What is a quality of earnings review, and how do you prepare for one?

The term sheet is signed and everyone is relieved. Then the investor's lawyers introduce an accounting firm that will carry out financial due diligence, including a quality of earnings review, and a data request arrives with well over a hundred lines. Monthly trial balances for three years. Customer-level revenue. Bank statements for every account. Payroll registers. GST returns. Contracts. A list of every one-off item in the P&L.

For a founder who hasn't been through it, the request feels like an audit, and in some ways it's harder. An auditor asks whether the accounts follow the accounting rules. A quality of earnings review asks a more commercial question: if we're paying for this company's earnings, or its growth in them, how much of what's reported will actually continue?

It's common in acquisitions and later-stage rounds, and increasingly shows up in growth rounds and some Series A processes, particularly where investors are pricing on revenue quality or a path to profit. Even when a full review isn't commissioned, the questions it asks show up in some form in almost every serious diligence.

What the review actually looks at

Firms structure their reports differently, but the core work tends to cover the same ground.

• ·Adjusted EBITDA. Reported EBITDA with one-off, non-recurring and non-market items removed or added back, so the number reflects the ongoing business.
• ·Proof of cash. Revenue and costs in the books reconciled to money actually moving through the bank, month by month.
• ·Revenue quality. How much is recurring and contracted, how concentrated it is, how it's recognised, and whether it can be traced from contract to invoice to cash.
• ·Run-rate costs. What the cost base looks like with the full-year effect of recent hires, price changes and contracts, rather than the historical average.
• ·Working capital. What normal levels of receivables, payables and inventory look like, and whether recent months have been unusually flattering.
• ·Debt-like items. Obligations that aren't borrowings but will cost cash: unpaid statutory dues, unfunded gratuity, customer advances, deferred revenue, disputed tax.

The output is usually a report that restates earnings and net cash on an adjusted basis, lists the issues found, and quantifies each one. Investors use it to confirm the valuation, adjust it, or add protections to the legal documents.

How adjustments work, worked through

Adjustments are where most of the negotiation happens, so it's worth seeing how they move the number. This example is illustrative, with an invented services company that reports EBITDA of ₹1.40 crore for the last twelve months.

• ·A legal settlement with a former co-founder cost ₹18 lakh during the year and won't recur. The review adds it back: plus ₹18 lakh.
• ·The two founders each take ₹12 lakh a year less than it would cost to hire someone to do their jobs. An investor paying for the business assumes those roles will eventually be paid at market, so the review deducts the difference: minus ₹24 lakh.
• ·An annual prepaid contract worth ₹22 lakh was recognised in full when invoiced four months before year end. Only four months of it was earned: minus ₹14.7 lakh.
• ·No gratuity provision has ever been booked. The review estimates the liability built up in the year: minus ₹9 lakh.
• ·The company calls its annual customer conference a one-off. It has happened every year for four years. No adjustment.
• ·A key account manager left and wasn't replaced for five months, saving ₹8 lakh the business will spend again once the role is filled: minus ₹8 lakh.

Adjusted EBITDA comes out at about ₹1.02 crore, roughly 27% below the reported figure. Nothing in the original accounts was dishonest. But if the deal were priced as a multiple of earnings, as acquisitions often are, the gap would be worth several crore. And even in a growth round priced on revenue, a finding that revenue was recognised early changes the growth rate the investor believed they were buying.

Notice which way most adjustments went. Founders tend to think of adjustments as add-backs that improve the number. Experienced reviewers find at least as many that reduce it. A company that arrives having already booked its gratuity provision and spread its prepaid revenue properly removes two of those findings before the review even starts.

Proof of cash: the test startups fail first

Proof of cash is simple to describe. Take revenue in the books for each month, adjust for movements in receivables, GST collected and TDS deducted by customers, and check that the result matches customer money landing in the bank. Do the same for costs and payments.

It's where problems surface quickly. Revenue booked for a customer who never paid. Refunds and chargebacks processed through the payment gateway but never recorded. Cash from a founder's personal account deposited and treated as a customer receipt. A related company paying some expenses. Gateway settlements netted with fees, so revenue is understated in one place and costs are missing in another.

None of these needs to be large to matter. A proof of cash that doesn't reconcile for several months tells the reviewer the books can't be relied on, and everything else in the review then takes longer and gets examined harder.

What's particular about Indian companies

Several items come up again and again in reviews of Indian startups, and most of them are about tax and statutory compliance rather than the business itself.

GST is the biggest. Reviewers compare output GST in the returns with revenue in the books, and input credit claimed with what suppliers have actually reported. Gaps become potential liabilities with interest. TDS is next: the TDS receivable in the books should match what customers have reported against your tax account, and TDS you were supposed to deduct and deposit on vendor payments should have been.

Statutory employee costs matter too. Unpaid PF or ESIC, contractors who look like employees, and gratuity that has never been provided for all turn into debt-like items that come off the price or go into an indemnity.

And related party transactions get close attention: payments to founders' other companies, rent to a family member, services bought from an investor's portfolio company. They're often perfectly legitimate. They need to be at market rates, approved properly and disclosed.

When the company isn't profitable yet

Most startups in a Series A or B process don't have earnings in the usual sense, so it's reasonable to wonder what a quality of earnings review is for. The answer is that the same questions get asked of a different number.

Instead of adjusted EBITDA, the focus moves to adjusted burn and the quality of revenue growth. Is the monthly burn the investor has been shown a fair run-rate, or did a few months of delayed hiring and deferred vendor payments flatter it? Is gross margin calculated with every cost of delivery included? Does recurring revenue really recur, and does it reconcile to contracts and cash? Would contribution margin look the same if discounts and free months were properly reflected?

The adjustments work the same way. A one-off legal cost comes out of burn. A vacancy that saved money for five months goes back in. Annual prepaid revenue gets spread across the months it was earned. The result is a cleaner picture of how much money the business really consumes to produce the growth the investor is paying for, which in a growth round is often the single most important number in the valuation.

How to prepare, starting six months out

The best preparation looks like good finance hygiene, done early and consistently.

1. 01Close the books monthly, on time, with every bank account, gateway and wallet reconciled.
2. 02Write down your revenue recognition policy and check that every significant contract follows it, particularly annual prepaid plans, milestones and implementation fees.
3. 03Build a customer-level revenue file that ties to invoices, to the books and to the bank.
4. 04Reconcile GST input credit against supplier filings, and output GST against revenue, every month.
5. 05Reconcile TDS receivable against what customers have reported, and check your own TDS deductions on vendors.
6. 06Book provisions you've been putting off: gratuity, leave encashment, bonuses, and expenses for bills that haven't arrived.
7. 07Prepare your own list of adjustments with evidence: one-offs, founder salaries, unusual months. Arriving with an honest list is far better than having one built for you.
8. 08List every related party and every transaction with them, with approvals.
9. 09Put contracts, board approvals, ESOP records and statutory filings into a data room before anyone requests them.

If that list looks long, it's because most companies do none of it until the request arrives, then try to do all of it in three weeks while also running the business.

During the review

Appoint one person to own the process on your side, usually whoever runs finance, with the founder available for commercial questions. Answer in writing, with evidence, and keep a log of every question and answer so nothing is asked twice or answered inconsistently.

Don't argue with findings before you understand them, and don't offer explanations you can't support. Reviewers are trained to notice when an answer changes. If something is genuinely wrong, say so, quantify it and explain what's been fixed. That's a far stronger position than defending a number that later falls apart.

And keep the business running. Reviews take weeks, and a company whose month-end close slips because everyone is answering diligence questions creates new problems for the next data request.

When the findings come back

Findings rarely kill a deal on their own. More often they change its shape. A valuation adjusted to reflect adjusted earnings. A portion of the price held back until a tax matter is resolved. Specific indemnities for statutory liabilities. Conditions to fix certain issues before completion.

Your negotiating position depends almost entirely on evidence. An adjustment you can show is wrong, with documents, usually gets reversed. An adjustment you simply disagree with usually doesn't. This is where having done your own reconciliations and your own adjustments list months earlier pays for itself, and where your lawyers and tax advisers need to be closely involved in how findings turn into contract terms.

The real point of preparing

A quality of earnings review is uncomfortable because it looks at a company the way an outsider with money at stake would. But the questions it asks are the same questions a founder should be asking anyway. How much of our profit or growth is real and repeatable? Do our books match our bank? What do we owe that isn't on the balance sheet?

Companies that can answer those questions calmly tend to raise faster and on better terms. More importantly, they tend to make better decisions in the years before anyone arrives with a data request.

Who wrote this

Shafneed is the founder of Simplify, a finance clarity and investment readiness practice working with founders across India. He writes about the questions founders bring before a decision, not after it.