As you might be able to tell by the adverts plastered all over the site, the Particulars alpha had it’s soft-launch yesterday (PS. [s2If !current_user_can(access_s2member_ccap_particulars)]you should totally buy it![/s2If][s2If current_user_can(access_s2member_ccap_particulars)]thanks for buying it!… is it creepy that I can tell that you’re logged in?[/s2If]). You might not know that we had a softer launch on Monday. Or why we had a soft-launch at all.
After this experience, I’m pretty convinced that soft-launches are the best thing ever.
Why, you might ask? Find out below in a 4-act morality play*.
*replace ‘morality’ with ‘practicality’ and ‘play’ with ‘blog post’. Yep, that’s probably more accurate.
Act 1: The Fallen Angle
The launch of Flatland: Fallen Angle.
The important thing about this failure is that due to a lack of experience and a lack of time, it was pretty likely that we were going to get the wrong payment provider. And having the wrong payment provider, it would only be after selling the game a few times that the problems would actually arise.
Act 2: The Stopped Bot
The launch of Unstoppabot.
Two things went wrong with our Unstoppabot launch. First, we accidentally launched the game two weeks early, obliterating any chance of getting noticed for being ‘new’. Second, we seriously screwed up the mail-out of our press release (to the point where we’re not even sure how many people got it).
Before we launched, we’d heard that studios sometimes do a launch in a smaller country (Australia or Canada usually) before going worldwide. We dismissed it, as we wanted to get the game out. While taking this strategy might not have fixed the first problem, it certainly would have highlighted our press problems before taking the game to the world stage.
Act 3: A Particular Problem
“But wait”, you might cry. “You only launched on the 10th!”. Ahh, but that’s where you’re wrong. We did a ‘super-secret early alpha’ launch 2 days earlier, by sending a special link to our mailing list and those who had filled out the Particulars survey, and giving them an even higher discount.
It turns out that this was probably a good idea. It turns out that there was a pretty big problem with our purchase system.
If someone had bought a game (ie. Flatland) from us before, they already had an account. If they didn’t login before purchasing, they wouldn’t be able to make a new account with the same email, and by the time they logged in, the purchase information would be gone. We could fix this for individual people, but I was sure glad that I only had to do 8 of these, rather than having to deal with all 100 or so people we already had on our system.
By the way, the solution to this was to put messages everywhere telling people to login. A tech solution is much harder (and possibly impossible), and I can deal with the 1 or 2 people who slip through the net.
The super-secret early launch also told us a plethora of details for our website design that made it hard for people to get to their purchased game. We also realised that unless we made it clear (and in bold, and in teal), people wouldn’t know what platforms the game was on. While this sort of thing would have survived the soft launch that we’re in now, I’d shudder to think what would happen if we waited till hard launch to try all these systems out.
Act 4: The Soft Launch
The SeeThrough Offices. Right now.
I think what I’ve learnt from all these experiences is that new things usually break. It often takes a couple of goes to get something right, and you simply don’t know what’s going to go wrong until you put something into the wild.
So soft-launch. You can spend forever trying to think through everything that might go wrong, or you could just get some people interacting with your system to find out what’s actually going wrong. You can then fix it, and fix it quickly, while there’s only a few people yelling at you about how broken your game is.
A second, less obvious reason for our soft-launch alpha is that we’ve never successfully run an alpha program before. We have no idea how to keep a user base engaged, and to ensure that they are continually upgrading the game and giving us good feedback. So we want a few weeks and an update or two with a smaller group, to learn a thing or two about keeping that engagement before we throw this game out to the world.
Bonus Act: That ‘rising cost’ thing
This is the final thing I kinda like about soft launches: I get to try crazy things with little risk. I think that by having a countdown of number of sales till a price increase, we can probably accelerate sales. I’m interested in comparing that to a simple timed countdown (which is what we’ve done with the super-secret early alpha – less than 6 hours to go guys!). So far, the timed countdown is winning, but it’s pretty early days for the number of sales model. I guess we’ll find out in the days to come.
I’d love to hear about any launch issues you might have had, whether it be game related or otherwise. Also any feedback on our Particulars launch is very much appreciated!