Inspiration

I've always enjoyed games like GeoGuessr, but the actual idea for Country Craze came from something much simpler. I was watching YouTube videos where people at carnivals were asked to guess countries just from a few emojis, and sometimes from a country's flag. It was surprisingly fun to watch because some answers looked obvious while others completely threw people off.

That got me thinking: instead of just guessing the country's name, what if players also had to know where that country actually is on the map? That one question turned into Country Craze—a game that mixes geography knowledge with quick thinking and a bit of pressure.

What it does

Country Craze is a geography guessing game where players identify countries from emoji or flag clues by clicking their location on a world map. With only three lives and a three-minute timer, every guess counts as players compete for the highest scores.

How we built it

I built the game on Reddit's Devvit platform using React and TypeScript. The game has two modes: Emoji Mode and Flag Mode. Players get a clue and have to click the correct country on a world map before the timer runs out. I also added a scoring system, lives, a countdown timer, and a scorecard so every game feels competitive and gives players a reason to come back.

Challenges we ran into

The hardest part wasn't the game logic—it was the map. I wanted players to click anywhere on the world map and have the game accurately figure out which country they meant. Getting that to work reliably took a lot more time than I expected. I also spent quite a while tweaking the UI because small changes made a huge difference in how the game felt to play.

Accomplishments that we're proud of

The accomplishment we're most proud of is building our own interactive world map from scratch instead of relying on an existing solution. We started by sourcing raw geographic JSON data, wrote custom scripts to clean and transform it into a format that fit our game's requirements, and then built a renderer capable of accurately detecting player interactions with every country.

This was by far the most challenging part of the project and took countless iterations to get right. The result is a fully customizable map that gives us complete control over gameplay and makes it easy to add new mechanics in the future.

One thing that makes this especially exciting is that we haven't come across another Reddit Devvit game with this level of custom map implementation. Building something this interactive and flexible on the platform felt like pushing the boundaries of what's possible, and we're incredibly proud of the final result.

What we learned

This project taught me that building a game is much more than writing code. A feature can work perfectly and still not feel fun. I ended up iterating on the gameplay, scoring, and interface many times until everything felt natural. I also learned a lot about working with geographic data, building interactive experiences on Devvit, and turning a random idea from a YouTube video into a complete game that people can actually enjoy.

What's next for Country Craze

This is just the beginning for Country Craze. The next big feature I want to build is a real-time multiplayer mode where players can invite their friends and compete head-to-head. Both players would receive the exact same set of questions, with 15 rounds and a 3-minute timer to score as many points as possible. The player with the highest score at the end wins the match.

Beyond multiplayer, I also plan to introduce more game modes, achievements, seasonal events, and additional clue types to keep the experience fresh. My goal is to turn Country Craze into a game that people don't just play once, but keep coming back to because every match feels competitive, fun, and different.

Built With

Share this project:

Updates

Submission history