Quick answer: Watch whether the issue stops occurring in the field on the version that includes your fix. Track issues per version, occurrences should drop to zero on the fixed build.
A fix that works on your machine might not work in the field, fixes address the wrong cause, miss edge cases, or get undone. Knowing a fix actually worked requires confirming with real data. Here's how to verify your fix did what you intended.
Don't Trust 'It Worked on My Machine'
A fix working once on your machine isn't proof, the bug may have depended on conditions you don't have, or your fix may handle only some cases. Real verification comes from the field, where the bug actually happened, not from a single successful test locally.
Bugnet shows whether an issue continues for real players after your fix, so you don't have to trust a local test. Knowing a fix worked means confirming it in the conditions the bug occurred in, which is the field, not your dev environment.
Watch Occurrences on the Fixed Version
The concrete test is per-version: after your fix ships in a build, watch whether the issue still occurs in reports from that build onward. If occurrences drop to zero on the fixed version, the fix worked. If they continue, it didn't, and you know to dig back in.
Bugnet tracks issues per version, so you can see whether a bug keeps happening on the build meant to fix it. Watching occurrences on the fixed version is the real verification, actual data from real players confirming, or refuting, that your fix landed.
Reopen If It's Still Happening
Verification can fail, and that's valuable to know. If the issue still occurs on the fixed version, your fix didn't work, and catching that early lets you address it before you've moved on assuming success. A bug isn't truly fixed until the data shows it's gone.
Bugnet surfaces continued occurrences after a fix, so a failed fix is visible rather than silently ineffective. Knowing your fix actually worked is not trusting a local test, watching occurrences on the fixed version, and reopening if it persists, the loop that lets you close bugs with proof instead of hope.
Watch whether the issue stops occurring in the field on the fixed version, occurrences should drop to zero. Don't trust a single local test; if occurrences continue, the fix didn't work and you reopen.