Steam's wishlist conversions - how hot are yours?

I know you love graaaaaphs.

[Hi, I’m Simon Carless, and you’re reading the Game Discoverability Now! newsletter, a regular look at how people find - and buy - your video games. Or don’t.]

Welcome - once again - to this nerdy as heck newsletter, looking at ‘interesting things that might help you make your game sell better’ for the 103rd week running.

It continues to be a deeply weird and disturbing time (and I hope you’re donating directly to anti-racism or pro-community causes). But here’s a small injection of helpful information, as we continue trying to parse the real world.

Steam wishlist conversions - getting granular.

Over the last couple of weeks, a few devs have been going public with various intriguing detailed Steam wishlist conversion stats.

Before we get started, I wanted to remind everyone that a recent GI.biz piece has some very useful official data from Valve, specifically graphs on ‘average for all wishlist conversions before release’…

..and ‘how much 12 month wishlist conversions differ from game to game’ (a lot!):

I still think the stat that most of us should look at is ‘how many wishlist conversions did I get in the first week?’. But some of these extra dimensions are very helpful.

To that end, Jake Birkett did some manual work in looking at how wishlists converted per month added for his RPG card battler Ancient Enemy:

Jake Birkett - Veteran Indie @GreyAlien

OK here is some pioneering research on Wishlist Quality By Age. If you find this useful, please become a Patron for more data: patreon.com/GreyAlien Conclusion, it's pretty bumpy. Month 1 had lots of real fans sign up. Mth 16 is month before launch, so fresh. Mth 17 is launch

Interestingly, there’s been a lot of speculation that older Steam wishlists convert more poorly. But Jake’s data shows that at least for his game - which has been adding wishlists steadily for 18 months - there wasn’t an incredibly big difference for ‘immediate post-launch’ conversions.

(Important caveat: Ancient Enemy’s wishlist additions were almost all organic, with no demos, search features, Steam features, etc. Which is good for getting a ‘clean’ reading on a good quality game.)

Conversely, Alina from Yes, Your Grace devs Brave At Night was kind enough to crunch her own numbers for the successful 2D kingdom-builder/story game, & got quite different results:

Alina Cebula @cebulaalina

Prompted by @GreyAlien I looked at some wishlist data for Yes, Your Grace this week! This graph shows the percentage of wishlist-to-sale conversion for each month since announcement. 📊 Month 6 is where the game was released. Thread 👇

So, yes, this shows wishlist conversions as low as 3% and as high as 40% (!). I just chatted to Alina about this, and she identifies Month 2 as being linked to two major things that potentially raised volume, but diluted wishlist conversions:

- a public Beta you had to sign up for (lots of ‘bookmarking’-style wishlists!)
- a spike in Steam ‘search suggestions’ traffic which I discussed at the time, and may have also contributed lower-converting wishlists.

Conversely, Month 4 was almost all ‘organic’ wishlists, which is to say - many less people wishlisted the game in that month, but they’d all proactively found the game.

Yes Your Grace’s ‘natural’ wishlist conversions being that high - albeit just for one 30-day period - hints towards the attractiveness behind the game that has led to its Steam success. (3,100 Very Positive reviews thus far!)

Finally, Tomer Barkan was kind enough to share some data from Judgment: Apocalypse, as follows (click through for the whole thread, it has lots of data!). Particularly wanted to highlight this graph:

Tomer Barkan - Suncrash @tomerbarkan

This graph only shows conversions that took place in May 2020 (the sale happened in May 2020), grouped by month in which wishlist was added. We can clearly see that recent adds are much more likely to convert than older adds. (2/6)

So it’s true that at launch, Steam wishlist conversions from older cohorts can be good - since the game is available for the first time. But now we’re all the way in May 2020, it’s the newly added wishlists that convert the best. (This… makes sense.)

I thought Tomer’s conclusions were useful:

“- Most conversions take place shortly after adding.
- Discounts are more likely to convert recent [additions].
- Graph 1 shows that despite the above, in the long run, wishlists added later (hence closer to large discounts) do not necessarily convert better.”

Conclusion: the ‘too many graphs’ problem.

[Apologies for unnecessary XKCD comic, but come on, it was sorta necessary?]

So, ending out this newsletter, should you care about this granularity of Steam wishlist conversion? Uhh.. that’s a tricky question.

By the time you see Steam wishlist conversions, your game will have done well, or not. So unfortunately, you can’t turn back time and say ‘oh, THOSE wishlists were good’.

And remember, your game could have a decent long tail & add non-wishlist purchases after launch too. (But again, good organic Steam wishlists are broadly indicative of success. So you can’t just ‘I’ll launch with 500 wishlists and wing it’…)

Overall, one thing I’m taking away from this is the following. Maybe it’s not a good idea to look at your largest pre-release wishlist-adding month, and presume those wishlists are high-converting. In fact, maybe if you have a major spike in wishlists, it’s probably related to lower-converting additions around promotions of some kind.

And under all of this is the natural buy-ability of your game - this isn’t purely a numbers game, your game needs to be good and attractive too. So… it’s a complex ol’ thing. That’s all.

Finally, good news in terms of more data - I ended up getting around 40 responses to ‘launch wishlists vs. first week sales’ survey I fielded recently. So I’ll be presenting the results to that in the next couple of weeks - and I hope that’ll clarify things further.

Take care, all, and be safe,
Simon.