See whether PeopleOS, ThynkBooks, ThynkFile, ThynkTax, ThynkAudit and ThynkReach are really answering, get a bell alert when a working connection breaks, and review background deliveries that failed.
Settings, Integrations health (admins only) lists every connected product with one of three states: Live (it answered correctly), Not set up (nothing is connected, which is not an error) or Not responding (it is connected but the last check failed, with the reason in plain words). A dashboard card that shows "not connected" because the connection is broken is no longer silent: the card, the background jobs and an hourly check all report here.
If a connection that has worked before fails 3 hourly checks in a row, every admin gets one notification in the bell. You are not alerted again for the same outage. When it answers again you get one "recovered" note. A connection that has never worked does not raise an outage alert (it shows as Not responding on this page). By default alerts are in the bell only; tick "Also email admins about outages" on this page to receive an email as well.
Below the connections, the page shows the outbox: business events such as invoice sent, invoice paid, estimate accepted and payment recorded are saved together with the change that caused them, then delivered in the background to automations, webhooks and your ThynkBooks books. If a delivery fails it is retried after 1 minute, 5 minutes, 30 minutes, 2 hours and 12 hours; after 8 failed attempts the event is marked dead and listed with its last error. Press Retry to queue a dead event again: only the deliveries that failed are repeated.
Each of the four event types above still works the way it always has until your administrator turns its background delivery on (the "Delivery in the background" column shows On or Off). The switch is a server setting, not a button on this page, so nothing changes by accident.