Problem
After deploying GA4 event tracking via GTM, there is no way to know if events stop firing until someone manually checks GA4 reports or re-runs the preview command. Website updates, CMS changes, A/B tests, or third-party script conflicts can silently break existing tracking at any time.
Proposed solution
Add a scheduled monitoring capability that automatically re-runs tracking health checks on a recurring basis and alerts when events break.
Core workflow:
- User configures a schedule (e.g. every Monday at 7:00 AM) and target artifact directory
- The skill automatically runs preview verification against the live site
- Compares results against the last known tracking-health.json baseline
- Generates a health delta report highlighting:
Events that stopped firing (regression)
New unexpected events detected
Health score changes over time
- Outputs the report to the artifact directory and optionally sends a notification (email, Slack webhook, or GitHub Issue)
Possible implementation paths:
- GitHub Actions integration: A reusable workflow that users add to their repo, triggered by cron schedule
- Built-in CLI command: e.g. ./event-tracking monitor --schedule "0 7 * * 1" that runs as a local cron job
- Agent prompt pattern: A documented prompt for AI tools to run periodically, e.g. "Every Monday, re-verify tracking for ./tracking-output/example_com and report any regressions"
Alternatives considered
Manual re-run: Users can already re-run preview manually, but this relies on discipline and is easy to forget
GA4 Realtime monitoring: GA4 shows real-time events, but requires someone to actively watch the dashboard and doesn't provide automated alerts
Third-party uptime monitoring (e.g. Pingdom, Datadog): These monitor page availability, not whether specific GA4 events are firing correctly
Google Tag Assistant browser extension: Manual testing tool, not automated or scheduled
None of these provide automated, scheduled, event-level verification with historical comparison.
Problem
After deploying GA4 event tracking via GTM, there is no way to know if events stop firing until someone manually checks GA4 reports or re-runs the preview command. Website updates, CMS changes, A/B tests, or third-party script conflicts can silently break existing tracking at any time.
Proposed solution
Add a scheduled monitoring capability that automatically re-runs tracking health checks on a recurring basis and alerts when events break.
Core workflow:
Events that stopped firing (regression)
New unexpected events detected
Health score changes over time
Possible implementation paths:
Alternatives considered
Manual re-run: Users can already re-run preview manually, but this relies on discipline and is easy to forget
GA4 Realtime monitoring: GA4 shows real-time events, but requires someone to actively watch the dashboard and doesn't provide automated alerts
Third-party uptime monitoring (e.g. Pingdom, Datadog): These monitor page availability, not whether specific GA4 events are firing correctly
Google Tag Assistant browser extension: Manual testing tool, not automated or scheduled
None of these provide automated, scheduled, event-level verification with historical comparison.