What is Microsoft Clarity: setup and CRO analysis

Microsoft Clarity is a behaviour analytics tool that shows what visitors do on your pages through heatmaps and session recordings. According to Microsoft's own FAQ, Clarity is free, has no traffic limits and collects data without sampling. Tools such as Google Analytics tell you how many people came and where they came from; Microsoft Clarity lets you see where those people clicked, where they got stuck and where they gave up. This article explains, step by step, how to set the tool up and how to use it for conversion rate optimisation.

What Clarity is and what it measures

Clarity has three core components. Heatmaps show visually the clicks on a page, scroll depth and the areas where attention concentrates. Session recordings let you watch anonymised replays of individual visits. The dashboard summarises metrics that automatically flag problematic behaviour.

The most valuable of these metrics for CRO are the ones Microsoft's documentation defines as "semantic metrics". A rage click is a user clicking the same area repeatedly in quick succession, and it usually signals frustration with an element that does not work. A dead click is a click that produces no visual response or navigation; it points to a broken link or an element that looks clickable but is not. A quick back is a user going to a page and coming back shortly after, which shows the navigation did not meet expectations. Excessive scrolling is scrolling beyond what is expected; it is a hint that the information being looked for could not be found. JavaScript errors and errors that occur after a click are added to these.

Setup: three routes and verification

Microsoft's setup guide describes three routes. The first is ready-made platform integrations: on platforms such as WordPress, Shopify and Wix, Clarity can be connected through a plugin or integration. The second is manual installation: you take the tracking code from the Setup section under Settings in your Clarity project and add it to the head section of your site. The third is the NPM package offered for JavaScript-based projects.

Verification after installation is essential. The simplest method is to wait for the live session count to appear in the Clarity dashboard. The more technical method is to open the Network tab in the browser's developer tools and look for requests going to Clarity's collection address. If you use a tag manager, check separately that the code fires on every page and in every language version; on multilingual sites, the code missing from one language's template is a common mistake.

The setup guide has another important note: Clarity cannot be used on sites aimed at users under 18. The tool also masks sensitive content by default. If you work in a sensitive field such as health, do not switch these masking settings off, and verify for yourself that information entered into form fields does not end up in recordings.

Consent management and data retention

Clarity uses cookies, so consent management is part of the setup. According to Microsoft's consent mode page, from 31 October 2025 Clarity enforces the consent requirement for users in the European Economic Area, the United Kingdom and Switzerland. If a site in Turkey receives visitors from these regions, this rule affects it too. Under local law, the rules on processing personal data also apply. The practical route is to use a cookie consent management platform and pass the user's consent status to Clarity.

On sites with a low consent rate, keep in mind that Clarity data represents only the users who gave consent; this partial view should not be forgotten when interpreting findings.

You also need to know the data retention period. According to Microsoft's FAQ, recordings are kept for thirty days from the date of recording; recordings added to favourites and a randomly selected sample are kept longer. This means that when you find an important session, you should add it to favourites.

CRO analysis: a one-week routine with Clarity

Clarity's value lies not in watching recordings at random but in starting with a question. The routine that works in practice looks like this.

• Read the numbers first. In Google Analytics, identify pages with a low conversion rate or a high exit rate. Clarity's Google Analytics integration makes it easier to join the two tools' data at the session level.
• Then look at the heatmap. On the page you picked, see where users click and how far down they scroll. If the main call-to-action button sits at a depth most users never reach, the problem is in the design.
• Next, filter with the semantic metrics. Filter the sessions on that page that contain rage clicks or dead clicks and watch a few of them. Behaviour repeated on the same element is enough evidence for a hypothesis.
• Finally, write a hypothesis and test it. A concrete sentence such as "users looking for price information scroll to the bottom of the page and cannot find it" can turn into a change and then a measurable comparison.

The most important rule of this routine is not to draw conclusions from a single recording. One user's odd behaviour is an example; the same behaviour repeated across many sessions is a pattern. Decisions are made on patterns. We collected the problems we often encounter with this method on clinic and aesthetic sites in our 12 critical CRO fixes article.

Common mistakes

The first mistake is installing the code and never looking at the dashboard. Clarity keeps collecting data, but a dashboard not opened with a question produces no decisions. The second mistake is watching desktop recordings only. On most sites mobile and desktop users run into different problems; heatmaps should be read separately by device type. The third mistake is not excluding internal traffic. If your team's own visits mix into the recordings, the picture gets distorted, especially on low-traffic pages; use filters to leave that traffic out. The fourth mistake is noting problems and leaving them without an owner. Every finding should have an owner and a date; otherwise the same rage click pattern will turn up again months later.

Fitting Clarity into your measurement setup

Clarity is not an analytics solution on its own; it is a qualitative layer that complements quantitative data. Google Analytics shows how many people converted, Clarity shows why they did not. That is why it matters that both tools are set up with the same conversion definitions, the same page groups and the same language split. We covered the foundations on the Google Analytics side in our GA4 setup and first 10 reports article.

We turn the problems found with Microsoft Clarity into an improvement list and a test plan as part of our CRO and UX service; combining behaviour data with ad and analytics data in a single dashboard is handled by our analytics and data service.

Sources

• Microsoft Learn — Clarity setup: installation routes, verification and usage restrictions.
• Microsoft Learn — Semantic metrics: definitions of rage clicks, dead clicks, quick backs and excessive scrolling.
• Microsoft Learn — Consent Mode: consent mode and regional consent requirements.
• Microsoft Learn — Clarity FAQ: free use, sampling and data retention periods.