This post isn't about how we can make the site be the best it can be. This post is about how to avoid the site being closed by the moderators like what happened to many private beta sites I have seen in the past.

Here is a link to the current stats that we may be marked on: http://area51.stackexchange.com/proposals/36296/earth-science.

As you can see, we are marked on both the number of questions asked and the number of answers provided, as well as the number of high reputation users.

I think we should regularly check where we stand on these statistics.

Currently the number of questions asked is very good, but our answer rate is only at 50% but we should be aiming for 90%. Therefore we should aim to answer a question for every question we ask if possible.

Finally the last private beta site I was a part of was shut down because there was not enough high reputation users. Therefore I also encourage you to not be conservative with your up votes.

  • 2
    $\begingroup$ Survival is the key. "To finish first, you must first finish." If a site doesn't reach "critical mass" because there isn't enough participation, it doesn't matter how good the (few) questions and answers were. $\endgroup$ – Tom Au Apr 20 '14 at 0:07

No, please stop. You are not being "marked" on "the number of questions asked and the number of answers provided." Rallying around statistics and urging folks to hit the numbers is not the way to build a healthy site.

Your job right now is to ask the interesting and intriguing questions you encounter in your every day work. Period. The Community Team takes a much more qualitative look to determine determine if you have a healthy, vibrant community that is going to create great content for a long time to come. But force-feeding dozens of questions in the opening hours or asking easier questions with the express purpose to "maintain a high answer rate" does not look like healthy participation to us.

This site has been live less than 24 hours and you're already quoting which stats are "very good" and which need to be worked on. Before you send everyone off to check their report card every day, please at least read this:

(I can't decide which post you need more, so it's best to be familiar with both.)

Your New Site: Asking the First Questions

Does this site have a chance of succeeding?

My fear is that hovering over that stats just to "hit the numbers" is one of the biggest failure modes that leads to poor performance and behaviors that makes this site look somewhat forced and uninteresting.

Yes, voting early and often to reward high-quality content is good. Remembering this site when you have interesting questions is good. Checking in regularly to see if anyone need help: good. But please enjoy the site and rally around the enthusiasm and prowess that got you this far. But obsessing over those numbers is not part of our process; it shouldn't be part of yours either.


  • 1
    $\begingroup$ To add this: it's better if a good question has no answer, than if a good question as a bad accepted answer. $\endgroup$ – gerrit Apr 16 '14 at 18:00
  • $\begingroup$ Truthfully, if we are asking questions which are so easily answered among the limited audience of a private beta, the questions themselves must not be very challenging; or more likely, they have simply been asked 100s of times before on every other site on the subject. In either case, that's not a terribly interesting site. Stack Exchange excels at the long-tail of intriguing questions that can only be answered by a few people. This may not be entirely 100% achievable in a private beta, but you should be thinking along those lines. You'll get there. $\endgroup$ – Robert Cartaino Apr 16 '14 at 19:16

[EDIT]I (not completely, but mostly) agree with Robert Cartaino - while stats and quantity are good too, quality is more needed now. So I editted the post to emphasiye this.[/edit]

One thing to have in mind: statistics don't tell us much on the first day of private beta.

Questions per day: at the first two days, everyone is spamming the questions they had in mind when the site was in commitment phase. Having less than 50 questions on day one means that the site would probably never graduate, because it starts too slowly. This is an average for two weeks, so now the number is very high (74/1 is much better than few hundred/14), but the average at day 15 will tell the truth. I hope we will have 50 questions a day someday again, but it will happen long after graduation to full site.
On the other hand, quality matters too - 50 good questions is much more than 100 bad questions.

Answered percentage and answer ratio: this should be low during private beta. One of the purposes of private beta is to gather some momentum for the day it ends. If all questions had accepted answer at that moment, how would new experts willing to help join? If it looked like this in the end of april, it would be bad.

Number of avid users: you can't have really high rep users now - the 200 rep (except for association bonuses, acceptations and bounties) maximum wouldn't allow this. But ten "avid users" now is good, and more than half of the 206 users here is not bad.

Visits/day - we can't expect too much visits during private beta. This stat will reach its local maximum at the start of public beta, and then it will decrease and it will take hard work to merit its increase. It is a big holiday when the beta sites makes a new record in visits per day!

Just to compare, here are stats for sustainability.SE, a site that started more than a year ago, and now is not close to graduation, but not endangered by closing. I think we have slightly faster start than Sustainability (@EnergyNumbers might confirm or refute this, he has been much more active than me there since the very start of that site).

So what to do?

  • If you can ask more good questions, ask them.

  • We don't need to ask lame questions just to cheat the stats. We need to balance between quality and quantity. I remember fiasco of Machine Learning - they wanted to be really professional from the very start, but high percentage of the users (including me) wasn't able to meet the requirements for "quality" Q/A, so the site didn't survive private beta. On the other hand, Astronomy's private beta was prolonged due to low level. Moderate level is the key to good start; we can increase our level to "high" later. As I see it now, we are below the "moderate" level - so no need for overkills, but we should try harder.

  • The same goes for answers - don't be extra strict, but don't spam clutter. The risk of clutter is slightly more dangerous now.

  • The first two days we should produce a lot of questions, then ask more questions, but focus on good answers. No answer is better than bunch of bad answers on the release of public beta.

  • on day 8, the site should have some community, some style, and some 100-200 questions (the better questions, the less is enough); >50% should be answered with answers showing our style.

  • again, private beta is a time to find basic ideas. The earlier we find a style that suits the community, the better. At least if our style wouldn't be totally flawed (like "everyone go rep-whoring!").

  • $\begingroup$ The first two days we should produce a lot of questions, then ask more questions, but focus on good answers. No answer is better than bunch of bad answers on the release of public beta. Robert said himself that we should not focus on quantity, but quality. We don't want to manufacture hospital food in large batches, we want a little culinary perfection to share with the world and demonstrate to new users how they should answer and ask. $\endgroup$ – Anonymous Penguin Apr 16 '14 at 21:34

The biggest problem I have seen to our stats is that many people who are asking questions are not answering questions. That isn't a problem when the website is established, but at this point it seems that there are a small number of people spamming questions faster than the people capable of answering said questions can answer. The quality of questions has already been brought up in another meta question.

  • $\begingroup$ Hi! That's quite typical for very new sites. You get many people who are really not aware of anything in the field. They're here just to ask couple questions that have bothered them since ever. (I'm such person too, despite I answered one question, but that's quite an exception.) It is fine, certainly in the first couple days, and it will very likely settle down soon to some reasonable answer rate. $\endgroup$ – yo' Apr 16 '14 at 15:43
  • $\begingroup$ Right, but a user I know can answer some questions, @InquilineKea, who have asked many questions. Not to call him out, but it would be nice that people who have studied this earth science formally both ask questions and answer others. $\endgroup$ – Neo Apr 16 '14 at 16:02
  • $\begingroup$ The @ doesn't work this way, so I.K. didn't see your message probably. Maybe you can approach him directly off-site and suggest him answering more questions? :) $\endgroup$ – yo' Apr 16 '14 at 19:57

You must log in to answer this question.

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