Ask HN: How long should we run private beta before opening it up to all?

7 points by JonathanWCurd ↗ HN
We are quickly approaching the point at which our MVP is ready for beta testing by the outside world. How long should we run a private beta (if at all) before opening up to all?

7 comments

[ 4.7 ms ] story [ 33.7 ms ] thread
You should run it for as long as is needed to test the effectiveness of your app. This period is one where code changes can be made to make your service better, decide pricing, test the market. e.t.c
Run it until you know there are no major snafus. I.e., system doesn't auto-purge your DB, assimilate all your users and nuke the planet.

Once things are stable, start scaling the beta up to a reasonable number of users. I.e., stress test, in increments. Once you meet a pre-set milestone, set it loose.

Goodluck!

Should we accept users first come first serve or filter them based on a simple survey and who we think would provide the best use cases.
It really depends. The survey is an interesting method of filtering people in of itself, to assure they'll use the service AND it has the added bonus of giving you some free market research. However, you don't want to be too intrusive at the risk of pushing people away.

Frankly, I think you'd be more than fine if you let it be on a first come, first served basis, with the added caveat of garnering a little market research. That way you're not turning away too many people.

You run the risk of taking on a lot of work if you start filtering everyone. And if you do that, you may miss out on a potential market you hadn't identified previously. Just a thought :)

One strategy would be to initially continue to offer it as a "private beta", while allowing X invites to each existing user to expand the beta userbase.