Midnight on a Tuesday may not be the ideal time to tackle this but I can always take a nap with the kids tomorrow I guess. I really want to get these thoughts down while they’re still fresh.
This will be the first post of many (well a few anyways!) talking about the way we’re building our first team here at Finji. We have never done this before, and I’m hoping that documenting it will make me more analytical about what we’re doing and maybe help others too.
Finji’s first big internal project is a game called Overland. We haven’t shown this game publicly much at all, so there’s probably not much point in me going on at length about it here, there will be plenty of that in the future. Suffice it to say for now that it is a Unity game with 3D characters and environments. Based on some previous experience we are pretty hopeful that we’ll be able to build this game in the way that we want with a team of roughly 3-5 people in various capacities. Of course, Finji is only 2 people, so we needed to find the other 3.
Even a year ago, my approach to build this team would be very different. In fact, it would look a lot like the way we found our programmer and brilliant co-designer Shay Pierce; I’d ask a friend, someone I trusted and knew already, if they wanted to make a game together. This is in almost every regard a relatively good approach to building a team, as long as you’re not just blind-hiring every friend you have or whatever (although I’m sure there is some project and some team out there for whom that approach could actually work somehow).
The reason I say “almost” is the pool of friends I have to hire from are… a lot like me: white middle-class cis-bros in their late 20s/early 30s. That sounds dismissive toward the geniuses I’ve been collaborating with for the last 5-6 years, which is definitely not my intent, but at the same time I have no choice but to recognize the… homogeneity of the teams I’ve built in the past.
This concern is motivated primarily by two things: the first is a purely practical concern for the final output quality of the game, and its ability to stand out from the pack. I have this theory that (call me crazy) a homogenous team with homogenous interests is going to produce a game that can’t help but reflect many if not all of the same inspirations as every other Steam game, even if it expresses itself in a different way. To put it another way, I feel like I have to acknowledge at this point in my life that maybe I’m not… maybe I’m not a brave enough artist to stand out, period. A good team of people with a broader base of inspirations should, in theory, lead to a more interesting game.
At the same time there is obviously a kind of moral or community motivation to show that even an idiot running a company out of his spare bedroom can build a virtuoso team that does NOT look and act exactly like every other game team out there. I think that has the potential to send a powerful message, and it’s something I would love to be involved in.
The thing is, it’s actually pretty hard to hire a diverse team, but not for the reasons everyone thinks.
It’s pretty chilling, but the main issue we ran into at first is that women simply wouldn’t apply for the jobs we were posting. I’ve seen people use this excuse to try to justify the ongoing imbalance in hiring in the games industry at large, or to excuse gross imbalances in speaker diversity at games conferences, etc. I have since learned this is… misguided, at best, and deliberately misleading at worst.
The main reason, as far as I could tell, that women weren’t applying for the jobs we posted, was because they assumed they wouldn’t even be considered for the position.
I’m having trouble putting into words exactly how much this disturbs me. Maybe I’ll be able to articulate my concerns here better in the future. Fortunately, this was not a significant obstacle for us. With help from the folks at Bento Miso I tried using less combative and intimidating language in my job posts, tried being more informative, and stressed the collaborative element of the work we have planned. I also emphasized the fact that we were specifically interested in building a diverse team and that we were encouraging POC and women and other under-represented groups to apply. Eventually we were able to convince people that normally would not have applied (for reasons too numerous to enumerate here) to go ahead and send in their info.
And I’ve been enormously happy with our results.
Others have said this with much more eloquence but I think it should go without saying that the idea that building a diverse team means sacrificing quality along any axis is offensively absurd, and absurdly offensive. While we have seen under-represented applicants generally have less experience than over-represented applicants, that’s an obvious systemic side effect of discrimination, and not a showstopper for us.
Something that also only became clear when we started the audio tests, is that hiring by portfolio is, perversely, a biased and unfair practice. This is the main reason we actually did the audio tests. Portfolios are primarily indicative of experience, and experience is something that is easier for some people to earn than others. Tests and demos are one way to help level that playing field, when done correctly.
Our hiring plan is a multi-step process that looks something like this currently:
1 - prepare a project-specific brief explaining what we’re looking for. This brief should be very carefully considered and constructed, to ensure that we are not unnecessarily intimidating artists
2 - solicit applications by publicly posting the brief to our company blog*
3 - work through the applications and select a subset of favorite applicants and invite them to do a paid practical demo or test
4 - based on the application, the applicant’s work experience (quality, not quantity), and the quality of the demo or test, select our most promising candidate**
5 - hire the candidate for a one month contract, in which they will work full-time on the project. if after one month it just isn’t working, both parties will be free to go their own ways, little harm done.
6 - if that one month trial period goes well, formally make the new hire part of the project team for the duration of the project.
There is one big problem (*) and one big caveat (**) with the list above. The big problem is with #2, posting the brief to our blog and twitter accounts. This is probably the greatest flaw in the way we’ve been approaching this so far, and is a difficult problem to solve. Basically, our efforts to reach out to under-represented groups and artists have been fundamentally passive in nature. We post to the blog, let the emails roll in, pick our favs. This is not an effective way to reach a wider community of artists. We need to find a way to do this more actively in the future.
The big caveat above, in #4, is that we are not just looking for the most promising candidate in general. We are looking for the most promising candidate for this specific project. Which of the applicants seemed to have the strongest connection to the material? Who created the work that resonated the most? It may not be the applicant with the most polished work or the most impressive portfolio.
THE STORY SO FAR
We’ve done one round of paid art tests, and one round of paid audio tests. 5 visual artists participated (3 women, 2 men) and 12 audio designers participated (4 women, 8 men), representing some 6 or 7 different countries on 4 continents. I think this was a good first step but I think we have a lot of steps to go still, especially when it comes to actively reaching out to outside collaborators.
We will be announcing the Overland team in our own special, Overland-y way, and I’m really looking forward to that. I’m also hopeful that we’ll be able to do some interviews with and publicly recognize some of the people that we ended up not bringing on for Overland specifically, but who made a huge impression on us during this process.
Ok, it’s getting late, time to wrap this up. I would love to know what you think of our process - it’s a work in progress, obviously, and we do intend to do it better next time. That said, we have the luxury of having pretty thick skin, so if you see ways for us to improve please do not hesitate to let us know, either on twitter or via email.