You closed a room type for sale, but the OTA is still accepting bookings for it, and this walks through why a stop-sell can fail to sync and the steps to actually fix it.
You apply a stop-sell for a room type — maintenance, an internal block, a date you want closed — and your channel manager shows it as applied, but a booking still comes through from the OTA for a date that should be closed. A stop-sell that doesn't actually sync is one of the few extranet problems that can turn into a real overbooking, so it's worth diagnosing properly rather than just reapplying it and hoping.
Your channel manager's own interface shows the room type as closed — no error, no warning. The OTA's extranet, checked separately, still shows it open, or a booking simply arrives for a date you believed was blocked. Because the failure is invisible on the sending side, most properties only discover it once a booking has already landed, which is the worst time to find out.
This is distinct from a stop-sell that was never applied in the first place — that's a process error. A stop-sell that was applied correctly in your system but didn't reach the OTA is a sync failure, and the fix is different.
Most connectivity setups send a stop-sell as a message to the channel the same way a rate or availability change is sent, and that message can fail to deliver, be delayed, or arrive and be rejected without a visible error on your side — each of those looks identical from your interface, which just shows the stop-sell as applied internally. A second cause is scope: a stop-sell applied at the wrong level — a specific rate plan rather than the whole room type, or vice versa — can sync successfully but simply not cover the booking that then comes through, which looks like a sync failure but is actually a scope mismatch. A third is sequencing: if a booking request and a stop-sell are processed close together, the order they're applied in can determine whether the stop-sell closes the room before or after that booking already went through.
Open the OTA's own extranet directly and check the actual current status of the room type and rate plan for the specific dates in question — this is the only way to know whether the stop-sell truly didn't sync, versus synced to the wrong scope. Check your channel manager's activity or change log for that stop-sell to see the exact scope it was applied to and the timestamp, and compare that timestamp against the booking's own timestamp if one came through. If the extranet shows the room correctly closed now but a booking still arrived earlier, that points to a timing collision rather than a sync failure.
This is exactly the case read-back verification is built for: every stop-sell is read back from the channel after it's sent and compared against what was intended, so a message that failed to deliver or was applied to the wrong scope shows up as a mismatch immediately rather than looking identical to a success. Combined with a physical no-double-booking guarantee at the database level, a stop-sell failure that does slip through still can't result in two overlapping reservations for the same room — it's prevented by construction, not by a check that could itself be skipped. You can see both on the public demo at try.primehostpartners.com without signing up.