Tag the patterns, save them as segments.
Signals turn a fuzzy hunch into a tracked pattern. Define a rule once, and every matching session — past and future — is tagged, counted, and ready to replay back-to-back.
Name the pattern once, find it everywhere.
2 stops, each one a real screen from the console.
LogClip proposes the rules from what it found
Suggestions come from what LogClip already found — vitals regressions and the errors behind the cause cards — each carrying how many sessions it would tag, so you adopt a rule knowing whether it matches a crowd or nobody. Below them, the signals already running.
A rule you can read
ALL of or ANY of, then the conditions — add one, nest a group, test it against recent sessions before you commit to it. No code, and the same builder makes a conversion when you flag the rule as a success.
Everything under signals
Define rules over a session's events, identity and shape — matching visits get tagged, filtered and replayed as a playlist.
Rule builder, no code
Eleven kinds of condition — events, requests, JS errors, console messages, pages reached, Web Vitals, frustration, device, location, identity and session shape — combined with all of / any of / none of, nested as deep as you need.
Suggested signals
Recurring failures and poor vitals come back as ready-made rules, each carrying how many sessions it would tag — so you adopt one knowing whether it catches a crowd or nobody.
Test before you save
Run the rule against recent sessions and see what it would have matched before it goes live.
Tagged as it happens
Most signals tag a session live; rules that depend on something not happening are decided when the visit ends, and say so.
Backfill the past
Adopt a signal and re-run it over the sessions you already have — the trend line starts with history, not from today.
A badge everywhere
Matching sessions carry the coloured badge in recordings, on the player, in funnels and in every filter.
Conversions are signals
The same builder, flagged as a win — that's what gives you conversion rate, revenue and cost-of-friction.
Trend per signal
A sparkline and count on every signal row, so a fix shows up as a line going down.
Segments and playlists
A saved filter over signals, device, friction and more — play every matching recording back to back.
Why it matters.
Rules over real behavior
Match on events, identity, page, error or friction shape. Matching sessions get a tag you can filter anywhere.
AI-suggested signals
LogClip spots recurring patterns — poor LCP, repeat 403s, a rage-clicked element — and offers them as signals to adopt.
Segments as playlists
Save a filter as a segment and play every matching recording back-to-back, like a highlight reel of a single problem.
Minutes, not quarters.
Describe the pattern
Build the rule, or adopt one LogClip suggested from what it already found.
Test it, then save
See the matches against recent sessions; backfill history if you want the trend.
Filter by it anywhere
The badge is a filter on recordings, a step in a funnel, a cohort for Ask AI.