Channel health and sync errors
What failed, on which channel, for which product — and what it most likely means.
Integrations fail. The question is whether you find out from a report or from a customer.
How it works
- Failures are recorded per channel and per product, with the message the channel actually returned.
- Retries are bounded and scheduled rather than looping indefinitely.
- Health is readable as a view rather than by reading logs.
Read the shape, not the count
Thousands of errors is less alarming than it looks once you group them. A large number of identical messages is usually one cause — an expired token, or a daily API allowance spent by mid-morning — and clears with one fix.
It is the handful of different errors that need attention. Those are individual listings with individual problems, and they are what a raw count hides.
The error that means something else
Worth knowing because it wastes days: an Amazon rejection reading “the product ID does not match the brand records” is almost always a brand problem, not a barcode problem. People go and check barcodes, which are fine.
What it does not do
It does not fix anything on its own, and it will not alert you by default. It is a place to look, and looking is still a habit somebody has to have.
Stop reconciling stock by hand
One catalogue, six channels, one stock figure that stays right.