Tuesday, February 1, 2011

Seeing the future - with charts!

Here is where I pretend to actually know what I am doing and look like a real businessman.

So you might have noticed that I announced a new game a few weeks ago.  It's called Gemini Rue and it's a neo-noir, sci-fi adventure game set in a dystopian future with lots of rain and gangsters and moody jazz music.  There's more to it than that, but you don't come to my blog for my marketing spiels.

(Or DO you?!  Hey, this *is* my blog, so I will just say that if you pre-order the game you'll get a nicely printed disc edition with DVD-style case and everything  It won't be available after the game launches, so be sure and nab it before February 24th.  There's a reason why I'm doing this, but that's a subject for another blog.)


This is the first game that I've thrown a lot of PR muscle behind, and it seems to be really working. As I wrote in an earlier post, I hired someone to do that for me because, well, I suck at it.  As a result, lots more gaming press than usual has picked up on the story.  So it's been quite exciting.  However, if I've learned anything it's that great press doesn't always translate into great sales, so I decided to take a look at my previous games and see how well Gemini Rue is measuring up.

Lots of indie devs are stat hounds, keeping track of every spike and valley of traffic.  I do have software in place that does that, but I never really studied it until now.  I decided to take some numbers and stack them up in certain ways and see if I could come to any interesting conclusions.  What I'm about to show you seems very logical and straightforward to me, although I'm sure you accounting types will probably laugh.

The bad news is that I no longer have the original sales information for Shivah and Blackwell Legacy.  I sold those games through a small service back in 2006 which went bust, taking all my sales information with it.  I now use a mainstream service called Plimus which everybody uses and doesn't seem to be going anywhere.  So I had all the information for Unbound, Convergence and Puzzle Bots.  I would have preferred to take data from all my games, but it was good enough.

Anyway, what I wanted to do was track how well each game did at specific points in the sales cycle.  Each game had a pre-order period before they launched, so checking the stats for both pre-order and launch seemed like good places to start.  Some games have been out much longer than others, so I figured the first month of sales for each was a good sample to take.

I divided up the sales cycles up like this:

1 day of pre-order
1 week of pre-order
Day before release
1 day after launch
1 week after launch
1 month after launch

I plugged in the numbers for the three games and I came up with this graph.  I am not comfortable with giving exact sales numbers so I removed those, but I see no problem with showing how the games sold when compared to each other:

Looking at this graph was an interesting trip down memory lane.  Unbound undoubtedly did the worst in its first month, but I was very green at the time with only one other game to my credit.  I barely did any marketing or PR at all and that obviously shows in this graph.  Puzzle Bots had very strong pre-order sales, but tapered off quickly after launch day.  Convergence is the most interesting, as it had very weak pre-order sales but then shot up like a rocket once it launched. 

What might account for Puzzle Bots' strong pre-order sales is that it had a very strong pre-order incentive - a limited-edition CD-ROM that you could only buy during the pre-order period.  Convergence didn't have that.  Convergence didn't have any pre-order incentives at all, which might account for the pre-order sales being kinda weak.

Anyway, I had three games worth of data over various points so I averaged them all out.  Then I calculated the average percentage increase of sales from one point to the next.   So based on the "Day 1" and "Week 1" pre-order sales figures of Gemini Rue, I was able to plug them into the graph and come up with this projection:

So, hey!  Check out that purple line.  Not bad. Using the sales figures of my previous games, I was able to create a reasonable projection of how I think Gemini Rue will perform, and it's looking pretty good. Of course, this is all theoretical and could end up being completely wrong, but it's a nice barometer.  Once the game has been out for a month, I'll be able to say for sure whether it sold above or below expectations.