ArcRaiders Build
Skill trees and a barter market
Arc Raiders, by Embark Studios, is one of the biggest game launches of the last five years. With one developer, I designed a site where players build skill trees, vote on them and trade the items that gate progression, plus many more tools to improve their runs. We launched just before the game, and reached 193k users in five months with no paid acquisition.
- Role
- Co-founder & Product Designer
- Team
- Me + 1 developer
- Timeline
- From the game's launch, 2025
- Scope
- Co-founder, Product Design, UX/UI, Marketplace, Brand
01 · The problem
One wrong point, for good
Arc Raiders is one of the biggest game launches of the last five years, made by Embark Studios, the Stockholm studio behind The Finals.
New players meet a skill tree with three branches and dozens of nodes, and every point they spend is permanent. Most didn't know where to put theirs, and had no way to tell a proven build from a guess.
Progression also asks for specific items. Players kept looting things they didn't need while missing the one they did.
We were two. Basile built the frontend and the backend. I designed the product, the brand and the design system, adapted from shadcn/ui. We launched a few days before the game came out.
02 · Skill trees
Let the community decide
Rather than write our own "best build", we gave players a planner and a vote. Anyone can create a skill tree, and the community rates it. Unsure players sort by top rated and follow a build others have tested.
The planner shows the tree point by point and level by level, so a build reads as a path, not only a final state. Each one shows its level range, its focus and who made it.
Streamers were the other half of the bet. A pro who shares their build on stream sends viewers to the site, and gives newcomers a source they already trust.

03 · Marketplace
A barter market for what gates progression
The marketplace lets players trade what they have already looted, or no longer need, for what they are missing. No money changes hands. Listings split into items for sale and items wanted, each item shows how many offers it has, and a trade closes in a conversation between two players.
Players sent us a lot of good feedback on its UX.
Around these two, we shipped many more tools to make each run better.
04 · Outcome
Streamers as the marketing team
The first five months, with no paid acquisition. Traffic came organically, carried by streamers sharing their builds, and players left millions of likes, ratings and trades.
- Page views
- 3.22M
- Unique users
- 193k
- Average session
- 9m 14s
- Visits
- 316k
- Pages per visit
- ~10
- Views on the peak day
- 53,768
05 · What didn't work
A week late, on a game in decline
We shipped the marketplace one week after a competing site. They also had features we didn't, and they took most of the traders. On a game launch, that week decided the market.
The second limit was out of our hands. A site like this grows and shrinks with its game, and Arc Raiders lost a large share of its players. Our traffic followed.