A driver's money, in columns
Driving is one of the few businesses where the operator genuinely does not know their own revenue. Money arrives weekly from a platform, already net of commission, mixed with tips and incentives that were earned under entirely different rules. Money leaves constantly and in small amounts: fuel or charging, the vehicle, insurance, cleaning, tolls, the phone.
All of it is printed in a PDF that no spreadsheet can read. FlowParse converts the statement with AI — every transaction with its date, description, signed amount and running balance — and validates it against the balance the bank printed, so nothing is missing before you start.
What you get is the thing a driver's accounts actually need: a categorised table of what came in, what the vehicle and the road cost, and what was genuinely left — from any bank, with no template to set up.
You earned more than your bank shows
Every driver knows the platform takes a cut. Fewer realise what that does to their books. The commission is deducted before the payout moves, so the money in your account is net — and your actual turnover, the figure a tax return asks for, is larger than anything the bank will ever show you.
This is not a technicality. Understating turnover means understating your expenses by exactly the same amount, and while the profit may come out the same, the turnover figure is what registration thresholds and credibility both hang on. Filing a return that says you earned only what landed is filing a wrong return.
Converting the statement gives you the banked side to the penny: every payout, dated, per platform, as a real number. Bridged against the platform's own weekly summary, gross becomes provable and the commission becomes a deductible cost with a figure attached instead of a shrug.
What FlowParse pulls from a driver's statement
Extraction is by meaning rather than by template, so it works with whichever bank you are paid into — including the app-based accounts most drivers now use: every transaction's date and value date, the full description including the platform's payout reference, the signed amount, the running balance, and the counterparty where named.
Descriptions matter because that is where the platform, the week number and the payout type hide. A tip payout and a weekly earnings payout may differ only by a word in the reference, and descriptions that wrap across lines are joined back into one field rather than truncated.
Every amount exports as a typed number with the correct sign, so a year of fuel and a year of payouts each total in one formula.
How to convert a driver bank statement
Upload the statements
Drop in one or many PDF statements from any bank or app account — digital or scanned.
Let AI extract them
Every transaction is read with its date, description, signed amount and balance.
Check the balance
Opening balance plus transactions is checked against the closing balance the bank printed.
Export and categorise
Download clean Excel or CSV — or an accounting-ready file for QuickBooks or Xero.
Tips, incentives and the weeks that were not typical
A payout is rarely one thing. It bundles fares, tips, and whatever incentive the platform was running that week — a quest, a surge guarantee, a referral. Each behaves differently: tips are yours in full, incentives are promotional and temporary, fares are the actual business.
Averaged together they produce a misleading picture. A great month built on an incentive that has since ended is not a great month to base a car purchase on, and a driver who cannot separate the two is planning against a number that no longer exists.
Structured payouts let the components be tracked as a series over time. When the incentive stops, you see it stop — as a line that flattens — rather than discovering it in the general feeling that work has got harder.
The cost of getting paid early
Most platforms will pay you today instead of on the weekly cycle, for a fee. It is a small fee. That is the point of it.
Taken once, it is trivial. Taken four times a week for a year, it becomes a real cost that no driver has ever consciously decided to incur, because it is never charged — it is simply deducted, invisibly, from money you were owed anyway. It is one of the purest examples of a cost that only exists in aggregate.
In a structured statement those cash-outs are countable, and the total is usually the most surprising number the exercise produces. Once you can see the annual figure, deciding whether early access is worth it becomes an actual decision rather than a reflex.
The honest boundary: we do the bank side
FlowParse does not connect to Uber, Lyft, Bolt or any other platform. There is no integration, no API and no login to your driver account — and we would rather say so than let you find out.
What we convert is the statement your bank produced. Your platform's weekly summary or annual tax summary is the other half, and you download it from them. The value is in having both as clean data, because the comparison between what the platform said and what the bank received is where the discrepancies live.
We also do not issue tax forms. We do not produce your 1099-K or 1099-NEC, we do not file returns, and we do not decide what is deductible where you live. We produce accurate, validated, categorised data — the raw material your return is built from, not the return itself.
Why the form says more than you were paid
Drivers in the US routinely get a shock in January. The 1099-K reports gross ride receipts — what passengers paid — while the bank received those receipts minus commission, minus fees, minus adjustments. The two numbers are supposed to differ, and the difference can be enormous.
Reporting only the banked amount against a form that says something much larger is precisely the mismatch that generates a letter. The correct treatment is to report the gross and deduct the platform's cut as an expense, which requires knowing both numbers exactly.
That is where converted statements earn their keep. The bank side becomes an exact, dated, validated series, and the gap to the form stops being alarming and becomes explainable — line by line, with the evidence attached. See also bank statements for a tax return.
Fuel, charging and the cost per mile
Fuel is the second largest cost in driving and the one most likely to be a mess in the books. It is bought in small amounts, dozens of times a month, at many merchants, often on the same card as everything else in your life.
Electric drivers have it differently but not more simply. Charging splits between home electricity — bundled inside a household bill that has nothing to do with the business — and public rapid charging that arrives as many small card transactions at a much higher unit price. Sorting one from the other by hand is a genuinely tedious evening.
Categorised structured rows turn that into a total per month with two keystrokes, and once fuel is a number and mileage is known, cost per mile becomes computable. That is the number that says whether the work is worth doing at current rates.
Finance, rental, insurance and depreciation
The vehicle is the business. It is also the most commonly mis-recorded thing in a driver's accounts, because how it is paid for changes what it is.
A rented vehicle is a straightforward weekly cost. A financed one is not: the payment is part interest and part capital repayment, and only one of those is an expense — the rest is you buying an asset that also loses value, which is a different deduction with different rules. Insurance is a private-hire policy, several times a standard one, and often paid monthly at a premium over the annual price.
In a structured statement each of those separates cleanly by counterparty and by pattern, which is what lets your accountant treat them correctly rather than lumping the lot into vehicle costs and hoping.
Two apps, one account
Very few drivers work one platform exclusively. You run whichever is busy, which means two or three payout streams landing in one account, each with its own commission rate, its own payout day and its own incentive scheme.
Which of them is actually paying you better is a question almost no driver can answer, because the comparison requires netting each platform's cut against its own earnings across the same weeks — arithmetic that is impossible while everything is text in a PDF.
Converted and categorised, each platform becomes a column. Earnings per platform per week, net of what that platform takes, is a straightforward comparison — and it occasionally reveals that the app you prefer driving on is the one paying you least.
When the business account is also your account
Most drivers never opened a business account. The payouts land where the shopping and the rent go out, which is entirely normal and makes the year end considerably harder than it needs to be.
Sorting business from personal in a PDF means reading every line. Sorting it in a structured export means filtering and tagging once, then reusing those rules for the rest of the year. It is the same job reduced from hours to minutes.
It also produces something worth having if HMRC or the IRS ever asks: a categorised, evidenced record showing exactly which transactions you claimed and why — rather than a reconstruction from memory.
The week-to-week reality
Driving pays weekly and costs daily. Fuel is bought before the fare is paid, the insurance goes out monthly regardless of how the week went, and the finance payment does not care that the airport was quiet.
That mismatch is why drivers who are earning perfectly well still end up taking an instant cash-out on a Wednesday. It is a timing problem, not an earnings problem, and it is invisible in an annual total.
A structured weekly series makes it visible: money in by week, money out by week, and the balance between them. Seeing the pattern is the first step to changing it — usually by shifting a payment date rather than by driving more hours.
A year of statements in one pass
Driver bookkeeping happens in one panicked session before a filing deadline, which is exactly when converting twelve PDFs individually is least tolerable.
Batch processing takes up to 100 statements at once and merges them into a single sheet, with duplicate detection for overlapping periods and a source-file reference on every row — so the whole year arrives as one sortable dataset.
Drivers who switched banks mid-year, or who keep an app account for payouts and a high-street account for everything else, get both consolidated into the same table.
The card the fuel goes on
Much of a driver's cost never touches the current account. It goes on a card: fuel, charging, tolls, car washes, the phone plan, the parking apps.
Card statements convert exactly the same way and merge into the same dataset, which matters because a driver's expenses reconstructed from the current account alone will be missing most of the actual spend.
See credit card statement conversion for the card side specifically — the two together are what make an expense total complete rather than indicative.
Straight into QuickBooks or Xero
If you run accounting software, or your accountant does, the converted statement goes straight in. FlowParse produces real bank-feed files — QBO, QFX and OFX — with a transaction ID per row, so a re-import does not double-post, and a Xero-ready CSV with the columns Xero expects.
That matters most for the accounts a feed does not reach: an app account, a closed bank, the months before you registered. Those are exactly the gaps that otherwise get typed in by hand at the worst possible moment.
Scanned and photographed statements too
Not every statement arrives as a tidy download. Older accounts post paper, and plenty of drivers have a year that exists only as photographs taken on a phone in a hurry.
OCR runs first on scanned and photographed statements, then the AI structures the recognised text and flags low-confidence figures for a quick check — so a phone photo becomes the same clean rows as a downloaded PDF.
Numbers you can file on
Around 98% field-level accuracy on standard layouts, with every low-confidence figure highlighted in an editable preview before anything exports.
And the statement proves itself: opening balance plus every transaction should equal the closing balance the bank printed, and FlowParse checks that every time. A missing payout or a duplicated fuel charge is caught arithmetically rather than being discovered by a tax authority. That is what validation means here — evidence, not a confident-looking table.
Your account, your data
A driver's statement is a personal document — your earnings, your rent, your life. Uploads run over TLS, processing is EU-hosted, the original PDF is deleted immediately after processing, and documents are never used to train AI models.
Nothing is retained once your export is produced.
Who this is for
Drivers filing their own return who want the turnover figure to be right, drivers running two or three apps who want to know which one actually pays, drivers deciding whether to buy or rent the next car, and the accountants who serve gig clients and would rather not receive twelve PDFs in March.
If you cannot currently say what you earned gross last year, or what a mile costs you, converting the statements is the step that produces both.
Convert your driving bank statements
Upload a statement and get clean, categorised rows — gross bridged to net, fuel and finance separated, cost per mile finally computable.