What happened in June of 2009 within the Fedora Project?

[Robyn Bergeron]

On Wed, Oct 14, 2009 at 8:12 AM, Larry Cafiero <larry.cafiero at gmail.com> wrote:
> Hey, all --
>
> I'm not sure I'm following this proposal correctly, so I may need a little
> help. So maybe a little clarification might help, please.
>
> On Wed, Oct 14, 2009 at 5:22 AM, Gregory Zysk <gz.int.project at gmail.com>
> wrote:
>>
>> Dear Marketing,
>>
>> One of the first things I see in order to help in marketing is the
>> establishment of measurement systems. Measurement systems allow us
>> internally to gauge how we are doing, and what needs to be improved. This is
>> true also for the those viewing the community from an external standpoint.
>> These measurements will provide us with more legitimacy and provide a
>> platform where we are transparent about our results and that our results are
>> measurable (and not some abstract way that no one can understand). I can
>> see https://fedoraproject.org/wiki/Marketing_research that we have begun to
>> start an formulate research that will be used to produce these statistics,
>> but fail to see any methodologies that are used to base these questionnaires
>> off of.
>

I'll have a more extensive response later this morning (like after I
replace my headlight on my car, ugh!) - but I just wanted to interject
here before we go running off on a tangent:

As the person heading up the market research "stuff," I would be
delighted to have more input / help in the process. I'm not quite sure
what you mean by methodologies, or if you mean something more like
goals, long-term strategy, etc.; at this point, given the lack of any
sort of user research being done in the past several years, the first
swag at doing surveys is to get a general baseline of who are users
ARE, so that we know who they are not, for the purposes of generally
catering to Fedora customers better.  Or to put it in marketing-speak:
determine who our market is, where our market could be, with the end
goal of figuring out how we get from point A to point B. The fact is,
we can speculate a lot about who Fedora users are, what types of user
groups they fit into (IT people? college students? predominantly
people not in the US?) - but without asking who they are, and getting
a general feel for that, and what their general satisfaction levels
are, we really can't start to ask more pointed questions.  In fact, it
becomes difficult to even set goals, or define strategies; we
certainly don't want to put ourselves in the position of setting goals
that are either impossible to obtain, or essentially already achieved.

And WRT goals / strategies, there has been a thread ongoing on the
fedora board mailing list discussing this very subject; Mel sent out a
mail last Friday referencing this.
https://www.redhat.com/archives/fedora-marketing-list/2009-October/msg00035.html

As an FYI, there is more coming on the market research front, which
I'll be sending out hopefully today; I've been saddled with being
sick, school stuff, and a lost cat :( so things haven't been moving
the way I'd like them to.  However, this discussion has been filled
with lots of goodness that I'll be taking into consideration, and of
course I'd like feedback when I send things out for people to review.

> OK, that's fine. I'm with you so far, until we get to here:
>
>> One thing I would like to start with to help all of you form a marketing
>> mindset is to ask the question of "What happened in June of 2009 within the
>> Fedora Project?
>>
>> As you can see: https://fedorahosted.org/fama/wiki/AmbassadorMetrics views
>> that we have had a steady increase since measurement began in January of
>> 2006. That is until June of 2009.
>>
>> Once we can answer this question, we can begin to answer these
>> sub-questions:
>>
>> 1) Who were these ambassadors?
>> 2) What specific contributor groups were they apart of?
>> 3) Where did they go after they left the ambassador group?
>
> Joerg can probably help out here -- please do, Joerg -- but wasn't June 2009
> when we implemented the mentor program for Ambassadors? Before this program,
> basically, the only requirement you needed to be a Fedora Ambassador was
> that you had an e-mail address, a regular pulse and steady breathing
> (although the breathing part didn't have to be steady, as long as you were
> breathing). Now there's a more detailed process to follow, which a.) scares
> off those who are not committed to Fedora and only want "free stuff," and
> b.) allows us a to cultivate a better quality of Ambassador.
>
> As Joerg may have mentioned also, I believe around this time the Ambassadors
> list was purged of non-participants. Messages were sent out and those who
> did not respond (or responded that they were no longer interested) were
> taken off.
>
> That could explain largely why numbers "dropped," providing a false negative
> when you look at the numbers without applying the changes.
>
> I could be wrong about the timing of the mentor project's initiation, but I
> would bet that's why numbers dropped.
>
>> Please feels free to give me your comments and suggestions regarding this
>> issue, so we can start to problem solve some issues to help us provide
>> better and more improved results, as we do technically with every release.
>
> Unfortunately, I don't have as much time to participate in the marketing
> group as I would like, but I'd be more interested in analyzing external
> developments, like why did record numbers of people download Fedora 11 and
> what are their experiences (good/bad/indifferent) and build a marketing
> scenario around that, rather than use the time and effort to look internally
> at how many people participate in Fedora and why.

This is precisely the kind of information I'd like to be getting from
doing user surveys in the future.  Figure out what we are doing right,
and do more of it; figure out what is going wrong, and fix it.

As for internal surveying, I think it would be useful to occasionally
gauge community members on their satisfaction levels, etc., but I
think this unfortunately becomes an exercise in scratching our own
backs; if they're here, they're probably happy, if they're not, they
certainly have the freedom to voice their opinions and
dissatisfactions (and likely are doing so), and if they are leaving,
there is probably a very low percentage that they will come back.  The
dissatisfaction level, I would suspect, would probably be low enough
to not warrant any huge changes to "the system"; it would be more
something that would need to be tackled on a one-on-one basis, which
is not something you want to do when you're surveying people (ie: I'm
going to get your name, and then I'm going to contact you directly -
this tends to turn survey participants off quite a bit).  Individual
group leaders could take on the task of figuring out why their people
are leaving, particularly if it seems to be systemic, but this may be
a lot to ask, but could certainly be valuable in preventing the issues
that caused community members to leave in the first place.

As for general statistics - I think it is certainly interesting, and
definitely something that could be automated/scripted, but we need to
be careful to not bombard users with endless requests for how they are
feeling.  Once a quarter, twice a year, focused on specific topics is
the way to go, IMO.  Looking at things like downloads, what countries
they are coming from, what architectures people are using can
certainly be valuable (if we have a crapton of downloads originating
in the Bahamas, and no ambassador there, then we might want to start
having the "who wants to move to the islands!!! contest) to ensure
that we are focusing resources in the right places.

>
> Maybe I misunderstand your proposal, Gregory, but you asked for comments, so
> here are mine.
>
> Larry Cafiero
> Regional Ambassador, U.S. West Coast states
> Fedora Project
>
>
> --
> Fedora-marketing-list mailing list
> Fedora-marketing-list at redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-marketing-list
>