Pinned Tweet
Welcome to BugSplat 🐛 Learn how we can help you more quickly and easily find and fix critical defects causing issues in your app --> bugsplat.wistia.com/medias/3…
You marked the bug fixed. Closed the ticket. Shipped the update.
Then it comes back silently - in a newer release.
This is a regression, and most teams only find out when users complain. @BugSplatCo has a smarter way. 🧵
Alerts land where your team already works: Slack, Discord, or Microsoft Teams.
Using Jira? BugSplat automatically updates linked issue statuses when a regression is detected. Your defect tracker stays in sync, no manual updates needed.
Stop learning about regressions from angry users. Set it up once and BugSplat watches every release for you.
Learn more about Regression Alerts in the docs 👇
docs.bugsplat.com/introducti…
Looking for some fun light reading? Our August Changelog Digest is live! changelog.bugsplat.com/welco…
In 1983 the US video game industry lost 97% of its revenue in two years. Atari buried truckloads of cartridges in a New Mexico desert.
The original video game crash, and what it still teaches about shipping quality: blog.bugsplat.com/great-vide…
"Users will tell us when it crashes."
They won't. They'll uninstall, refund, or leave a one-star review that says "buggy mess" with no repro steps.
Do you still need crash reporting in 2026? blog.bugsplat.com/do-you-nee…
You fixed it. You closed it. Three releases later it's back, and nobody noticed for a month.
Regression alerts: BugSplat flags a closed defect the moment it reappears in a new version. docs.bugsplat.com/introducti…
20% of your bugs cause 80% of your crashes.
Which means most teams spend most of their time fixing crashes almost nobody hits.
The 80/20 rule of bug fixing: blog.bugsplat.com/the-80-20-…
A crash report tells you the state at the moment of death. A log tells you the story that led there.
Teams argue about which to invest in. The answer is both, wired together: blog.bugsplat.com/crash-repo…
NASA ships code where a crash can cost $300 million and there's no "send report" button on Mars.
What their process looks like, and which parts are worth stealing for teams with actual deadlines: blog.bugsplat.com/why-nasa-c…
Your crash stacks went blank mid-sprint. Nobody changed the crash reporter.
Somebody changed the build.
Upload symbols automatically on every build so this never happens: docs.bugsplat.com/education/…
EXCEPTION_ACCESS_VIOLATION is your program touching memory it doesn't own. Reading a null, writing past an array, using something already freed.
What it actually means, and how to track one down: blog.bugsplat.com/access-vio…
Your crash report says the app died at 0x00007ffb2a4c1e37.
That address means nothing without one thing: the symbol file from that exact build.
A short thread on symbols, and how teams lose them.
The fix is boring: upload symbols automatically at build time, every build, and let your crash system store and match them.
BugSplat does this with symbol-upload in CI. Set it once and stop thinking about it.
Full write-up, including what to do when stacks come back blank: blog.bugsplat.com/symbol-fil…
