Bank reconciliation for an automotive business
An independent review of four bank and fintech accounts for a Nigerian automotive business. Every line was rebuilt from the statements, proven against the bank's own totals and sorted into plain categories in Google Sheets.
- Client
- Nigerian automotive business
- Accounts
- Commercial, microfinance and fintech
- Scope
- Over ₦3 billion across four accounts
- Delivered
- Google Sheets workbooks
What was getting in the way
Money moved between customers, car suppliers, investors, a financing partner and daily expenses, across several accounts. The records had not been categorised properly, so the owner could not see what came in, where it went, or what was still owed.
Many kinds of money
Customer payments, supplier payments, investor money, a financing partner and daily expenses all ran through the same accounts.
Spread across accounts
Money moved across several bank and fintech accounts, including a related personal account.
No clear picture
Without proper categories, revenue, borrowed money and money held for others all looked the same.
Check the books yourself
Sort a few lines from the evidence, run the summary checks and match a transfer across two accounts.
Illustrative recreation. The client's records are confidential, so every line, name and amount here is a sample built for this page. The scope and results quoted on this page are real.
Under the old labels, a customer's car payment, an investor's money and a payout to a car's owner are hard to tell apart.
How it was done
Pull out every line
Extracted every transaction from four long statements, each in a different bank layout, into clean spreadsheets.
Prove nothing is missing
Matched the opening balance, money in, money out, closing balance and payment counts against each bank's printed figures, and checked every running balance against the line before. All of it matched exactly.
Explain the bank quirks
Documented the habits that would otherwise look like errors. One bank counted its opening balance as money in, another left failed transfers out of its totals, another moved balances between internal ledgers on the same day.
Trace money between accounts
Matched over 700 transfers on both the sending and the receiving statement, by date and amount, allowing for transfer fees. This showed one bank printing the client's own account under another institution's name, which made internal transfers look like payments to an outside company. Reversed transfers were marked so they were not counted as money paid out.
Sort it in plain words
Built a category list around the client's own codes (INV for investment, UPK for upkeep, OTH for other account). Every line was categorised from what the statement showed, never guessed.
Ask, don't assume
Payments with no description went into their own "no reference" categories and onto a list for the client to confirm, as the client asked.
Build workbooks people can use
One workbook per account: a Summary with "does it add up" checks and monthly totals, a Transactions tab with dropdown categories, a People tab, a Cross-check tab and a Labels guide. Totals are live formulas, so they update when a category changes.
Check the checker
Reviewed every final workbook line by line. This caught names spelled two ways being counted twice, money out labelled as money in, typos in categories and categories missing from the summary. Automatic check lines now confirm the totals still match the bank.
- Google Sheets
- Formulas
- Dropdown validation
- Conditional formatting
- PDF data extraction
- AI-assisted analysis, checked against the bank
What changed
- Every statement was proven complete, with all totals and balances matching the bank exactly.
- The money moving between the client's accounts was fully traced and confirmed.
- Thousands of transactions now sit in plain categories, each one explained, so the owner can see where the money went.
- Unclear items became a short list for the client to answer, instead of being hidden or guessed.
- The workbook format and category system can be reused month after month.
Procurement automation platform
Got a task on your list? Send it over.
A report, a design, a build or a set of docs. You'll get a fixed quote and a deadline before anything starts.