3 ms·

Ok, here's a specific question I've been mulling over for a little while. We're launching our college in about 10 days and have been discussing the best way for

by padwiki 15y ago

Ok, here's a specific question I've been mulling over for a little while. We're launching our college in about 10 days and have been discussing the best way for prospective students to try out the system. We have limited server resources (limited as in we probably shouldn't let in more than 1000 guests at once until we know our true capacity) but want to show the system off to as many people as possible. Questions we need to answer: Should we require any personal information in order to try the system? If so, should we verify that information before signing in. If we don't collect personal information up front, is it worth bugging them when their session is over to be put on a mailing list? How should we handle limiting simultaneous users? Options: 1. Allow unlimited user signups but only allow logon if total active sessions less than our max. 2. Only allow signups if there is space in the system, logon is immediate. 3. No signup at all, just rotate through a fixed number of guest accounts for preview. If full, no entry allowed, but can be put on mailing list when slot opens up. 4. ????? 5. Profit? Our target audience is proficient programmers and our primary system focus is speed, simplicity and usability. Any input is very appreciated.

mikegreenberg 15y ago
Your top priority should be to be available for as many people as possible. If you can get more than 1000 people on your server at once then you need to take advantage of that momentum to build a buzz about your product. Putting artificial roadblocks in the way is fine, but at least create the CORRECT roadblock. Instead of a "Server unavailable" when the user attempts to try things out, maybe have a separate server running a static version of your product which can be served quickly from cache that a user can play around with while they wait for availability. Or find a way to improve the caching on the website so support a higher number of individuals. You could offer them access to the live application if they leave their email address with you. And then you could stagger the invitations for live accounts over a longer period of time without losing any potential leads. Do you collect personal information upfront? I personally think you should let the customer play with as much of the application as possible without forcing them to commit personal information. The fewer barriers there are, the more time the user will have to explore your value proposition. If you have limited server resources, it might be worth it to invest a few bucks for a couple of days on a beefier server just for your launch. You probably won't need it longer than that and you can probably afford an AWS instance for a few days.

by padwiki 15y ago

mikegreenberg 15y ago

Your top priority should be to be available for as many people as possible. If you can get more than 1000 people on your server at once then you need to take advantage of that momentum to build a buzz about your product. Putting artificial roadblocks in the way is fine, but at least create the CORRECT roadblock. Instead of a "Server unavailable" when the user attempts to try things out, maybe have a separate server running a static version of your product which can be served quickly from cache that a user can play around with while they wait for availability. Or find a way to improve the caching on the website so support a higher number of individuals. You could offer them access to the live application if they leave their email address with you. And then you could stagger the invitations for live accounts over a longer period of time without losing any potential leads. Do you collect personal information upfront? I personally think you should let the customer play with as much of the application as possible without forcing them to commit personal information. The fewer barriers there are, the more time the user will have to explore your value proposition. If you have limited server resources, it might be worth it to invest a few bucks for a couple of days on a beefier server just for your launch. You probably won't need it longer than that and you can probably afford an AWS instance for a few days.