How to Set Up and Watch Session Recordings in statpx

Numbers tell you a form has a 60% abandonment rate. They don't tell you that visitors are getting stuck on a validation error that never explains itself, or repeatedly clicking a button that silently fails. Session Recordings in statpx close that gap by replaying real visitor sessions — cursor movement, clicks, scrolling, and rage clicks — so you can watch exactly what a real person experienced. This guide covers turning recordings on, protecting sensitive input, filtering the ones worth watching, and using the player itself.

Turning On Session Recording

Recording is off by default and has to be enabled in two places. First, go to Site Settings for the site you want to record and turn on the recording toggle. From there you control two important settings:

• Sampling % — the percentage of sessions to record. On high-traffic sites, sampling 10-20% is usually enough to find patterns without storing every session.
• Retention days — how long recordings are kept before they're automatically purged.

Second, add this line before your statpx tracking snippet on the pages you actually want recorded:

Both switches have to be on: the Site Settings toggle governs whether the site records at all (and at what sample rate), while window._stRecord opts a specific page into being eligible for capture. This lets you record just your signup flow or checkout pages instead of every page on the site.

Masking Sensitive Fields

Typed input values are never captured by design — statpx does not record what visitors type into any field, so passwords, card numbers, and personal details never reach a recording. But some pages display sensitive text that isn't typed input, like an account balance, an email address rendered on the page, or a customer's name in a dashboard header. To blur those out of the replay, add class="st-mask" or a data-st-mask attribute to the element:

Any element carrying either marker is redacted in the recorded replay, while the rest of the page records normally. Audit any page with account data, order details, or user-generated content before turning recording on for it.

Filtering by Rage Clicks, Long Sessions, and Errors

Once recordings start coming in, the list on the recordings page can be filtered so you don't have to watch everything to find what matters:

• Rage clicks — sessions where a visitor clicked the same spot repeatedly in frustration, usually because something appeared clickable but wasn't responding.
• Long sessions — sessions that ran unusually long, which can mean deep engagement or, just as often, a visitor stuck trying to figure something out.
• Errors — sessions where a JavaScript error occurred, letting you watch exactly what the visitor was doing the moment your site broke for them.

Start with the rage-click filter after any redesign — it surfaces the exact spots where your new layout confused people, faster than reading through a random sample would.

Using the Replay Player

Click Replay on any session in the list to open the player. From there you get standard playback controls: play/pause, a speed selector (1×, 2×, 4×, 8×) so you can skip through the quiet parts of a long session, and a scrubber to jump directly to a specific moment. Rage clicks are marked visibly on the timeline so you can jump straight to the moment of friction instead of watching the whole session from the start.

The Bottom Line

Session recordings turn an abstract drop-off rate into a video you can actually watch. Instead of guessing why visitors abandon a form or bounce off a page, you can replay the exact session and see the moment things went wrong. Enable recording with a conservative sampling rate, mask anything sensitive with st-mask, and use the rage/error filters to jump straight to the sessions worth your time. Pair replays with the click heatmap for the full picture — the heatmap shows you where the friction is happening across many sessions, and the recording shows you exactly what it looked like for one real visitor. If your site throws client-side errors during a recorded session, cross-reference them with JavaScript error tracking to fix the root cause.