How we continuously verify our delivery pipeline.
Most providers ask you to trust a badge. We run a live canary through our own production pipeline every sixty seconds and publish the results — this page documents exactly how that works, and what it does and does not prove.
Probes run continuously against production · incidents open automatically on failure.
How it works
A real email, every minute
Every 60 seconds, we send a real message through the same production API, queues and sending infrastructure used by customer traffic. No synthetic shortcuts, no staging environment standing in for the real thing.
Verified end to end
Each probe is tracked through five hops: API accepted, handed to the sending infrastructure, recipient mail server accepted the message, delivery webhook received, webhook signature verified. Every hop is timed against its own SLA — a slow hop is flagged before it becomes a slow email.
Cryptographically verified
The webhook that closes the loop is Ed25519-signed. Verifying that signature confirms the event was generated by wemail and wasn't modified in transit — the same signing scheme that backs every customer webhook we send.
Alarms, not press releases
Results feed the status page and our internal alerting directly. A failed or slow hop opens an incident automatically — and a dead-man switch alerts the team if the monitor itself ever goes silent.
What this is not
It is not a marketing number. We publish the mechanics — what is probed, how often, and what counts as a failure — so the claim is checkable, not just quotable.
It is not a guarantee for your domain. End-to-end verification covers our sending infrastructure. Your own reputation — authentication, volume, complaint rate, recipient engagement — still shapes where your mail lands.
It is not the whole story yet. Independent inbox-placement measurement across major mailbox providers is on our roadmap. Until it ships, the right way to validate placement is a pilot: run your real traffic side by side and compare.
Common questions
Why send a probe email every minute?
Because "the API is up" is not the same as "the full delivery pipeline is healthy". A minutely end-to-end probe catches queue stalls, provider slowdowns and webhook failures within sixty seconds — usually before any customer notices.
What happens when a hop fails or slows down?
The monitoring pipeline opens an alert episode immediately, notifies the on-call team, and the affected hours are marked degraded or outage on the status page. Recovery is verified the same way the failure was detected — by the probes themselves.
Do you measure inbox placement?
Not publicly yet. Today we verify delivery mechanics end to end and publish the system live; an independent seed-panel measurement of inbox placement is on the roadmap. Meanwhile, the free tier makes a mirrored pilot with your real traffic the most honest test.
Judge it on your traffic.
3,000 sends per month, forever. No credit card. Run the pilot and compare the numbers yourself.
