Impact score or Priority of exceptions

Hey,

First up, we’ve been using Sentry for over a year now and it’s great. Thanks!

I’ve been doing some digging but can’t seem to find anything in your github issues or on here - so forgive me if this has been discussed before.

Are there any plans to introduce either a concept of priority or impact score to Sentry? I can see here that it’s been mentioned once before quite a while ago What does sort by priority mean? .

Basically, previously we have just fixed all bugs and now we’ve reached a scale where we want to be a bit more refined in our approach. Something like a “Do, defer or delete” concept. We want to track priority to measure and hopefully trend towards bugs being both fewer (we can do this now) but also more minor.

Thoughts?

Cheers
Morgan

1 Like

Absolutely. Right now we’re focused on removing some limitations of querying data, but we’re hoping to have more signals to better create basic impact scoring early next year. The big (easy) missing one for us is connecting events to session impact. For example, if we can just let you sort “errors by % of requests impacted in the last 48 hours” that’s going to be a lot more meaningful than our time decay scoring (which is what the current priority sort attempts).

@zeeg Any update on this?

Alternatively, having at least a way to sort errors by number of affected users would be a good-enough proxy.

Please, any updates on this thread? There are a lot of minor issues in my errors list, so would be nice if we have better algorithm to sort by prioriity. The suggestion by @the21st is good. In addition, I think that would be useful if consider something like security vulnerabilities/criticality level.

@EduardoLima806 Sentry recently released an update and now you can sort issues by the number of users impacted.