TL;DR

I shipped a generic expense app (Receiptix) in July 2024. After seven months of effort it plateaued at $100 MRR, and I hardly used it myself. In February 2025, after a year of working on the project, I decided to reflect on what to do next and I also realised that I actually do use the app but only while traveling. So I niched down and rebuilt it as Spentrip. First week live: $100 in sales - already matching Receiptix’s peak.

Last July I shipped Receiptix, a track-anything expense app. I thought that, by entering into a very packed market, I could capture a slice of the pie. Nothing to validate, and the users were obviously out there. The reality check:

* Launch (July 2024) - first paid user a month later for $8.

* Seven months of iterations - max $100 MRR. Worse, I wasn’t even using the app myself which should've told me something.

In Feb 2025, after a whole year of working on the project, I decided it's time for self-reflection. Should I continue? Should I start a new project? Why am I not using my own app? Then I realized it's not completely true: I do use the app but only while traveling. I like to have a log of all my trips to have a sense of how expensive it is to travel to certain places to help me plan my future travels. It was the moment I decided to try one more time.

So I niched down. New app was called Spentrip. It is a trip-focused expense tracking app that helps travellers track their expenses on-the-go. The main features:

* AI-powered receipt scanning - users can simply snap a photo of their receipt in any language and the app would do the rest (OCR recognition, categorizing, etc.)

* Voice expense tracking - users can just speak expenses to their phone (the app would categorized everything automatically as well)

* Multi-currency support - the app tracks spending across multiple currencies with automatic conversion

* Trip-based orgnization - all the expenses live inside "trips"

* Export - one can export all their expenses as CSV later

First week after launch - $100 in sales, which matches the old app's MRR after 7 months. Still a whole journey to go but it's much more inspiring anyway :)

Tech

* Flutter (iOS & Android)

* OCR: Google Document AI

* Backend: Firebase

* Data extraction and categorization: Claude 3 Haiku

What I’ve learned so far

1. Niche > broad. Solving a smaller problem beats “does everything.”

2. Build for yourself, honestly. If you don’t open your own product, figure out why.

3. Super-crowded markets are tough for solo indie hackers.

Next steps

* Add group-trip splitting

* Offline mode

Happy to answer anything.