I have read a number of posts that give the target number of questions per day for a site to graduate from Beta status as ten. (Another said fifteen but apparently that is out of date.) I do realize that graduation decisions depend on much more than this, but still, I would like to know: when the stats are calculated, do questions posed in the associated Meta site count?
Child meta participation is an important part of getting to the point of being ready to graduate but I'm pretty sure that the questions per day asked there are not part of the ten questions per day metric used to determine whether a site is ready to graduate.
I'm a moderator on a relatively active new beta (under a year at this point). We have about 5-6 questions per day in general. But meta is really hit and miss. Some days we have people posting 2-3 questions per day and others we have only one for an entire week.
There are two factors that will influence a meta question being asked:
- There's something to discuss that needs to be hashed out by users.
- There's someone who sees that this is the case and is willing to take the time out to ask the question.
Meta questions are spurred by need and posed by people who have reached a threshold of wanting to express that need. We only ask them when there's an issue on the site that someone feels needs to be discussed.
Low Meta Participation
As such, there are two possibilities in relation to low meta usage:
- There's not much else that needs discussion.
- There's not anyone interested in starting such a discussion.
The first of the two is fine. If you look at some mature sites, many of them have very little meta discussion because there's not much else to figure out. So, as a site ages, they may actually have very little going on on their meta site - because they know their scope and what's going on. And that's not necessarily a bad thing.
When it becomes a bad thing is when we do have discussions that need to happen but there's no one who will take the time to ask the question (or answer it). This low meta participation by users is a sign of poor engagement and may lead to difficulties finding users willing to maintain the site - which would be very bad for its longevity.
A few sites, recently, have had problems finding users willing to stand up to moderate and I see this as a symptom of this second point. If all we have are people who want to use the site but not who want to care for it, the site has a problem and that needs to be discussed and addressed before a site leaves beta.
High Meta Participation
Following from the above, high meta participation could be caused by a similar two factors:
- There's still a lot of stuff that needs to be discussed.
- There are lots of users willing to participate in meta.
One could argue that a site that still has a lot of meta activity still has a lot to figure out, and may not be a good candidate for graduation for that very reason - but it's really going to depend on what sort of things are actually happening on meta. If there's a lot of meta questions (but few answers or votes) then that indicates that a site needs help. Questions are being asked but no one is helping determine what should come of those questions.
Alternately, if there are a lot of active meta users, they may be doing all sorts of "fun" meta activities that don't really relate to the health of the site (e.g. challenges/contests, announcements, getting-to-know-you posts). Or, they may be working out tagging minutiae or figuring out why a question is open/closed or was reopened/reclosed. These have little to do with whether the site is graduation-ready.
So, this requires really looking at the questions being asked rather than looking at a number. There's no way to know - based solely on a number - whether the activity is indicative of a site not ready to graduate because they still are figuring out their scope, or one that's got its ducks in a row and is comfortable and ready to go.
I discuss a lot of this from the other direction in my feature request to change the "Graduation" process.
So, to recap, I think the above shows a strong argument for why the team wouldn't use meta questions per day as part of the 10 QPD metric required for the removal of the beta label.
Both low and high meta participation can be a sign of a site that's ready for graduation and a sign that it's not. As such, the raw number is a poor indicator of readiness.