I’d be lying if I’d say I don’t check Steamspy on a regular basis. As a dev, I’m curious to see how other games are performing to get a sense of what is doing well and what isn’t. In fact, for my next project I used Steamspy to analyze games of the same genre to draw some non-scientific conclusions as to what might help a game in that genre to succeed.

What I just said right there is a terrible thing to do if you only use Steamspy to draw your conclusions. It doesn’t say anything about what kind of marketing was done, who are the devs who released these games (were they high-profile devs or newcomers) or how many copies were actually sold at full price.

Most people don’t know how to analyze data on Steamspy

Each time Steamspy is discussed, someone needs to point out that Steamspy shows an approximation of the number of owners and not sale numbers. Steamspy is clear about this yet we all know people don’t read or bother to make sure they understand what they see. Opinions need to be posted ASAP!

This makes it hard to have a discussion about Steamspy in a wild environment like the Web. Even other devs need to be told from time to time that Steamspy needs a few days before showing results that are a bit accurate. You just can’t check Steamspy the day a game is released. Seems easy enough to remember yet it needs to be repeated over and over.

The other thing people often don’t understand is that you can’t do “number of owners” * “full price” = “revenue. It doesn’t tell you how many copies were acquired at a discount, in a bundle or even for free. It can make appear a poor selling game like a hit and a success like a failure when put side by side. A game with 10,000 owners that was never discounted beats a game with 50,000 owners that was put in numerous bundles.

Finally, these stats don’t say anything about net revenue but I regularly see people using them to speculate on that topic. How much did it cost to make that game? How many ways money need to be split? A game with 25,000 owners is more successful than another one with 10,000 right? But what if the money the game with 25,000 owners makes need to be split 4 ways and the other game was released by a solo dev?…

Because of these misinterpretations, I usually don’t engage in a discussion involving Steamspy because half of it is about remembering people to be careful about what they make of the data they see.

It’s really none of your business to know about mine

Yet the choice to reveal data related to my business is made for me. I don’t have a say in it. Is it really that bad? Probably not. I just feel uneasy this is done without having any power over it. Oh Steamspy isn’t revealing any direct financial data but hold on a bit.

I said the data on Steamspy doesn’t tell the whole story but the number of owners is pretty accurate. When a game is released, it becomes easy to get a relatively good idea how much money it made if you check Steamspy after a week if the release process was a “normal” one as you know exactly at which price that game was sold for these 7 days.

If I release a game at $14.99, you know I worked alone on it and see Steamspy saying 100,000 owners after a week then it becomes easy to speculate that my financial situation is pretty sweet right now. At least sweeter than if Steamspy is showing you 5,000 owners unless I managed to give away 95,000 copies for free before release.

If I’d win the lottery, the number of people aware of this would be pretty limited as that’s when you start having sharks circling you. Do devs with games with that many owners are really getting harassed? Maybe, maybe not but if it was me I’d still like to decide myself how to handle this without a third-party deciding for me.

But it’s public data and it was already possible to draw conclusions before Steamspy

Steamspy isn’t stealing anything and anyone could build their own Steamspy. It just organizes what anyone could access and there was already a way to come up with wild speculations before Steamspy… The number of reviews.

Oh sure, the number of reviews is far less accurate but you don’t need a fancy degree to compare two games released at the same time when one has 30 reviews and the other one 15,000. If you have your own game on Steam it’s not that foolish to compare your number of reviews to another game to know how many people have bought that game (again by being careful about the history of these games so it’s best done soon after release).

An example of how confusing it all can be

The game Volume was recently featured as a daily deal on Steam so I took some time to check its stats as I wanted to know what kind of success a game needs to be featured that way on Steam. I went on Steamspy and it’s saying about 86,000 owners. Wow! I heard this game didn’t do so well so my info must have been wrong. I went on the game’s page on Steam and then saw the game had “only” 243 reviews… What? (note that it went up a bit recently probably because of that daily deal)

To give you an idea of why I was surprised, let’s compare it to March of the Living. Steamspy will tell you that is has about 6,000 owners (that number is actually weird as it was higher before so right now it’s not quite accurate) and the Steam’s page shows 387 reviews. How can my game have more reviews when it has about 14 times less owners? A quick search revealed that Volume was in a Humble Bundle at some point which might explain how Steamspy and the number of reviews don’t seem to match.

So what’s the truth here? It’s hard to tell for sure without knowing the stats before the Humble Bundle happened but I’d be tempted to consider the number of reviews here instead of Steamspy to get an idea of many owners this game had before the bundle. This would mean that MotL probably sold more copies than Volume before the bundle happened even though Steamspy seems to say otherwise.