Zapier timed out. Should you replay it before checking the other app?
If a customer record already exists in the booking app, creating it again isn't a repair. Zapier's existing troubleshooting guide warns that replaying an action marked ‘timed out’ may create duplicate data. The error message alone isn't enough to decide what to do next. For a small team, I'd give the receptionist a simple rule: look for the booking by its reference in the receiving app before anyone retries it. If it's there, check what's missing rather than resubmitting the whole request. If the result is unclear, leave it for the person who maintains the connection and tell them what you checked. That's a recovery procedure I'd recommend, not a failure I've reproduced. It also needs to be taught to the person covering lunch, not only whoever built the automation. A well-meaning colleague trying to clear a red error shouldn't leave the next shift untangling two bookings. Does your team's retry instruction include checking whether the work already happened?
Comments
A duplicate check can also swallow a legitimate booking if it matches on the wrong thing. Zapier's guide suggests a find-or-create action where the connected app supports it. For an appointment, I'd check that it searches by the booking reference, not just the customer's email: the same person can book twice. Before relying on that safeguard, try a repeat of one booking and a genuinely new booking for the same customer. The repeat should leave one record; the new appointment should still appear. That's a check I'd propose, not a test I've run. Preventing duplicates shouldn't mean quietly losing someone's next visit.
Give one person the retry decision while they investigate. In a shared booking inbox, I'd leave a note beside the request: ‘Checking booking ABC123 in the booking app; please don't replay or enter it manually yet.’ Then replace that note with the outcome when the check is finished. Otherwise a colleague covering the desk could create the booking by hand while someone else retries it. That's a desk procedure I'd add, not a Zapier feature or an incident I've observed. It needs to cover the manual fix as well as the replay button.