More detailed stats

We’re currently focusing on decreasing the total number of errors our application throws into Sentry, and were wondering if there are plans to create a more detailed stats page that can be used to see the number of events/ we sent over a long period of time? It seems that the existing stats page only shows events over the prior week, and we’d be interested in seeing more long-term trends - has our aggregate number of bugs increased or decreased over the past month, couple months, etc. After that, being able to drill down and see long-term trends for a given event type would also be very useful.

Alternatively, if this can be accomplished somewhat easily via API, that would also be good to know.

Thank you!
Jonathan

1 Like

We do have plans to expose a lot more of this, but in the meantime you can use the API.

Open up Chrome and look at the request we send to the server there. Anything we’re doing you can do yourself (barring a few special cased requests).

https://docs.sentry.io/api/organizations/get-organization-stats/

In addition to @zeeg’s comments, the new organization weekly report email is one of our first steps in that direction: Weekly Reports for Early Adopter Organizations

The problem that I have with the current weekly reports and reporting in general for Sentry is that the numbers (e.g. 40% less events than previous week) do not correlate with how good we’re doing as a team - there can be a minor exception that is happening 2k times per day and then there’s one important exception that has a big impact on our business but it only happens once.

A more accurate reflection of how we’re doing would be the number of new distinct events (issues). We’d be able to say we have resolved 20% more bugs than in the previous week.

1 Like

We are interested in more detailed stats as well. Ideally we could request something like
https://sentry.io/api/0/projects/sentry/project-name/environment-value/?stat=generated&since=1500029367&until=1500029967 through the API so we can distinguish the amount of events per 10 or 15 minutes, not only per project, but also per environment of that project.

This way we can create alerts if project ‘important’ and environment ‘production’ starts to generate a lot of eventes, while for other projects and/or environments things are probably less worrysome.