Crash and freeze figures for Term, from AppGallery Connect’s monitoring of real installs — not from our test devices. A script refreshes them week by week, including the weeks that look bad.
Across the last 64 days, on real devices in real use.
Over the same 64 days. A freeze is the system killing the app for not answering.
Of the last 9 weeks, no crash and no freeze at all.
| Week | Freezes | Crash-free |
|---|---|---|
| 09/21 – 09/28 | 0 | 100.00% |
| 09/14 – 09/21 | 0 | 100.00% |
| 09/07 – 09/14 | 0 | 100.00% |
| 08/31 – 09/07 | 0 | 99.98% |
| 08/24 – 08/31 | 0 | 99.96% |
| 08/17 – 08/24 | 0 | 100.00% |
| 08/10 – 08/17 | 0 | 100.00% |
| 08/03 – 08/10 | 0 | 100.00% |
| 07/27 – 08/03 | 0 | 99.79% |
The figures come from AppGallery Connect’s crash and app-freeze metrics — the same monitoring the store shows us, counting what happens on real devices in real use. Nothing on this page was measured on our own phones.
Crash-free sessions is 1 − crashes ÷ app launches. AppGallery reports a period’s launches only when that period contained a crash, so most single weeks have no denominator at all and their rate reads “—”; the headline asks for the whole quarter in one query, where it is reported.
A week with no crashes is 100% crash-free whatever its launch count was, so those rows carry a rate even though nothing was there to divide. What this page also leaves out is the individual crash signatures: they name internal functions and would tell a reader nothing.
Pulled 2026-09-29.