The story
Why I built FinMan
FinMan was not built as a startup. It started as a tool I made for myself, because I simply wanted to know where my money goes — without spending evenings on spreadsheets. This is the honest story of how it grew, feature by feature, out of real everyday problems.
The beginning
Nothing on the market fit
I wanted a clear picture of my spending, so I tried the existing finance apps first. Every one of them failed me in the same way: too much manual work. Some required entering every expense by hand, others needed endless configuration, and almost none of them could pick up my online card payments automatically.
At some point I realized I was spending more time feeding the tracker than actually understanding my finances. So I did what developers do — I started writing my own. The very first version was gloriously unfashionable: Perl on the backend, plain HTML and JavaScript on the front. It was ugly, but it was mine, and it did exactly what I needed.
First automation
Expenses that record themselves
The first real breakthrough was the Monobank integration. The bank offered an API with webhooks: pay with the card, and a second later the transaction lands in my app — amount, merchant, category. No typing, no forgetting. For the first time my spending picture was complete without me doing anything.
Then came other banks, which meant importing statements. Each bank had its own format, so support grew bank by bank: CSV here, PDF there, different columns, different date formats. Eventually I got tired of writing a parser per bank and taught the app to import any statement using AI — today it recognizes the structure of a statement it has never seen before.
The car question
What does one kilometer really cost?
I always wanted to know what my car actually costs — not a vague total, but a breakdown: how much goes to fuel, how much to repairs, how much to scheduled maintenance, and how all of it translates into the price of one kilometer.
So I built a mechanism around refueling details: price, liters, and the odometer reading. From those three numbers the app calculates fuel consumption and cost for the distance driven, and together with repair records it answers the real question — what a kilometer costs me. Today you just send a photo of the fuel receipt and the odometer to the bot, and the rest happens automatically.
Utilities
A photo instead of a calculator
Utility bills were another monthly ritual that quietly ate my time: read the meters, calculate the difference from last month, multiply by the tariff, repeat for every service. Simple math, repeated forever.
Now it works like this: you take a photo of the meter and send it to the bot. AI reads the digits, recognizes which utility service the meter belongs to, and records the reading. The app computes the difference, applies the tariff, and shows what to pay. What used to be an evening with a calculator became three photos from a phone.
Honest numbers
Currencies and categories that tell the truth
The local exchange rate was not always stable, and that distorted the picture: expenses from different months simply could not be compared in the national currency. So FinMan became multi-currency — every payment keeps its exchange rate, and analytics can be viewed in a stable currency to show real spending across any period.
I also wanted to separate business projects and investments from everyday household spending. That is how excluded categories appeared: they do not mix into your daily totals, but you can still analyze them separately whenever you need the full picture.
Today
From a personal tool to a product for everyone
Somewhere along the way I noticed that the tool had become genuinely convenient — the routine was automated, the analytics answered real questions, and I was no longer doing any manual bookkeeping. So why keep it to myself?
FinMan was rebuilt as a modern web application with family groups, a tenant mode for landlords, Telegram and WhatsApp bots, AI categorization, and support for dozens of languages. The core features are free, because the tool was born free — I made it for myself first, and that is still how every new feature gets tested.
Try the tool that was built to be used, not sold
Start with one bank import or one meter photo — and let the routine take care of itself.