Measuring use of new features (was Re: Bulk cancelling of jobs in Treeherder)

[Lawrence Mandel]

I've seen a number of emails that end with something along the lines of
"hope this is useful" or "I expect this will be useful". This does seem
like a useful feature and I think our gut is often correct but I also think
we can measure feature uptake/use to determine how often certain features
are actually used. This information can help guide feature and UI decisions
in our tools. Do we collect metrics on feature use like "cancel all the
things" today?

Lawrence

On Tue, Feb 7, 2017 at 3:23 PM, William Lachance <wlac...@mozilla.com>
wrote:

> I just landed a feature in Treeherder which lets you bulk cancel a group
> of jobs (but not necessarily the whole push). I wrote up a blog post about
> it here (includes a tutorial video):
>
> http://wlach.github.io/blog/2017/02/cancel-all-the-things/
>
> This is a relatively new feature so please file bugs (needinfo'ing me) if
> you run into problems.
>
> Hope people find this helpful,
>
> Will
> _______________________________________________
> dev-tree-management mailing list
> dev-tree-...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-tree-management
>

[William Lachance]

On 2017-02-07 3:28 PM, Lawrence Mandel wrote:
> I've seen a number of emails that end with something along the lines of
> "hope this is useful" or "I expect this will be useful". This does seem
> like a useful feature and I think our gut is often correct but I also think
> we can measure feature uptake/use to determine how often certain features
> are actually used. This information can help guide feature and UI decisions
> in our tools. Do we collect metrics on feature use like "cancel all the
> things" today?

We do not. Ideally we'd wire up google analytics (or equivalent) to
Treeherder so we could measure that.

Will