Microsoft Clarity in 2026: AI Heatmaps, Copilot Insights, and Why It Is Now a Serious Analytics Tool

Microsoft Clarity launched in 2020 as a free alternative to Hotjar and FullStory — offering heatmaps and session recordings at no cost, with no data sampling and no traffic limits. In 2026, Clarity has grown into something more significant: a tool used on over 100 million websites, with AI-powered features that generate automated insights, answer questions in natural language, and surface UX problems without requiring manual session review. Clarity is now a serious behaviour analytics platform — and the fact that it remains completely free makes it arguably the best value in the analytics market. This guide covers every major feature added in 2025 and 2026 and how to use them effectively.

Clarity captures user behaviour on your website and makes it visual and searchable. The three core features are heatmaps, session recordings, and the dashboard.

• Heatmaps show aggregate user behaviour across all sessions for a given page. Click heatmaps show where users click (including rage clicks and dead clicks). Scroll heatmaps show how far users scroll down the page. Move heatmaps show where users move their cursor. All three views are available for desktop, tablet, and mobile separately.
• Session recordings capture individual user sessions as playback videos — every mouse movement, click, scroll, and navigation action recorded and replayable. Recordings are filterable by country, device type, browser, entry page, exit page, time on page, and dozens of other dimensions.
• Dashboard aggregates key metrics: sessions, engaged sessions, pages per session, scroll depth, rage click rate, dead click rate, quick back rate (users who return to the previous page almost immediately, indicating the page did not meet their expectation), and JavaScript error count.

The most significant new feature added in 2025 is Copilot integration within the Clarity interface. Microsoft's Copilot AI is embedded directly in the Clarity dashboard and can answer natural language questions about your session and heatmap data. Instead of manually reviewing dozens of session recordings to identify a pattern, you can ask Copilot: 'Why are users rage clicking on the checkout page?' or 'What is the most common exit page for mobile users who land on the blog?' and receive a generated answer drawn from your site's actual session data.

The Copilot insights feature works by analysing patterns across your full session dataset — not just a sample — and identifying statistically significant behaviours. This is meaningfully different from manual session review, which is subject to selection bias (you tend to watch sessions that confirm what you already think). Copilot surfaces patterns you would not have thought to look for. Example queries that Copilot can answer from Clarity data: 'Which pages have the highest quick back rate this month?', 'Are rage clicks concentrated on a specific element?', 'What percentage of checkout sessions have a JavaScript error?', 'Which traffic source produces the highest scroll depth?'

Clarity added Smart Events in 2025 — automatically tracked interactions that require no code implementation. Smart Events captures: button clicks (with button label), form submissions, form abandonment (started but not submitted), video play/pause/completion, file download clicks, outbound link clicks, and scroll milestone events (25%, 50%, 75%, 100% of page scroll depth).

These events are tracked automatically for any site using Clarity's standard script tag, with no configuration required. In GA4, equivalent event tracking requires either manual implementation with gtag() calls or the use of GA4's automatic event detection (which is less comprehensive). For sites that have not yet implemented custom GA4 events, Clarity's Smart Events provide immediate behavioural data with zero additional setup. The data is filterable in session recordings (e.g., 'show me sessions where the user clicked the checkout button but did not complete the form submission') enabling direct investigation of conversion drop-off.

The heatmap feature received significant updates in 2025, including segment overlays and comparison views.

• Segment comparison. You can now overlay two segments on the same heatmap — for example, mobile vs. desktop, or organic search traffic vs. paid traffic — to see how different user groups interact with the same page. This was previously only available in paid alternatives like Hotjar.
• AI-flagged zones. Clarity now automatically annotates heatmaps with AI-identified areas of concern: zones with unusually high rage click density, dead click clusters (where users click on non-interactive elements, suggesting a UI misunderstanding), and low-engagement zones on pages where engagement is expected.
• Dynamic element support. Heatmaps now correctly render clicks and interactions on dynamically loaded content — elements that are injected into the page via JavaScript after the initial load, such as modal dialogs, accordion content, infinite scroll items, and AJAX-loaded components. Previous heatmap implementations often lost this data.
• Historical comparison. You can compare heatmaps between two time periods to see how user behaviour on a page changed after a redesign, a copy change, or a promotional period. This was one of the most-requested features and was added in early 2026.

Clarity masks all text input fields by default — passwords, credit card numbers, and any content typed into form fields are never recorded. Clarity also provides automatic PII masking, which uses pattern recognition to identify and mask content that looks like email addresses, phone numbers, and similar personal data in visible page text. For GDPR and UK GDPR compliance, Clarity processes data in a way that Microsoft describes as compliant, but organisations should ensure their privacy policy discloses the use of session recording software.

The key GDPR consideration: session recordings capture real user behaviour, which is personal data under GDPR when it can be linked to an identifiable individual. Most legal interpretations treat session recording as requiring either a legitimate interest basis (balanced against user privacy) or consent. If you are operating under a strict consent framework, Clarity should be loaded only after user consent is granted. Clarity supports conditional loading — the tracking script can be loaded via a consent management platform trigger rather than in the page's initial load.

Microsoft Clarity and Google Analytics 4 are complementary, not competitive. They answer different questions about the same users. GA4 answers 'what happened' — sessions, conversions, traffic sources, page views, goal completions. Clarity answers 'why it happened' — what specific interactions led to or prevented conversion, where users are confused or frustrated, what the experience actually looks like for different user segments.

Clarity integrates natively with GA4: in the Clarity dashboard, you can filter sessions by GA4 segments, goals, and custom dimensions. In GA4, Clarity provides a Session Recording URL custom dimension that links directly from an individual GA4 session to the corresponding Clarity session recording. This means you can identify a specific segment in GA4 (e.g., users who started checkout but did not complete it) and watch their actual Clarity recordings in two clicks.

Clarity captures Core Web Vitals (LCP, INP, CLS) for each recorded session, using the browser's PerformanceObserver API. This means your Clarity session data includes actual field performance data — LCP and INP values experienced by real users in their browsers — which can be filtered and cross-referenced with behavioural data. Sessions with poor LCP (over 2.5 seconds) can be filtered to examine whether slow load times correlate with higher exit rates, quick back events, or reduced scroll depth.

This session-level Core Web Vitals data complements the aggregate field data in Google Search Console and the lab data from tools like PageSpeed Insights. Use the PageSpeed Checker at /tools/pagespeed-checker for technical diagnosis of what is causing slow Core Web Vitals, and use Clarity to see how those performance issues translate into observable user behaviour changes.