API
CheckingEverything the apps, the till and your own integrations read and write through.
This page does not read a dashboard we maintain. It calls the live services from your browser, on your connection, while you are looking at it — so what you see is what your own network can actually reach.
Everything the apps, the till and your own integrations read and write through.
Live chat, presence, huddles, and documents several people edit at once.
This site, the marketing pages and the sign-in entry point.
Checks run in your browser and measure round-trip time from where you are sitting. A slow or failed reading can mean the service is struggling — or that your own network, VPN or DNS is. If a service looks down here but fine for a colleague, that difference is itself the useful part.
The moving parts, in the terms that matter to you rather than to us.
One API serves every client — phone, tablet, till, screen and browser. They all talk to the same place and see the same records, so there is no version of the truth that only one device has.
Each business is its own tenant. Every record carries its owner and every query is filtered by it before it reaches storage, so one customer's data is never one mistaken parameter away from another's.
The till, the kitchen screen and the queue display keep serving through a wobbly connection and reconcile when it comes back. A brief outage should cost you a sync, not a shift.
Releases roll out gradually rather than all at once, so a bad change is caught while it is still small. We do not take the platform down to deploy.
If something is wrong and we know before you do, it goes out in the app and by email to account owners. We would rather tell you about a problem you had not noticed than have you find it at the counter.
Account owners get a health view inside the app for their tenant specifically — sync state, device check-ins and anything queued. That is the one to open when it is only you seeing the problem.
Tell us what you were doing, on which device, and roughly when. That is usually enough to find it in the logs without a back-and-forth. Production-down issues on an Enterprise plan go to the 24/7 line.