NatWest Statement to Excel Converter
Convert NatWest bank statement PDFs into clean Excel spreadsheets. FlowParse reads the NatWest layout — the Date, Type, Description, Paid in, Paid out and Balance columns — and rebuilds every transaction as an editable, signed row. NatWest personal current accounts, Select, Reward, and NatWest business banking statements are all supported.
Scanned or digital · multi-page · balance-validated · duplicates removed
NatWest gives you a PDF; bookkeeping, Self Assessment and loan applications want a spreadsheet. Instead of retyping by hand, upload the statement, let the AI extract it in seconds, review the editable preview, and export to Excel, CSV or a QuickBooks-ready file — with balances validated so it reconciles before export.
AI extraction, no templates
Reads this bank's layout with no per-bank setup.
Every column preserved
Date, description, amount, balance and every extra field, 1:1.
Balance-validated
Opening + transactions = closing, checked on every file.
Scanned & multi-page
OCR for image PDFs; long statements stitched into one list.
What FlowParse extracts from a NatWest statement
A NatWest statement carries Date, Type, Description, Paid in, Paid out and Balance columns. FlowParse reads every row and extracts each field, merging paid in and paid out into a single signed amount while keeping the type code and description.
The balance is captured per row so the engine can confirm the statement reconciles from opening to closing across all pages.
| Field | Captured in the export |
|---|---|
| Date | Transaction / posting date, normalised |
| Description | Payee, merchant or transfer detail, verbatim |
| Amount | Single signed value (debits −, credits +) |
| Balance | Running balance per transaction |
| Reference / type | Check number, transaction type, reference |
NatWest statement formats and account types
NatWest, part of the wider NatWest Group, uses a consistent layout across personal current accounts (Select, Reward) and business banking, with type codes distinguishing payment kinds. Credit-card statements use a charges-and-payments format.
FlowParse reads all of them without setup, identifying columns by meaning so each statement yields the same clean structure.
New to this? Read the guide on how to convert a bank statement PDF to Excel.
Why NatWest PDFs are hard by hand
Separate paid in and paid out columns, type codes and a running balance make manual re-keying slow and error-prone — one wrong figure breaks the reconciliation.
FlowParse reads the whole statement at once, signs amounts correctly, and validates the balance, turning the job into a 30-second conversion.
How a NatWest statement is laid out on the page
NatWest uses the standard UK grid: Date, Type, Description, Paid in, Paid out and Balance, as one continuous table rather than a set of named blocks.
Direction is carried by which of the two money columns a figure occupies, not by a sign on the number, and on most rows one of the two is empty. That is the single most important thing to preserve during extraction — merge the columns and every amount survives while direction does not, producing a file of entirely plausible numbers and a reconciliation out by exactly twice the affected total.
The Balance column gives a per-row anchor, which makes the row-to-row check available: each balance minus the previous one should equal that row's signed amount, pinning a misread figure to an exact row. The Type column carries a short code for the payment method, and it is worth keeping as its own field because it distinguishes transactions the description alone cannot separate.
FlowParse merges Paid in and Paid out into one signed amount, keeps the Type code distinct, excludes the Balance column from the transaction list, and runs both the row chain and the document-level check against the opening and closing balances NatWest printed.
Getting the statement out of NatWest
Log in to Online Banking at onlinebanking.natwest.com and select Statements & transactions from the main menu. NatWest's own guidance describes downloading recent or historic statements going back up to seven years, with a PDF version shown in Online Banking as soon as a statement has been issued, which you can save and print.
Seven years is generous by international standards, and it depends on the account still being open. Once an account closes the self-service route goes with it at every bank, so if you are switching, closing a business account or winding a company up, download the full history before the closure completes rather than afterwards.
Verified against NatWest's own support-centre pages on accessing bank statements online, August 2026. Confirm the current path before relying on it — banks reorganise their apps without notice, and one of NatWest's older how-to URLs already redirects to a fallback page.
The statement as evidence, not just as data
NatWest makes a point on its own support pages that is worth repeating, because it changes how you should treat the file: a downloaded PDF statement can serve as proof of identity, address or income.
That is a different job from bookkeeping, and it explains why the PDF matters even once you have converted it to a spreadsheet. A lender, a landlord or a compliance check will accept the bank's own document; they will not accept a spreadsheet you produced from it, however accurate. The two artefacts are not interchangeable, and only one of them can be reissued by the bank.
So keep both, deliberately. The converted data is what you work with — filter it, categorise it, reconcile against it. The PDF is the evidence behind it, and it is what you hand over when somebody official asks. If the two ever disagree, the statement is what settles the question, which is precisely the reason for keeping the original unmodified rather than only the convenient derivative.
This also affects how you store a year. Filing only exported spreadsheets and deleting the source PDFs feels tidy and quietly removes your ability to evidence anything. Keep the documents, organised by account and period, and treat the spreadsheets as working files derived from them.
What usually goes wrong with a NatWest conversion
Four failures, each leaving output that looks entirely reasonable.
Paid in and Paid out merged into one column. Direction disappears, payments read as receipts, and the arithmetic misses by exactly twice their value.
The Type code folded into the description. Nothing looks broken and categorisation downstream gets quietly worse, because the most stable signal about what a transaction was has been buried in a text field that banks reformat without notice.
The Balance column imported as an amount. A position treated as a movement, inflating the file by the size of a balance — usually large enough to be noticed, which makes it one of the more forgiving errors here.
Rows lost at a page break. Column headings repeat at the top of each page, and where they are not recognised as repeated furniture either they enter the data or the genuine first rows of the page are dropped with them. The row-to-row balance chain cannot see this, because it closes over the gap and remains perfectly continuous; the document-level check is what catches it.
Using the statement to find what leaves the account every month
Once a year of NatWest transactions is in a spreadsheet, one review pays for the whole exercise: listing everything that recurs.
Recurring outflows are where money quietly leaks. Subscriptions nobody cancelled, services renewed automatically, insurance that rolled over at a worse rate, a standing order to a supplier you stopped using. None of it shows up as a problem in any individual month, because each amount is small and each looks exactly like it did last time. It shows up when twelve months are in one place and sorted.
The Type code is what makes the review quick, which is the practical reason for keeping it as its own column rather than folding it into the description. Filter to the payment types used for recurring instructions and you have the candidate list in seconds, rather than reading a year of descriptions looking for repetition.
The distinction between the two main recurring types matters when you act on the list. A standing order is an instruction you gave your bank, for a fixed amount, and you can change or cancel it yourself. A direct debit is an authority you gave the recipient to collect varying amounts, so cancelling it without telling them leaves an obligation unpaid and usually a failed payment. The type tells you which conversation you need to have.
Take a full twelve months rather than a sample. Annual charges — insurance, licences, professional subscriptions, domain renewals — appear exactly once, and a three-month extract misses every one of them. That is also why this review is worth doing from converted statements rather than from memory: the statements contain the things you have forgotten, which is precisely the point.
One caution on acting from the list: a payment that stops appearing is as interesting as one that keeps appearing. A direct debit that quietly failed, a standing order that lapsed when card details changed, a supplier who stopped collecting — each leaves a gap in the pattern rather than an entry, and gaps are invisible when you read a year looking for what is there. Sorting by counterparty and checking that each recurring item appears the number of times it should is the version of this review that catches both.
From PDF to clean data in three steps
1 · Upload the PDF
Drop one statement or many — digital or scanned, any number of pages.
2 · AI extracts & validates
Every transaction is read, amounts signed, balances checked, weak pages retried.
3 · Review & export
Check the editable preview, then download Excel, CSV, QBO, QFX, OFX or Xero.
Convert to the format you need
The same extraction powers every export — convert once, then choose the output your next tool expects.
| Format | Best for |
|---|---|
| Excel (.xlsx) | Analysis, totals, sharing |
| CSV | Importing into any tool |
| .QBO | QuickBooks Online & Desktop |
| .QFX | Quicken |
| .OFX | Most accounting tools |
| Xero CSV | Xero statement import |
Manual work vs FlowParse
Re-typing a statement by hand is slow and error-prone. FlowParse is near-instant, accurate, and scales to any volume.
| What happens | By hand | FlowParse |
|---|---|---|
| Reading the data | Copy-paste line by line | AI extracts every transaction |
| Scanned statements | Re-typed manually | OCR reads them automatically |
| Debits & credits | You fix every sign | Pre-signed, normalised |
| Balance check | Manual reconciliation | Validated automatically |
| Time per statement | 20–40 minutes | Under 30 seconds |
| A full year | Hours per account | One Smart Merge upload |
Import straight into your accounting software
Need it in your books? FlowParse builds a real bank-feed file. The .QBO imports directly into QuickBooks with no CSV column mapping, each transaction carries a unique ID so re-imports never duplicate, and Quicken users get .QFX.
Validated, not just converted
One extraction powers Excel, CSV, a real .QBO/.QFX/.OFX file that auto-imports into QuickBooks or Quicken without mapping, or a Xero-ready CSV — whatever your next step needs.
More on the validation and reconciliation engines.
Your bank statements stay private
Bank statements are among the most sensitive documents you own. Converting one never puts it at risk.
Encrypted transfer
Every upload and download runs over TLS.
Deleted after processing
Your original PDF is removed as soon as it's converted.
EU-hosted
Processed on EU infrastructure.
No AI training
Your documents are never used to train models. GDPR-aligned.
Read the full security overview.
Who converts these statements
Accountants
Bring client statements into Excel or QuickBooks without re-keying.
Bookkeepers
Catch up months of statements in minutes.
Small businesses
Get PDF-only statements into the books, ready for tax time.
Loan & mortgage applicants
Turn statements into a clean spreadsheet for underwriting.
Finance teams
Standardise statements from many banks into one format.
Anyone with a PDF statement
Skip the manual typing — convert and move on.
