There’s this thing called the Pareto Principle, which says:
roughly 80% of the effects come from 20% of the causes
You can quibble about the specific number values. Maybe 80 and 20 aren’t exactly right. But as long as you have customers that aren’t perfectly evenly distributed across bugs, you should consider that maybe some of your bugs aren’t worth fixing.
Here’s a contrived example: Imagine you got a product XYZ, and you got 100 users. They’re all mad because of five bugs (bug A through bug E).
- 80 of your users are mad because of bug A. (80% of 100)
- 16 other users are mad because of bug B (80% of the remaining 20)
- 3 users hate bug C (80% of the remaining 4 users)
- User #100 filed two bug reports: D and E. He won’t be happy until both are resolved.
If you add up 80 + 16 + 3, you’re at 99 users. In other words, if you fix 3 out of 5 bugs, 99% of your customer base would be satisfied.
However, making that last customer happy is probably not worth it! You can satisfy 99% of your market by doing 60% of the required work.