--- title: Measuring AI Traffic Is More Than Just Counting Clicks — isla Studio url: https://isla-stud.io/sv/ratgeber/ai-traffic-messen/ date: 2026-08-13 --- # Measuring AI Traffic Is More Than Just Counting Clicks Cloudflare now shows how brands appear in a recurring AI prompt panel. Google tracks organic link impressions in AI Overviews and AI Mode. This reflects an ad impression, not human attention. Your web analytics only begin when a user clicks. Those who clearly distinguish between these signals won’t get an overall AI visibility score, but they will receive a useful diagnosis. Cloudflare’s new AEO Visibility Dashboard (AEO stands for Answer Engine Optimization) displays four percentage values. Its true editorial value lies in the measurement chain behind it: Cloudflare brings together technical agent readiness, synthetic tests of AI responses, and real-world crawl and referral signals into a single interface. Since June, Google has expanded Search Console to include a dedicated report for generative AI impressions. This report tracks something different: how often Google has recorded an organic impression for a link from your website in AI Overviews or AI Mode. It doesn’t reveal whether a human consciously noticed the link. And then there’s GA4, Plausible, or your own analytics setup. These tools track the visit to the website and what happens afterward. All three perspectives are useful. Things get murky when a synthetic prompt test becomes supposed reach, a bot request becomes a citation, or a few clicks lead to a diagnosis of „invisible.“ A measurement chain and five questions I wouldn’t treat AI Visibility and AI Traffic as a single metric, but rather as five consecutive questions: Is your website accessible to relevant systems, and is it being crawled? Is your brand or website mentioned or cited in tested responses? Has Google recorded an organic impression for a link? Did someone click and visit your website? What happened next—leading up to an inquiry, purchase, or another meaningful goal? This sequence is not a universally applicable AI funnel. Not every platform discloses all stages, and a click does not necessarily stem from a previously observed citation. As a verification model, however, it prevents you from comparing metrics that come from different contexts. Measurement Point | Typical Data Source | What You Really Know | What Remains Unclear | Accessibility and Crawl Agent Readiness Checks, Server/Edge Logs, local bot analyticsA system was allowed or able to retrieve a URLWhether the page was used in a responseMention and CitationRepeated, documented prompt panelThe brand or domain appeared in the tested responsesHow often real users saw itLink impressions recorded by GoogleSearch Console, „Generative AI“ reportGoogle recorded an organic impression for a website link in AI Overviews or AI Fashion-conscious perception, isolated Gen-AI clicks, queries, and conversionsClickGA4, Plausible CE, other web analytics: A recognized visit reached the website; all unrecognized or referrerless visits; post-click impact: events, shop/CRM data, funnel analysis; what happened after the recognized visit in your own measurement system; which unobserved influences contributed to the decision 1. Accessible does not mean selected The first measurement point occurs before any response. Can a crawler read your robots.txt and sitemap? Does it reach the page? Does it receive content or a 403? Does a link lead nowhere and end with a 404? Cloudflare bundles such technical checks under „Agent Readiness.“ The associated „AI Operator Activity“ also shows real crawl and referral activity on the Cloudflare network, broken down by operator. There, for example, you can see that a provider is fetching many pages but isn’t sending any visible referrals, or that its requests are failing due to 403 and 404 responses. These are genuine signals from the website. They document what a system requested there and what response the server provided. They do not document what was later included in a private assistant’s response. Regardless of Cloudflare, you can find the same set of evidence in server logs or your own bot analysis. If WordPress is your stack, citelayer® can analyze AI bot requests and detected AI referrals separately on-premises. I’m deliberately not specifying a fixed number of supported bots because detection patterns vary by release. 2. Cloudflare AEO Visibility Does Not Count Responses Completely Cloudflare introduced the AEO Visibility Dashboard on August 6, 2026, as an Early Access feature. Access is requested via the Overview tab of the Cloudflare Dashboard. It is therefore not yet a standard feature that has been rolled out to everyone. The interface answers the question of how a website performs in a response panel generated by Cloudflare. To do this, Cloudflare derives the industry and category from the website and formulates likely questions for unbranded discovery, recommendations, comparisons, and general advice. The official pipeline diagram also mentions custom prompts. Currently, Cloudflare queries Claude from Anthropic and GPT from OpenAI. Four metrics are key: Citation Rate: What percentage of the tested responses within the category cite the website as a source? Mention Rate: How often does the response mention the brand—regardless of whether the website is linked as a source? Prominence: When the website is cited, how early does it appear, and what portion of the response is attributed to it? Share of Voice: What percentage of citations in the panel are attributed to the company’s own website compared to competitors? Cloudflare’s AEO view shows Citation Rate, Prominence, and Mention Rate broken down by Anthropic and OpenAI. The percentages are derived from a repeated prompt panel—not from a complete count of real user responses. Source: Cloudflare, „AEO: Are AI assistants recommending you?“ (August 6, 2026). The method goes beyond an occasional manual prompt because Cloudflare explicitly accounts for model variance. According to the methodology blog, prompts are run multiple times across different models via the AI Gateway. The response text and cited sources are then evaluated. To do this, Cloudflare combines deterministic text analysis with an assessment by Workers AI when a nuanced classification is required. The official Cloudflare chart separates prompt queries and response evaluation from industry and edge signals. It is precisely this separation that is important: visibility in responses is tested, while crawl and referral activity is monitored on the network. Source: Cloudflare, „AEO: Are AI assistants recommending you?“ (August 6, 2026). For market comparison, Cloudflare generates an industry panel once per category. This snapshot serves as a reusable baseline for domains in the same category, rather than posing all comparison questions anew with every scan. This makes results available more quickly and establishes a common frame of reference. However, it also means that the timeliness and relevance of this panel are part of the measurement method. What These Percentages Are Not Citation Rate, Mention Rate, Prominence, and Share of Voice are not a complete count of real user responses. Instead, Cloudflare observes questions it generates itself and the responses from selected models. The panel can be a useful, repeatable sample. It is neither an AI ranking nor a measure of absolute truth for overall market visibility. Even Cloudflare’s real-world network data does not fully bridge this gap. It tracks crawls and referrals to the website. It does not see every response a user has read in a private assistant window. That’s exactly why I still find the combined interface exciting: it juxtaposes technical readiness, synthetic answer visibility, and real-world network activity. As long as these three types of signals remain distinguishable as such, this constitutes a meaningful triangulation. What I Still Want to Know During Early Access The public launch materials explain the general workflow but leave important questions unanswered: How many questions are included in the panel, and how are custom prompts weighted? Which languages and regions are covered? Which specific model versions are running, and with how many iterations? How often are the category panel and baseline updated? How exactly are Prominence and Industry Fit calculated? How can a misidentified category be corrected, and what is recalculated afterward? Although the sample UI shows a „Change market“ feature, this does not yet constitute complete public method documentation for correction and recalculation. These points are verification questions for Early Access, not an invitation to guess. 3. Google Counts Actual Generative AI Impressions Google Search Console extends the measurement chain in an area that usually remains invisible to other tools: Google can track when it has actually displayed a link within its own Generative AI feature. The new Generative AI report currently covers AI Overviews and AI Mode. Google records an organic impression when a link to your website was displayed to a user there. This is based on the link’s appearance in the supported feature, not on the user’s conscious awareness of the link. Experiments in Search Labs are explicitly excluded. Unlike Cloudflare’s Citation Rate, this metric does not come from prompt monitoring. It describes an organic link impression recorded by Google within Google Search. However, it is not cross-platform: The report does not include data on Claude, ChatGPT, or other assistants. Google is initially rolling out this view to only a subset of properties. If it’s missing, this may be due to the limited rollout or an insufficient number of impressions. Exclusion via the separate Search Generative AI control panel can also prevent participation. In the dedicated view, you can analyze impressions by page, country, date, and device. There is currently no query tab available there. Likewise, this special report does not currently show isolated Gen AI clicks, a dedicated CTR, or position. The new Generative AI report in Search Console shows organic link impressions recorded by Google in AI Overviews and AI Mode. These indicate display, not conscious awareness. The view includes pages, countries, devices, and dates—but no search queries and no isolated Gen-AI clicks. Source: Google Search Central, „Introducing Search Generative AI performance reports in Search Console“ (June 3, 2026), CC BY 4.0. The relevant data is also included in the Web Search section of the standard Search Performance report. There, it remains aggregated with other web search results. The specialized report thus provides a separate view of Gen-AI impressions, but does not yet offer a complete Gen-AI funnel. Four limitations are important to note when analyzing the data in detail: By default, the chart aggregates at the property level. If two results from the same website are shown in a generative AI feature, they are counted together as a single impression. Using the URL filter switches the aggregation to the URL level. In the table, pages are aggregated at the page level; countries, devices, and dates are aggregated at the property level. Therefore, the totals in the table and chart may differ. The usual limits of the Search Performance report also apply here, including a maximum of 1,000 table rows and the standard time period limits. The most recent data may be preliminary and subject to change. The chart and table can be exported. In practice, this means: The report answers the question, „Was a Google link displayed in a supported generative AI feature?“ It does not yet answer separately, “Which question led to this, who clicked, and what happened afterward?” 4. The click begins in your web analytics Since May 13, 2026, Google Analytics has been assigning recognized referrers from AI assistants to the “AI Assistant” channel. GA sets the medium to “ai-assistant” and the campaign to “(ai-assistant).” Google lists ChatGPT, Gemini, Deepseek, Copilot, and Grok as examples, not as a complete list of referrers. Clicks from Google AI Overviews and AI Mode, on the other hand, are categorized under “Organic Search” in GA4. As a result, the same umbrella term “AI” appears in different analysis paths depending on its source. A small number in the AI Assistant channel therefore does not disprove the existence of generative AI impressions in Search Console. Here’s how to check the click layer: In GA4, open an Acquisition report and use the Session – Default Channel Group dimension. Check whether AI Assistant appears as a row. In an exploration, filter by the medium “ai-assistant” and add the landing page. Compare the results with referral sources and organic Google search. Not every AI domain is necessarily included in Google’s maintained list, and not every visit carries a usable referrer. GA4 is just one example. With a self-hosted Plausible Community Edition, you can track referrals, campaigns, and custom events within an infrastructure of your choice. This gives you more control over the data flow. It’s not a blanket “get-out-of-data-protection-or-consent-free” pass: tracking, additional properties, hosting, and the legal basis must align with your specific setup. 5. The Real Business Question Begins After the Click A referral is interesting, but unfortunately, it’s not yet a result. Only the post-click layer reveals whether someone reaches the appropriate landing page, takes a meaningful next step, makes an inquiry, or makes a purchase. Plausible CE can map this phase using its own events. For a deeper, data-sovereignty-sensitive analysis, I’m developing a standalone analytics integration for Plausible CE—under the working title Funnelboard—(a free analytics software that can be run on your own server). At the currently documented stage of development, Funnelboard runs alongside Plausible CE, reads Analytics data in read-only mode, and uses it to calculate funnels, revenue by source, and common user journeys. In the setup described, the Analytics data remains within the user’s own infrastructure for this analysis; the EDD license check is intended to be performed externally. At this time, this statement regarding the development status neither replaces a legal review nor constitutes a guarantee of any future operational model. In this model, WordPress can serve as a store, a content management system, or a source of additional local signals. The five metrics work just as well for other CMS platforms, online stores, and custom-developed platforms. Three Evidence Categories Instead of an AI Visibility Score The five metrics can be traced back to three evidence categories: Synthetic response observation: Cloudflare’s AEO panel or your own documented prompt monitoring shows what happens in a defined sample. Real system signals: Edge/server logs, Cloudflare Operator Activity, and Google’s generative AI impressions show observed processes within the respective systems. In-house usage and impact data: Web analytics, events, and e-commerce or CRM data show the detected visit and what happens afterward within your own measurement scope. No single category replaces the others. Together, they help provide a useful diagnosis: Unreachable or many 403/404 errors: Check technical access, URL structure, and crawlability. Crawls but no citations in the panel: Examine intent match, source value, market classification, and prompt set. Panel citations but no actual Google impressions: Be aware of platform limitations; prompt samples and Google display are not the same thing. Google impressions but barely any detectable clicks: Check link presentation, message type, landing page, and measurement gaps—without assuming a zero value based on the absence of isolated GSC click counts. Clicks but no corresponding impact: Examine the landing page, offer, funnel, user journey, and event concept. Cloudflare’s new dashboard therefore replaces neither Google Search Console nor web analytics, logs, or post-click analysis. It complements them by offering an interesting new perspective. Google now provides a separately trackable, recorded organic link impression for its own generative AI sections. Whether it garnered attention and what business outcomes result from it must be verified using other data sources. If you want to get the basics straight first, read “What Is AI Visibility?”. I describe how to set up repeatable tests in “How I Approach AI Visibility Audits with citelayer®.” For technical crawlability, the next step involves AI crawlers, robots.txt, and content signals. A citelayer® AI Visibility Audit can consolidate the metrics for your website into a reliable baseline—without any promises regarding traffic, citations, or rankings. Sources Cloudflare: How the New AEO Visibility Dashboard Works Cloudflare: Press Release on Early Access Google Search Central: New Reports for Generative AI Impressions Google Search Console: Generative AI Performance Report Google Analytics: AI Assistant in the Release Notes Google Analytics: Definition of Standard Channel Groups Plausible: Self-Hosting the Community Edition Plausible: Analyzing Referrers and Campaigns Plausible: Tracking Custom Events