Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

MDN Metrics: Definitions and needs

1 view
Skip to first unread message

Jeremie Patonnier

unread,
Nov 25, 2016, 7:32:00 AM11/25/16
to dev-mdc, MDN (Public)
Hello Everybody,

For quite sometimes we are regularly discussing on how to measure success
an track activities on MDN. Based on the recent evolution in Mozilla
internal process such question become more and more important.

Recently, Janet, Kadir and I worked on defining the various metrics we wish
to track to measure community success:

- Definition:
https://docs.google.com/spreadsheets/d/1nHk6-NfBuy4GvVDzB0eKGE06OYs_6fGKWM40gSlb8Lk/edit?usp=sharing
- Measurement:
https://docs.google.com/spreadsheets/d/1UXz8OZWzGY6KeIpwqXLg6eGBAYo7H5GmgQABIAAK4XE/edit?usp=sharing

The next step is to logically automate the tracking of those metrics. After
some investigation, it appears that we have good tools to do this at
Mozilla. However, using such tools require to get into a full project work
to be able to properly aggregate all the necessary data.

Because of that, it has been suggested to directly work at a higher level,
about all metrics necessary to measure MDN success. Regardless of that I'm
planning to refine the work we've done for community metrics in order to
move forward. It would not be a bigger work to work at a larger scale.

That said, this were I need you. Like we did for community we need to
figure what are the relevant metrics worth tracking for the other area of
MDN (content, platfom dev, marketing action, etc.).

So in short: What do *you* think would worth tracking about MDN? (and why)

Thanks for you're help and insight
Best,
--
Jeremie
.............................
( ̄(エ) ̄)ノ Lovely Wild Bear
Mozilla Developer Network <https://developer.mozilla.org/>
Twitter : @JeremiePat <http://twitter.com/JeremiePat>

Luke Crouch

unread,
Nov 26, 2016, 11:24:00 AM11/26/16
to Jeremie Patonnier, dev-mdc, MDN (Public)
Hopefully everyone is sick of me showing this by now. If not I haven't been
as annoying or passionate* about it as I should be ...

This is a Lean Analytics metric visualization for User-Generated Content
platforms like Reddit, Stack Overflow, MDN, etc.:

http://content.screencast.com/users/groovecoder/folders/Jing/media/c3a96549-5e06-4d30-b25a-ce348842c310/00002195.png

The bottom line of all those metrics is good quality content. That is, only
work on a metric (e.g., viral invites) when it produces better quality
content in the end.

For MDN, I've always considered "quality content" to be composed of:

Positive:
* Content helpfulness ratings by users
* Site uptime
* Site performance

Negative:
* Spam

The existing referral metric could be seen as a high-level proxy for
content quality, assuming that high-quality content makes more readers
refer back to MDN. I also like referral metrics more than NPS because NPS
is typically measured with a terrible question like, "How likely would you
be to promote ..." while referral metrics are saying how much users are
*actually* promoting.

-L

* Though apparently I *am* passionate enough to type this out on the
Saturday after US Thanksgiving.

On Fri, Nov 25, 2016 at 6:31 AM, Jeremie Patonnier <jer...@mozilla.com>
wrote:
> _______________________________________________
> mdn mailing list
> m...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/mdn
>
0 new messages