Market analysis
Marketing analytics, first half of November 2026: ChatGPT starts keeping score
In the first half of November an AI assistant began reporting its own mentions and clicks to the businesses it names, while studies showed AI-shaped visits landing in search and direct, agents filling the logs and rank trackers struggling through a core update.
Signal readout
Topic mix
- AI: 31%
- Google & Search: 31%
- PPC: 8%
- Analytics: 26%
- Brands: 4%
- AI31%
- Google & Search31%
- PPC8%
- Analytics26%
- Brands4%
Share of the issue's words about each topic, counted section by section.
Why it matters
Stop reading the AI referral line as the size of AI's effect. Put three numbers side by side each month: AI referrals in GA4 (segmented on both referrer and utm_source), branded search clicks in Search Console, and direct visits to deep pages. When an AI surface grows, the second and third usually move before the first does.
The short answer
- What
- A reading of marketing measurement in the first half of November 2026: ChatGPT's Business Hub and its mention and click counts, how much traffic AI assistants send by eight different counts, where AI-shaped visits are filed, Search Console's API update and the Google tag's merger with Tag Manager, AI agents in the logs, rank tracking during the November 2026 core update, and the results and claims of measurement vendors.
- Why
- Because for the first time an AI assistant reports on itself, while the reports businesses already rely on still file AI-influenced visits as search or direct, count agents as visitors, and read rankings through tools Google is actively slowing down.
- Who
- Analysts, marketers who report AI or search performance, agencies that sell or report AI visibility, and teams that buy measurement, rank-tracking or panel data.
- Where
- Global, with platform features as Google, Microsoft, OpenAI and Amazon released them, and US data where only US data exists.
- When
- 1–18 November 2026, with late-October items where the November reporting depends on them.
- How we know
- From the three weekly GetLoopLoop digests of November and research notes for 14–18 November, which rest on Google's own pages, OpenAI's announcements as reported by the trade press, vendor research (Ahrefs, SE Visible, Similarweb, Graphite, Semrush, HUMAN Security, Peec AI, SparkToro, Seer) and independent studies. Where a claim was read in a write-up rather than the original, the write-up is named. Every source is listed at the end.
In plain words
For two years, the only way to know how often ChatGPT mentioned your shop was to ask it questions yourself and count. In November ChatGPT itself started showing businesses how often it named them and how many people clicked. That is useful, but a click there is not proof that anyone reached your site. Other studies found that when an AI recommends a shop and the person visits later, the visit is usually counted as coming from Google or typed in directly. Robots that browse for people now fill a lot of website logs, and the tools that watch Google rankings had trouble seeing clearly while Google reshuffled its results. The safest numbers are still the ones you collect on your own site.
October's analysis of this topic ended on the line that the numbers you collect yourself held up best. The first half of November tested that from the other side. For the first time an AI assistant began counting for the businesses it names: OpenAI opened Business Hub on 10 November, with mentions, profile views and clicks from ChatGPT. It is the number the AI visibility market has been estimating from outside for two years, now supplied by the assistant itself — with thresholds it does not disclose and a tooltip that says its clicks are not visits.
Around it, the rest of the fortnight showed how much of AI's effect still lands in the wrong column. An industry blueprint cited Similarweb data showing that most visits after an AI recommendation are credited to search. Google published a page on identifying its Search traffic that cannot tell AI Overviews from ordinary results. Agents, from Meta's Muse to Google's newly documented Google-Agent, filled more of the logs. And the November 2026 core update started while Google was slowing the result links rank trackers rely on and, in one spotted test, asking searchers to sign in.
This analysis takes the half-month by theme: the assistant that counts itself, how large AI traffic is, where AI-shaped visits are filed, what Search Console and the Google tag changed, the agents in the logs, rank tracking in a core update, and the vendors who sell the measurements. Then it checks October's watch list item by item and says what to watch in the rest of November.
ChatGPT starts counting for itself
The first sign was a screenshot. On 6 November Search Engine Roundtable showed an image Juozas Kaziukėnas shared on LinkedIn: a tool to "track how often your business is mentioned in ChatGPT conversations", with mentions and clicks as a daily chart, for businesses that verify ownership, and without prompt data or competitor comparison. Four days later OpenAI announced it as Businesses in ChatGPT, in beta. OpenAI's own page refused our requests, so what follows is via Search Engine Roundtable and PPC Land.
A business claims its profile, verifies it by phone or DNS, edits its details and product feeds, and sees in Business Hub its mentions, profile views, clicks to its site and "discovery topics" — the themes of conversations in which it was cited. Prompts and conversations are not shared, and numbers appear only above thresholds OpenAI does not disclose. US ecommerce and online-service businesses can also publish a Business Agent and "sponsor their Business Agent and use it as the destination for ChatGPT Ads".
Clicks don't confirm website visits or conversions.
Business Hub tooltip, OpenAI, via PPC Land, 11 November 2026
That tooltip is the most honest line in the product, and the numbers PPC Land could see show why it matters. OpenAI's own example business turned about 35% of its mentions into clicks. Two live dashboards turned about 3.5%.
| Dashboard | Period | Mentions | Clicks | Clicks per mention |
|---|---|---|---|---|
| OpenAI's example business | 7 days | 15,356 | 5,432 | about 35% |
| BuildUp Bookings (live) | 28 days | 690 | 24 | about 3.5% |
| Zianvo (live) | 28 days | 28 | 1 | about 3.6% |
Business Hub arrives as ChatGPT puts many more links in front of people. Peec AI's David Konitzny, in a LinkedIn post written up by PPC Land on 16 November, charted the share of ChatGPT conversations whose answer carries at least one clickable external link, sidebar citations not counted. It sat around 10% before early October and jumped two days after GPT-6 became the default.
ChatGPT chats with an external link in the answer, % (Peec AI)
more external links don't automatically mean more clicks or referral traffic.
David Konitzny, Peec AI, via PPC Land, 16 November 2026
Third-party trackers moved the other way. PPC Land reported on 29 October that Semrush would stop tracking Microsoft Copilot in its enterprise AI visibility product on 9 November, because "Recent changes to Microsoft Copilot's access requirements prevent AI visibility platforms" from reading it.
Our reading: the assistant's own count and the outside sample now answer different questions. Business Hub says how often ChatGPT named you and how often someone clicked a link, for one assistant, above an unknown floor, with no prompts. A sampled tool says which questions name you and which name your competitors, for any engine it can still reach. Neither says whether anyone arrived. A business that can claim a profile should, and should file the number beside its own analytics, not in place of them.
How big is AI traffic? Eight answers in two weeks
No question was measured more often in the first half of November, and no two answers measured the same thing. On 3 November OpenAI's Adam Fry said ChatGPT's "volume of search-like queries is now roughly one-fifth the volume of queries of Google", without saying which Google figure he used. The same day Search Engine Journal set Ahrefs' traffic tracker beside it: across 118,671 sites in October, Google sent 24.74% of tracked traffic and ChatGPT 0.36%. Queries are not visits.
| Who measured | Sample | Finding | What it counts |
|---|---|---|---|
| OpenAI (Adam Fry), 3 Nov | Not stated | ChatGPT's search-like queries about a fifth of Google's | Queries |
| Ahrefs, via SEJ, 3 Nov | 118,671 sites, October | Google 24.74% of traffic, ChatGPT 0.36% | Referral visits |
| Graphite and Similarweb, via SEJ, 9 Nov | Six assistants; US sites | 76 bn monthly AI sessions; organic search sent 49× AI's visits in Q2 | Panel-modelled visits |
| SE Visible, 12 Nov | 28,500 sites | 79.2% got some AI traffic in September; typically 2–5 per 100 organic visits | Analytics referrals |
| Semrush, 12 Nov | 50,000 sites | AI 0.14% of visits | Referral visits |
| Seer Interactive, 12 Nov | One agency, two years | AI 0.3% of sessions, 0.2% of signups | Own GA data |
| Tripadvisor, via SEJ, 8 Nov | One company | AI under 1% of traffic | Company reporting |
| SparkToro, 17 Nov | 118,163 US prompts | 6.9% show purchase research; 12.1% of responses name a brand | Prompts |
Read together, the table resolves into two statements that are both true. AI traffic is now nearly universal — SE Visible counted 39.7% of sites with any AI referral in January 2025 and 79.2% in September 2026, with "the sharpest increase" in summer 2026 — and it is still small, at under 1% of the average site's traffic by Graphite and Similarweb's count. Graphite and Similarweb also stopped comparing AI sessions with searches at all, saying they "can no longer reliably compare" the two, partly because of about 11.8 billion monthly US Google sessions in the Android app that their March edition did not count. When the vendor that builds the comparison withdraws it, the ratio quoted in a pitch deck deserves the same caution.
What ChatGPT is used for limits what it can send. SparkToro's panel of 12,321 US users, published on 17 November, found purchase research in 6.9% of prompts, and Rand Fishkin estimates Google's rate of purchase-journey queries at 16 times ChatGPT's or more. His conclusion: "AI visibility matters a huge amount!" — "But NOT BECAUSE OF CHATGPT." He means Google's AI features.
Try it on your own page
See what Voicescope finds on a page of yours
Everything above is measurable on your own work, and the check takes about a minute. Paste one address and read what comes back.
Open VoicescopeFree · no account · one at a time
Where AI-shaped visits are filed
If AI referrals are small, part of the reason is what gets counted as one. The IAB's attribution blueprint of 12 November, written up by PPC Land, cites Similarweb on what happens after an AI assistant recommends a brand and the person visits later. The visit is rarely credited to AI.
Visits after an AI recommendation, credited to, % (Similarweb via IAB)
Google's own help left the same gap. A new Search Central page on identifying Google Search traffic, reported on 12 November, lists the referrer values Search traffic arrives with — https://www.google.com/ or a local Google domain, and an Android app value — and says the header carries "only the scheme and host". AI Overviews and AI Mode are not mentioned at all, so a visit from either looks like any organic visit, and Google itself says the method "isn't exact." In GA4, as October's edition described, the AI Assistant channel recognises chatbots and leaves both Google surfaces in Organic Search.
That matters more as AI Mode spreads, because the best evidence of its effect on clicks arrived this fortnight. PPC Land reported on 17 November a preregistered field experiment by researchers at the University of Pennsylvania and Northeastern, posted as an unreviewed preprint: about 1,100 US users, a browser extension, and a week in one of three arms. Users whose searches were sent to AI Mode clicked through to external sites 18.8 percentage points less often than users of current Search, and the share who clicked an ad fell 42.7 points; use of Bing, DuckDuckGo or Yahoo rose 11.2 points. Removing AI Overviews showed no significant effect, because the hiding filter broke mid-study. PPC Land notes that the paper does not address the shopping ads Google added to AI Mode in February 2026.
The site-level evidence points the same way and shows why referral reports cannot see it. Cyrus Shepard took his site out of Google's generative AI features in September and estimated, in Search Engine Journal's write-up of 9 November, that AI Overviews and AI Mode had sent about 21% of its clicks, with a click-through rate of about 0.9% against 0.4% for the site; in his earlier note of 30 October, ordinary organic clicks "ALSO DROPPED" when the AI features were off. One site, one month. Ziff Davis, per PPC Land on 7 November, lost about 37% of its organic Google referrals in the third quarter while Search Console counted its pages in AI Overviews and AI Mode about 2.7 billion times — exposure measured to the unit, outcome only as a loss.
Search Console and the Google tag: more plumbing, same blind spot
Search Console got its most useful change of the period on 16 November, and it was not about AI. Google brought three features from the interface into the Search Analytics API: a brandedClassification field that splits branded from non-branded performance — only for eligible properties, with no rows for the rest — a webMultimodal search type for Google Lens, Circle to Search and image uploads, also in the BigQuery bulk export, and filterLists for several countries or devices in one request. Google said it acted "following ecosystem requests to make this data available for custom reporting and automated workflows". Branded against non-branded, exported every day, is the closest thing a site has to its own brand-demand series.
The AI report did not change. On 2 November Gary Illyes answered a suggestion that clicks and queries were coming: "The team is super interested in giving more data, data that makes sense, but we don't have anything to announce yet." Search Engine Journal notes that the UK competition regulator's requirements, which include clicks for AI Overviews and AI Mode, apply from 3 December. Meanwhile the report broke: Google confirmed that "a logging error caused a decrease in impressions on the Generative AI performance report" on 9–11 November, affecting "data logging only", on days that overlap the start of the core update. On 16 November Barry Schwartz added that "Google Search Console's performance reporting has been a bit laggy."
The tag changed shape. On 17 November Google posted the release note for the Google tag's merger with Google Tag Manager: every Google tag becomes a Tag Manager container, so a gtag-only site gets GTM's interface, debugging and version control, with on-page behaviour unchanged and "you'll receive the upgrade in January 2027" for installations Google considers unsupported. The same month brings a second dated change to server-side containers.
| Date | Change | Status |
|---|---|---|
| 31 Oct–4 Nov | Search Console click spikes, up to 10× for some UK clients, likely bots | Spotted; no Google comment |
| 5 Nov | Performance report: several devices at once | Live |
| 9–11 Nov | Generative AI report impressions undercounted by a logging error | Confirmed by Google |
| 16 Nov | Search Analytics API: branded classification, webMultimodal, filterLists | Announced by Google |
| 17 Nov | Google tag becomes a Tag Manager container; unsupported setups upgraded in January 2027 | Release note |
| 17 Nov | Google Ads Search Journey Awareness help page for Smart Bidding, Search only | Help page |
| 11 Jan 2027 | Server-side GTM passes full IP addresses to Google Ads, DV360, SA360 and CM360 tags; Analytics stays redacted | Opt-out; known from a Google email on LinkedIn |
| Mid-Nov | Amazon Ad Tag adds a 30-day first-party cookie in North America, with consent | Release note, 14 Nov |
Agents in the logs
October ended with Cloudflare's finding that automated traffic had overtaken human traffic. November put names to it. HUMAN Security, in a report written up by PPC Land on 13 November and Search Engine Journal on 14 November, found that Meta's Muse sent 59% of the AI agent requests it saw in October, and Meta and OpenAI together about 90%. Agent traffic was about eight times August's; 85% of requests went to product and search pages and 1.9% to checkout. Agents "still browse far more than they buy."
Share of AI agent requests, October 2026, % (HUMAN Security)
Agents behave differently from people once they arrive, which is why they distort a conversion report. Microsoft Clarity, published on 29 October and written up by PPC Land on 1 November, counted more than 5.2 million sessions it classified as agents and found checkout in 25.21% of them, against 14.65% of human sessions; PPC Land notes that the counts imply a smaller base than 5.2 million.
Two companies then asked sites to let their agents in. On 13 November OpenAI launched ChatGPT Wallet for agent purchases, for US Pro users, and its merchant playbook, as reproduced by PPC Land, says "websites need to ensure agents are not inadvertently blocked", asking merchants to recognise requests signed with Web Bot Auth. On 17 November Google published a help page for Google-Agent, which "is the name used by agents hosted on Google infrastructure to navigate the web and perform actions upon user request": Google-Agent in the user agent, IP ranges in user-triggered-agents.json, an experiment with Web Bot Auth signatures, and six troubleshooting checks for CDNs and firewalls. PPC Land notes that Google's list of user-triggered fetchers says they "generally ignore robots.txt rules."
The names are not proof. SenseGap showed on 5 November that Google's reverse DNS instructions accept any googleusercontent.com hostname, which every Google Cloud machine gets by default; over eight days its own site saw 89 requests from such machines claiming to be Googlebot that passed the loose check, and across four sites 32% of requests claiming to be GPTBot were fake. For analytics the lesson is concrete: an agent that runs JavaScript can land in GA4 as a session, an agent that does not lands only in the server log, and neither is labelled as an agent by default.
Rank tracking through a core update
Google's Search Status Dashboard lists the November 2026 core update as started on 10 November at 10:06 Pacific, "The rollout may take up to 2 weeks to complete." When the dashboard was read on 18 November it had no completion entry: the update was still rolling out. Search Engine Roundtable expected it to finish about a week after 16 November.
Reading it from outside was harder than in any recent update. Some tools showed volatility on 8 and 9 November, before the announcement, and Barry Schwartz wrote that "it is hard to trust the tools since they seem a bit off." Glenn Gabe believes "the update landed on Friday the 13th." SISTRIX's radar note for 12 November reads "No usable data." On 16 November Schwartz described quieter chatter than usual, which he partly put down to third-party trackers struggling.
The tools, in aggregate, also seemed to show more volatility on Saturday, November 14th, than other days.
Barry Schwartz, Search Engine Roundtable, 16 November 2026
The reasons are on record. From 30 October Google limited how fast its /goto result links resolve: on a page of 18 links "the 7th link may result in a 1 second wait" and by the 15th "there could be a 40 second wait", according to DataForSEO, Serper and SerpApi, with no Google announcement. SE Ranking published on 3 November its count of September's unrelated results — 57.3% of 10,000 US keywords on 17 September — and said results became unstable again at the end of October. On 17 November Brodie Clark reported being made to sign in before seeing any results, even in incognito; Google's Rajan Patel said "I don't believe it's meant to happen on the first page of results like this but we may be testing something here..." It is a spotted test, not a confirmed change. And Google's case against SerpApi was dismissed on 10 November with leave to amend by 30 November, without a ruling that the scraping is lawful.
Our reading: during this update a tracker line is a hypothesis, and Search Console is the check — itself laggy and, for AI impressions, wrong on 9–11 November. Compare the same pages before and after, wait for the dashboard's completion entry, and treat a tracker's date of change with a margin of a day or two.
The vendors who sell the numbers
Measurement is a good business at the moment, and the vendors' own results say why. Similarweb reported its first GAAP profit on 17 November, via PPC Land: third-quarter revenue of $83.0 million, up 15.6%, net income of $3.5 million, and AI-related revenue at almost 20% of the quarter, up from 11% at the end of 2025 and about two-thirds of new annual recurring revenue. One executive put the model plainly: "When you start selling data to AI and to agents, you mostly sell data." Similarweb's panel stands behind several numbers in this analysis — the IAB's attribution split, Graphite's 76 billion sessions, the assistants' traffic shares — and every one of them is a modelled estimate.
AI visibility is now sold as a campaign, not only as a report. Digiday described on 16 November how Bustle Digital Group sells articles written to be cited by ChatGPT, Gemini and AI Overviews, briefed with Profound's data and promoted with Reddit ads; Innisfree's visibility across 200 tracked prompts rose 30%, per BDG's case studies on Profound data. Profound's Josh Blyskal: "You've got to be able to connect analytics into action". The same vendor measures the input and the result.
The research side kept pointing back to organic search. Semrush and performance.io found on 16 November that 83.8% of AI Overview citations across more than 14,300 US pharma keywords came from domains in the organic top ten, and that the number-one domain was cited 68.8% of the time an AI Overview appeared; citations were matched at domain level and clicks were not measured. Ahrefs found on 3 November that Reddit mentions correlate with AI visibility at 0.75 across 10,000 brands — "These are correlations", it says. Each is a useful sample. None says how many prompts would be needed to see a change in your own brand, which is the question October's SISTRIX study answered: about 70 prompts for ±5 points.
October's watch list, checked
The October edition ended with eight things to watch in November. Here is what happened by 18 November, as far as our sources show.
| October's item | What happened by 18 November | Verdict |
|---|---|---|
| 12 Nov: Meta's on-device conversion counting on by default | The date passed; none of our November sources reported on the switch or its effect | Not yet visible |
| Gemini's tags and iOS referrer | No Google statement and no new report | Not yet |
| ChatGPT's referrer | No public OpenAI statement on the October bug; Business Hub now reports ChatGPT clicks itself, and links appear in 47.8% of chats (Peec AI) | Partly, by another route |
| Search Console AI clicks or queries; Page Indexing | Illyes, 2 Nov: "we don't have anything to announce yet"; a logging error on 9–11 Nov; branded data reached the API on 16 Nov. Nothing on Page Indexing | Not yet |
| Mid-Nov: Microsoft search term landing page reporting | No confirmation in our sources | Not yet |
| 20 Nov: Meta's crawl terms | Meta paused its updated business and ad terms, due on 20 November (SEJ, 10 Nov) | Went the other way |
| Rank trackers and link indexes | Altered results continued: /goto throttling from 30 Oct, unstable results at end of October, a sign-in test on 17 Nov; trackers struggled in the core update. No news on Majestic | Happened, worse |
| AI-visibility vendors publishing prompt counts | Business Hub shares no prompts; Semrush dropped Copilot on 9 Nov; one BDG case names 200 tracked prompts | Mostly not |
Two items deserve a sentence more. The ChatGPT line was framed as a question about referrers, and the answer came from a different direction: OpenAI did not explain October's lost tags, but it started counting clicks on its own side. That is the pattern October described for Meta and Google — counting moves to the platform — now extended to an assistant. And the rank-tracking line went further than expected: in October the problem was altered pages, in November Google slowed result links and was seen asking for a login, which turns a data-quality problem into an access problem.
What it adds up to
This section is our reading, not a measurement.
- The assistant now reports on itself, and on its own terms. Business Hub is the first first-party AI visibility number. It covers one assistant, hides its thresholds and prompts, and says its clicks are not visits. Useful, and not an audit.
- AI traffic is everywhere and small, and both statements will stay true for a while. Eight counts in two weeks agree on that and on little else. Before quoting any of them, ask what it counts: queries, sessions, referrals or prompts.
- Most of AI's effect is filed under other names. Search 55.9%, direct 19.9%: the referral report sees the smallest part of it, and Google's own help cannot separate AI Overviews from organic visits.
- Agents are a segment, not noise. They reach checkout more often than people, they are concentrated in two companies, and they are asking to be let in. A report that does not separate them will misstate conversion.
- Third-party ranking data is fighting for access. A core update read through throttled links and a sign-in test is read with a margin. The data that does not move under you is the data you collect yourself.
Our tool: measure from your own side
This fortnight's measurement problems point the same way as October's. AI-shaped visits arrive filed as search or direct, and agents arrive filed as visitors; the third-party panels that size AI traffic disagree with each other by the method they chose; and rank trackers lost sight of Google in the middle of a core update. The answer is data you collect yourself: your own crawl of your own site, your own Search Console export, AI-readiness checked page by page, and AI visibility judged over many questions rather than one check.
That is what GLLmolfar is for. It is GetLoopLoop's desktop crawler, in preview, for Windows and Mac, and it crawls your site on your own computer. It sorts findings by impact — indexability, redirects, canonicals, internal links, structured data — and checks how pages read to AI search: an AI-readiness score, which AI bots may fetch what under robots.txt, and llms.txt. It groups pages by meaning with an on-device embedding model, for topics, cannibalization and semantic search, and does not send the crawl to a cloud AI service. It reads Search Console data read-only, from Google straight to your computer, runs planned crawls with alerts, compares crawls over time, and exports to Excel, Google Sheets and Jira. It sends getlooploop.com only anonymous usage statistics, which can be switched off. And it runs an MCP server, so an AI agent such as Claude Code or Codex can drive it.
What to watch in the rest of November
- The core update's end. Up to two weeks from 10 November points to about 24 November. Wait for the dashboard's completion entry before comparing before and after.
- Business Hub. Whether OpenAI discloses its thresholds, opens it beyond the first businesses, and whether independent users publish how its clicks compare with their own ChatGPT referrals.
- Search Console's AI report. The UK regulator's requirements, including clicks for AI Overviews and AI Mode, apply from 3 December; watch for anything Google ships before then, and for a note on the 9–11 November gap.
- Google-Agent and Web Bot Auth. Whether signed agent requests become something analytics and CDNs can segment by default.
- The sign-in test. Whether Google confirms login walls for searchers and what that does to rank tracking and SERP data vendors.
- 30 November: the deadline for Google's amended complaint against SerpApi.
- Meta's conversions. Whether advertisers report a jump in Meta-reported conversions after 12 November, which our sources have not yet shown.
To see what assistants actually say about a brand, across many questions rather than one, Brandscope reads the answers directly.
Questions and answers
What does ChatGPT Business Hub measure?
For a business that claims and verifies its profile, it shows mentions in ChatGPT, profile views, clicks on links to its site and the topics of conversations in which it was cited. It does not share prompts or conversations, shows numbers only above undisclosed thresholds, and its own tooltip says clicks don't confirm website visits or conversions.
Why do visits influenced by AI show up as search or direct traffic?
Because a person often reads an AI recommendation and visits later by searching or typing the address. Similarweb data cited by the IAB credits such visits to search 55.9% of the time, direct 19.9% and referrals 13.5%. Google's help on identifying Search traffic also cannot separate AI Overviews and AI Mode from ordinary results.
How much traffic do AI assistants send to websites?
Some to almost every site and little to most. SE Visible found 79.2% of sites got AI referrals in September 2026, typically 2–5 per 100 organic visits; Graphite and Similarweb put AI under 1% of the average site's traffic; Ahrefs measured ChatGPT at 0.36% of tracked traffic in October against Google's 24.74%.
What did the Search Analytics API add in November 2026?
On 16 November Google added a branded classification for eligible properties, a webMultimodal search type for Google Lens, Circle to Search and image uploads, and filter lists for several countries or devices in one request. The generative AI report still shows impressions only.
Can I trust rank trackers during the November 2026 core update?
With a margin. The update began on 10 November and was still rolling out on 18 November. Google has slowed its result links since 30 October, SISTRIX reported no usable data for 12 November, and a sign-in test was spotted on 17 November. Check tracker changes against Search Console on the same pages.
Sources
Every claim above is dated and linked. Where a number comes from a measurement rather than an announcement, whoever measured it is named; where we read a write-up rather than the original, the write-up is listed.
- Marketing analytics in October 2026: the month the labels came off — GetLoopLoop, 31 October 2026
- ChatGPT tools for business — Search Engine Roundtable, 6 November 2026
- ChatGPT Business Hub — Search Engine Roundtable, 10 November 2026
- Businesses gain ChatGPT click and mention counts in OpenAI's Business Hub — PPC Land, 11 November 2026
- ChatGPT links to websites in nearly half of chats, up from 10%, Peec AI finds — PPC Land, on David Konitzny's LinkedIn post, 16 November 2026
- Semrush drops Copilot from AI visibility tracking on November 9 — PPC Land, 29 October 2026
- Google's search volume is about 5x ChatGPT's, OpenAI says — Search Engine Journal, 3 November 2026
- AI reaches 76 billion monthly sessions — Search Engine Journal on Graphite and Similarweb, 9 November 2026
- AI traffic adoption study — SE Visible, 12 November 2026
- Is AI traffic higher quality? — Semrush, 12 November 2026
- ChatGPT named 21 competitors in 2025 and us 0 — Seer Interactive, 12 November 2026
- Tripadvisor says SEO will still weigh on growth in 2027 — Search Engine Journal, 8 November 2026
- Claude takes website traffic share in October — OfficeChai on Similarweb, 10 November 2026
- We Analyzed 118,000 ChatGPT Prompts from 12,000+ Users to Answer: How Do Americans Use AI Tools in 2026? — SparkToro, 17 November 2026, and Search Engine Journal
- AI-shaped sales often get credited to search and direct, IAB warns — PPC Land, on the IAB's blueprint of 12 November 2026
- Similarweb books first GAAP profit as AI-related sales near 20% of revenue — PPC Land, 17 November 2026
- Identify Google Search traffic — Google Search Central, reported by Search Engine Roundtable on 12 November 2026
- Google's AI Mode cuts clicks to websites by 18.8 points, field test finds — PPC Land, 17 November 2026, on an arXiv preprint we did not open
- Google's AI features sent one site an estimated 21% of its clicks — Search Engine Journal on Zyppy, 9 November 2026, and Search Engine Roundtable, 30 October 2026
- Ziff Davis loses 37% of Google traffic as publishers count the AI bill — PPC Land, 7 November 2026
- Search Analytics API update — Google Search Central, 16 November 2026, and Search Engine Roundtable, 16 November 2026
- Click data in Search Console AI reporting — Search Engine Roundtable, 2 November 2026, and Search Engine Journal
- Generative AI performance report data bug — Search Engine Roundtable, 12 November 2026
- Search Console clicks spike and multi-select device filters — Search Engine Roundtable, 3–5 November 2026
- Google tag and Tag Manager unification — Search Engine Roundtable, 17 November 2026, and Tag Manager release notes
- Google ad tags gain full visitor IP addresses in server-side GTM in 2027 — PPC Land, 2 November 2026
- Google Ads Search Journey Awareness — Search Engine Roundtable, 17 November 2026
- Amazon Ad Tag adds a 30-day first-party cookie — PPC Land, 14 November 2026
- Meta's Muse leads AI agent requests — Search Engine Journal on HUMAN Security, 14 November 2026, and PPC Land, 13 November 2026
- Checkout appears in 25% of AI agent sessions vs 15% for humans — PPC Land on Microsoft Clarity, 1 November 2026
- OpenAI hands its AI agents a wallet — PPC Land, 14 November 2026
- Allow Google-Agent — Google, with Search Engine Roundtable and PPC Land, 17 November 2026
- Googlebot, reverse DNS and Google Cloud — SenseGap, 5 November 2026
- November 2026 core update incident — Google Search Status Dashboard, read 18 November 2026
- Google November 2026 core update — Search Engine Roundtable, 10 November 2026, and ranking volatility, 9 November
- November 2026 core update: ranking volatility — Search Engine Roundtable, 16 November 2026
- November 2026 core update radar — SISTRIX, 11 November 2026, updated 16 November
- Google goto result links throttling — Search Engine Journal, 5 November 2026
- Unrelated Google results hit over 57% of keywords overnight — SE Ranking, 3 November 2026
- Google sign-in protections after the first search — Search Engine Roundtable, 17 November 2026
- Google loses again against SerpApi — Search Engine Roundtable, 10 November 2026
- How Bustle Digital Group is helping Pampers and Innisfree show up in AI recommendations — Digiday, 16 November 2026
- Semrush and performance.io research on pharma brands in AI Overviews — Semrush News, 16 November 2026
- Which types of mentions correlate the strongest with AI visibility? — Ahrefs, 3 November 2026
- Measuring AI visibility — SISTRIX, 28 October 2026
- Meta pauses updated business and ad terms — Search Engine Journal, 10 November 2026
- GLLmolfar — GetLoopLoop, desktop crawler, preview
The product
Watch it on every page, every day
One reading tells you where a page stands today. The product asks the same questions of the same assistants continuously, so a change is something you are told about rather than something you go looking for.
Request an invitationInvite-only while we keep the readings honest
On this topic
How do we measure any of this?
Every day Every figure tagged Measured, Derived or Modeled, and a weekly board pack that keeps the three apart. How the product works
Author
Olexandrauthor
15+ years in marketing and SEO. Dozens of products, mostly the US market. Research and measurement.
LinkedInPass it on
Read in your language
Opens a browser translation of this English article. The original source stays in English.