What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

Community size

Not all communities/target-audiences are of the same size. There are a lot more programmers than quantitative finance specialists for example.

That does not necessarily mean that the StackExchange format is unsuitable for those communities, but it will result in less questions per day, less visitors, etc.

Metrics: Quantity vs. Quality

The way success is measured for beta sites has a significant bias towards quantity (daily visits, questions per day) instead of quality (% answered, answers per question). For specialized communities it might be impossible to reach the desired quantities even with great dedication from every potential member. Even without quantity, a stackexchange site may still become the authoritative source for such communities because of the superior format.

Fixed Costs

I understand that there is a certain fixed cost to create a new site. Graphical design would be a part of that cost for example. These fixed costs would require a certain critical mass. However, it is difficult to imagine that a topic such as 'home improvement' should be struggling to meet those quantity requirements, while quality seems excellent.

Discussion

I'd like to see the success-metrics of a beta-site become more quality-oriented. Is this desirable, and how can this be achieved? Please discuss.

NB: For context, see this question.

share|improve this question
    
Got to pay the bills... –  C. Ross Feb 11 '11 at 14:54

3 Answers 3

Well, ideally it is based on both.

Honestly from day one very, very few of our sites have a quality problem -- they tend to produce, in my not-so-humble opinion, amazing quality as a general rule. And those that do not, we shut down:

http://blog.stackoverflow.com/2010/12/no-artificial-intelligence-in-area-51/

Therefore, if you assume that we don't tolerate extremely low quality, and assume quality standards, then the only metric left is quantity -- size of community.

There are certain things that you simply don't get in a community until it reaches a tipping point, and that takes time and reasonable growth. It's not so much "you must be this big" but "you must show signs of growth, otherwise you stagnate and the community gets bored and eventually moves to some other more happening city."

share|improve this answer

I am an active user of the WordPress Stack Exchange, and it frustrated me too that we were "stuck" in beta for such a long time, while other sites were "passing" us. We were doing our best to provide quality answers, and we seemed to have happy users, but not so many upvotes (or at least not spread out enough to get 150 users with +200 rep).

But then I realized: does it matter? Does it matter that your site stays in beta for a longer time? We can operate as a full site: we have the same domain name as we will have after graduation, we can get arbitrary migrations of questions to our site (they must be done by a moderator, but we wouldn't make the SO shortlist anyway - unless that system is changed). The only difference is that you get a new design, and the reputation requirements go up for a lot of actions.

And that final issue is the most important one. You need a large group of +200 users to handle the basic moderation of the site. You can't rely on the moderators alone to keep the site clean. Robert indicated multiple times that the number of +200 users is the most important metric when deciding which site to launch next.

But it is not the only factor. We still only have 117 "avid" users, but it seems there was a huge traffic increase, which means we are currently preparing our design for graduation.

share|improve this answer

There is a lot less a "size" component to those metrics than you might suspect. The numbers you see listed in Area51 (questions/day, # avid users, visits/day) are a rough estimate of how much a site has to be used to remain interesting. They are fairly modest requirements. So what you are seeing is that "popular" subjects zoom past those numbers relatively quickly while niche-y subject simply take more time.

But in no way are those numbers used to indicate that a smaller communities are unsuitable for the Stack Exchange format.

Consider the #1 most important requirement for a sites survival: The Answer Rate. As long as a site provides a good end-user experience (i.e. questions get high-quality answers consistently), the site is allowed to continue.

A site graduates into a fully-fledged site when it has pretty much guaranteed its continued success. But there's nothing inherently troubling if a site needs much longer than those 90 days to accumulate more "excellent" ratings. That "90 days" number is an arbitrary minimum to assure that wildly-popular sites are not simply one-hit wonders that take off in popularity as a fluke.

The tiniest site can be completely awesome providing the highest quality answers to important questions. But if there are only a small handful of users providing those answers, the site is vulnerable to losing those users and failing. So we say "hang out in beta awhile until you've had the time to accumulate more questions and answers, which in turn will attract more users."

When the site has a bit more momentum, that will provide for a better graduation launch to a much stronger site. There's no rush and nothing unusual or troubling about waiting a bit longer until you can fill all the seat for an awesome launch.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .