AI search and Google traffic: what brands should measure
A practical framework for separating Google search, AI answers, human referrals, conversions, and crawler evidence.
AI features are changing how people discover information, but one answer, referral, or crawler request cannot describe the whole search journey. Measure each surface directly before deciding what changed for the brand.
Start with Google search evidence
Use Search Console for queries, impressions, clicks, index diagnostics, and pages shown in Google Search. Google's AI features guidance says AI Overviews and AI Mode remain part of Search and that their links are included in Search traffic reporting. Google also says eligibility does not guarantee crawling, indexing, or serving.
Google announced on June 3, 2026 that dedicated generative AI performance reports were rolling out to a subset of sites. Treat that report as a Search Console view, not as a universal measure of every AI answer on the web.
Keep answer evidence separate
Run a defined set of questions on the platform you want to inspect. Save the exact wording, date, location, response, linked sources, and model context when available. A response is a sample under those conditions. It is not a universal ranking, market share, or recommendation guarantee.
Measure human visits and outcomes
Use analytics for referrers, campaign parameters, landing pages, sessions, goals, and revenue. An AI referral is an observed human visit from an AI product. Compare it with other sources using a defined period and attribution rule, while keeping missing or unattributed sources visible.
A referral does not prove that a citation, crawl, or recommendation caused the visit. A conversion is an analytics event under the site's event model, not an inference from an answer or a request.
Inspect crawler evidence separately
Use server or infrastructure logs to identify crawler requests, requested paths, timestamps, source coverage, and HTTP outcomes when available. This can show access and delivery conditions. It does not prove indexing, citation, recommendation, or conversion.
SearchSeal combines privacy-friendly website analytics with a separate crawler evidence module. It keeps pageviews, visitors, sessions, acquisition sources, goals, revenue, and human AI referrals apart from observed crawler requests.
Use a small evidence table
- Search: Search Console impressions, clicks, queries, and index state.
- Answers: exact platform samples and linked sources.
- Referrals: observed human sessions from AI products.
- Outcomes: defined goals and recorded revenue.
- Crawlers: requests, paths, and HTTP outcomes.
Label the source and date beside every metric. This makes a traffic change easier to investigate without turning separate observations into one visibility score or a causal story.