Interview: Tuning promotion to make Academia a success

Ryan Sumo talks specifics on methods to break through.

[Some of you may recall a recent Game Discoverability Now! newsletter which discussed Squeaky Wheel’s Steam game sales - $1 million gross - for their studio’s game Academia: School Simulator.

Well, I had some follow-up questions based on the data (which is re-pasted below in part, along with the game trailer!) Squeaky Wheel co-founder Ryan Sumo was kind enough to answer them. I think his answers illuminate his analytical approach to games - one of the reasons I believe his studio is doing well.]

Q: Leading up to and after alpha launch - what was most important to sales: working with streamers, working with the traditional online media (journalism/fansites), or working with your Discord community?

A: So keep in mind that we launched Early Access in 2017, and with the acceleration of change these days this “advice” is probably not super valid.

When we launched the most important thing for us was our YouTube video devlogs. Our Pre Alpha devlog was released into the wild on Feb 13, 2017, and got about a thousand views (now 1908) and our first official devlog got maybe 5k or 6k views on launch (now 8k).  That was with very little fanfare and just sharing it on social media and letting the youtube algo do its thing.  

In contrast, our published game Ruinarch released its first devlog late last year and up to now has clocked only 307 views.  There could be a lot of things behind this. Maybe the name is not good or easily searched.

Maybe people aren’t looking for the kind of game that Ruinarch is.  Or it could be that people’s viewing habits and Youtube’s algo has changed such that it’s simply not driving people to our videos.

Streamers were very important to the launch, much more so than traditional media.  It’s really hard to tell because there’s not a lot of ways to track how people are finding out about the game, but what I can say is we never once experienced a spike of interest from traditional online media.  They mostly give the team a morale boost because they’re what we used to read back in the day.

I don’t think Discord was even a thing back then.  And even now we’re not super keen on Discord. I think that Discord is a great tool if you have someone on your team that is super into it and into engaging with your fans. It’s something that has to be managed. 

But it also feels a little blown out of proportion. It’s entirely possible to be successful without having a Discord server. We mostly engage with our players through social media, email, and the Steam discussion boards, and that’s worked fine for us till now.

Q: I see your Alpha updates has pretty decent sales spikes - did you theme them in some way (like around certain game features) and how did you publicize them to have that happen?

A:We’ve been starting themed updates recently, but the Alpha updates have been focused more around marketing them as a milestone in development.  It’s like saying hey, we’ve improved a lot of stuff since Alpha 3, so try Alpha 4 now!

In terms of publicizing them, I want to keep some “secrets” up my sleeve for now.  But I will say it’s not rocket science, and it’s just using what is available for any dev to do on Steam to tickle the algorithm. That combined with some advertising and outreach to streamers is all we really do.

Q: We saw in your data that you got big sales spikes from a couple of specific streamers, but other uplifts were less obvious. What's your opinion on why certain streamers 'make a difference' and others don't?

A: I would say a lot of this is based on cultural norms and Steam demographics.

So for example we had one Latin American streamer that made a video with 2.3 million views but the sales on that day were only 56. Whereas a UK streamers’s video which had 1.8 million views netted us 787 sales that day.  There’s a lot to unpack here.

First I need to establish that I am painting with a broad brush here and that I am from what is generally understood to be a developing country, the Philippines.

My limited knowledge of the Latin American market is that it is very similar to the Philippine market, in that there is an acceptance of pirated games (side note : while growing up game piracy was so rampant I had a favorite pirate brand that I trusted). In short, people probably pirate a lot of games there, including ours. 

I would also assume that for a variety of reasons Steam has not penetrated that market very well, and so a culture of buying games on Steam is not well established.

Lastly, like the Philippines, the majority of the population is playing F2P mobile games on their phones.  We had a lot of messages from players asking us where they can get the game for free and whether it would be on mobile. 

People also seemed to be very confused about how to play it, with little understanding or knowledge of Steam or the PC platform, generally speaking. I’m learning Spanish right now though so it was nice to try to decipher their messages before giving up and using Google Translate.

[SIMON’S NOTE: I’ve also observed that there are sometimes different types of streamers. I see both ‘novelty streamers’ - where people just enjoy watching them play but aren’t necessarily gamers themselves, vs. ‘core game streamers’ that have a lot of dedicated game players waiting to buy.]

Q: Do you have a way of estimating how a streamer contributes to your bottom line?

A:Yes I do.  Sometime last year I took a look at the peaks generated by streamers and did some spreadsheet-Fu to try to grapple with the numbers. 

A few caveats. I am not a statistician or mathematician so there may be mistakes here. There is no way to know which sales came specifically from streamers vs through other means. But for simplicity I am assuming that the majority of the sales were due to streamer impact.  I don’t want to mention the specific streamers here, since I am sensitive to the fact that this is their livelihood.  

So broadly speaking what I can say is that for the UK streamer, each view was worth $0.008 .  For the Mexican streamer, each view was worth $0.0002. For context, that means for 10k views we would earn $80 from the UK streamer and $2 from the Mexican streamer. This does not take into account the possibility of ripple effects like other streamers noticing the game and then playing it for their audiences, but it’s a good number to meditate on.

Q: Would you ever pay a streamer to cover your game?

A:So this is a complicated question.

Ethically, I used to be a little troubled about paying streamers, because it feels like paying journalists to give good reviews.  I’m a little more flexible now, as I see streamers as entertainers more than critics (although they do take a huge shit on our game sometimes for comedic effect.) I waffle on this a lot, but in general letting them play the game online and letting them earn money that way seems ethically cleaner.

In terms of business, I guess the answer is maybe. Taking the example I gave above, if the same UK streamer contacted us and said “hey love your game, we do sponsored content for $4k a video.  I’d look at the gross revenue we generated before ($15k), make some assumptions for net revenue (say $8k). Paying $4k to make $8k is a pretty good deal, given the possible ripple effects the video may have.

If they asked for $8k, I might still agree because then we wouldn’t be losing money but hoping that the ripple effects are worth it.  If they asked for $10k I’d probably say no, since I’m now losing money, and there’s no way to predict whether they can replicate their past numbers (ie they can’t really promise a set number of views).

Each streamer has a different audience though, and two UK streamers may be speaking to entirely different audiences, one of whom is more likely to buy a game on Steam than the other. 

Tl:dr maybe?

Q: Relatedly, do you ever think paid promotion for a game (Reddit, editorial sites, etc) make sense? 

It’s hard to say what kind of paid promotion makes sense.  I know people who are very much anti and people who are gung ho about it.  I would say that we do put a little bit of money into Reddit ads because we get a fairly good CPC and CPM for the money. 

I haven’t seen huge effects on our sales, but then I only ever spend about $300 a month, hardly enough to really make a dent.

Q: Your daily deals were important - did you go after Valve for them or did Valve approach you, and do you think there's a threshold for timing?

All of our Daily Deals were decided by Steam.  I think once you’ve proven your game sells up to a certain threshold (only Gabe knows what) they have a team that does this for you.  I’ve pitched them on a Daily Deal before but they’ve turned me down.

Tl:dr Steam’s gonna do what Steam’s gonna do.

Q: Do you feel like your game has been successful just because it has the right theme and execution, or do you think how you've explained it or promoted it had made a significant difference? (i.e - does 'marketing' or 'community' majorly help?)

A: On the question of success, I would say that our game hasn’t quite been as successful as other similarly made games (I wrote about that here).  I think the theme is catchy, but also really difficult to make into a good sim game.  And in terms of execution, I look back after 3 years and all I can see is the mistakes we made and how much better our next sim game will be.

I think that marketing and community are two separate things.

With marketing, I would say only a fool would try to release a game without any form of marketing and expect to be successful (success being defined as making back your money and time spent on the game, and then some).  Our marketing efforts through Youtube, Twitter, Facebook, and Reddit were absolutely vital to getting the word out and getting people interested in the game.

With community, I think it’s a more complex thing.  I see community as a force multiplier for your marketing efforts.  IE if you have an engaged community that really loves your game they’ll multiply your marketing efforts by creating their own content.  But community is something you really have to be constantly engaged with and growing.

You can’t just release a couple of gifs and say “hey guys spread this!”.  I would say you need at least one person on your team who is actively engaged in community building, or hire someone to do it, otherwise put your efforts elsewhere.  We managed to survive for 3 years without a Discord so it’s not entirely necessary.  

I do think there is value to community, so recently I hired a fellow gamedev to help us engage our community on Twitter and Facebook.  It’s too soon to say what kind of results that will bring, but that may be fodder for another article!

[One final thing before we go - Ryan notes “From 7AM, Thursday, April 2 to 7AM Friday April 3 (Pacific Time) we're selling Academia for 40% off if you buy the game on the Humble Widget on the linked page, and donating 40% of net revenue from Humble widget sales to various covid-19 related causes and groups.” A wonderful gesture. So if you like this interview… you know what to do!]