Status
Stated plainly, measured before published
Status pages in this category tend to open with a green dot and a percentage nobody can audit. Ours opens with a sentence, because a sentence is what we can currently prove.
Current state
No incidents to report — and no measured uptime to show yet. This page gets live numbers when the measurement pipeline ships; until then we refuse to print a percentage we didn't measure.
What this page will publish
When the measurement pipeline ships, every figure below arrives under the same rule as the rest of the site: measured from our own systems, shown with its sample and window, or absent — see how we source every figure.
- API uptime, 90 daysMeasured by our own external probes, plotted over a rolling 90-day window with the probe interval stated. Not a badge — a chart with its method attached.
- Delivery-pipeline uptime, 90 daysThe path from supplier webhook to your inbox, measured end to end over the same rolling 90-day window. If the pipeline was down, the chart says so.
- Time-to-first-code, per routeThe p50 and p90 of the measured TTFC distribution for each route, always with the sample size and the time window inline. Routes without enough orders show the same honest sentence they show everywhere else on this site.
- Incident historyTimestamped, written at the time, and left up afterwards. An incident log that only contains flattering entries isn't a log.
Get told when the numbers exist
One email when the first measured uptime chart goes live, and incident notices after that. Nothing else.