Steps to Diagnose Windows 11 Crashes and Find the Cause

Start with the crash pattern, not random fixes. This walkthrough helps you separate app crashes, system crashes, and BSODs, then test the most likely causes in the right order so you can narrow the fault without losing clues.

Share
A Windows 11 laptop sits open on a desk in a home office.

The fastest way to diagnose a Windows 11 crash is to classify it first: one app, several apps, or the whole system. That tells you which layer to inspect next, instead of changing settings at random. Start by recording the exact crash pattern before you touch anything, because a reboot or restart can wipe out clues.

Start by classifying the crash

Before you change anything, write down the pattern while it’s still fresh. That gives you a baseline and stops you from chasing the wrong layer.

Record these details:

  • What crashed: one app, File Explorer or the desktop shell, or the whole PC.
  • What it looked like: the app closed, Windows froze, the screen went blue, or the machine restarted.
  • When it happened: on launch, after sleep, during video playback, while gaming, or under heavy load.
  • What else was open: background tools, overlays, syncing apps, antivirus, or remote access software.
  • What changed recently: a Windows Update, driver update, new app, new peripheral, BIOS update, or a setting change.

That first split matters. A single app crash often points to the app itself, a plug-in, or an overlay. Crashes across several apps usually point to a shared driver, a Windows file problem, or unstable hardware. A full-system crash, especially a BSOD or reboot, pushes you toward drivers, memory, storage, GPU, or power instability.

If the crash only happens after waking from sleep, treat that as a different pattern from a launch crash. If it happens only under load, such as video editing or gaming, that narrows the field again. A machine that crashes while idle is telling you something different from one that falls over only when the GPU is working hard.

Before you move on, decide which bucket it fits best:

  • one app
  • several apps
  • the whole system

That choice determines the next check, and it keeps the rest of the process organized.

Check the last change and the easiest causes first

Start with the low-risk checks that often explain a new crash streak. In practice, the most useful clue is usually what changed right before the crashes began.

  1. Think back to the last change. Look for a Windows Update, driver update, new app, new antivirus tool, new dock, USB device, or a setting you changed.
  2. Remove anything newly added. Unplug recent USB devices, docks, hubs, adapters, expansion cards, and external drives, then test again. Microsoft’s support guidance also calls out newly added hardware as a first thing to remove.
  3. Check for heat. If the fan is roaring, the case feels unusually hot, or the crash happens under load, overheating moves up the list. Clean vents, make sure the PC has airflow, and avoid blocking intake or exhaust. You do not need a perfect temperature reading to notice a bad thermal pattern.
  4. Check free space on the system drive. Windows needs room for updates, paging, logs, and temporary files. Microsoft says to aim for at least 10–15% free space on C: so Windows can create swap files and perform maintenance.
  5. Note whether the crash tracks a peripheral. If it only happens with one headset, webcam, controller, or dock attached, that device or its driver becomes a stronger suspect.

If the crash started right after a change, undo that one thing first. That is usually more productive than stacking five other fixes on top of it.

A Windows 11 laptop sits next to disconnected USB peripherals on a work surface.

Use Windows tools to see what actually failed

Windows already keeps two useful records: Reliability Monitor gives you the timeline, and Event Viewer gives you the detail. Use both, because they answer different questions.

Open Event Viewer. Search for Event Viewer, then check Windows Logs > Application and Windows Logs > System.

Event Viewer > Windows Logs > Application
Event Viewer > Windows Logs > System
Settings path

For an app crash, look for entries that name the application and an exception code. For a system crash, look for bug check, kernel, disk, display, or driver-related events. If the crash is a BSOD, a minidump may also be present in C:\Windows\Minidump.

Open Reliability Monitor. Search for Reliability Monitor in Start, then open View reliability history.

Control Panel > Security and Maintenance > Maintenance > View reliability history
Settings path

Look for red X entries around the time of the crash. Note the date, the failing app or component, and whether the same problem repeats.

  1. Match the pattern to the likely layer. - Repeated entries for the same app point toward that app, its add-ins, or overlay tools. - Display driver, storage, or kernel entries point more toward a driver fault than a pure app bug. - Sparse or unclear logs do not clear hardware. They just mean you need more isolation.

A useful example is a media app that closes only when a GPU overlay or capture tool is running. That can look like the app is broken, but the log often points to the driver or overlay instead. In Microsoft Q&A, one Windows 11 user described crashes around media playback and background apps and asked how to log the running processes to diagnose it.

Separate software trouble from hardware trouble

Safe Mode and Clean Boot help you separate Windows and third-party trouble. Safe Mode loads a minimal set of drivers and services, while Clean Boot disables most non-Microsoft startup items and services so you can test a cleaner Windows session.

  1. Try a Clean Boot if Safe Mode is too limited for your test. This helps when the PC must run closer to normal, but you still want to strip away background software. If the crash stops in a Clean Boot, re-enable items in small groups until the culprit returns.

Try Safe Mode first if the crash is broad or hard to explain. If the PC stays stable in Safe Mode, the problem is more likely a third-party driver, startup app, overlay, or service than core Windows itself. Microsoft’s troubleshooting flow uses Safe Mode for that reason.

WinRE > Troubleshoot > Advanced options > Startup Settings > Restart
Windows Recovery Environment path

Then choose Safe Mode.

  1. Watch what still crashes. If the same app still crashes in Safe Mode, that app or a core dependency is still suspect. If the whole system still reboots or BSODs in Safe Mode, I start thinking less about user-mode software and more about drivers, memory, storage, GPU, or power.
  2. Use the crash trigger to guide the next branch. A crash under load points harder at GPU, thermals, or power. A crash during sleep or wake points toward power management, chipset, storage, or a driver issue. A crash that hits several unrelated apps points toward Windows files, a bad driver, or unstable hardware.
  3. Separate likely hardware causes. - RAM: random BSODs, app corruption, and crashes that seem to move around. - Storage: hangs, file corruption, bad reads, or crashes tied to disk activity. - GPU: crashes during games, video, browser acceleration, or external display use. - Power instability: sudden reboots or crashes under load, especially with no useful app-level error.

Microsoft has also been testing a memory-related diagnostic scan that runs after a BSOD restart in some Windows 11 Insider builds, which is another sign that memory corruption remains a common target when crashes look system-level. If Safe Mode and Clean Boot both fail to make the machine stable, hardware rises on the list fast.

Do not keep forcing repeated crash cycles if the machine is getting worse, corrupting files, or rebooting in the middle of your work. Back up anything you can first.

  1. If the crashes keep happening after those checks, note whether they still occur with the same trigger, the same app, or the same peripheral. That distinction tells you whether you should focus next on a driver, the app itself, Windows files, or hardware.

What to save and when to stop

Before you ask for help or send the PC for repair, save the evidence while you still have it. A good crash report is more useful than a vague “it keeps crashing.”

Collect these items:

  • The exact wording of any error message.
  • A screenshot of the crash screen or BSOD.
  • The date and time of the crash.
  • Reliability Monitor entries around the event.
  • Event Viewer entries from Application and System.
  • Any stop code from a BSOD.
  • The steps that reliably reproduce the crash.
  • A list of recent changes, including Windows Update, driver updates, new apps, and new hardware.
  • If present, the minidump file location.

Stop troubleshooting and protect data first if the crashes are frequent, getting worse, or putting files at risk. Back up anything you cannot afford to lose before you keep testing. If the machine is rebooting while you work, showing signs of storage corruption, or crashing so often that you cannot finish a backup, pause the diagnosis and preserve your data first.

When you hand the problem to someone else, give them the pattern, the logs, and the last change. That usually saves a lot of guesswork.

Conclusion

If the crash stays tied to one app, focus on that app, its add-ins, or anything that runs on top of it. If Safe Mode or a Clean Boot changes the behavior, the problem is usually in third-party software rather than Windows itself. If the crash follows load, sleep, disk activity, or persists even in Safe Mode, the stronger suspects are drivers, Windows files, RAM, storage, GPU, or power instability. The result you want is a short, evidence-backed list of likely layers, not a long list of guesses.

Frequently Asked Questions

Should I check Reliability Monitor or Event Viewer first?

Start with Reliability Monitor because it gives you the timeline and shows repeated failures at a glance. Then use Event Viewer to get the technical details for the exact crash, driver, or service that failed. Together, they tell you both when it happened and what Windows thinks broke.

If Safe Mode stops the crashes, what does that usually mean?

It usually means the problem is outside core Windows, such as a third-party driver, startup app, overlay, or service. That does not prove which one yet, but it does tell you to focus on software you added or enabled rather than assuming the whole operating system is bad.

When should I suspect bad RAM or storage instead of an app?

Suspect RAM or storage when crashes affect several different apps, when BSODs keep returning, or when the machine is unstable even after Safe Mode and Clean Boot tests. Random corruption, hangs during disk activity, and crashes that do not stay tied to one program are stronger hardware clues.

Sources