading a top-of-funnel educational article may not be ready to book a sales call. But they may download a template, subscribe to a practical email series, or read a related case study. Someone reading a comparison page, on the other hand, may be much closer to a demo or consultation.

Your blog should not rely on one CTA for every reader. It needs intent-matched next steps.

You have no product-led content

A lot of blog content is useful but commercially detached.

What do I mean by this?

Well, it explains the problem, gives advice, but then stops before connecting the issue to the product, service or solution. This often happens because teams are afraid of sounding too sales-led. They want to be helpful, so they avoid mentioning what they sell.

That is a mistake.

Product-led content does not mean turning every article into a sales pitch. It means showing how your product, service, framework or expertise helps solve the problem being discussed.

So, if you are a private clinic writing about knee pain, explain how your assessment process works and when someone should seek specialist help. If you’re a mental health provider writing about anxiety, show how different therapy options support different needs.

Remember, readers do not just need information. They need to understand why you are credible, how you think, and how you can help.

A blog that never connects expertise to commercial value will struggle to generate qualified leads.

You’re publishing isolated articles instead of topic clusters

This one is a classic.

Many companies publish content as a list of disconnected ideas. One week they write about trends. The next week they write about tools. Then they publish a founder opinion piece, followed by a generic how-to article. But one-off blog posts rarely build authority on their own.

Each article may be fine individually, but together they do not create a clear content architecture.

Bots and buyers both need structure. They need to see that you understand a topic deeply. That means building clusters around important themes.

For example, instead of publishing one article on “skincare,” a beauty brand might build a cluster around:

• What a good skincare routine looks like

• How to build a skincare routine for your skin type

• Skincare routine examples for morning and evening

• Skincare for dry, oily, sensitive or combination skin

• Cleanser vs toner vs serum vs moisturiser

• How to layer skincare products correctly

• Skincare ingredients to look for, such as hyaluronic acid, niacinamide or retinol

• How to choose the right products for your skin concerns

• Common skincare mistakes

• How to measure whether your skincare routine is working

These pieces should link to each other and support a core commercial page. That is how you build topical authority and guide readers through a journey.

A scattered blog creates scattered results.

You are missing bottom-of-funnel pages

Many blogs over-invest in educational content and underinvest in bottom-of-funnel content.

Bottom-of-funnel, or BOFU, content targets readers who are actively comparing options, evaluating providers or preparing to buy.

Examples include:

• “Best X software for Y”

• “X alternatives”

• “X vs Y”

• “How to choose an X agency”

• “Content marketing agency for SaaS”

• “SEO consultant for B2B companies” - you don’t need to look too far for that one ;)

• Case studies

• Service comparison pages

These pages may not always have huge traffic potential, but they often attract people with much stronger commercial intent.

If your blog only answers broad educational questions, you may be helping future buyers without giving current buyers a reason to choose you.

A balanced content strategy needs content for every stage of the journey: awareness, consideration and decision.

There’s no conversion path

Even strong content will underperform if there is no clear path from reading to action.

Ask yourself the following questions:

• What happens after someone lands on a blog post?

• Can they easily find a relevant service page?

• Is there a related case study?

• Is there a useful lead magnet?

• Are internal links guiding them to the next logical step?

• Is the CTA visible?

• Does the page explain who you help and what you do?

A page should NEVER live in isolation.

Every key article should have a clear purpose. Some should bring the right people to your site. Some should build confidence in your expertise. Some should help readers understand their options. Others should encourage them to take action.

Once you know what each page is meant to do, you can create a stronger next step — whether that is reading a related guide, requesting a quote, booking a call or speaking to an adviser.

The real issue: your blog is not connected to revenue 💰💰💰

A blog creates REAL business value when it is built around buyer intent, commercial relevance and clear conversion journeys.

That means choosing topics because they matter to your audience and your business. It means creating content clusters instead of random articles. It means publishing BOFU pages, not just educational posts. It means using calls to action that reflect the reader’s stage of awareness. And it means connecting your expertise to the problems your product or service solves.

The goal is not simply to get more traffic- it’s to attract the right people, answer the right questions, build trust, and make the next step obvious.

If you would like help with any of the issues raised above, please reach out to me via me@stevenwilsonbeales.com

Related articles

• Competitor research starts with your own product

• Austen, Eliot and the Messy Business of AI Visibility & Prompt Tracking

• The Truth About AI Crawlers and Publisher Content

• As a Publisher, Should You Be Investing More in PPC or Organic?

• Commodified News is Dead. Long Live Commodified News!

I know I’ve already shared my highlights about The Podcast Show, but we’re about to experience a heatwave here in the UK and there’s no way I’m going to look at a laptop for the next 48 hours.

So, I thought I’d share with you a bonus edition about two AI-related sessions I went to. Both of these sessions involved Jeanine Wright, CEO and Co-Founder of Inception Point AI which over the last three years has uploaded more than 15,000 AI-generated podcasts and now has over a million listeners every month. These are all voiced by synthetic characters and pretty much cover any topic you can think of. Go on, I dare you.

How about bald eagles?

Or maybe pasta?

You listened right? And you’d be forgiven for thinking that all of this was harmless, inconsequential podslop. And the ironic thing is, Jeanine would probably agree with you.

You see, Jeanine knows that we are quickly approaching a moment when AI-generated content is going to rapidly outscale human content. She also knows that while a large volume of Inception Point AI’s podcasts are, as she admitted, ‘terrible’ she is relying on three core principles:

1. The quality of AI-generated content is only going to increase over time
   

2. People will become more comfortable with AI-generated content over time.
   

3. The output of the product is ultimately more important than the input.
   

I think, if pushed, Inception Point AI might argue that people might say they have aversions to consuming AI-generated content but, at the end of the day, they are always going to choose a product that is the most reliable, relatable and cost-effective. Trusting a product is really about recognising great service.

Jeanine described the podcast market as inherently ‘broken’ because it relied too much on delivering big ‘hits’. The scale required to produce a hit necessitated high production values and expensive distribution plans which greatly pushed up the barrier to entry. AI should be viewed as a helpful disruptor in the podcast space, giving listeners the opportunity to hear a vast amount of information that is currently locked away in text.

Sensing a restless crowd, Nicky Birch, Innovation Lead, BBC Gen AI Programme, tried to soothe things by crediting Jeanine for being brave enough to join the debate on stage and suggested that perhaps they needed to think of a different category for AI-generated audio - perhaps not ‘podcasts’. This wasn’t really a consideration for Jeanine who proceeded to point out that we shouldn’t treat audiences like children. They could make their own decision about what they wanted to enjoy.

All in all, this was a fascinating debate orchestrated by Katie Prescott, co-host of The Times Tech podcast. I think what the panel lacked was representation from one of the podcast platforms. Inception Point AI is only responding to current market conditions and that means we’re all going to have to plough through lots more podcasts about pasta before we get to the good stuff.

Soz.

PS.James Cridland also wrote about the session here and has been covering Inception Point AI far longer than a whipper-substacker like me.

Related articles

• Five Things We Learned At The Podcast Show 2026 About The Future Of Audio

• How to Grow a Podcast: Product, People and Promotion

• The Truth About AI Crawlers and Publisher Content

• Austen, Eliot and the Messy Business of AI Visibility & Prompt Tracking

• What does the Reuters Journalism and Technology Trends and Predictions 2026 report tell us about where newsrooms are heading?

In short: A true keyword-stuffed bonanza!

This was my first time at the conference, which was a great place to swot up on practical tips and get a sense of where the industry is heading. There was even the obligatory pair of conference socks which I politely declined in favour of a rather tasty sherbet dip. And if that’s not eye-watering enough, here are five of my key takeaways:

1.Not every podcast has to be a video
I arrived slightly late for a live recording of Matt Deegan’s The Media Club (apologies, Matt) but just in time to hear BBC Audio’s Head of Production & Development Tom Curry discussing the merits of video. Tom pointed out that even though a lot of video-first podcasts are launching it’s not necessarily an absolute requirement. Video is expensive so always consider if it’s the best use of your resource. Perhaps it might be better allocated to refining the script or improving the recording process? Do what you think best swerves the audience.

Regarding vertical video, Jim Waterson had an interesting anecdote. He’d recently invested in a video clip campaign to try and boost audience to his London Centric newsletter. He didn’t quite get the return on investment he’d wished for, but he did get sources for future stories.

Elsewhere, Francesca Hunt and Mark Pearson presented Ofcom’s Audio listening in the UK report. I was intrigued by this slide that reaffirmed that we don’t always need cinematic productions for our video podcasts, although I will always repeat that poor video (and especially poor audio) can do an incredible amount of damage to your brand - so get the basics right.

2. Experiment your way towards better monetisation
With the launch of Sky Insider’s ad-free listening and bonus episodes this week, it was an excellent opportunity to hear about podcast monetisation with Andrew Thomson, Executive Director of Business Operations at Sky News. Alongside Noiser’s CEO Pascal Hughes and Tortoise Media/The Observer’s Jack Riley the key theme here was finding the perfect price point for your product. Jack acknowledged audience retention was now become a key focus - part of a wider media trend as brands look to super serve their current subscribers now that top of funnel activity has become harder to convert. All three panellists agreed that creating a sense of community around podcasts was essential as was immediately adding value as soon as the subscription kicked in. Podcasts should dovetail nicely with audience retention methods because they are ‘habit-forming’.

Need some ideas to encourage subscriptions? Pascal pointed to the power of the ‘padlocked’ podcast - the greyed out podcast you can see, but can’t touch until you sign up. Also, try lowering your annual membership costs and increasing your monthly costs - but keep iterating to find what works for your audience.

3. Don’t forget to warm up your audience with the cold open
AudioUK’s Chris Baughen hosted a well- attended session on basic storytelling techniques with Willard Foxton from Storydriver. Willard explained why the cold open was so critical to any story, and why writing teams spend an enormous amount of time redrafting until they landed on the best opener. It essentially has to be a microcosm of your overall story, setting up expectations that can be delivered upon or upset. Also, try and make sure you have a few ‘Oh shit!’ moments in your podcasts - moments that are so impactful to an audience, they feel compelled to tell someone else. The gist of the session? Basic story telling techniques remain essential!

4. Think about the relationship you want with your audience
The Pod Save The UK session was hilarious with comedian Nish Kumar and journalist Coco Khan. They gave us a great insight hugely popular and irrelevant podcast on politics. As Nish said: ‘We really spent a long time deciding the kind of relationship we wanted with the audience. And eventually decided we ARE the audience. We’re not trying to be experts, we are just trying to be more informed.’ Nish also made a interesting side note about politics in general. It’s not necessarily true that the older you get, the more you move to the right of politics. ‘The less you give people to conserve, the less they are going to swing conservative.’ That’s interesting, as well as their conversation about what would make the ultimate woke packet of crips.

The answer is, of course, Space Rangers.

5. Authenticity doesn’t have to be human
I was pleased to see a session on Generative Engine Optimisation (GEO) at the show and we had an interesting panel discussion with Jeanine Wright from Inception Point AI, Grant Durando from Right Side Up and Benjamin Doyle from Voy. It’s definitely something I’m going to drill into further as some of the comments from Jeanine raised more than a few eyebrows in the room. Without simplifying her argument too much she explained that we’re reaching the stage where there’s going to be more AI-generated content online than human. At the same time, that AI content is becoming more useful and valuable to the audience every day. At the moment we think of trusted content as essentially human, but as we begin to rely more on AI, that is going the change. Audience will trust useful information, speedily delivered and in formats that are convenient. Paraphrasing Voy’s Benjamin Doyle, that’s all very well, but what if your trying to source information about a medical injection? He thinks people would prefer a human to be involved somewhere in the loop for that particular answer.

This particular session stayed in my mind for quite a while afterwards so I’ll do a follow up. The AI-tsunami isn’t going away so what safeguards, if any, should be put in place to protect audiences from the free market of (mis) information? And who should manage those safeguards - the regulator? the platform? the podcaster?

Whatever the answer, it’s still evident that podcasts will play a vital role in our media consumption because people crave connection. We just need to be aware that AI is disrupting that connection and we still need to figure out if its ultimately a ‘benefit or a hazard’.

Which is, if you didn’t know, a Blade Runner quote.

Related articles

• How to Grow a Podcast: Product, People and Promotion

• How to Grow a Podcast: Product, People and Promotion – Beers & SEO episode

• Mum, Are Podcasts Created By Humans Better Than AI-Generated Podcasts?

• Newsrewired 2026: In Opaque Times, Let Experimentation Shine Through

• What does the Reuters Journalism and Technology Trends and Predictions 2026 report tell us about where newsrooms are heading?

Related articles

• How to Grow a Podcast: Product, People and Promotion

• Five Things We Learned At The Podcast Show 2026 About The Future Of Audio

• Mum, Are Podcasts Created By Humans Better Than AI-Generated Podcasts?

• Newsrewired 2026: In Opaque Times, Let Experimentation Shine Through
  --------------------
  
  Connect with Barry & Steve
  SEO Summit 2026 - https://newsseo.io
  Barry's Consultancy service: https://www.polemicdigital.com/
  Steve's Consultancy service: https://www.stevenwilsonbeales.com/

The key theme this year was experimentation with a real focus on practical application - be it via TikTok, Claude Co-Worker or Reddit. I’m going to list some of the highlights below but please check out Journalism UK’s live blog for more details.

So in no particular order.

• We had a great ‘state of the nation’ keynote from Ezra Eeman (Strategy & Innovation Director at NPO) who brought us up to date with the impact of AI on newsrooms and what to expect in the future. He referenced Good Tape created by journalists at Zetland, Semafor’s vibe-coded charts and WSJ’s Story Maker tool as all examples of how newsrooms are innovating through the use of AI. He also spoke about how agentic AI will improve our current workflows and surface content faster to us via apps like Huxe. But if this means less audience going to publisher owned-domains then how can we adapt? Cue liquid content and the News Atom project from Sannuta Raghu - which is amusing because we used to call this approach ‘adaptive content strategy’ - but a great read nonetheless. The NLWEB protocol was also referenced, possibly helping publishers on their path to API ‘nirvana’.

  

  

• We had a great debate about the merits of creator journalism from Kassy Cho, Alessandro Accorsi and Valeriia Voshchevska. Some publishers might be reluctant to engage with creators because they feel that they may not be impartial. But as the panel pointed out, with so many ‘crisis chasers’ out there capitalising on the latest trends and spreading misinformation - newsrooms need to engage with creators to fight back. Newsrooms can no longer be so impartial that they remain invisible in the feeds, they need to captivate an audience with personality, tone and opinion, as well as a commitment to values. The way to do this isn’t by hitting people with explainers, it’s about also being in that same viral moment as their audience and only dropping contextual insights once the audience has warmed up to your brand.

• Continuing the theme of experimentation and innovation, Grace Miller from Flight Story gave us a great insight into how their company (who produce the second biggest podcast in the world - Diary of a CEO) approach innovation. Their approach to their podcasts, as well as experimentation is to try and maintain that curiosity we all have as children in everything they do. They take an MVP approach to experimentation, accepting that 70% of their experiments are probably going to be inconclusive, 20% will fail, but 10% will be successful. I thought that was a great attitude - acknowledging acceptable mistakes. They tend to only test with one variable at a time - e.g. one A/B test of a thumbnail - because if you have two many successful variables, which ones do you actually pick? They also experiment with paid media and synthetic audiences across their team of 120 staff. It was a great presentation and I think we’re going to hear a lot more Grace and the successes they are having over at Flight Story.

  

  

• Luke Bradley-Jones also spoke about experimentation at The Economist, specially their Insider video product. Interestingly they made forty pilot versions of the Insider before launching the project and had a panel of thousands to help with the initial testing. Luke says they have a completion rate of over 40% which is pretty good for video and they are super-focused on the first ten minutes of any video to maintain attention. The show is also spliced into three main sections, chapters almost, to help that retention. Luke also mentioned that the Economist will be launching a new streaming product this July called ‘The Economist Play.’ Targeted at younger audiences (the ‘future subscribers’) this will be broadcast with three main shows: a ten minute news briefing, a twenty minute round up of the biggest stories and then a forty minute deep dive into a specific topic or issue. They will also be launching a series of games as well as a free tier to warm up the audience to habitual usage.
  

• Tom Calver from The Times Data team showed us how he worked to introduce a number of data charts on their website and app based on the simple premise of showing how Britain is doing on a number of key (election) topics. It’s such a simple but brilliant idea. Kudos to him and the team for making that happen

  

• Jim Waterson, Isabelle Roughal and Daniel Ionescu gave us honest accounts of launching their own ‘indie’ publishing businesses. The message that resonated with me was ‘sell a vision’, be laser-focused on delivering brilliantly original stories and be prepared to work 80 hour weeks. If you’re starting up yourself, then start lean and never think you need to purchase the ‘best’ or most expensive. Building up a new product takes time, and you will need every penny at the beginning just to keep going. Interesting fact from Isabelle - 50% of podcasters stop after their seventh episode. I better tell Barry that we need to keep going!
  

• When you hear the words ‘Live Journalism’ you probably think of teams reporting back on the news in real-time. But it is also the name given to a new (ish) form of news communication via live performance and theatre. Verónica Muñoz Martínez, Glenda Cooper and Annie Slemrod gave a great overview touching on forced homelessness in Gaza and the 2024 Spanish floods. As an A-Level Theatre Studies student (which I almost flunked) this was right up my street. I think many journalists would crave the type of audience engagement and feedback they would get if they took their stories to the stage.

So that’s a wrap. Another key event in the calendar for journalists ticked and many new connections forged. I’ll be back at the next Newrewired event in November.

Related articles

• How Global South World Is Taking on Brainrot, Algorithms and News Avoidance

• What Newsrooms can Learn from Duolingo – The ULTIMATE audience engagement and retention app

• Five Things We Learned At The Podcast Show 2026 About The Future Of Audio

• Google’s New AI Dashboard – A Pyrrhic Victory for Publishers? PLUS World News Media Congress Highlights

• “Stop renting your audience from other platforms”: What publishers can learn from Good Housekeeping, The Guardian and The Times

It’s conference season here in the UK, so I’m treading the fine line between post-event networking drinks and next-day breakfast-event croissants. In my twenties, I would have zipped from venue to venue, sipping aperitifs like some petulant rock star, caring not for written notes or presenter quips. But now, all I do is update Substacks and socials with my post-event analysis.

What went wrong?

As someone who has mainly worked in B2C all my life, it’s been a while since I entered the B2B domain. But with the renewed focus on paywalls and subscriptions in the news industry, I thought I’d pop along to Stationers’ Hall in London to find out how the biggest players in the B2B information and events space were growing and maintaining audiences.

I could have scraped the transcripts that were available from the event. I could have used AI to craft you a masterly-looking report. But instead, you’ve got me, three drinks down, typing wildly with off-the-cuff observations. If you want a more sensible approach, talk to me tomorrow morning over a floppy-looking pain au chocolat.

Let’s dive in.

It was refreshing to hear from delegates that many do not have the fear around AI that we often hear in the publishing space. This is primarily due to the fact that many B2B businesses are more diversified, with live events and adjacent premium information products. The lesson here for news publishers is perhaps: don’t put all your eggs into one content basket.

How can B2B publishers monetise podcasts? We had a fabulous masterclass with Peter Houston and Esther Kezia Thorpe, who took us through their observations. The answer is: yes, you can monetise niche podcasts, but you really need to drill down into the exact value of your podcast and what kind of package you can offer potential sponsors. Some great learnings here.

Immediately following that session was another masterclass on squeezing maximum value from your live conference footage. Rich Hewes from Subthread did a great job showing us that video production and distribution don’t have to be expensive — you just need to be intentional about what you are going to record. How many events have you been to where you see cameras rolling but very little footage online afterwards?

“Quality journalism comes at a price and should be paid for.” That was Kevin Costello, Haymarket CEO. I know we hear this a lot in news publishing, but it was great to hear it again in the spectacular surroundings of the livery hall. One thing B2B figured out years ago: there is no long-term business model for commodified news.

“AI is moving from promise to performance.” That’s Kevin again.

“People don’t go to events anymore for commerce and networking. They want to have fun!” That was Douglas Emslie, chairman of Cuil Bay Capital and Raccoon Media Group. It was really interesting to hear his take on the “festivalisation” of events, where people want to mix work with leisure. This sounds a lot like a running event my brother goes to every year, where you — yes — run, but also drink beer along the way.
Sounds fun, but also a little precarious.

Moat, moat, moat. Perhaps the biggest buzzword of the day, but no less critical. If you don’t have a point of difference in the market, or you are just very bad at communicating that point of difference, you might as well pull up the drawbridge and turn the knights out. That’s it, medieval metaphors exhausted.

“Trust” was hammered a lot — but I couldn’t remember any funny quips, so let’s just accept that it remains extremely important.

There was a lot of talk about MCPs and investing less in the desktop GUI experience in favour of APIs. In the future, when you switch on your laptop and your preferred agentic workflow AI is ready to get crackin’, this seems like an entirely sensible approach. But agreeing on a properly monetisable model for MCPs doesn’t seem fixed yet. There were also concerns that, although MCP seems to be the future, there will be fewer opportunities to track what the audience actually does with your data than with a GUI interface.

“Everyone has a plan until they get punched in the mouth.” That was Julie Harris, CEO Expana, using a wonderful Mike Tyson quote in reference to the impact of AI.

“Trust is a compounding currency in B2B.” I really liked that.

Someone referred to the future “unicorn” of a business as someone who heads up AI but also understands all areas of the business, from sales to marketing to product. They will be a “Swiss Army knife”.

The Warren Buffett quote was used: “You can buy anything with money except the one asset people value most — time.” In other words, if you can save people time, they will pay for it. This was used in reference to live events, where you can offer paid extras like queue-jumping — otherwise known as speedy boarding — or spaces at events where attendees can work.

“As a live event organiser, we are an outcomes business, bringing buyers and sellers together. The role of our publication is to turn that experience into a 360-day experience.” I loved that way of thinking.

That’s it for now, folks. I even got to wear my very own Stationers’ tie, which comes out of the wardrobe once a year. I also have many other ties, so if that’s the kind of premium content you desire, please consider becoming a paid Substack subscriber.

Related articles

• “Stop renting your audience from other platforms”: What publishers can learn from Good Housekeeping, The Guardian and The Times

• Newsrewired 2026: In Opaque Times, Let Experimentation Shine Through

• Why Your Blog Isn’t Driving Conversions

• Competitor research starts with your own product

• What Newsrooms can Learn from Duolingo – The ULTIMATE audience engagement and retention app

Why start a podcast?

There are many reasons why you might want to start a podcast, and with more than 4.7 million listeners worldwide, that’s a pretty impressive audience! But with 342,000 podcasts being uploaded in the last 30 days alone, how can you possibly expect to cut through? And with the advent of AI and podslop you might be forgiven for just calling the whole damn thing off.

However, and somewhat ironically, now that AI summaries and chatbots are here to stay, it’s never been more important for brands and publishers to prove their authority. A good podcast can do something an AI-generated summary can’t: build trust, demonstrate expertise and create a direct relationship with an audience.

Audio audiences are also fiercely loyal, but discoverability remains a major challenge. So how do you build a podcast that people can actually find, understand and return to?

The Three P’s of Podcast (and Video) Optimisation

If you’re already seeing flat downloads, poor retention or limited discovery, this is usually where I start with clients: Product, People and Promotion.

Product: why the show exists, what it offers and how it is structured

People: how audience insight shapes the proposition, format and content

Promotion: how you make the show discoverable across podcast apps, search, social, video and owned channels

Each of these three areas is important, and they often overlap. A lack of focus in any area will result in diminished success.

Product

I’ll be blunt: promotion cannot rescue a fundamentally unclear podcast proposition. No amount of social media promotion, clever cover artwork or SEO’ing is going to elevate your podcast if the way the ‘product’ is conceived is fundamentally flawed. Well, actually, that’s a bit of a lie. Some successful podcasts do have obvious flaws but many are supported by major PR, talent or marketing budgets. Do not copy what they do blindly just because they are already at the top.

Setting Goals

I asked a number of podcast producers why some podcasts fail and they all agreed - it’s because the why and the who of the product was never identified. In other words the following was never asked:

• Why are we launching this podcast now?

• Who are the listeners and what needs are we satisfying?

• Why should the audience want to listen to our podcast above our competition?

Setting a clear goal for your podcast is the No.1 priority before you do anything. Of course, not all goals are good goals, a poor goal can cause a lot of confusion and a waste of resources. An example of a poor goal might be ‘raise as much awareness of our brand as possible’ or ‘generate as many leads as possible.’ They are weak goals because they are vague — and vague goals usually lead to vague results.

A good goal should state the problem and how you are going to resolve it. The more specific you are when creating goals, the better the outcome will be for everyone, including the listener.

This might sound all rather complicated but you can quickly start addressing these questions by simply reframing the problem like this:

Instead of writing: ‘We want to publish a podcast about X because….’

Write this: ‘Our audience needs a podcast about X because….’

(Thanks to Chris Stone for that reminder. Our UX brethren would say LEAD with the NEED.)

That’s why I always say that creating a podcast is not about creating a marketing ‘add-on’, it’s all about creating a brand with its own identity and purpose. Having a clear idea of that will really help with all the messy questions that soon rear their head when you start producing podcasts. Questions like…

What is the ideal length of a podcast?

Well, the general rule of thumb is the more frequent the podcast, the shorter it can be, whereas less frequent shows can often be longer. But that’s all poppycock without a clear understanding of what your audience exactly wants and are used to. The key really with podcasts is to create consistency with your output so you become part of your audience’s habitual routine. That’s why changing the format or schedule can really annoy listeners - unless you are transparent and explain why. If your interview with a guest takes half the amount of time a guest usually takes, don’t feel obliged to pad out your podcast for the sake of padding it out. That way ends in madness and dwindling audience retention. Which brings us neatly to…

What are the metrics that matter in podcasting?

Many publishers cite downloads as a key metric for their podcasts but if they are not careful, this can soon become a vanity metric. A download doesn’t equate to a listen. There’s quite a few metrics you could look at but nothing is more humbling than audience retention. If 60% of your audience are dropping off after 5 minutes, you have a problem which could be down to many things including the lack of…

…Story Arc

Having a narrative structure in some shape or form is one of the best ways of attacking low audience retention. You need a clear Beginning, Middle and End and in amongst that a whole series of promises and pay-offs. Clarify the problem you are going to resolve at the start and then launch yourself into the main section. Maintain audience attention by introducing something different at any point you think the audience might zone out. Hang that open loop in the air so listeners can’t wait to reach the end when the whole story arc is resolved. End with a recap, a big thanks and a possible tease to a future episode. Simples?

Of course, I’m doing 3,000 years of storytelling a huge disservice here but you get the idea. How you tackle your story arc/s is flexible, but you need structure otherwise you’re going to ramble, meander and ultimately bore.

Speaking of which…

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

A quick story-arc detour: I once realised while listening to “Light My Fire” by The Doors that the song works almost like a podcast structure: a clear opening promise, an extended middle bit (question and answer solos) that maintain attention, and a return to the original motif for resolution. The lesson for podcasts is simple: even when you wander, the audience needs to feel there is a shape.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

When podcast formats can work against you

Something I’d also mention is that while structure and format are essential, I’ve often seen podcast growth plateau because the format becomes too restrictive. A show might take off because it has two funny people talking about funny, interesting topics — what’s not to love? But if the team is unwilling to explore new subjects, bring in different guests, or broaden the creative scope because they have a rigid idea of what the format “should” be, that can quickly lead to stagnation.

This often becomes more obvious when a podcast moves onto YouTube. Video opens up a whole world of cross-promotional opportunities, especially around guests, collaborations, clips and visual formats. But teams can struggle to take advantage of that if they’re too attached to the original audio-first structure.

Of course, there’s a difference between “can’t” and “won’t.” Another reason podcasts stagnate is good old-fashioned office politics. Sometimes it’s a power play; other times it’s restrictions from the brand police. If you hit that kind of roadblock, my suggestion would be to influence through data so your arguments feel less subjective. Do plenty of competitive research, and be ready to explain why other brands in your niche have been able to grow by taking a more flexible approach.

Most importantly, take the time to understand the objections. Maybe the team simply wants to focus on audio. Maybe they’re not interested in video, or they don’t have the resources to support it properly. And that’s fair enough. But understanding the real reason behind the resistance will help you work out whether the format is genuinely protecting the show — or quietly limiting its growth.

People

By people, I mean audience and by audience I really mean ‘What do people really need in their lives?’ There’s lots of podcasts that I listen to that give me what I want - an update on this or that - and they have their place. But the podcasts that sit in my library and are often re-listened to are those that always answer the ‘Why’. Essentially, I’m not looking for breaking news updates, I want comment and analysis - ultimately making sense of what’s going on and how it impacts my life. So how do you find out what people actually need?

This may be a little repetition of Product but essentially, you want to be very clear about the audience and your niche, as well as look for a gap in the market. Ask yourself:

• Who is the audience?

• What do they struggle with?

• How will your show help them?

• Why should they listen to you over anyone else?

If I was going to launch a show tomorrow on, say, ‘Gardening Tips 101’ I would be spending time researching what challenges the audience faces in this area. Reddit, Google Trends, YouTube, X are all good ways of really drilling in along with a multitude of SEO research tools. But you do need to try and understand the audience’s need behind the questions being asked online.

For instance, with enough time I would soon realise that although ‘gardening tips’ might make an excellent podcast search optimisation title (more on that later), the real issue facing gardening in the UK (where I live) is the changing climate and how it’s throwing traditional horticultural timelines out of whack. With more random weather patterns, how do you adapt as a gardener? That’s the real need which your podcast must address. It doesn’t have to be explicit, you can still talk about the best time to plant a daffodil, but you must identify the underlying need to make your podcasts compelling and relevant.

I’m mentioning Audience Needs a lot and that’s because I’m a big fan of the Audience Needs model proposed by Dmitry Shishkin. The basic idea is that audiences do not simply want “content”; they have specific needs — to be updated, informed, inspired, connected, distracted or helped. Digital Newsrooms have been using this model for yonks and is a great way to think about the various reasons why someone might want to listen to a podcast in the first place. Adapting this model somewhat you could arrive at the following needs for podcasts which might include:

• I want the latest updates

• I want the wider context

• I want to be inspired

• I want to be part of a like-minded community

• Sigh, I just want to be distracted from what’s going on in the world

You might be able to think of many more needs when it comes to podcasting. You could add different audience ‘moods’ for instance or consumption habits i.e. what the listener is actually doing when listening to a podcast…which, for me, is often running up some random hill and back again.

You might also have audience needs that are less abstract and more data-driven. Your ideal audience might exhibit certain behaviours that you’ve identified through your first party data, for instance. It’s not uncommon for businesses to adapt the definitions here based on their own internal language or unique audience insights. Just make sure that you are actually identifying a real audience need and not projecting your own unconscious bias which sometimes happens when companies attempt audience personas. The answer, at the end of the day, is always speak directly to your audience to get the best insights.

In addition to audience needs you also need to do a decent amount of competitor research. Do you know what the most popular shows are in your niche? Go into the podcast sections of Apple and Spotify right now and type in the keywords you think are most applicable. You will see quite a few surprises. Like any form of competitive research, try and work out why the top podcasts in your niche rule the roost. Look at the cover art, titles, descriptions, formatting, guest interviews etc. But most important of all - take time to listen to them. You’d be surprised how many people don’t actually do this. Which brings us to…

Promotion

There’s no point in creating a podcast if no one is going to listen to it, and if you’ve followed everything above then you will have mitigated many of the problems that sometimes happen when it comes to podcast promotion. However, discoverability is still quoted as a major challenge to podcasters, so let’s look at some of the ways we can make this easier.

Like any digital marketing campaign, defining precise attribution to a specific acquisition channel can be more than a little tricky particularly when it comes to podcasts. Potential listeners might have seen a video on TikTok, followed a recommendation from a friend or read a review before they decide to actively search for a podcast on a platform. That’s why it’s useful to keep track of the latest audience consumption reports so you can understand which platforms to prioritise depending on your own unique circumstances.

For instance, according to Riverside, 50% of podcast listeners find new podcasts through their podcast listening app; 31% of listeners find new podcasts on YouTube, followed by Spotify (24%) and Apple Podcasts (12%). Additionally, 55% of podcast listeners say that they get podcast recommendations from people they know (word of mouth).

The recent Reuters Institute report on podcasting is also a worth-while read when it comes to consumption trends:

Based on this, be prepared to target a wide range of promotional opportunities which might include (and are not limited to): social media clips, community engagement (Reddit etc), YouTube, podcast app discovery (otherwise known as PSO or Podcast Search Optimisation), SEO, PR and paid ads.

As I started this whole guide initially by looking at PSO, I’ll start with that.

Podcast Titles and Descriptions

In podcast audits, one of the most common issues I see is that the show title, episode title and description are all doing the same vague job. None of them clearly explains who the episode is for, what problem it solves, or why it deserves attention now.

When launching a new pod I think it’s entirely reasonable to tweak and experiment with titles and descriptions - don’t assume you can’t re-edit once you’ve published. There are tools out there that can help you with this but it starts by understanding how the two main podcast platforms - Apple and Spotify - actually promote podcasts.

Those two platforms are slightly different in how they surface podcasts. Apple still places significant emphasis on editorial curation, while Spotify leans heavily into personalised and algorithmic discovery. It’s always well worth understanding how a show appears and is promoted in each environment.

Always check how the title and description are displayed, how the text gets truncated or cut off. Just like classic SEO, identify the primary and secondary keywords. Make sure the primary one is included in the title where possible. Ditch the wit and pinpoint the key topic/entity/person being discussed. Make sure your descriptions fill more than one line and don’t overboard with keywords. Be kind to your audience and take time on this part. You won’t be disappointed.

Podcast titles need to create a curiosity gap that doesn’t disappoint. A bad curiosity gap is really just a form of clickbait which overpromises and under-delivers. Clickbait titles are probably the quickest way of killing a show so proceed with caution.

There are a couple of decent ‘PSO’ tools out there that can help but do you REALLY need them? Well, I think it might depend on your objectives. I’ve certainly found they have helped highlight competitors I wouldn’t have checked out before so, in that way, they had sped up my workflow.

Once you are in this world of optimising podcast titles and descriptions then you might find yourself a mere skip and jump away from using tools like Claude or ChatGPT to assist you. Just be warned that AI generated text can really suck. In most cases, I’ve ended up writing copy myself.

Treat cover art in the same way as you might treat a social media thumbnail. It must be simple and easy ‘to get’. Does the image support the show’s main thesis even if that thesis is ‘A podcast on tax avoidance can be fun!’ Look at what your competitors are doing, create a Miro board with all your options to discuss with your team.

Should my podcast also be on YouTube?

I get asked about this a lot and my answer is ‘yes’ - but you need to understand what it entails. YouTube offers so many great benefits, extending your audience reach and providing you with all the data you could ever want. Just be mindful that video and audio audiences are not the same. There is a reason why audio CPMs are much higher than YouTube CPMs - and that’s because engagement is longer in audio for podcasts. In addition, once you start optimising for YouTube, you’re going to start doing things you may not have considered when you were primarily working with audio. For instance, if you’re not careful you can get quickly sucked into creating endless YouTube Shorts with no apparent benefit.

As Grace Jones once sung, don’t be a ‘Slave To The Algo-Rhythm.’

Sorry.

Another thing you may need to consider is switching on a camera is not the same as switching on a mic. Recording audio between two people can unearth wonderful intimate moments of absolute honesty - a key reason why people are drawn to podcasts in the first place. But switch on a camera and all of that can quickly evaporate as the dynamic changes. So be intentional when recording video and adapt if the conversational tap begins to run dry.

Lastly, think about the visual identity of your show which is as significant as the actual content. It’s an old adage but it’s true - the setting can often be the ‘main character’ of your production. Not that every video podcast needs to look like a scene from Tolkien’s Rivendell, you just need to consider how the surroundings can play a part along with any graphics that might accompany the show.

So is video essential? No, but it can help, as well as send you into a complete tangent.

I’ve previously mentioned that video clips are a great way of promoting your show. The cost of producing these clips has massively decreased with tools like Riverside, just make sure that the output is exactly what you want. Automated clipping tools don’t always select the best segments of your podcast conversation. Only YOU are the judge of that so don’t outsource that decision to a tool.

As DigiDay reported recently be careful not to alienate your audio listeners if you are also embracing video. There’s been far too many examples of presenters holding up something to the camera and talking in a way that excludes an audio listener. I should know, I’ve done it myself. Never make your audio listeners feel like an afterthought.

Does my podcast need a website?

I would say, all things being equal, yes. A website gives your podcast a single owned destination: a place for show notes, search visibility, email capture, analytics, listener feedback and conversion journeys. If resources are tight, even a simple landing page is better than scattering every call to action across different platforms. You might even want to use linktree. Just make it’s easy for your audience to find you and give feedback.

In Summary

Thanks for reaching this far. I hope I haven’t taken all the fun out of podcasting with talk of titles and descriptions but I hope I’ve demonstrated that Podcast optimisation is not a checklist you apply after publication. It is a way of designing the show so that the right audience can find it, understand it, trust it and keep coming back.

If you’re planning a new podcast, struggling to grow an existing one, or trying to understand how audio and video should work together in your content strategy, I can help.

I work with brands, publishers and content teams on podcast strategy, podcast search optimisation, audience insight and promotional planning.

Get in touch at me@stevenwilsonbeales.com and I’ll happily talk through where your podcast may be losing audience — and where the biggest growth opportunities might be.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

I’ll also be joining the Spring Community call to discuss all things podcast SEO on Wednesday, May 20 at 11 a.m. ET/4 p.m. GMT. Register here to receive a reminder and ask a question. See you there!

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Related articles

• How to Grow a Podcast: Product, People and Promotion – Beers & SEO episode

• Five Things We Learned At The Podcast Show 2026 About The Future Of Audio

• Mum, Are Podcasts Created By Humans Better Than AI-Generated Podcasts?

• Newsrewired 2026: In Opaque Times, Let Experimentation Shine Through

• What does the Reuters Journalism and Technology Trends and Predictions 2026 report tell us about where newsrooms are heading?

Related articles

• Google to Let Publishers Opt Out of AI Overviews: What It Means for Content Owners

• Google’s New AI Dashboard – A Pyrrhic Victory for Publishers? PLUS World News Media Congress Highlights

• Austen, Eliot and the Messy Business of AI Visibility & Prompt Tracking

• Why Your Blog Isn’t Driving Conversions

• Will AI Kill SEO? Publisher Growth Tactics for 2026
  --------------------
  
  Connect with Barry & Steve
  SEO Summit 2026 - https://newsseo.io
  Barry's Consultancy service: https://www.polemicdigital.com/
  Steve's Consultancy service: https://www.stevenwilsonbeales.com/

This week Barry and I had a crackin’ exchange with Luke Halls. It was great to hear about the very healthy subscription model at The Times of London, in addition to their having one of the most engaged news apps in the UK.

So let’s dive in.

----------------------
🎧 Listen to the episode now:
🎙️ Spotify:

🎙️Apple Podcasts:

----------------------

Steve: Let’s find out a bit more about The Times. For those who may not know, The Times and The Sunday Times have around 659,000 subscribers. One of the reasons I’m such a fan is that I noticed years ago how often The Times would rank particularly high in search. What became clear was that it wasn’t just a search play, it was an app play as well, because you were creating very targeted articles that gathered people in via search and then added value inside the product. Before we go into the SEO strategy and how you might be dealing with our lovely AI overlords, it’d be good to know how you got into journalism in the first place.

Luke: Yeah, of course. I think, as with most people on my team and possibly in the industry as well, my route was quite twisty, bendy, windy. I’ve always been a music guy. I’ve been playing in bands for half my lifetime. When you’re a teenager, anything editorial to do with music really captures the imagination, so I was very into titles like NME, Q, Mojo, and Guitarist magazine. That was probably my first real introduction to journalism in a meaningful way. It only really kickstarted when I was about eighteen and managed to land a week’s internship at The Telegraph in 2013. I was on the SEO team, which is funny given where I am today. During that week I learned the fundamentals: internal linking, pitching content ideas, how taxonomies work, that kind of thing. Then I went to Liverpool Hope University, where I studied English and communication studies. As part of that, I did modules on newspaper journalism and digital publishing, and on the side I was writing music and concert reviews for the student radio blog. The copy was awful, but it taught me so much about the process of journalism and what good looks like.

Steve: In SEO, everyone has a lightbulb moment when they figure out that this is for them. Did you have one of those?

Luke: I definitely did. That came after university. I’d graduated and was still writing for music websites on the side, but I also landed an internship at Wallpaper magazine, which is an art and design-focused publication. I moved in as the digital intern and got involved in everything digital: production workflow, social, homepage management, and SEO. I spent a lot of time discovering the different pockets of digital publishing. A couple of years in, when I was given a permanent position, I realised I had a much better understanding of search and what meaningful traffic looked like, especially because we were a non-paywalled brand and search was a key referrer. This was in the days of AMP, so that was all the rage then. After that I moved over to HELLO! magazine, where I worked as an SEO executive, then moved into an audience editor role, which gave me the freedom to experiment and really focus on search as a referrer. I’ve been at The Times for almost five years now, exclusively in SEO roles. I came in as an SEO journalist, then moved into a tech SEO role as an analyst, then deputy head of SEO, and as of about three weeks ago, Head of SEO.

Steve: Congratulations. Tell us a bit more about the role of SEO at The Times. What are your objectives, how has it changed, and where are you focused at the moment?

Luke: I lead a team of seven spanning editorial and technical SEO. Fundamentally, my job is to make sure we’re meeting quality audiences via search with the aim of converting them into subscribers. The SEO team exists to support the editorial output of the newsroom across all desks through a search lens. In practice, that means identifying trends, capitalising on opportunity gaps, building up the newsroom’s understanding of SEO, and meanwhile striving for really squeaky-clean technical SEO performance. Basically, staying as 360 on search as possible.

Steve: The first thing that struck me there is that SEO is very much part of the machine to drive loyalty and subscribers, right?

Luke: One hundred per cent. We use data religiously in this newsroom, and we look at it through two different lenses: audience and subscribers. A huge part of our success is thinking about how we can add our voice to a conversation in a meaningful way that people want to read and want to subscribe to read. Because we have such a strong history as the paper of record, we’re also very focused on leveraging that expertise and elevating the perception of us as a trusted brand. We’re mindful of where we can add ourselves to the conversation in a way that makes sense for the brand.

Steve: Without putting you on the spot — which I am — are there examples where you wouldn’t cover a story, or where you’d say, “We’re not going to cover that in that way. We’re going to wait, or take a different angle,” because there’s more correlation with subscription or loyalty?

Luke: My team is always very mindful of the trends we raise as potential avenues to drive search traffic. We’ve got heaps of data points that tell us where our core readership and our subscribers are showing interest, and where something might be fleeting or not that interesting. I can’t think of a perfect example off the top of my head, but essentially we place subscribers at the heart of what we do editorially and we don’t want to do anything to compromise that core readership. So we’re very mindful to target trends that cater to both sides.

Steve: Based on the audience information you have, has the introduction of AI changed the kinds of questions you’re asking? There’s obviously a lot of revision happening in news organisations with AI Mode, AI Overviews and so on.

Luke: We understand that AI summaries, particularly AI Overviews, are cornering certain areas of output, so we’re very mindful of how we package content with that in mind. We’re careful not to commission or pitch stories that we think are going to be swallowed up immediately on publication. We’re focused on identifying the gaps where we can tangibly make a difference. We know AI summaries are grabbing the attention top of fold, so we challenge ourselves to identify what the right opportunities are, what the best keywords are, and what is actually going to move the needle in a meaningful way.

Steve: I mentioned the ski holiday example because The Times came up for me as a really useful resource. That feels like one topic where you’ve always performed particularly well, but I’m sure there are others. Are you mindful in your weekly or monthly planning that you need to super-serve the topics that have already done well for you?

Luke: That’s exactly it. We continuously scrutinise our data, and understanding keyword intent is one of the main things we think about. It’s all about analysing things on a URL-by-URL level, monitoring search to see whether something has been absorbed by summarisation, and asking: if something has worked, how can we build on that and develop it as an area of expertise?

Steve: That leads into the current landscape with AI. How are you adjusting to the new search landscape?

Luke: We’re very aware of the parts of our website that are being most tangibly hit. Lifestyle content is probably the area that’s more likely to be swallowed up, and the rise of zero-click search is obviously something we’re continuing to talk about internally. So if we’ve commissioned a piece of content, we ask how best to serve it, where it’s going to drive the most traffic, or more importantly, the most meaningful traffic. We’ve always viewed Google Discover as secondary to organic search because it’s not as active or intentional as someone going out of their way to Google something.

Barry: That’s interesting because a lot of publishers end up being very Discover-first. They chase the spikes.

Luke: Yeah, and we’re quite careful with that. Discover can be useful, and we have seen encouraging growth in impressions, clicks and CTR, but because it is volatile, we don’t want it to become the core strategy. We’re quite sceptical of it in that sense. We look at whether something has a clear trend angle and whether there is obvious keyword intent, but even then, the fundamentals matter most: the headline, the lead image, and making it enticing without tipping into sensationalism.

Steve: And I suppose that comes back to being disciplined around the subscriber proposition.

Luke: Exactly. We’ve got very strict standards on headlines because we want the subscriber experience to remain front and centre.

Barry: One thing that strikes me with The Times is that it’s never felt like a brand that chases traffic for its own sake. It feels much more like you decide what is worth doing for the audience and then apply SEO to it, rather than the other way round.

Luke: I think that’s fair. We place subscribers at the heart of what we do editorially. If we are looking to commission something, it has to meet the demands of a readership that is paying for it. It has to provide value to those who are going to go out of their way to pay to read it.

Steve: That’s a really important distinction because in some organisations SEO became the tail wagging the dog.

Luke: Yes, and that’s not the position we want to be in. Search supports the journalism. It helps us reach people. But it doesn’t define the editorial mission.

Steve: How do you think about original journalism in that context? Because if AI summaries are swallowing more commodity content, then surely the moat becomes the things only you can do.

Luke: Absolutely. Readers associate us with premium original journalism. That’s where our authority is strongest. We know people come to us for trusted reporting, expert-driven analysis and distinctive editorial products. Those are the areas where we think we can continue to add real value and where audiences will still go beyond the summary because they want the depth.

Barry: Do you think that applies particularly to things like guides and recurring editorial franchises as well?

Luke: Definitely. We have areas where we know we already have authority and established demand, and the question is how we deepen that. If something works, we ask how we build on it, how we develop that expertise, and how we make it even more useful for the audience.

Steve: Let’s talk about paywalls. One thing I’ve always found interesting is that The Times has stayed very firm on the idea that the journalism is worth paying for.

Luke: It’s always been a very conscious decision that our journalism is worth paying for. We have run a couple of open-paywall weekends in the last few years to measure the interaction with that, and it was a useful reach exercise. But fundamentally we’re very self-assured in the quality of our content. Across referrers, if people find us, they find high-quality, expert-driven journalism that we believe is worth the paywall.

Barry: Do you experiment with things like gift links, like The New York Times and The Financial Times, where subscribers can gift an article for free to a limited number of readers?

Luke: That’s a very good question. I don’t have the answer to that — that would be something for our marketing team.

Steve: I think you should pick that as your idea, because I do think it helps with subscription rates. Actually, I just subscribed for £1 on The Times and I did get an alert saying I could share that to two or three people under a friends-and-family approach.

Barry: Beat me to it. Well done.

Steve: It’s an interesting point about paywalls. Some publishers have them, some don’t. In the conversation around AI-resilient content or original journalism, there’s a lot of language and a lot of exercises you can do with editorial teams who don’t have paywalls to get them to invest more time in original research. Obviously that has to be supported higher up so teams are given the freedom to experiment. But sometimes it might just come down to asking: if you had a paywall, what kind of content would you produce? Sometimes simple prompts like that can elicit great ideas. I’ve found with teams I’ve managed in the past that when traffic has been challenged, asking “What would you write if there weren’t any targets?” produces a lot of great original ideas. Sometimes being too wedded to a target can lead to playing to the algorithm instead. I don’t know if you’ve got any response to that, Luke.

Luke: That’s a really good question. I think it does come back to confidence in your journalism and understanding what your audience actually values. If you know what your readers come to you for, then it becomes easier to prioritise the ideas that will matter in a subscription context and not just because they look attractive in a dashboard.

Steve: On the subject of Discover, do you actively optimise for it or do you just recognise when something has the right shape for it?

Luke: More the latter. We don’t necessarily monitor Discover trends explicitly, but we do look at whether a story is tangibly anchored to a trend and whether there’s clear keyword intent behind it. That shapes how we package the story for the audience we want to reach. For us, success in Discover still comes down to fundamentals that haven’t changed for a long time: the headline and the lead image, presented in the most enticing way possible without being sensationalist. We’ve seen strong results from headlines that are first-person led, expert driven, or curiosity inducing without drifting into clickbait.

Steve: On your point about headlines, I’ve recently done some work advising on content management systems, and one of the recurring questions is how many headline fields you should have. Years ago you might have had one headline for the article, an SEO title or page title, and an OG title for social. Now you also have Google Discover in the equation. Some publishers have decided to go with just one headline field. What’s your view on that?

Luke: It’s very topical, because it’s something we’re discussing internally at the moment. I think The Times has maybe been a little late to the party with multiple headlines. We currently have two: the homepage headline and the article headline, which is what you see in search and across social. We’ve been scoping out the idea of SEO-only titles for organic search because we know there are situations — live blogs, for example — where the article headline may be anchored on one point while other publishers are ranking in search on slightly different topical points. We’re looking to develop more flexibility there. But more broadly, we’re very confident in our headline construction. Our production staff are wizards at their craft. They can take a keyword and turn it into something reader-friendly, accurate, interesting and eye-catching.

Steve: Do you use AI in any way, shape or form for that? I’ve seen some AI integrations in CMSs recently, and some of the headline suggestions are actually quite impressive.

Luke: We would never, strictly speaking, write an article exclusively using AI. We do have a tool in our CMS that can look at the copy and provide suggestions, but if that happens it is rigorously sense-checked and always for the human reader. We don’t want something that’s just garbled language or that removes the meaning we want the story to have. The functionality exists, but we’re very mindful about how it’s applied and we tread carefully.

Barry: They are getting better, though. I’m a bit of an AI sceptic in this regard, but I’ve seen some really interesting newsroom integrations where AI alleviates a lot of the busywork that journalists and editors have to do — things like OG headlines, alt attributes for images, even internal linking to some extent. I’m not sure how much you can say about this, but is that something happening at The Times as well?

Luke: I can’t talk too much about it, but it’s definitely something we’re thinking about. From my team’s perspective, those monotonous tasks are absolutely areas we think about. But if we do pursue anything, there always has to be a human at the end of the chain to sense-check it and make sure it’s logical, sensible and ultimately in service of the reader.

Steve: Barry, on the question of headlines, there’s been some talk in the press about Google rewriting headlines. Do you have a view on that?

Barry: Yes. There was an article in The Verge saying Google is rewriting headlines using AI. I read it, saw the screenshots, and thought: that’s basically what Google has been doing for years, just now with AI involved. I haven’t seen evidence of Google rewriting Top Stories headlines, which is the context where news publishers should really care. In the standard blue-link results, Google has always tweaked titles to some extent. In Top Stories, it generally picks from one of the existing headline variants — the visible headline, the structured data headline, sometimes the OG title or SEO title. I haven’t seen it invent a headline out of thin air. In many cases, when Google does rewrite a headline, it actually helps the website because it improves click-through rate. So rather than objecting to it, I’d say watch what Google does and learn from it.

Steve: That’s an interesting one, because we don’t want reporting like that to discourage people from investing the time in headline optimisation. Headlines really are the windows of your shop. Luke, if we finish with a look ahead, what are the key things you’re thinking about now?

Luke: Understanding what makes our subscribers tick and leveraging that more is a big one. SERPs are changing so much, and I’m keen to understand better how both Discover and organic are evolving, especially with things like Reddit and social bleeding over. I’m also excited by the idea of more cross-collaboration between different audience teams internally. YouTube is another area I’d like to spend more time learning about, because when we look at search behaviour week to week, we see so much YouTube carrying through. It’s a search engine in its own right, so it would be silly not to leverage the expertise of a team that works on search engines to benefit performance there too.

Steve: Let’s try to get the YouTube version of this podcast ranking for some classic keywords for you, Luke.

Luke: I already hope so.

Steve: Brilliant. It’s great to have you on. We’ll have to check in with you again in six to eight months’ time. Barry, I believe you have an announcement regarding an event happening towards the end of 2026.

Barry: Yes — we’ve launched the 2026 News and Editorial SEO Summit, already the sixth edition. The first batch of speakers has been announced, most of the agenda is nailed down, and I’m very excited about the line-up. We’ve expanded the event a little as well. It used to be a two-day online event; now it’s three days, each with a specific theme. Day one is SEO in the context of news. Day two is broader SEO and AI. Day three looks at audience growth beyond search and Discover — podcasts, video, newsletters, that sort of thing. Tickets are now on sale at super-early-bird pricing.

Steve: It’s a great event. It really opens this space up to a wide range of specialists, and it caters to the whole spectrum of people who are starting their journey in audience growth, because it’s not just exclusive to SEO. There’s video optimisation in there too, and it’s moved into more of a search-everywhere approach. Lastly, I have my own Search and AI Fundamentals Visibility course in London on May 19, aimed at people trying to make sense of how to move forward as a journalist or newsroom manager in this space. It very much builds on the kinds of conversations we’ve been having today. I think it’s “the wood for the trees” — this beer is beginning to take effect. But let’s leave it there. Thanks so much, Luke, for coming in, and we’ll get you on a future pod.

Luke: My pleasure. Thank you so much for having me, guys. Great chat. Take care.

Related articles

• The Secrets Behind The Times’ SEO Success with Luke Halls

• Commodified News is Dead. Long Live Commodified News!

• The Power of Google Discover for Publishers with Nicola Agius

• As a Publisher, Should You Be Investing More in PPC or Organic?

• “Stop renting your audience from other platforms”: What publishers can learn from Good Housekeeping, The Guardian and The Times