Bug severity vs priority: the difference, with examples
Severity is how much damage a bug does. Priority is how soon it should be fixed. Here is how to set each one, with examples of bugs where the two disagree.

Two bugs are reported on the same morning. In the first, the app crashes when a user uploads a photo larger than 50 MB from a tablet in landscape mode. In the second, the company name is spelled wrong on the home page.
Which one gets fixed first?
Most teams fix the spelling mistake, and they are right to. The crash is the worse bug, but almost nobody will hit it. The typo is harmless, and every visitor will see it.
That gap between how bad a bug is and how soon it should be fixed is the difference between severity and priority.
Severity: how much damage the bug does
Severity describes the bug's effect on the product when it happens. It is a technical judgement, and the tester who found the bug usually sets it.
A common four-level scale:
- Critical. The product cannot be used, or data is lost or exposed. Checkout fails for everyone. One user can see another's orders.
- High. A main feature is broken and there is no way around it. Password reset emails are never sent.
- Medium. A feature misbehaves, but users can still finish what they came to do. A filter returns the wrong results, though search still works.
- Low. Cosmetic or minor. A misaligned icon, a spelling mistake, a wrong colour.
Severity does not change with circumstances. A crash is a crash whether it happens to one user or a million.
Priority: how soon it should be fixed
Priority describes the order of work. It is a business judgement, and the product owner or project manager usually sets it.
- Urgent. Fix now, and release as soon as it is ready.
- High. Fix in the current cycle.
- Medium. Fix when the higher items are done.
- Low. Fix when convenient, or perhaps never.
Priority does change. A low-priority bug on a page nobody visits becomes urgent the week that page is featured in a campaign.
When the two disagree
The four combinations are easiest to understand through examples.
| High priority | Low priority | |
|---|---|---|
| High severity | Payment fails for all cards | Crash in a feature used by a handful of users on an old browser |
| Low severity | Wrong price shown on the home page banner; logo broken on launch day | A typo in the footer of a rarely visited page |
High severity, high priority. Nobody argues about these. Everything stops until it is fixed.
High severity, low priority. A serious failure in a place few people reach. It should be fixed, but not ahead of problems that affect everyone.
Low severity, high priority. Technically trivial, but visible, embarrassing or legally important. A wrong phone number on the contact page is a one-word fix that costs you customers every hour it stays.
Low severity, low priority. Real, harmless, and reasonably left for a quiet week.
How to decide
For severity, ask about the bug:
- Does it stop someone completing a main task?
- Is data lost, wrong, or shown to the wrong person?
- Is there a way around it?
For priority, ask about the business:
- How many users will meet it?
- Does it cost money or damage trust while it stays?
- Is something depending on it, such as a launch, a demo or a deadline?
- How hard is the fix, and how risky?
The last question matters more than people admit. A medium bug with a five-minute fix often goes ahead of a high one that needs a week.
Common mistakes
- Marking everything critical. When every bug is urgent, the label stops meaning anything and the team starts ignoring it.
- Letting one person set both. Testers tend to rate by damage, managers by visibility. The two views are both needed.
- Treating severity as priority. This fills the schedule with rare crashes while visible, simple problems stay live for weeks.
- Never revisiting. Priorities set a month ago may be wrong today.
What a tester should write
A good bug report gives the team what it needs to set both values without guessing:
- What happened, and what should have happened.
- How often it occurs: every time, sometimes, or once.
- Who is affected: all users, one browser, one account type.
- Whether there is a workaround.
- A suggested severity, with the reason.
Our guide on how to write a bug report has a full template.
Why this matters for cost
Severity also affects what testing costs if you pay per bug. In our pay-as-you-go pricing, a critical bug and a cosmetic one are priced differently, because they are not worth the same to you.
When we report bugs, each one comes with a severity and the evidence behind it, and you decide the priority. If you would like your product tested that way, start a project.