Building Game Scout: Mistakes I Made and What I Learned
The first game-scout app - a mobile app where you could ask questions to AI and get game recommendations
I started working on game-scout early 2024. At first my idea was having a mobile app with game database together with AI so you can chat with it and get games you wish for. Since then I did a lot of mistakes and now I can say I learned lessons in the journey and I can share my mistakes and maybe this will help me to avoid them next time.
The Beginning: Building Without Validation
In the beginning I had the idea of finding co-op games that I can play on couch with my friends. After talking with a few friends I saw that I can also add more information like stealth games or medical games so it will be easier to find niche games. My biggest mistake here was trying to get everything. I tried to fetch all of the games and tried to implement AI capabilities and spent a lot of time. I can see that I had a dream but I did not clearly identify the problem and work with it. I focused on solution, I tried to build something.
Lesson learned: Should have focused on problem instead of solution.
First Pivot: From AI to Simple Filters
First realization was AI was harder to use than logical filters. AI implementation was a shiny feature that I dreamed about but in practice it was just time loss. I spent a lot of time to get games to context in optimal way and tried to create a response of games that actually made sense. I realized filters were easy to implement, debug and also easy to use for me.
Lesson learned: You don't need shiny product to start, it is enough if you can solve the problem.
Second Pivot: From Mobile to Web
When I was trying to show my friends and trying to get some feedback I realized that it is hard to deploy an app and share with friends. Feedback loop was too long because I had to wait and resolve issues at app store and I could not send the app to my fellow Android users. I realized that it will be easy to just convert it to web app and send a link. Everyone can check it in a few seconds.
Lesson learned: Feedback loop should be as fast as possible.
Third Pivot: From Features to Getting Users
I spent more than a year to get all the games and implement extra features and filters. When I finished I was proud with my app but the issue was no one was using it. After first feedback loops even my friends stopped looking for it. I stopped using it. I started to feel like I should move to another project and this was a big failure but at the same time I did not want to move to another project again because I am always changing projects and never got users or live product at all. So I decided to work on marketing, just try to reach out to more people.
Lesson learned: Start with users, not features.
So I started to share my application on Reddit, I started a blog to share what I do, and started to build SEO pages so I can get some clicks.
You can read more about this journey:
Looking Back: What I Should Have Done
On the first day I should have just created SEO pages with a few games. I did not have to fetch all games and implement features, AI chat etc. I could have just created some SEO pages and watched search engines to see if I can validate my idea. After I got validation I could have worked on features and tried to get these users to my website and actually I am still on this step right now. Right now I feel like I validated my idea, but now I need to show that I can engage with users and help them to find new games.
Lessons Learned
Here are the key lessons from my journey:
- Focus on the problem, not the solution - I jumped to building features without clearly validating if the problem existed for others
- You don't need a shiny product to start - Simple filters solved the problem better than complex AI chat
- Keep the feedback loop fast - Moving from mobile to web made iteration much faster
- Start with users, not features - A year of features meant nothing without people using them
- Validate early with SEO pages - A few landing pages could have tested demand in days instead of months of building
What's Next
Now I would like to focus more on users. Getting real users not just views from Google and talk with them to understand how I can help more. I would like to decrease bounce rate and help people to decide to buy a game.
If you're building something, I hope my mistakes help you avoid the same pitfalls. Start small, validate early, and focus on users from day one.
Check out game-scout.app to see where it is now.
Let's connect!