AI search upends SEO metrics as brands seek new measurement paths

The old SEO scoreboard no longer fits

Clicks used to be the cleanest proof that search was working. Now AI answers can introduce a brand, recommend it, and even help move a buyer toward purchase without sending the same signals rank trackers were built to count. That is the core problem the funnel query pathway is trying to solve: not how to chase one perfect number, but how to measure visibility in a search environment where the familiar evidence trail is breaking apart.

The practical value of that framework is not hype, it is discipline. It gives AI visibility teams a way to stop collapsing every mention into one bucket and start separating awareness-stage exposure from bottom-funnel influence. If a brand appears in an AI-generated summary, that is not the same thing as being shortlisted, and a shortlist mention is not the same thing as a conversion-ready action. The framework matters because it forces those distinctions into reporting.

Why traditional SEO metrics are losing their grip

Traditional SEO metrics were designed for a visible results page, stable rankings, and trackable clicks. Those signals still matter in classic search, but they fall short when assistive and agentic systems answer the query directly, summarize the answer, and compress the user journey. In that world, a brand can gain influence without the tidy trail of impressions, position changes, and click-throughs that used to anchor monthly reporting.

Google’s rollout of AI Overviews in the United States made that shift impossible to ignore. The company said the feature had already been used billions of times in Search Labs, that it would reach hundreds of millions of users that week, and that the rollout path aimed to serve over a billion people by the end of 2024. OpenAI pushed the same conversation forward with SearchGPT, introduced as a temporary prototype that would provide fast, timely answers with clear and relevant sources, with the best features later folded into ChatGPT. By January 2026, Microsoft Advertising was already framing GEO around three data pathways, feeds, crawled data, and offsite data, which is a useful reminder that AI discovery does not begin and end on a single webpage.

What the funnel query pathway actually tries to measure

The funnel query pathway treats visibility as a macro system made up of strategy, measurement, and analysis. That is a smarter posture than trying to force AI search into a standard rank-tracking dashboard, because the unit of analysis is no longer just the keyword. The real question is how a buyer moves through discovery, shortlisting, retrieval, and action across search, assistants, and agents.

That shift sounds abstract until you apply it to reporting. A brand can now be visible at the top of the funnel without earning a click, visible in the middle of the funnel without ranking number one, and visible at the bottom of the funnel through offsite signals that shape an AI system’s recommendation. The framework is useful because it tells teams to measure the path, not just the endpoint.

How the framework helps separate awareness from influence

This is where the methodology earns its keep. Awareness-stage mentions are the broadest signal, the brand name appears in a generated answer, a summary, or a recommendation set. Bottom-funnel influence is narrower and more meaningful, because it shows up when a brand is shortlisted, chosen, or acted on after the AI surface has already done its work.

That difference should change what goes into the KPI deck. Instead of leaning on a single visibility score, teams need layers: mention rate in AI outputs, inclusion in category comparisons, presence in recommendation lists, and downstream behaviors such as branded search lift, direct visits, leads, or sales. The point is not to measure everything equally, but to stop pretending that one metric can explain both discovery and demand.

A more disciplined reporting stack would look something like this:

• Awareness metrics: brand mentions in AI summaries, answer boxes, and assistant responses
• Consideration metrics: inclusion in “best,” “top,” or shortlist style outputs
• Action metrics: assisted conversions, branded search growth, referral traffic from AI surfaces, and revenue tied to downstream behavior
• Coverage metrics: completeness of feeds, crawlable content quality, and offsite authority signals

That structure is what turns AI visibility from a vague brand-health story into something a performance team can actually manage.

Why data pathways matter more than old keyword lists

The article’s skepticism about prebuilt keyword lists is well earned. Keyword lists are convenient, but convenience is the problem. They tend to reflect what is easy to track, not what users actually ask when they interact with AI systems. If the query set is too narrow, the measurement model will miss the very moments when a brand is being introduced to a buyer.

Microsoft Advertising’s three-pathway view is helpful here because it broadens the data conversation. Feeds matter when a system can ingest structured product information. Crawled data matters when AI needs to read and interpret web content. Offsite data matters when live experiences, reviews, and other external signals help shape the answer. Together, those pathways explain why brands cannot treat AI visibility as a pure on-site SEO problem anymore.

The user journey is already changing

The behavioral evidence backs this up. Pew Research Center found that 58% of respondents made at least one Google search in March 2025 that produced an AI-generated summary. In its browsing analysis, which used data from 900 U.S. adults and 68,879 unique Google searches, 12,593 searches produced an AI summary. Pew also found users were less likely to click result links when a summary appeared, and that source links inside those summaries were clicked very rarely.

That matters because it shows the funnel itself is moving. When the summary answers the question, the click is no longer the default proof of attention. For measurement teams, that means the old habit of treating traffic as the main proxy for visibility is too blunt. AI search can create awareness without traffic, and it can create preference before the user ever reaches a traditional landing page.

The publisher and brand stakes are getting louder

The economics behind this shift are hard to ignore. Columbia Journalism Review reported that Similarweb estimated worldwide search traffic had fallen 15% over the previous year. It also said that in the year after AI Overviews launched, the share of Google news searches with no click to any result rose from 56% to nearly 69%. At the same time, the Independent Publishers Alliance filed an antitrust complaint with the European Commission, and the News Media Alliance warned that Google’s AI Mode would further reduce publisher traffic and revenue.

That is why the funnel query pathway should be read as more than a measurement theory. It is a response to a market where AI search is changing discovery faster than most reporting stacks can adapt. Brands that cling to rank and click as the whole story will miss the portion of influence that happens before the visit. The teams that win will measure visibility as a sequence, not a snapshot, and they will build KPIs that can tell the difference between being noticed and being chosen.