Two admin screens keep you ahead of data problems instead of behind them:
Admin → Sync status — the connector's health and every feed's freshness: what arrived, when, and how much. Each feed has its own rhythm (some minutes, some nightly), and the screen judges freshness against that feed's schedule — a nightly feed isn't "stale" at lunchtime.
Admin → Feeds — the configured feeds themselves: what's synced and on what schedule.
When something goes wrong:
- One feed errored, connector fine — open the feed for its error detail; most settle on the next scheduled run once the cause (usually a source-side hiccup) passes.
- Connector hasn't checked in — the issue is on the server it runs on: restarts, credentials changed, or network. It reconnects on its own once it can; nothing needs re-installing.
- A day is missing — it stays visibly missing (the product never backfills a gap with a neighbouring day). Once the source syncs again, history that can be re-read is caught up automatically; a snapshot that never existed is reported as a gap, honestly.
The nightly feed-health summary email tells you about problems without you checking — if it's quiet, things ran.