In partnership with
👀 A new look
Welcome to Franc Talking!
As any good marketer should, I listened to the feedback of the survey I sent out recently and have changed the design as it just wasn’t working for some. I hope you like it.
I do listen and really value your feedback, so if there is something you want to see in 2026, make sure to submit your input here.
An action-packed edition today, including my discussion with Mark Williams-Cook.
Turn AI into Your Income Engine
Ready to transform artificial intelligence from a buzzword into your personal revenue generator?
HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.
Inside you'll discover:
A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential
Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background
Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve
Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.
What happened this week
🗣️ Top Stories
If you work in marketing, product, or tech - and want to keep pace with the shifts AI is driving across search, software and strategy - here’s what you need to know this week.
🏠 Rightmove in the news for the wrong reasons
AI search is starting to expose how vulnerable marketplace “gatekeepers” really are. Platforms like ChatGPT can already pull together property listings, local data and next steps in seconds, reducing the need to trawl portals like Rightmove.
While the threat is still early, it’s existential: if AI becomes the interface, the middleman loses leverage. Rightmove knows this, committing £60m to AI-powered search, chat and discovery tools to defend its position - but investors are uneasy, margins are set to fall, and long-standing underinvestment is catching up with the business.
It’s a familiar Franc Talking theme that is endangering many brands. In an AI-first world, distribution and attention shift to the interface that answers the question, not the platform that once owned the traffic.
The Rightmove story highlights why marketplaces are at risk when you have a platform that can go out and do the research and heavy lifting for you.
If I want to find a four bedroom house in North London that is near a park, has a local village vibe (with some independent coffee shops) and good schools, I can prompt ChatGPT to help me find this information, without having to visit an Estate Agent or one of the big property platforms.
This is of course existential risk for Rightmove as traffic numbers dwindle, Estate Agents fees drop - suddenly the platform isn’t as appealing for investors.
One area I have been working with affiliates on is how often they are used in citations to help AI generate the answer to the user. For some affiliates, their business model is moving from a pure play CPA to a flat fee retainer, using Obsero data to back up their influence and AI Search presence.
Brands like Rightmove have the option of developing apps for the likes of ChatGPT and follow suit of Zillow, the US based property marketplace, but you are then handing over invaluable data to a platform that could eventually swallow you up.
🚀 ChatGPT approaches 900 million WAU (but Gemini is closing in)
The AI search race is tightening. While ChatGPT still dominates, Google’s Gemini is growing fast across both app downloads and web usage, prompting OpenAI to declare an internal “code red” and refocus on its core product.
Gemini’s momentum matters because Google’s true AI reach goes far beyond its standalone app - billions already encounter AI answers through Search, AI Overviews and AI Mode.
For Franc Talking, this reinforces a key point: AI search isn’t a single product battle, it’s an interface war embedded across ecosystems. ChatGPT may still win on engagement and raw usage, but Google’s distribution advantage means visibility, not just users, will decide who ultimately controls attention - and where ads, commerce and influence eventually flow.
Following Sam Altman’s ‘Code Red’ a couple of weeks ago, the OpenAI team will be thrilled to be nearing 900 million WAU, although will be fully aware that Google’s Gemini is growing in popularity.
Google of course has AI Mode, which has been pushed via TV, Chrome and YouTube recently resulting in a surge in demand for the platform, and AI Overviews, which is now under the scrutiny of the EU for anti-competition, so has always been the main threat, it was just more of a surprise it took Google this long to ‘catch up’.
With OpenAI forecasting 2.6 billion WAU by 2030 (est ~220million paying a subscription), they still have a long journey ahead of them to convince users to ditch Google in favour of making ChatGPT their AI platform (to act, search and discover) of choice.
🚀 ChatGPT 5.2 is released
GPT-5.2 is a meaningful step forward, but not a clean breakout. Developers testing the new models point to stronger reasoning, fewer hallucinations, better long-running tasks and genuine gains in complex coding and maths - albeit with higher costs, slower performance and tighter guardrails.
Crucially for AI search, the context isn’t just model quality but competition: Gemini 3 is forcing OpenAI to move faster, and Google’s distribution advantage across Search, AI Overviews and Android keeps the pressure on.
The takeaway for Franc Talking readers is familiar - model upgrades matter, but the real battle is where these systems surface answers at scale. Accuracy, reasoning and economics are improving, but control of the interface is what ultimately decides who wins attention, trust and downstream value.
I do feel that the likes of OpenAI, Anthropic and Google should focus more so on the customer outcome from the model usage, rather than the model name - which is alien to most users.
That said, ChatGPT enhanced its model this week, with the launch of GPT 5.2, which it claims will “enhance productivity across real-world tasks” and “better reliability & fewer errors”.
Over the past six weeks I’ve shifted towards Google Gemini, but am still a subscriber to ChatGPT. I value both platforms and use either for different tasks, however - ChatGPT has frustrated me (given I’ve used it longer), not so much with the hallucinations, but always attempting to provide an answer, even if it doesn’t have full context of what is being asked.
I have noticed follow up questions to define the criteria more, which is something that I value over spouting nonsense and aiming to just to answer the question.
Fewer errors, hallucinations and clarifications is a welcome addition.
The best articles I’ve been reading this week on AI Search
📖 What I’ve been reading this week
The best articles this week on AI Search and how it affects business, marketing and tech.
🤖 AI
🔎 Google & Search
Google Zero is under investigation by the EU (Sub required)
Google Tells Advertisers It’ll Bring Ads to Gemini in 2026 (Sub required)
Learn about Mark’s take on AI Search and what you need to know to stay relevant
📹 My discussion with Mark Williams-Cook
There are a handful of SEOs that I always look out for to get their take on the latest developments in search. Mark is one of those people, usually with a humerous or insightful take.
Mark has that ability to explain complex systems without theatrics, call out nonsense without ego and still make you laugh while doing it. I like the fact he challenges and asks for evidence of claims, rather than blindly following what others do.
This conversation was less about chasing the latest acronym and more about understanding how search, trust, brands and incentives really work - whether that’s Google, ChatGPT, Reddit or any other platform where you are trying to grow your brand.
In this discussion we covered:
GEO vs SEO (and why the argument misses the point)
Why “GEO” doesn’t represent a clean break from SEO, but an evolution of the same fundamentals - intent, trust, authority - playing out in new interfaces and behaviours, not a brand-new discipline invented to sell playbooks.Reddit, reputation and why brands can’t fake it
ChatGPT citing Reddit doesn’t mean brands should rush in with corporate accounts. Real visibility comes from being worth talking about, not trying to game communities that will happily tear you apart if you do.The myth of GEO ‘tactics’ (schema, llms.txt and other comforts)
A clear, technical dismantling of why many so-called GEO tactics lack evidence, don’t align with how LLMs are trained or retrieved, and often repeat the worst habits of early SEO - tricks without durable value.Hallucinations aren’t mystical - they’re just bullshit
Why softening the language around AI errors hides real risk, especially as trust shifts from websites to generated answers - and why marketers need to understand this properly, not hand-wave it away.Why brand still beats manipulation (even in AI search)
From site quality scores to brand demand, the same truth keeps resurfacing: systems reward entities people actively seek out and trust. Trying to reverse-engineer shortcuts is fragile; building equity compounds.
You can follow Mark on LinkedIn, Bluesky and subscribe to his newsletter, Core Updates.
He also runs AlsoAsked.com, his well known marketing tool that helps brands understand what questions are being asked in search.
For marketers keen to grow visibility in AI Search - here are five tips to get you started
💪 What you can do to drive growth in your business
Mark Williams-Cook joined me to discuss GEO, why SEO is still fundamental and a warning for brands that try to fast track growth on platforms like Reddit.
If you want some actionable tips that you can pick up today, check out the top five from what we spoke about in our chat:
Optimise for follow-on questions, not single answers
Map the next question a user is likely to ask and answer it on the same page or journey. Reducing “time to result” beats ranking for isolated queries, especially as conversational search becomes the norm.Measure visibility where clicks no longer exist
Start tracking where and how your brand is referenced in AI answers, forums and summaries - not just traffic. Impression-style visibility and sentiment are becoming leading indicators of future demand.Use Reddit as a signal, not a channel
Analyse what real users complain about, recommend and compare - then fix or reinforce those things elsewhere. Let customers create the conversation; don’t try to manage it directly unless you’re prepared to lose control.Kill low-value content that dilutes trust signals
Thin pages, scattered help articles and half-answers weaken overall credibility. Consolidate, merge or remove anything that doesn’t clearly satisfy intent - especially content written purely to “exist”.Build brand demand deliberately, not accidentally
Align PR, partnerships, product messaging and content so people actively search for you by name. Systems consistently favour entities people already recognise and seek out - this is now a growth lever, not just a brand one.
Obsero new feature launch
⚒️ Obsero launches advanced citation analytics
Obsero now includes Advanced Citation Analytics that enables brands to be laser focused on engaging with the sources that are being used in answers by ChatGPT, Perplexity and Google.
It has been a busy quarter at Obsero and I’m delighted to announce the launch of advanced citation analytics available to all customers.
With advanced citation analytics, brands are able to drill down on the top risers, fallers and understand which publishers are the most influential in their category. I’m sure you’ve seen the headlines about how important Reddit, YouTube and Wikipedia are (stating the obvious, right?) - but how does that apply to your category?
Advanced Citation Analytics enables you to view the top risers in your category. This means those that are gaining traction in citations and thus are becoming more influential.
Screenshot from Obsero showing the top citation riser in the ‘get fit’ space.
Likewise, brands can also see the biggest fallers that are declining in influence (note - Reddit is taking a tumble here):
Screenshot from Obsero showing the biggest citation fallers in the ‘get fit’ space.
You can also view the most influential sources based upon number of citations, prompts and unique pages:
Screenshot from Obsero showing the most influential sources in the ‘get fit’ category. Interesting to see that Reddit has a much larger number of pages that are sourced, but tomsguide.com has far fewer, but is reference more often (just!).
All data provided is for the ‘get fit’ category in the UK.
If you’d like to see a demo of this in action, feel free to reach out to me via email or through Obsero’s contact form.