One of the blind assumptions I’ve always made about Divinity: Original Sin is that many women play it. I never had any doubts about this because I saw my partner play D:OS for many hours. She’s quite picky about her games and likes RPGs, so I figured that if the game would be unappealing to her (or womanhood in general for that matter), she would’ve told me right away.
My feeling that many women play D:OS is further reinforced by the people I meet at trade-shows. For every two or three male fans, there’s always at least one woman. Based on that, my lazy mind concluded that at least 25% to 33% of our audience is female, if not more (given that trade-shows are not exactly gender balanced).
I never really questioned this number because it fit with the split I saw for events like PAX. According to this report for instance, 35% of PAX visitors is female, and I think of PAX players as players who would typically enjoy D:OS.
But then this morning, I saw a number that didn’t compute at all. According to the Google analytics page for our Kickstarter campaign, only 4,19% of the page’s visitors is female versus 95.81% male! WTF?
That didn’t match at all what I thought. My instant reaction was – “oh really, so there’s not a lot of women on Kickstarter. Interesting, I always thought ….”
At which point I did the sensible thing and googled “kickstarter gender distribution”.
The first result I found showed that indeed, Kickstarter is more popular among men (70% male vs 30% female) but that still didn’t really explain the 5% I was seeing. So I reasoned that perhaps there are other Kickstarter categories that are more popular among women, and that games just aren’t that attractive to them, at least as something to crowdfund.
So I went to our Facebook page.
Turns out only 9% of the people following us there are women! I went WTF again! That really didn’t match my expectations. Where were my at least 25% to 35% women? What about my scientifically sound and completely unbiased sampling results from those fan encounters at trade shows? Could it really be so low?
So I delved into the Steam analytics.
Sadly, there I couldn’t see any demographic data. This seriously sucked, because it could’ve validated (or invalidated) an idea that had formed on our group chat i.e. The women who play our games will play the hell out of them, but they will only take an active interest when the games are released or about to be released. Our male audience on the contrary might be more interested in backing us sooner, for what I guess will be a variety of reasons.
This at least fit with my own observations. My sampling subject at home certainly won’t bother reading a lot of gaming previews or watch you tubers, at least, I think she doesn’t. But she will buy a game that comes highly recommended and looks like she’ll enjoy it.
After some more discussion, our hive-mind came to the conclusion that even if we’d know the gender distribution on Steam, we wouldn’t have been able to make any conclusions, because many couples share their Steam accounts (at least, we assume they do). And so any conclusion based on Steam stats would be wrong.
At which point someone else said we shouldn’t based important decisions on data as shaky as Facebook or Google analytics.
So I started thinking: I’ve never cared for analytics before, so why do I care now? The most I’ve used it for in the past was to see if the PR people we hired in different territories were doing their jobs. To give you an idea of how much I use it, I actually had to ask someone to tell me how to access my gmail account before I could see the current data because I was curious about how people discovered our kickstarter page.
At which point I reassured the team that we weren’t going to let some fancy numbers affect any decisions. We’ll happily continue making games we care about and would like to play ourselves. Which reassured them because they were starting to worry after I didn’t stop asking questions about this
But of course, that didn’t mean I wasn’t curious and actually also a little bit worried.
Because if we indeed have as many women players as I think we do, does that mean we are communicating on the wrong channels about our games? Obviously we’d like them to know about what we’re making. So how do we talk to them? According to this report, we should at least be seeing a 50/50 split, not a 5/95 or 10/90.
So, since quite a few among you work on the dark side, I figured I might as well just ask you. You boys & girls must have data about this. I’m particularly interested in learning if other CRPGs have the same pre-release gender split like the one we are seeing on Facebook & Kickstarter, and if that then settles out into a 50/50 when the game goes on sale. Or is the analytics thing just broken, and am I making a big fuss over nothing? Or, are we really doing something wrong? And if so, what?
All input more than welcome!
So based on the feedback I’ve been receiving, here’s a couple of insights that may help explain things. First off, Google Analytics itself – Google’s adwords page explains how they determine gender. It’s quite primitive apparently.
“Sarah’s favorite hobby is gardening. Many of the gardening sites and blogs on the Display Network that she visits have a majority of female readers. Based on this, Sarah’s browser could be added to the “female” demographic category.
As a result, Google may show Sarah ads from advertisers who have chosen to show their ads to women.”
Kind of puts things in context.
Next up, most of the reactions I got from women really do indicate they won’t engage in the communities around games for a variety of reasons. This brings up interesting questions about how they learn about what games they’ll like. My guess would be that critic scores might be quite important here.
Another type of comments puts the fault in our shoes.
We had a few commenters bringing up the “nostalgia” factor that applies to CRPGs. This factor does not appeal as much to women, as there were less women gaming at the original peak of CRPGs in the 1980s and 1990s, and so us talking about a old-school RPG or reviewers calling it a modern Baldur’s Gate didn’t help.
Some commenters also suggested that we didn’t reach media regularly consumed by women and suggested we highlight female Larian game dev staff to do that.
And then there’s all this feedback that really has nothing to do with the subject and which reassures me the world will remain an interesting place for a long time.
If anything else pops up, I’ll add it here.