How Measurebase monitors your domains
DNS records get changed by whoever manages your registrar. Tag configs get edited by whoever's in GTM that week. Scripts get copy-pasted onto the wrong site. None of that shows up as an error — it just quietly stops your data from arriving. Measurebase's monitoring runs a set of checks across every domain and account, continuously, and opens an alert the moment one of them fails.
What gets checked
- DNS drift — re-verifies that every active domain's A record still resolves to the expected target. Domains only get checked once at creation otherwise, so nothing today would catch a registrar record changing, or a CDN getting put in front of your subdomain, months after go-live — this closes that gap.
- No data — flags a domain that's gone quiet: no real tracking request seen for longer than its configured threshold. Catches the cases DNS can't — a container republish that broke something, a removed snippet, an ad-blocker rollout — anything that zeroes out traffic without DNS ever being wrong.
- Elevated errors — flags a domain whose error rate spikes well above its normal baseline, catching a broken container config or a bad deploy before it silently eats a chunk of your traffic.
- Foreign traffic — catches a specific, real mistake: another Measurebase customer's site accidentally using your container's tracking script, so their visitor data lands in your account instead of theirs. Detected by checking whether any other domain registered in the system shows up as a Referer on yours.
- 80% of quota / Quota reached — account-level, not per-domain. Fires from the same usage calculation that drives auto-upgrade, so the alert and the real trigger never disagree about how close you actually are.
Setting your no-data threshold
Every other check runs on fixed logic with nothing to configure. No-data is the one exception: each domain has its own threshold, set on that domain's Monitoring panel, from 1 to 168 hours (a full week) — 24 hours by default. A newly created domain also gets a grace period before this check applies at all, so a domain you haven't finished installing yet doesn't alert on day one.
Raise it for a low-traffic B2B site where a quiet Sunday is normal; lower it if you'd rather know within the hour that something broke on a high-traffic property.
Where alerts show up
- The alert bell — in the topbar on every dashboard page, with a badge showing how many are open and a dropdown listing the most recent, whichever domain or account they belong to.
- The Alerts page — opened from that dropdown's "View all alerts" link: every alert across the account in one table — domain, type, message, open/resolved status, and when it opened — paginated with a Load more button.
- Email — sent automatically the moment an alert opens (subject: "New {severity} alert — …") and again once it resolves ("Resolved — …"), so you don't have to be watching the dashboard to find out.
- Slack — click "Add to Slack" on your account's Settings page to connect up to 3 channels, and every alert posts there too, alongside email rather than instead of it. Alerts tied to a specific domain name that domain directly in the message; account-wide alerts (quota) name the account instead.
An alert clears itself automatically once the underlying problem is gone — DNS resolving correctly again, traffic resuming, usage dropping back under the threshold — no need to dismiss anything by hand.
Also part of the same system
The same monitoring loop backs the 24-hour staleness check on Auto Grant Consent — a testing-only power-up that's dangerous to leave on, checked the same way a domain going quiet or DNS drifting is checked: continuously, with an alert the moment it's been on too long.