Welcome to the world of Search

The one thing I love about working in digital marketing is the pace that it changes and the fact it is a very dynamic market. Every few months the industry has another "search is dead" moment. This summer gave us plenty of material for one. AI Overviews expanding to fill more real estate, ads arriving in ChatGPT in the UK, and Google telling Reuters that its new EU results are the worst quality drop in its history. Reading the period properly, though, and I think the story is different. The biggest changes between August and mid-September were not ranking algorithms. They were about control and measurement , who can see AI visibility, who can switch it off, and what metrics should brands actually be reporting to the board. Here is my thoughts of what mattered, what didn't, and what I'd do about it. 1. The algorithm story was mostly noise There was exactly one confirmed ranking update in the period: the August 2026 spam update, which ran from 18 to 21 August. It was short, with no new policies and no blog post, but it hit sites relying on scaled and automated content hard. What there wasn't was an "August core update". Several agency blogs published recovery guides for one. Google never confirmed it. The trackers did spike in early and mid-August, and there was a strange drop around 4 September that Glenn Gabe saw reverse on 13 September for some large sites. That's volatility, not an announced update. There have also been issues with some tool providers that track visibility due to Google blocking crawlers that scrape SERP results. My view ... if your reporting uses industry blog posts to explain traffic changes, you're telling clients stories. Use the Search Status Dashboard, check first-party data, and wait before changing any content. 2. Search now works differently depending on where you are Two regulatory changes deserve more attention from UK than they've had. From 30 August, Google's site reputation abuse manual actions no longer affect rankings for searchers in the EEA, following pressure from the European Commission. UK searchers are outside the EEA, so enforcement here is unchanged. Less noticed: Google expanded its site reputation abuse guidance to roughly ten times its previous length, and that applies to everyone. On 8 September, Google launched a redesigned EU results page for travel, hotels and restaurants under the Digital Markets Act. Aggregators now get the most prominent position, and direct supplier listings lose features like date filters and live pricing. Again, the UK doesn't get this layout. My view ... As we look to understand how audiences differ across markets we also need to understand and differences in the final platform interface. Multi-market brands need EU and UK reporting separated, and travel brands with EU exposure should capture baseline data now. 3. AI visibility finally became measurable, with limits On 31 August, Google's generative AI performance report and its AI opt-out control became available to every site in Search Console. It's worth remembering this started in the UK, after the CMA required Google to give publishers a way out of AI features. It's a real step forward, but the data is thin. You get impressions across AI Overviews, AI Mode and Discover. You don't get clicks or queries, and a logging error undercounted impressions from 13 to 17 August. Google says opting out won't affect standard rankings, but it will remove you from AI answers completely. My view ... for most commercial brands, opting out on impressions data alone would be a mistake. Baseline for several months first. Combine it with Bing's AI Performance report, which shows citations and grounding queries, and you now have the start of a proper AI visibility dashboard. 4. AI is taking over the page, and it's being monetised The change that concerns me most for organic performance came on 27 to 28 August: for some queries, AI Overviews now expand automatically into a full AI Mode-style answer with the follow-up box already open. Google confirmed it. Organic listings get pushed well below the fold. So tracking pixel depth should be a key consideration. The paid side is moving just as fast. Google is testing traditional exact and phrase match campaigns in AI Mode, as well as text link ads. And on 11 August, OpenAI launched ChatGPT Ads in the UK, largely to mixed reviews. My view ... AI Mode and ChatGPT now need to be in Total Search planning for both paid and organic. For financial services and other regulated brands, FCA financial promotion rules apply to conversational ads, so compliance needs to be involved early. 5. So is AI replacing search? The best data says no This is the most important finding of the period, and it comes from Similarweb's 2026 Generative AI Landscape report. Generative AI web visits grew 70% year on year to 9.5 billion a month. On its own that sounds like AI eating search. But the share of ChatGPT users who also use Google was 95% in September 2025 and still 95% in May 2026. If people were switching, that figure would be falling... But it hasn't moved. Two more figures support this. ChatGPT's share of generative AI traffic fell from around 76% to 53%, but the share went to Gemini, Claude, Perplexity and DeepSeek, not back to search. And people aren't leaving search; they're clicking less inside it. AI Overviews now appear on more than 40% of US searches. I'd add three caveats before anyone puts that 95% on a slide: It measures audiences, not queries. Someone can still use Google while sending particular types of question to ChatGPT. It excludes API use, desktop apps and in-app AI, which is where replacement of search tasks is most likely. It's vendor research from a company that sells AI visibility tracking. The dataset is credible, but the conclusions suit Similarweb commercially. My view ... this supports what I've argued for a while, and it aligns with WPP finding that most purchases go to brands people were already predisposed towards. Search is where preferences formed elsewhere get confirmed. AI hasn't replaced that. It has added a new layer before the click where those preferences are shaped. 6. "GEO is just SEO" is true only inside Google This is a new and not a new one. On 25 August, John Mueller repeated Google's position that there's nothing special to do for generative AI responses in Search. For Google's own AI surfaces, I largely agree: they draw on the same index and quality systems. Outside Google it doesn't hold. Orbit Media's citation study found four AI models cited the same domain for the same question just 1.7% of the time. Visibility across ChatGPT, Gemini and Claude is fragmented, and it depends heavily on earned media and third-party mentions rather than your own site. What I'd do now Only diagnose confirmed updates. Ignore unconfirmed "core update" claims. Baseline AI visibility before touching the opt-out. Revise organic click forecasts down for informational queries. Treat Google and the other AI engines as separate problems. Track each engine and invest in earned media. Put ChatGPT Ads into UK plans, with compliance review built in, to test, but there is a minimum investment. Change the KPI, not the channel mix. Keep investing in search, and report citation frequency, share of model and branded search lift alongside traffic. The brands that get this wrong will either cut SEO because they believe search is dying, or ignore AI visibility because search looks fine. The data supports doing both: keep search strong, and measure what's happening before the click. Sources Google August 2026 spam update: Search Engine Journal , Search Engine Roundtable September volatility and reversal: Search Engine Roundtable Site reputation abuse in the EEA: Search Engine Land EU DMA results page: Search Engine Land Generative AI report and control: Search Engine Land AI Overviews expanding: Search Engine Roundtable Search campaigns in AI Mode: Search Engine Land ChatGPT Ads UK launch: OpenAI Is AI replacing search: Similarweb Mueller on generative AI: Search Engine Roundtable LLM citation study: Orbit Media Bing AI Performance: Bing Search Blog

I sat in a client review earlier this year where two AI visibility dashboards, both paid for by the same organisation, reported the brand's share of voice in ChatGPT eleven points apart. Nobody in the room could explain why. The head of digital asked which number she should put in the board pack. I told her neither, and that the honest answer was a range with a confidence note attached. That went down about as well as you would expect. Thirty years in this industry and I have never seen a measurement category grow this fast on foundations this thin. The IAB now counts more than twenty companies selling AI visibility measurement, each with its own methodology, each capable of producing a different answer for the same brand. Only sixteen percent of brands systematically track AI visibility at all. Meanwhile Fractl data reported by Digiday puts roughly twenty four percent of search and content budgets into AI visibility work. We are spending like the measurement is solved. It is not. What still holds up, for the forceable Let me set the table before I clear it. Branded search volume is now one of the most commercially useful signals a brand has, because it is where AI influence surfaces. Similarweb work and our own at WPP Media found that roughly fifty six percent of AI influenced traffic arrives as branded search rather than an AI referral click, days after the conversation happened. Impression share in Google Search Console still matters, arguably more than clicks do. Paid search conversion and quality score still matter. Crawlability, structured data and site speed still matter, because a page a model cannot read is a page it cannot cite. What has stopped working is the click as the primary unit of account. Ahrefs measured a fifty eight percent reduction in position one click through rate when an AI Overview appears, and Pew's browsing data found users clicked eight percent of the time with an AI Overview present against fifteen percent without. If clicks are still your headline KPI, and Kantar research suggests they are for around seventy eight percent of brands, you are measuring the shadow rather than the object. What actually changed, and why it matters now Three things happened in quick succession this year that should reset how CMOs think about this. First, Google gave us a real, if partial, measurement surface. On 3 June 2026 Google launched dedicated Search Generative AI performance reports in Search Console , rolling out to UK site owners first under CMA pressure. For the first time you can separate visibility inside AI Overviews and AI Mode from ordinary organic. The catch is significant: impressions only. No clicks, no click through rate, no queries. It answers "did I appear" and refuses to answer "what was it worth". Second, the scale question stopped being theoretical. Semrush analysed 126 million US AI search prompts between January and April 2026 across ChatGPT, Gemini, AI Mode and AI Overviews, covering 1,200 brands in 22 industries. Only 36 of those brands held a top 100 position on every platform in every month. The same study found ChatGPT citing an average of fifteen sources per response against Gemini's three. Per platform measurement is not a nice to have. A blended visibility score across engines that behave that differently is an average of incompatible things. Third, and this is the one that should worry vendors, Rand Fishkin ran 2,961 controlled tests across ChatGPT, Claude and Google AI with 600 volunteers . Ask an AI for brand recommendations a hundred times and you get the same list fewer than one time in a hundred, and the same list in the same order fewer than one time in a thousand. His conclusion, which I share, is that visibility percentage across many prompts run many times is defensible, and any tool selling you a ranking position in AI is selling you nothing. The frameworks I would build on, and where I would push back The IAB's Measuring Visibility in the AI Era is the most useful thing published on this in eighteen months. Its four P's give us a causal hierarchy rather than a metrics soup: Presence, whether you appear at all Prominence, where and how visibly Portrayal, in what context and with what accuracy Persuasion, whether any of it moves anything Its real contribution is the split between directional and decision grade data, and a floor that should embarrass a lot of vendor decks: fewer than fifty queries in a measurement programme is classed as exploratory, not even directional. Two honest caveats. Caroline Giegerich told AdExchanger the IAB deliberately avoided calling this a standard, because standards need stability and the market has none. And the framework measures single query, single response visibility. That is not how people use these tools. Real decisions unfold over four, six, ten turns of a conversation, and a brand can be named confidently at turn one and gone by turn five. That gap is why I use the PLUS framework with clients alongside the IAB vocabulary. Four factors the IAB does not fully price in. Personalisation: responses adapt to account history, so universal rankings are a fiction and you should audit by persona instead. Location: outputs are hyper localised, so measure local citation accuracy across your top territories. Uniqueness: models are probabilistic, so batch test the same prompt fifty times and report mention probability, not a snapshot. Stateful ness: context carries, so track brand decay rate across a thread. I want a brand holding above fifty percent mention visibility at prompt four and above forty percent by prompt six. That number tells me more about commercial resilience than any single answer share ever will. Traditional metrics and their AI era equivalents