Dominion Akinrotimi
Home / Writing / O-Sight
Product · FinTech

O-Sight: see where every naira actually went

How a bank statement becomes three honest figures, on your phone, without being uploaded.

By Dominion Akinrotimi · October 2026 · 6 min read
O-Sight: see where every naira actually went. Real income and net savings shown over a photo of two women crossing a Lagos street
LiveFree to use today
0 uploadsYour statement never leaves your phone
Every rowChecked against the balance
3 figuresIncome, spent, saved. No score.

Money in is not income

Open any bank statement and add up the credits. That number feels like what you earned, but it is not. It includes money you moved back from your own savings, refunds for failed payments, and transfers from your other accounts. On the sample statement O-Sight ships with, ₦3,255,750 came into the account, but only ₦2,877,550 of it was real income.

I built O-Sight so that anyone can drop in their statement and see three honest figures: what you really earned, what you spent, and what you kept. It began as a personal bank statement analyser, and it grew from one simple aim: financial literacy, people seeing and understanding their own money.

Campaign post: 3,255,750 naira came in, only 2,877,550 was income
Figures from O-Sight's invented sample statement.

How it works, in four steps

1. Your statement is read on your own device

The Excel statement is opened by code running inside your browser tab. It is never uploaded, and there is no account. You can prove it: load the page, switch on airplane mode, and it still works. Close the tab and everything is gone.

2. Every row gets exactly one type

Each transaction becomes one of: income, spending, fee, savings deposit, savings withdrawal, transfer between your own accounts, reversal, or unclassified. A transfer counts as your own account when its name clearly matches the name on the statement. Anything the engine does not recognise is counted and shown as unclassified, never guessed.

3. The statement is checked against itself

This is the part I am proudest of. For every row, O-Sight checks that

previous balance − debit + credit = balance printed on the row

and that the totals match the statement's header. If rows are missing or misread, or the bank left a whole month out of the file, the dashboard says so before anything else. Real exports are messy: a bank can list a payment a few seconds before the savings withdrawal that paid for it, so O-Sight puts rows back in bank order (without adding, dropping or changing any) before checking.

4. Plain figures, plain facts

Each headline figure is defined once, with its formula shown in the app:

There is no "financial health score" and no grade. Insights state facts with their numbers and dates, and do not judge you or tell you what to do.

The part people share: money cards

O-Sight makes pictures you can save or post, built on your phone for any period you choose. "My money" sums up a period. "My growth" sets one period against an earlier one. And "My pace" is the blunt, funny one: at your current savings pace, how long until your next ₦1,000,000? No card shows your amounts unless you switch them on, and none ever names a person.

Proof of income without a payslip

Many Nigerians earn through transfers, not payslips. When a landlord, an embassy or a lender asks for proof of income or proof of funds, the evidence is already in the statement. O-Sight can produce a one-page income report or a statement of funds for a visa, made on your device. It names missing months, lists the checks it ran, and says plainly that the figures are not verified by the bank. Seeing your own figures is free; a report is a small one-off payment.

Campaign post about income reports for rent, visas and no-payslip earners

Privacy, as rules

Where it stands

O-Sight is live and free. It reads OPay statements today, including the OWealth savings sheet, and more banks are planned as real exports become available to test with. A research paper on extracting personal financial insight from Nigerian bank statements has been submitted for review.

What it taught me

Need this kind of work done?

I take on data engineering, analytics, machine learning and research projects, and I speak and train teams on data and AI.