The problem
I own hundreds of games I’ll never play. I’m not the only one: the industry is designed for you to buy, accumulate, and move on to the next thing — not to enjoy what you already have. The tracking apps that exist don’t help much — they tell you about the next release, they recommend the new stuff, but they don’t answer the question I actually ask myself every night: what should I play today, out of everything I already own?
That question was the starting point, before Backlog as it is now even existed.
The first idea: an Akinator-style recommender
My first prototype wasn’t a backlog-management app. It was something much smaller: a recommender I installed directly on my Steam Deck.
The mechanic was inspired by Akinator: instead of giving you a list or asking you to filter by genre, it made you choose between two games from your own library. Each choice halved the rest of the options, until you reached a single game. No infinite scrolling, no analysis paralysis — just quick, binary decisions until you got an answer.
The problem with Steam’s categories
For those binary choices to make sense — so it wasn’t literally heads or tails — I needed to compare games by real characteristics: tone, pacing, length, mechanics. Steam’s tags aren’t enough for that.
The solution was to pull tags from an external video-game database and build my own categorization layer, richer than Steam’s, so the binary decision tree had real criteria behind it.
What didn’t work (and what stayed)
This prototype didn’t quite take off as a standalone product. But it wasn’t wasted time: the categorization layer I built then is, right now, the foundation I’m reusing in Backlog, the project I’m working on with my friend Carlos. The idea is to use those same tags, further down the line, to feed a kind of pet or Tamagotchi inside the app that recommends or comments on your library.
It’s funny how a “failed” prototype ends up being the foundational piece of something else.
What’s next
In the next post in this series I tell the story of how we went from this solitary recommender to Backlog: an app where you can connect your Steam library, triage your games Tinder-style (swipe yes, swipe no), and even — with a fairly creative workaround — import your Nintendo history from screenshots and an LLM.
This is the first post in a series where I document, in real time, how we’re building Backlog.
[!NOTE] Editorial brief Idea: anti-consumerism in games — enjoy what you own, not buy-buy-buy. From an Akinator-style Steam Deck recommender to
Backlog(with Carlos). Series: part 1 of the Backlog series (series_part: 1). Next: Tinder-style triage + Nintendo history import. Audience: indie/product hustle; links to thebacklogcase study. 🟢 consumption. Note: there’s a mis-titled duplicate of this undercampus 3d.md(cleaned separately).