Automated Amplitude Event Health Checks
Every morning, WebRun opens Amplitude, reads yesterday's volume for the events you rely on, compares each against its recent normal, flags anything that stopped firing, dropped sharply, or turned up with no warning, writes the findings into Airtable, and pings you in Telegram.
How do I know when product tracking breaks after a release?
WebRun checks your Amplitude events every morning, comparing yesterday's volume for each watched event against its recent average. It flags events that stopped firing, moved sharply, or appeared unannounced, logs them in Airtable, and pings you in Telegram, so a broken release is caught the same day.
- A dead event is caught the same morning, not a week later
- Dashboards stop being read on numbers that are wrong
- Every anomaly is tagged stopped, spiked, dropped, or new
Built for product analysts · data teams · growth teams · engineering managers
What does WebRun do on every run?
The exact actions WebRun takes, in order - in plain language, so you can adjust anything.
-
WebRun signs in and gets to work
Opens
analytics.amplitude.comin a real browser with your saved login - no setup, no API keys. -
1
Amplitude - compare event volumes
amplitude.com
WebRun opens Amplitude to compare event volumes. - Open Amplitude and select the project and the events you have asked it to watch
- Read yesterday's volume for each event and compare it against the recent daily average
- Flag events that fired zero times yesterday but were firing before
- Flag events whose volume moved sharply in either direction beyond the threshold you set
- List event names seen for the first time, since an unannounced new event usually means a release shipped
- Check the user properties attached to the flagged events for values that suddenly went blank
Done when Every watched event has a volume comparison and the anomalies are listed with their size.
-
2
Airtable - log the anomalies
airtable.com
WebRun opens Airtable to log the anomalies. - Open your tracking health base in Airtable and add a row per flagged event
- Record the event name, yesterday's volume, the recent average, and the percentage change
- Tag each row as stopped, spiked, dropped, or new, so triage is quick
- Keep the history, so a repeatedly flaky event is visible over weeks rather than argued about
Done when Airtable holds a tagged row per anomaly with its numbers attached.
-
3
Telegram - ping you the same morning
telegram.org
WebRun opens Telegram to ping you the same morning. - Send a short Telegram message naming the events that stopped firing first
- Follow with sharp drops and spikes, each with yesterday's number against the average
- List newly seen event names at the end, since those usually explain the rest
- Say nothing on a clean morning, or send a one-line all clear if you prefer
Done when You have this morning's tracking health message in Telegram.
How is each run configured?
Secure by default
Connect once, stays signed in
WebRun signs in once and keeps each session in a persistent environment, so every run picks up right where it left off.
Every action is checked against this policy before it runs.
Questions, answered
Will it change anything in Amplitude?
No. WebRun reads event volumes and charts and reports what moved. It never edits a chart, blocks an event, or changes a taxonomy setting, so your workspace is untouched.
How does it know what normal looks like?
It compares yesterday's volume for each watched event against that event's own recent daily average, so a naturally quiet event is not flagged just for being small.
Will weekends set it off?
No. Set the threshold and the comparison window you want and WebRun uses the same weekday pattern, so a normal Sunday dip is not reported as broken tracking.
Related templates
Put this on autopilot.
Turn it on in minutes - or have our team set it up for you.