Steam contains an absurd amount of information about how we play games.
Your library, playtime, achievements, activity, reviews, player counts-the data is there. But a lot of it is fragmented, difficult to compare, or simply not presented in a way that tells you much about your own gaming habits.
That’s what pushed me to build Coilr.
Coilr is a Steam analytics and game discovery platform that tries to turn raw gaming data into something easier to explore and actually useful.
The idea
The original question was pretty simple:
What if your Steam profile could tell you more than just how many hours you played?
I wanted Coilr to answer questions like:
-What does my gaming activity actually look like over time?
-Which games dominate my library?
-What games am I overlooking?
-How active is a game right now?
-Is its player base growing or shrinking?
-When are people actually playing it?
-What should I play next?
That gradually turned Coilr into something larger than a simple Steam profile viewer.
Today it combines personal Steam analytics, game statistics, discovery features, historical data, comparisons, and social features in one place.
The stack
The core stack is intentionally pretty straightforward:
JavaScript
Node.js
Express.js
Steam Web API
Vercel
I like keeping the stack relatively simple because the difficult part of Coilr isn't really rendering pages. It's dealing with data. Steam provides a lot of information, but turning that information into useful insights introduces a bunch of interesting problems.
Raw data isn't the interesting part
Showing that somebody has played a game for 312 hours isn't difficult. The interesting part is figuring out what that number means when combined with everything else. A useful gaming analytics platform has to think about things like:
-historical changes
-different types of games
-player-count fluctuations
-missing data
-games with tiny communities
-games with millions of players
-comparing players fairly
-deciding which statistics are actually meaningful
A giant wall of numbers isn't analytics. The goal is to give those numbers context.
Tracking games over time
One part I've especially enjoyed building is historical game tracking. A current player count is useful, but by itself it doesn't tell you much.
Imagine seeing:
12,400 players online
Is that good?
Maybe.
If the game had 2,000 players a month ago, that's huge growth. If it had 150,000 players six months ago, that's a completely different story.
So Coilr doesn't just focus on snapshots. The idea is to make it possible to understand how games change over time.
That opens the door to things like:
-player-count trends
-peak activity
-historical comparisons
-popularity changes
-long-term game health
And once you collect enough data, you can start asking much more interesting questions about Steam as a whole. Building pages that are useful on their own Another thing I've tried to avoid is creating thousands of pages that exist only because they can exist. A game page should actually answer questions.
Instead of simply displaying a title, image and current player count, I want Coilr pages to gradually become useful resources containing things like player activity, historical trends, review information, playtime data and other context. That also makes the project much more interesting from a technical perspective. You're effectively taking multiple pieces of gaming data and trying to turn them into a coherent story about a game.
Game discovery is surprisingly difficult and Steam has an enormous catalogue. That's great, but it creates another problem: discovery.
Most people probably don't need access to more games. They need better ways to find the handful of games they'd genuinely enjoy. Traditional recommendations are usually heavily influenced by popularity.
But someone's Steam account contains much richer signals:
-the genres they repeatedly play
-how long they actually stick with games
-which games they finish
-what they abandon quickly
-multiplayer vs. single-player habits
-similarities between libraries
There's a lot of room to experiment here. That's one of the areas I want to push further with Coilr.
Adding a social layer
Gaming is obviously also social. A surprisingly common problem is owning plenty of multiplayer games but not having anyone available to play them with. So Coilr also has a player-discovery side. Rather than simply throwing people into a generic LFG feed, the idea is to make compatibility more useful by considering things like shared games, region, availability and communication preferences.
It's not supposed to replace Steam.
It's more about building tools around the data and social connections Steam already makes possible.
One of the biggest lessons so far
Building Coilr has reinforced something I've noticed with a lot of side projects: The feature that sounds impressive isn't always the feature people actually use. Sometimes the smallest feature suddenly gets traffic. Sometimes something you spent days building barely gets touched.
And sometimes users start using a page in a way you didn't originally expect.
Analytics become incredibly useful at that point-not just analytics inside Coilr, but analytics about Coilr itself. I've become much more interested in watching how people navigate the product and then improving the parts that already show signs of demand.
Where I'm taking it next
There's still a lot I want to build.
Some of the areas I'm particularly interested in are:
-deeper Steam profile analytics
-better historical game statistics
-more useful game discovery
-comparisons between players and games
-original research using aggregated gaming data
-better ways to discover compatible players
-yearly gaming summaries
The broader goal is to make Coilr a place where your Steam data feels less like a database and more like an actual picture of your gaming life.
It's still evolving, and that's honestly the fun part.
If you're building something with the Steam API-or you've worked on gaming analytics before, I'd love to hear what problems you ran into.
Top comments (0)