Financial diligence
Revenue reconciled to cash, month by month.
Earnings, debt and working capital rebuilt from bank data, and from tax-return data where the market provides it, both pulled with the company's consent. Each month is matched to the ledger. Your committee reads a reconciliation instead of management's EBITDA (earnings before interest, tax, depreciation and amortisation).
Funds and family offices
Term sheet signed, exclusivity running. The deck says 40% growth. The tax returns may say something else, and the committee will ask which one is cash.
Private credit lenders
You are sizing a facility on stated turnover. You need verified inflows, existing debt service, and the month the borrower's cash actually turns.
Founders and CFOs
You are raising. Someone will tie your numbers out eventually. Better that you see the breaks first and explain them yourself.
Bank data and, where the market provides it, tax-return data are matched to the ledger for every month in scope. Where a market has no such feed, we use statements the company supplies. Each figure is tagged source-pulled (taken directly from the bank or tax portal) or company-supplied. No sampling. The summary memo goes to the deal partner or credit head. The workbook goes to whoever has to defend the number at committee.
Revenue in the books is 21% above bank customer receipts for the year. Two thirds of the gap is in the final month, on three invoices to one customer that shares an address with a promoter entity.
Illustrative finding. Not a real company.
What lands on your desk
Summary memo
Findings, adjustments and open questions, each tied to source rows. Short enough for the partner or credit head to read before committee.
Reconciliation workbook
For the associate and the CFO. Bank against ledger against tax data, month by month, live formulas, every variance listed with its rows.
Earnings and net debt
Reported to adjusted EBITDA, each adjustment sourced or marked unsourced. Debt from bank servicing and registered charges, set against the debt the management accounts show.
Evidence file
Each pulled document with its retrieval time and a hash, a digital fingerprint that shows it was not altered after retrieval.
How the work runs
Company consents
The company gives consent for bank data through the RBI Account Aggregator framework, and for tax data on the taxpayer's own authority where the market provides it. We never take passwords. The clock starts when the set is complete: consent given, data pulled, ledger export received.
Pull the source
Bank transactions, tax returns where available, the ledger export from the company's accounting software, and filed accounts from MCA21. Four sources for the same year, none of them prepared for this review.
Match month by month
Bank credits are split first: customer receipts, loan inflows, promoter money in, transfers between the company's own accounts, refunds. Only customer receipts are set against booked revenue and, where available, tax-return turnover, after allowing for tax collected, advances, and the gap between invoice and receipt. Each variance above threshold becomes a flag, listed with the source rows behind it.
Check counterparties
Top 20 customers and suppliers: whether counterparties file their GST returns, registry status, and links to promoters by director, shareholder or address. Net debt is rebuilt from bank servicing, charges and filed borrowings. Receivable and payable days are computed from the books and tested against bank timing.
Analyst review
A named analyst checks every flag against source rows, drops the false ones and writes the summary. A second reviewer reads the draft before release.
Revenue against cash
Filed accounts against bank customer receipts, month by month, for 12 months, and against tax returns where the market provides them. Every gap is shown with its size and the rows behind it.
Bank analysis
Inflows by counterparty. Round-number and same-day in-and-out patterns. Credits from related parties. Cash deposits. Returned cheques. Loan inflows kept apart from revenue, so borrowed money is never counted as sales.
Earnings bridge
Reported to adjusted EBITDA, each adjustment sourced. One-offs, capitalised costs, related-party pricing, and the items that cannot be sourced, which are listed and not guessed.
Debt and liabilities
Net debt from bank servicing, charges and filed borrowings. Guarantees and contingent items from the notes. Statutory dues in arrears where visible.
Working capital
Receivable and payable days from the books and, where the market provides them, tax returns, tested against bank timing. Ageing where supplied. Signs of stretch, such as payables days climbing in the quarter before a raise.
Concentration
Top 10 customers and suppliers by bank data and, where available, tax returns. Counterparties that are unregistered, inactive, newly formed, or that share an address or director with the company.
Cash conversion
Operating profit to operating cash, and the gap explained or flagged.
Red-flag register
Each item ranked by size and confidence, with its source.
Questions for management
A ready list, in the order that matters, each with the figure behind the question.
Extension by agreement: 24 months, journal-entry testing, cohort and unit economics, tax and capex review.
Sign and pay. Send the company a consent link or a request for statements.
Gives one consent link where bank data is automatic. Otherwise sends statements and accounts. Tax returns are added where the market provides them. The clock starts when the set is complete.
Data is cleaned and matched: accounts against bank credits, and against tax returns where available.
A named analyst works through every break and adjustment. A second reviewer reads the draft.
Pack delivered with every figure tagged to its source. If we miss the 2 to 5 days, the fee for that pack is reduced.
Usually (our estimate)
Usually 7 to 8 days: request lists, wait for documents, chase, then reconcile by hand on a sample of months.
Kubr
2 to 5 Business Days from a complete data set.
Typical times are targets we are still testing in the pilot. The 2 to 5 day window is the commitment. Terms in Disclosures.
What the cases taught us
Bank inflows that did not match the books would have shown up in many of the 483 cases. Revenue inflation is in many of them.
Fake or shell customers and suppliers appear in many cases. Checking whether counterparties file their GST returns is quick and mostly automatic.
Operating-metric claims that could not be verified feature in many cases, and unit-economics failure in many of them. We test the metric against a source, and not against the deck.
Counts come from our first review of the public cases in the casebook on this site. We mapped each case, on hindsight, to the checks that could have surfaced it. It is our reading, not a finding about any company. Many cases describe allegations that are contested. See the cases
ChecksWhat we check, in plain words
| Check | What it means | An illustration (invented) | It can mislead when |
|---|---|---|---|
| Bank credits vs books | We compare the money that actually moved through the bank with the sales and costs in the books, month by month. | The books show ₹4 crore of sales in a quarter. Bank credits from customers add up to ₹2.5 crore. | AA only sees the accounts the company links, so an undisclosed account stays invisible. Cash-heavy businesses and timing differences create false flags. |
| Counterparty tax / GST | We look up the main customers and suppliers in the tax system and see whether they are active, filing, and report the same trades. | A customer the company calls its largest has not filed a return in a year. | Needs taxpayer consent or GSP access. A registered, filing shell still passes this check, which is why C3 and C4 run beside it. |
| Ownership / related-party links | We look for hidden ties between the company's counterparties and its owners, through directors, shareholders, addresses, phones and auditors. | A supplier paid ₹3 crore shares a registered address with the promoter's brother. | Nominee directors and offshore layers hide links. Name matching throws up false positives, so every flag shows its evidence path. |
| Counterparty existence | We check that each material customer, supplier or payee is a real, working business. | A vendor with ₹6 crore of invoices turns out to be a one-room office with no staff. | Real companies can be used as pass-throughs. Existence does not prove a real trade, so C6 and C7 still apply. |
| Document authenticity / direct pull | We get statements, returns and certificates from the issuer, not from a PDF the company sends. | A bank statement sent by the company shows a balance the bank does not hold. | The company can refuse consent, and a refusal is itself a flag. Some banks and co-op banks are not on AA yet. |
| Unit economics & burn | We rebuild margin per sale, customer acquisition cost, payback and cash burn from verified numbers, not from the deck. | The deck shows a profitable order. Rebuilt from invoices, each order loses money. | Cost allocation is a judgement call. Kubr shows its assumptions and lets the user change them. |
| Operating-metric verification | We test claimed headcount, stores, vehicles, units and users against independent traces. | The company claims 800 staff. Provident fund filings show 140. | Contract staff do not appear in EPFO. Vahan access needs licensed APIs. Some businesses have no physical trace. |
| Books edit log (Ledger DNA) | We check whether the accounting software shows entries backdated or edited after a period closed, especially just before a raise or a loan renewal. | A burst of edits to last year's sales a week before the investor meeting. | Logs can be disabled at database level (auditors qualified this in year one) or tampered with on export. Kubr needs the company to cooperate. |
| Drawing power vs books | We compare the stock and receivables statements the company gave its bank with its own books. | The bank was told stock was ₹20 crore. The books say ₹12 crore. | CARO is annual, so the signal comes late. It covers only borrowers above the ₹5 crore limit. |
What it will not tell you
- Bank and tax checks need the company's consent. If it is refused, they do not run, and the refusal goes in the memo.
- Cash that never touched a bank is out of view. Cash-heavy businesses get a stated coverage limit.
- We see the accounts we are given. Any account that appears in charge documents or ledger codes but was not supplied is listed by name.
- We do not audit, attest or certify, and we do not value the company. Adjusted EBITDA here is a sourced bridge from management's figure, not an opinion on the accounts.
Not in this service: audit, value the company, count stock or give an opinion.
What it can do
- Tie bank customer receipts to booked revenue for every month in scope, and to tax-return turnover where the market provides it.
- Find money with no entry in the books, and entries with no money behind them. The second kind clusters in the last month of the year.
- For top customers and suppliers, check whether counterparties file their GST returns, and flag those with no trace of operations.
- Compute receivable and payable days and test them against bank timing. Read the accounting-software edit log, where the software keeps one, for backdated entries, with the company's cooperation.
Why it differs from the usual way
Usually
A reviewer works from bank PDFs and a ledger the company chose to send.Kubr
Where the market allows, we pull bank data at source with consent and tag every figure source-pulled or company-supplied. A PDF can be edited before it is sent.Usually
Bank-to-books work runs on a sample of months or of large items.Kubr
Every month and every transaction in scope is tied across bank, ledger and tax data. Revenue inflation tends to sit in the months nobody sampled.Usually
Counterparties are taken as named on the invoice.Kubr
Top counterparties are checked for existence and for links to promoters. A registered, filing customer that shares a director with the seller is a finding.Records we examine
| Record | What we take from it |
|---|---|
| Bank data, with consent (the RBI Account Aggregator framework) | |
| Tax-return data, where the market provides it (with consent) | |
| Ledger export (company-supplied) | |
| Accounting-software edit log (with consent) | |
| Company registry (MCA21) | |
| Charge and security registrations (Form CHG-1 charge registrations) | |
| Public employment records, where they exist |
Some records need the company's consent. The report says which were pulled from source and which the company supplied.
In India
- What we pull here
- GST return checks (GSTR-1, GSTR-3B, GSTR-2B), e-invoice data, bank data through an Account Aggregator, filed accounts (AOC-4), and income tax records (ITR, 26AS, AIS) where the company uploads them.
- What is thinner here
- Consent-based layers wait for the company. Under the e-invoice threshold there is no e-invoice layer.
What it costs
Pick the size of the deal or exposure. The fee is fixed before work starts.
₹2 lakh
Above the top band the fee is quoted individually. Indicative base, in local currency, before tax. All services and what moves the fee
Shape the scope
Time horizon
Entities
Speed
The fee shown for your deal size is an indicative base. A longer look-back, more entities or a faster turn moves it. The quote fixes the number before work starts.
Questions
What if the company won't consent?
Bank and tax checks do not run. We can work on company-supplied statements, tagged as such, and the report records the refusal as a finding.
Do you need portal passwords?
No. Where a tax feed exists, it comes through an authorised channel on the taxpayer's own approval. We never ask for passwords.
How many months do you cover?
Twelve months is standard. Twenty-four is an extension. Twelve shows the pattern. It rarely shows seasonality in a business with one large annual cycle, so say so at scoping.
Is this a quality of earnings (QoE) report?
It covers the parts of one that can be tied to source. The 24-month extension goes further. We do not issue an accountant's report. If your committee or lender requires one signed by an accounting firm, this does not replace it.
Often paired with
Start with one company.
We reply within one business day with scope, price and a date. Indicative prices in INR, before GST at 18%. Scope and price are fixed in the engagement letter.