Microsoft Clarity, the company’s analytics product for site behavior, launched agentic analytics on Wednesday, November 4, 2026. The release lets site owners see which AI agents visit their pages and test whether those agents can finish tasks such as checkout. Two other features shipped alongside it: Ask Clarity and an expansion of Brand Agents.
Microsoft framed the launch around growth in automated visitors. In its announcement, the company said, “AI agents nearly doubled in three weeks to almost 2 million.” Search Engine Roundtable, which reported the launch, flagged the three-week span as the striking part. The excerpt available does not say what Microsoft counted, where the count was taken, or how it defines an agent. Treat the figure as Microsoft’s own claim until those terms are published.
The new Agent Activity dashboard covers three questions. Microsoft says it shows “how agentic traffic changes over time, which agents visit your site, and what they do once they arrive, ranging from searches and form submissions to logins, downloads, appointment bookings, and checkout.”
Simulation assessments go a step further. Microsoft says they let teams test agent task completion “before a real customer journey fails.” Each simulation reports whether the task finished, the time it took, the number of interactions, and where the experience could improve. The first scenarios cover commerce and travel, with more journey types planned. Microsoft gave no date for those.
Ask Clarity accepts plain-language questions about website behavior and returns answers and recommendations backed by session evidence. Search Engine Roundtable’s Barry Schwartz compared it to Ask Analytics in Google Analytics. Brand Agents, which Microsoft positions as a way to surface products in Copilot, Bing, and MSN, now extend to local businesses that want to capture leads and book appointments.
Analysis: the practical value here is measurement that most teams lack. Server logs and standard analytics tags were built around human sessions, and separating agent behavior from ordinary bot noise has been guesswork for many sites. A dashboard that names agents and logs their actions at least gives a first reference point. Whether Clarity’s classification holds up against a team’s own logs is untested, and the source offers no independent validation.
The simulation feature changes the question being asked. Crawlability checks ask whether an agent can reach a page. A checkout simulation asks whether an agent can complete a transaction, which is where form markup, login walls, and script-heavy flows can fail without any visible error for human shoppers. That failure mode never appears in a crawl report.
The local Brand Agents expansion adds a separate consideration for multi-location businesses. If Microsoft routes lead capture and appointment booking through agents on Copilot, Bing, and MSN, the business profile data those agents draw on becomes more consequential. The source does not describe what data the agents use, so that point remains an open question for local teams.
Teams that sell online or take bookings should turn on the Agent Activity report, record a baseline of agent visits and actions over the next thirty days, and run the checkout or travel simulation on their most valuable journey before drawing conclusions from Microsoft’s growth figure.
Search Engine Roundtable, Barry Schwartz, reporting on Microsoft’s Clarity announcement of November 4, 2026.
