... And that's what I get for trying to be helpful but forgetting that this particular site is run by people who actually *respond to questions* and do so very quickly and informatively.
(Sorry for the spam. Just had to say it.) ----- Original Message ----- From: "Denise Paolucci" <[email protected]> To: "Jäääärne" <[email protected]> Cc: [email protected] Sent: Tuesday, March 31, 2009 12:02:26 AM GMT -05:00 US/Canada Eastern Subject: Re: [DW Discuss] Invite codes On Mar 30, 2009, at 10:54 PM, Jäääärne wrote: > When the site launches on April 30, account creation will be > available either with an invite code or via buying a paid account. > If I understand the earlier information correctly, invite codes are > going to be issued to existing beta testers. The question is: will > some small amount (say, one for each month of paid time) of invite > codes be available to paid users upon buying an account? > > I would like to create an account on DW and invite a couple of > friends. Purchasing an account seems like the simplest solution. > But then, I cannot suggest that my friends do the same if I'm > actually inviting them. So, it would be neat if buying a paid > account meant getting some invite codes. Would that be possible at > all, or would doing such thing mean losing control over initial > userbase growth? The answer to your question is: Yes and no! (I know, I know, that's helpful.) Mark's already given you the basic answer, so here I come (as usual) to give the tl;dr version ... We haven't yet decided if there'll be any "automatic" invite codes issued (so, like, the "paid accounts get one code per month of paid time" thing that LJ used to have, when LJ had invite codes). We may; we may not. We probably won't. However! We *will* be releasing new invite codes on a regular basis. (We haven't defined 'regular basis' yet. I suspect it will be either once every two weeks, or once every month.) When we go to do that, we'll look at the current site usage statistics: * How many active accounts do we have? * What percentage of those active accounts are paid vs. free? * What does the paid account churn look like? [Churn is defined as: (total lapsed paid accounts that don't renew within 7 days of lapsing / total paid accounts) * 100 -- so, that'll tell us if people are paying the $3 for one month of paid time to create an account and then letting the paid time lapse without renewing, or if they pay to create an account and then decide to keep paying us.] * How many "outstanding" (ie, unused) invite codes exist? * What does our network performance look like? (ie, how 'fast' is the site sending out data) * What does our hardware performance look like? (ie, how overloaded are the machines, both memory and CPU) * What does our current financial state look like? Do we have the cash on hand to add more hardware if we'll need it? Once we look at all that, we'll roll the bones, apply the historical formulas we'll develop for how likely it is that a paid user is going to stay paid, project what kind of income we'll have in the upcoming time period, and use all of that to determine how many more free accounts we can support. Call that number X. X will be the number of invite codes we'll give out for that particular round of invite release. (Remember, we'll be doing this every two weeks or every month.) We can choose to release those X invite codes in a number of ways. (Afuna and Pauamma, who programmed our invite code system, were incredibly thorough.) We can say "distribute these X codes to all paid users", or "distribute these X codes to all people who don't have any invites left", or "distribute these X codes to all people who have invited someone who then went on to pay for their account", or "distribute these X codes to all people who invited someone who then went on to be active", etc. This is because we want to make sure we're rewarding people who use their invite codes "well" -- invite lots of people, invite people who are active on the service, invite people who go on to pay us money, etc. We are going to be insanely paranoid about how quickly we grow; we want to keep the site usable for everybody during our buildout phase. We also know that people are going to want to move their entire social circles over to Dreamwidth, and we *like* that, because historically speaking, a new user who has strong social ties on a service is far more likely to be an active, long-term user than someone who doesn't. We *also* know that if people aren't able to bring their friends with them, they're going to find Dreamwidth less useful and relevant. So, it's going to be a very delicate balance. (I've pondered a bit about offering blocks of invite codes for sale, priced to take into account that we set our pricepoints so that each paid account subsidizes about 20 free accounts, but haven't even gotten far enough to the point of 'let me talk to Mark about this'. It'll all depend on what our numbers look like after launch.) One thing that we *have* done to make sure that people who can't get an invite and can't afford even the $3 for one month of paid time, or can't make a payment to us for whatever option, can still use the service: we've seriously cleaned up a lot of our OpenID support. Anyone with an account on nearly any other LJ clone site can sign into Dreamwidth with their OpenID, and do everything but post entries or join communities: they can maintain a reading list, comment on others' entries, and otherwise do most functions a 'real' account can do. (You can do that on LJ and on most of the other LJ clone sites, but we've done a lot of work to make it way more usable.) --D -- Denise Paolucci [email protected] Dreamwidth Studios: Open Source, open expression, open operations. Coming April 30! _______________________________________________ dw-discuss mailing list [email protected] http://lists.dwscoalition.org/cgi-bin/mailman/listinfo/dw-discuss
_______________________________________________ dw-discuss mailing list [email protected] http://lists.dwscoalition.org/cgi-bin/mailman/listinfo/dw-discuss
