Another big issue is that the average discount rate varies a lot and I don't have data for discount history to correct this. For instance if a game has reduced its price since its launch my formula will underestimate the revenue. Bear in mind this method is quite simplistic, and is prone to a number of issues. So that is how I approximate and project the revenues in my data. Side note - I'd be very grateful to any dev with older titles who can let me know how this sales curve holds up to their data! Thanks to Jake Birkett's article I'm pretty confident about the accuracy of the first year but the later years of this curve was hard to collect data for, and I'm not super confident about the accuracy yet. So let's say a game is 7 months after release, we can use our graph to see that statistically this might be around 40% of its eventual 5 year revenue, so we would plug that into "current percentage of lifetime" and get a multiplier of 2.5. This calculation goes like this:Ĭurrent approximate revenue * (1 / current percentage of lifetime) We can use these numbers to project the lifetime revenue of a given game, as long as we know its current approximate revenue (we can use reviews at mentioned before) and its release date. I've settled on these numbers based on sales data I've been able to gather from other games (including our own), and also by looking at the review graphs of various games on steam, to see how the incoming reviews taper over time. Some games have a larger long tail, making revenue well beyond 5 years, and some peter out more quickly. These numbers, like many of the numbers in this article, are approximate. for simplicity's let's call a game's lifetime revenue 100% after 5 years (sales generally trickle very slowly at this point).On average, a game makes the following percentage of its lifetime revenue (cumulative): To guess at the lifetime revenue, here are some numbers that are helpful: Games which released recently have only made a portion of their lifetime revenue. 0.8 is our game's ( Eastshade) average discount after 8 months, our deepest discount so far being 40%.Īnother important factor in guessing a game's eventual revenue is how far the title is into its sales curve. Also the average discount multiplier tends to decrease throughout a game's lifetime, as the discounts get deeper. This is of course another approximation, because a game's actual net revenue will vary quite a bit based on how deeply its discounted, and the geographic composition of Its sales (which effects average regional price as well as percentage of sales subject to VAT). Owners * US price * 0.93 (VAT) * 0.92 (returns) * 0.8 (avg regional price) * 0.8 (avg discount) * 0.7 (platform cut) Once you have the owners, you can multiply it by the US list price, and then multiply it by about 0.38 to guess the net revenue. One of the simplest ways to guess at how many owners a particular Steam title has is to multiply its total review count by 50 (according to Jake Birkett the actual range of this number varies between 30 and 100, but 50 is a nice conservative number). Approximating A Game's RevenueĪpproximating a game's revenue is a useful tool for an indie developer. Throughout the article I use the words interchangeably, since I kind of see tags as a more granular form of genre. Note: The title of the article says genre, but technically what I look at in this article is tag, which is more granular and I feel more useful. So one fateful day I got to thinking: Wouldn't it be nice to have a spreadsheet of every game on steam, its tag, genre, release date, and approximate revenue? I thought it might be fun to look for patterns in the data, particularly in the tags.