How to Spot a Fake or Altered Bank Statement (and Verify the Real One)

Bank statements are among the most requested documents in finance. Landlords use them to judge affordability, lenders to verify income and assets, employers and immigration offices to check funds, and accountants to support records. That makes them a target for fraud. Editing tools, templates and online "statement generators" make it easy to produce a convincing fake, and altered statements are a known component of loan fraud and rental fraud.

This guide is written for people who receive statements and need to rely on them: underwriters, property managers, accountants, auditors, HR teams and family law professionals. It covers the red flags, the checks that catch most fakes, and the verification steps that remove the need to trust a document at all. It deliberately does not explain how fakes are made.

Types of bank statement fraud

Understanding what fraudsters do helps you know where to look:

• Altered genuine statements. A real statement with edited figures: inflated deposits, removed withdrawals, changed balances, or a different name or address. This is the most common type, because most of the document is authentic.
• Fabricated statements. Documents created from scratch using templates, often imitating a large bank's layout.
• Borrowed statements. Genuine statements belonging to someone else, with the name changed or used outright.
• Selective statements. Genuine but incomplete: missing pages or months that would show overdrafts, gambling, other loans or returned payments.
• Window dressing. Genuine statements showing a temporary deposit, borrowed for the purpose, to inflate the balance on the statement date.

Fraud is also not limited to applicants. Insiders can alter statements to hide misappropriation from employers or auditors, which is why auditors often request statements directly from banks rather than from management.

Each type calls for different checks. Arithmetic and formatting checks catch alterations and fabrications; cross-checks and direct verification catch borrowed, selective and window-dressed statements.

Check 1: The arithmetic

Genuine statements are produced by bank systems, so the numbers always add up. Edited statements frequently do not, because changing one figure correctly requires changing many others consistently.

Summary check. Opening balance + total credits - total debits = closing balance. Recalculate totals from the transaction list rather than trusting printed totals.

Running balance check. If the statement prints a balance after each transaction, recompute it row by row: previous balance plus credit or minus debit. The first row where your figure and the printed figure differ is where an edit is likely.

Continuity check. Each statement's opening balance should equal the previous statement's closing balance. A gap suggests a missing statement or a fabricated sequence.

Page totals. Some statements carry balances forward from page to page. These must agree too.

Doing these checks by hand is tedious for long statements. Converting the PDF into structured data makes them instant. Our balance reconciliation check recomputes every running balance on upload and flags the exact rows that break, which is the same test a careful analyst would run.

Worked example: a running balance that gives the game away

Consider these rows from a statement submitted with a rental application:

Date	Description	Paid out	Paid in	Balance
01 Mar	Balance brought forward			1,204.18
03 Mar	Card purchase, supermarket	62.40		1,141.78
05 Mar	Salary, employer name		4,850.00	5,991.78
06 Mar	Rent standing order	1,150.00		4,841.78
10 Mar	Card purchase, fuel	48.20		3,793.58

The first three movements are consistent: 1,204.18 - 62.40 = 1,141.78, plus 4,850.00 = 5,991.78, minus 1,150.00 = 4,841.78. The fuel purchase is where it breaks: 4,841.78 - 48.20 should be 4,793.58, not 3,793.58. The difference is exactly 1,000.00.

The most likely explanation is that the salary was edited from 3,850.00 to 4,850.00 and the person editing updated the balances near the change but not the later ones. A check of the employer's payslip, which shows net pay of 3,850.00, confirms it. Without the running balance check, the edit would have been easy to miss, because the document looked perfectly normal.

Check 2: Visual and formatting red flags

Look at the document at full size and zoomed in. Common signs of editing:

1. Inconsistent fonts. Digits in a different typeface, weight or size from the surrounding text, especially in amounts and balances.
2. Misaligned numbers. Amounts that do not line up with the column, or decimal points that wander.
3. Uneven spacing between characters in names, addresses or amounts.
4. Blurry or pixelated areas around specific numbers on an otherwise crisp document.
5. Logo problems: low resolution, wrong colours, outdated branding or slightly wrong proportions.
6. Spelling and grammar errors in headers, disclosures or the bank's own text.
7. Wrong or missing details: no page numbers, no statement period, missing sort code, routing number or IBAN where the bank normally prints them.
8. Round numbers everywhere, particularly in salary deposits, which in real life are usually net of tax and deductions and rarely round.
9. Too-perfect regularity: every deposit on exactly the same day, every bill the same amount.
10. Formatting that does not match the bank, such as date formats or column names the bank does not use. Our bank directory describes typical statement layouts and balance labels for over a hundred banks.

None of these proves fraud by itself. Banks redesign statements, scanned copies look blurry, and some salaries really are round. Visual flags tell you where to look harder.

Check 3: Content plausibility

Read the transactions as a story. Do they make sense for the applicant?

11. Income without tax. A salary deposit equal to the gross salary stated on an application suggests either a misunderstanding or an edit.
12. Missing everyday spending. A real current account shows groceries, transport, utilities and subscriptions. A statement with only a salary in and rent out is unusual.
13. Employer name mismatch. Payroll descriptions should match the employer named in the application.
14. Large unexplained deposits shortly before the statement date, which may be window dressing or an undisclosed loan.
15. Dates that do not exist (31 June), or transactions on bank holidays where the bank would not process them. Card transactions can occur on any day, but some transfer types only settle on business days.
16. Transaction codes that do not belong to the bank's country. A UK statement with US-style ACH codes, or the reverse, is suspicious. Our bank statement abbreviations reference lists codes by country.
17. Merchant descriptors that look invented, or local merchants in a city the applicant does not live in.

Check 4: The PDF file itself

Electronic statements downloaded from online banking are generated by the bank's systems. Their file properties can reveal tampering.

18. Document properties. In a PDF reader, open the document properties. Look at the creator and producer fields, and the creation and modification dates. A statement whose producer is a consumer editing tool, or whose modification date is long after the creation date, deserves scrutiny.
19. Fonts embedded in the file. The list of fonts can show a typeface that does not appear in the bank's genuine statements, used only for a few edited characters.
20. Text layer anomalies. In a genuine electronic statement, you can select and search text. If some numbers are images while the rest is text, or selecting a figure reveals a different value underneath, the document has been altered.

Two caveats are important. First, metadata can be edited too, so clean metadata does not prove authenticity. Second, legitimate actions such as printing to PDF, compressing or scanning change metadata. Use these checks as signals, not verdicts, and always request the original download if you have doubts. A genuine applicant can almost always provide one within a day.

Check 5: Cross-checks against independent evidence

The most effective document checks compare the statement with something the applicant did not create:

• Payslips and payroll records: net pay on payslips should match deposits.
• Tax returns and tax authority records: annual income should be consistent with statement deposits.
• Employer verification: confirming employment and pay directly with the employer.
• Previous statements and references: a prior landlord's records of rent paid, for example.
• Invoices and contracts for self-employed applicants: client payments should match invoiced amounts and client names.
• Credit reports: loan and card payments on the statements should match accounts on the credit file, and vice versa.

Inconsistencies between these sources are far stronger evidence than any single visual flag.

Check 6: Verify directly with the source

The most reliable approach removes the applicant from the data path:

• Bank-connected verification. In many countries, open banking or account aggregation services let the applicant authorise read-only access to their account data, which arrives directly from the bank. Many lenders and letting agents now use these services as standard.
• Verification of deposit. In some markets, lenders send a formal verification request to the bank, with the applicant's written authorisation, asking it to confirm balances and account history.
• Direct download in your presence. For smaller transactions, asking the applicant to log in and download the statement while you watch, or to forward it directly from the bank's secure messaging, adds assurance.
• Calling the bank using a phone number you find independently, not one printed on the document, to confirm details the bank is permitted to confirm with the customer's consent.

Direct verification is not always available or proportionate, but for high-value decisions it is the standard of care.

Training staff to spot fakes

Most fake statements are caught by people who know what normal looks like. A short internal training programme pays off quickly:

• Build a reference library of genuine statements from the banks you see most often, with personal data redacted, so staff can compare layouts.
• Share real examples of detected fakes, again redacted, and walk through how they were caught.
• Teach the arithmetic checks first; they are objective and need no specialist knowledge.
• Set clear escalation rules so junior staff know when to pass a file to a senior reviewer instead of making a judgement call alone.
• Review the process annually, because fraud techniques and bank statement designs both change.

A practical verification workflow

For organisations that review many statements, a tiered approach balances effort and risk:

Tier	When	Checks
1. Basic	Every statement	Completeness (all pages, all months), arithmetic and running balances, name and address match, obvious visual flags
2. Enhanced	Any tier 1 flag, or above a value threshold	PDF properties and fonts, plausibility review, cross-check with payslips or invoices
3. Direct	Any tier 2 concern, or high-value decisions	Bank-connected verification, verification of deposit, or employer and bank confirmation

Document the checks performed for each file. If a decision is later challenged, a record of a consistent process is valuable.

Converting statements to structured data makes tier 1 fast. Upload the PDF, let the reconciliation check recompute balances, and review the flagged rows and summary. Our income verification and fraud red flags pages describe these workflows, and lenders use the same approach at volume.

Beyond bank statements: other documents that get faked

The same principles apply to the other documents commonly submitted alongside bank statements.

• Payslips. Check that gross pay, deductions and net pay add up, that tax and social contributions are plausible for the salary, that year-to-date figures progress correctly from one payslip to the next, and that net pay matches the bank deposit.
• Employment letters. Verify the employer exists, find its contact details independently, and confirm the letter with the employer.
• Utility bills used as proof of address. Look for consistent account numbers across bills, plausible consumption, and a provider that actually serves that address.
• Invoices from the self-employed. Match invoice amounts and client names to deposits on the statements, and check the invoice numbering sequence for gaps or implausible jumps.
• Tax documents. Where possible, obtain transcripts or records directly from the tax authority with the applicant's consent rather than relying on copies.

In every case, the pattern is the same: check internal consistency, check consistency with other documents, and verify with the issuer when the stakes justify it.

What to do if you suspect a fake

• Do not accuse. There may be an innocent explanation, such as a scan that looks odd or a bank's new layout.
• Ask for clarification or the original. Request the statement downloaded directly from online banking, or use a bank-connected verification method.
• Escalate internally according to your organisation's fraud policy.
• Keep records of the documents received, your checks and your communications, including the original files exactly as submitted, so that any later investigation can examine them.
• Report where required. Regulated firms may have obligations to report suspected fraud; others can report to the police or the relevant fraud reporting service.
• Do not alert the subject if your obligations require confidentiality, such as suspicious activity reporting by regulated businesses.

A note for applicants

If you are providing statements, provide complete, unaltered documents, every page, for the periods requested. Do not remove transactions you would rather not explain; explain them instead. Altering a bank statement to obtain a loan, tenancy or benefit can be a criminal offence, even if you could have afforded the commitment. If a statement contains information you consider irrelevant and sensitive, discuss it with the requester rather than editing it.

Frequently asked questions

What is the quickest way to detect an edited bank statement?

Recalculate the balances. Check that opening balance plus credits minus debits equals the closing balance, and recompute every running balance row by row. Edited statements often break at the row that was changed.

Can PDF metadata prove a bank statement is fake?

It can provide strong evidence, for example if the producer is an editing tool or the file was modified long after creation, but it cannot prove authenticity on its own. Metadata can be altered and legitimate actions also change it. Use it alongside other checks.

How do lenders verify bank statements?

Lenders combine document checks (arithmetic, formatting, plausibility) with independent evidence such as payslips and tax records, and increasingly with direct verification through bank-connected services or verification requests sent to the bank with the applicant's consent.

Is it illegal to edit a bank statement?

Submitting an altered bank statement to obtain credit, a tenancy, employment or a benefit can amount to fraud or forgery in most jurisdictions. Even small edits can have serious consequences.

Can software detect fake bank statements?

Software helps by automating arithmetic checks, recomputing running balances, extracting transactions for plausibility review and flagging anomalies. It does not replace independent verification. StatementPilot's reconciliation check identifies rows where printed balances do not follow from the transactions, which is one of the strongest signals of editing.

What should I ask for if I doubt a statement?

Ask for the original electronic statement downloaded directly from online banking, or for consent to verify through a bank-connected service. Genuine applicants can usually provide these quickly.

Summary

Fake and altered bank statements are caught by layering checks: arithmetic and running balances first, then visual and file-level red flags, then plausibility and cross-checks against independent records, and finally direct verification with the bank. Automating the first layer frees time for the judgement calls. Try StatementPilot free to recompute balances on any statement in seconds, and read how to read a bank statement for a refresher on what genuine statements contain.