Quick answer: To do a bug bash: gather people to intensively test the game over a focused period, capture every bug and crash they hit automatically, and triage and prioritize the findings afterward.

A bug bash is a concentrated effort to find bugs fast. These are the steps to run one that pays off.

Step 1: Gather People to Test Intensively

Start by gathering people, your team, community, or testers, to intensively test the game over a focused period (a few hours or a day), trying to find as many bugs as possible, including unusual things and edge cases. A bug bash concentrates testing effort to surface many issues quickly.

Bugnet works during the bug bash: by capturing crashes automatically from everyone testing, it turns the bug bash participants into a source of real crash data across their varied hardware, so the bash surfaces device-specific crashes you could never find on your own machine.

Step 2: Capture Every Bug and Crash Automatically

Next, capture every bug and crash the participants hit, automatically, not just what they remember to write down: people in a bug bash will not report every crash they hit, so automatic crash capture ensures you get the full set of technical issues, alongside the bugs they do report. Capturing everything is what makes the bash thorough.

Bugnet captures the technical findings automatically: it records crashes from all bug bash participants with full context (stack trace, device, version, breadcrumbs) and groups them by impact, so you get every crash they hit, even the ones they would not report, with the context to fix them, maximizing what the bash surfaces.

Step 3: Triage and Prioritize the Findings

Finally, triage and prioritize the findings afterward: a bug bash produces a lot of bugs, so deduplicate them, rank by impact and severity, and decide what to fix. A bug bash only pays off if you act on what it finds, so the post-bash triage that turns the pile of findings into a prioritized plan is essential.

Bugnet helps you triage the technical findings: it groups the bash's crashes by signature (deduplication) and ranks them by impact (prioritization), so the many crashes the bash surfaced arrive organized into distinct, prioritized issues, making the post-bash triage fast and ensuring you fix the most important findings first.

To do a bug bash: gather people to test intensively over a focused period, capture every bug and crash they hit automatically (not just what they write down), and triage and prioritize the findings afterward, a bug bash only pays off if you capture and act on what it finds.