Due Diligence: What It Is, What It Is Not, and How Deep It Goes

Due diligence is the verification a careful person carries out before committing to a decision: checking that what has been claimed is true, that what should exist does, and that nothing material has been left unsaid. Due diligence is neither research nor audit. Research forms a view; audit is a formal opinion. Its depth is sized to what is at stake and how reversible the decision is.

A decision is only as good as the facts underneath it, and the facts in front of a decision-maker are almost always supplied by someone who wants the decision to go a particular way. The seller describes the flat. The borrower describes the business. The distributor tips the share. None of them is necessarily lying, but each is interested, and an interested party's account needs checking by a disinterested one before money moves. The check a disinterested party makes is due diligenceThe care a reasonable person takes to confirm the facts before entering into a commitment. Diligence means careful, sustained effort; due means the amount the situation reasonably calls for.. Due diligence runs under four heads, at a depth sized to the decision, and a warehouse purchase of Rs 34,00,00,000 carries the whole exercise below, from first check to final decision.

What is due diligence, in one sentence?

Start with a household buying a second-hand scooter. The seller says it has done 18,000 kilometres, has never been in an accident, and is registered in his name. A careless buyer pays and rides off. A careful buyer looks at the odometer, checks the registration certificate against the seller's identity card, has a mechanic look at the frame, and asks why the front mudguard is a different shade from the rest. None of that is distrust exactly. A reasonable person does exactly that before handing over Rs 45,000 for something they cannot easily give back.

Due diligence is verification before commitment: the work of confirming, from sources the other side does not control, that what has been claimed is true and that nothing that matters has been left out. Hold on to the three parts of that. VerificationEstablishing that a statement is true by checking it against evidence independent of the person who made it, rather than by accepting it because it was said. means checking against evidence, not re-reading the claim more carefully. Before commitment means the work is done while the buyer can still walk away; a check performed after signing is an inspection, not diligence. And "nothing left out" means the exercise looks for the absent fact as much as the false one. The most expensive surprises are usually the things nobody mentioned rather than the things somebody misstated.

Aravalli Agro Foods, an invented listed maker of packaged snacks and staples, has a board decision in front of it that will carry the rest of this guide. Devika Rathore, the promoter and Managing Director, brings a proposal to buy a warehouse near the northern plant for Rs 34,00,00,000. The selling company is controlled by her cousin. The seller has supplied a description of the property, a valuation report and a copy of the title deed. Every one of those documents is a representationA statement of fact made by one party to another to induce a decision: this is the area, this is the title, this is the condition. A representation is what diligence checks; it is not itself evidence. by the seller. The board's diligence is what turns representations into verified facts, or shows that they cannot be verified.

How is due diligence different from research and from audit?

Three activities look alike from a distance because all three involve someone reading documents about a company. Research, diligence and audit are different jobs, done by different people, for different reasons, and confusing them is where most misuse of the word begins. Consider buying a house. Research is the months of reading about localities, price trends and what a household needs; it forms a view about what to buy. Diligence is what a buyer does about one specific flat once the decision to want it has been made: the encumbrance certificate, the approved plan, the society dues, the water pressure at seven in the evening. Audit is nothing the buyer does at all; it is a chartered accountant's formal opinion on the builder's accounts, produced for everyone, on a fixed cycle, to a published standard.

Research forms a view, diligence verifies before committing, and audit gives a formal opinion; the three answer different questions and none can substitute for another. Research asks "is this worth pursuing?" and produces a judgement. Diligence asks "is what has been claimed true, and is anything missing?" and produces a list of verified facts, open points and findings. Audit asks "do these accounts give a true and fair view?" and produces an opinion in a fixed form under professional standards. Each is done by a different party. Research is done by anyone with a view to form: an analyst, an investor, a household. Diligence is done by, or for, the party about to commit: the buyer, the lender, the board. Audit is done only by an independent statutory auditor appointed under law. So when a board member of Aravalli Agro Foods says "the accounts are audited, so we have done our diligence on the seller", two of the three have just been collapsed into one, and the one that was actually needed has not been done.

What does due diligence cover, in the usual heads?

Go back to the scooter for a moment and notice that the careful buyer's checks fell into four kinds without anyone planning it. Is it really his to sell (the registration certificate)? Is the price right (what similar scooters go for)? Does it actually work as described (the mechanic, the mudguard)? And who is this seller (why is he selling, does the story hang together)? The four kinds recur in every diligence exercise from a scooter to a factory, and practitioners give them names.

Diligence usually runs under four heads, legal, financial, commercial and people, and each head pairs a claim the other side has made with a check the other side does not control. Legal diligence asks whether the seller can deliver what is being sold: title, authority to sell, encumbrances, permissions, pending disputes. Financial diligence asks whether the numbers are right: the price against independent valuations, the taxes and dues attached, the accounts if a business is being bought. Commercial diligence asks whether the thing does the job claimed: the size, the location, the condition, the fit with what the buyer actually needs. People diligence asks who is on the other side of the table: who controls the selling company, what their record is, and whether any of them is connected to the buyer's own directors or promoter. For Aravalli Agro Foods' warehouse the four heads become four concrete tasks: a title search at the registrar, two independent valuations, a site visit with an engineer, and a check of who controls the selling company. Look at the figure below and notice that every right-hand box is something the seller did not write.

How deep is deep enough?

How much of this is enough is the question everyone silently asks. Nobody hires a lawyer to buy a Rs 45,000 scooter, and nobody buys a flat on a handshake. The household already knows the answer instinctively; the discipline just names it. Two things set the depth. First, how much is at stake, not only in rupees but as a share of what the buyer has: Rs 34,00,00,000 is 47 per cent of a year's profit before tax for Aravalli Agro Foods, and 3.6 per cent of a year's revenue, so it is a large decision for this company even though a much larger company would shrug at it. Second, how reversible the decision is. A machine that can be returned within 30 days can be checked lightly. The market will do the rest of the diligence: if it does not work, it goes back. A warehouse, once registered in the buyer's name, is the buyer's, roof and all.

The depth of diligence is set by proportionality: it rises with what is at stake and falls with how easily the decision can be undone, so a Rs 34,00,00,000 irreversible purchase earns far more than a Rs 3,40,000 reversible one. ProportionalityThe principle that the effort spent checking should match the size and consequences of the decision: heavy checks for large or irreversible commitments, light ones for small or easily undone ones. is why the same board can wave through a Rs 3,40,000 packaging machine on the purchase manager's say-so and insist on a full exercise for the warehouse without being inconsistent. Proportionality also explains why the scopeThe agreed list of what a diligence exercise will and will not examine: which heads, which documents, which questions, and how far back. Set before the work begins. Everyone then knows what "done" means. is agreed before the work starts: on the warehouse the board decided in advance that all four heads would be examined and that the exercise would end in a written sign-offA written statement, from a named person, that a defined set of checks has been completed and what they found. The sign-off fixes who verified what. Responsibility cannot evaporate later. from the company secretary listing what was verified and by whom. The position map below sets the two questions as its axes, and depth rises towards the upper right.

What is a red flag, and what follows from one?

Back to the scooter and the mudguard that is a different shade from the rest. The seller said the vehicle has never been in an accident. The mudguard says something was replaced. A repainted mudguard is not proof of anything; a mudguard can be replaced for a dozen innocent reasons. But it is an inconsistency between a claim and a fact, and a careful buyer does not proceed until it is explained. Perhaps the answer is a parking scrape, in which case the check has cost one question. Perhaps the answer does not come, in which case the buyer has learned the most useful thing diligence can teach: what the seller's story does not cover.

A red flag is an unexplained inconsistency between what has been claimed and what has been found, and its effect is to stop the process until it is explained and, if it holds, priced in. Three things about that definition earn their place. A red flagA finding during diligence that contradicts a claim, cannot be reconciled with the other facts, or reveals something the other side did not mention. A red flag calls for an explanation before the process continues; it is not, by itself, a verdict. is not an accusation, and treating it as one is a beginner's mistake that sours a negotiation over what turns out to be a filing error. A red flag is a stop, not a verdict: the process pauses, the question is put, the answer is checked. And it is judged by materialityWhether an item is significant enough to change the decision of the person relying on it. A finding too small to alter the price or the choice can be recorded and passed; one that would alter either cannot.: an unexplained Rs 40,000 of unpaid property tax on a Rs 34,00,00,000 warehouse is noted and settled at closing, while an unexplained Rs 2,00,00,000 of roof repair changes the price. On the warehouse the site visit produced exactly such a flag, and the exercise set out below follows what the board did with it. The decision map below shows the only two exits a red flag has.

What did diligence on the warehouse purchase find?

A definition only settles once it has been seen at work. The exercise here runs from the first check to the decision. Aravalli Agro Foods' board, with the audit committee chair Suresh Menon taking the lead, set the scope: all four heads, a written sign-off, and a fixed timetable. Legal went first. A title search at the sub-registrar's office, an encumbrance certificate and a lawyer's opinion all agreed: the selling company holds clear title, there is no mortgage on the property, and the company has authority to sell. First head verified. Financial next. Two valuers appointed and paid by Aravalli Agro Foods, working separately, came back at Rs 32,50,00,000 and Rs 35,00,00,000; the seller's asking price of Rs 34,00,00,000 sits inside that bracket, so the price is not out of line with the market. Second head verified, with a bracket rather than a single number. Honest valuation usually produces a bracket.

Then commercial, and here the sequence stopped. The site visit with a structural engineer found that the roof over the main bay needs Rs 2,00,00,000 of work, an item that appeared nowhere in the seller's description or in the seller's valuation report. The unmentioned roof is a red flag in the exact sense already defined: a fact exceeding a claim, unexplained. The board did not cancel and did not accuse. The board put the question, obtained the engineer's costed report, had the seller's representative confirm the finding on site, and priced it. And then people: the shareholding of the selling company, pulled from the registry, confirmed what Devika Rathore had already declared to the board, that her cousin controls it. Nothing about that is wrong in itself. Relatives buy and sell property from companies all the time. But it makes the purchase a related-party transaction, and a related-party transaction travels a separate route: disclosure of the interest, the interested director stepping back from the vote, and approval by the directors who have no interest in it. The diligence found clear title, a fair price bracket, an unmentioned Rs 2,00,00,000 roof, and a related-party fact, and its result was to proceed at a reduced price with the repair reflected, through the related-party route.

Now the arithmetic, and it can all be redone from the table. The seller asked Rs 34,00,00,000. The buyer's two independent valuations bracketed that figure at Rs 32,50,00,000 and Rs 35,00,00,000, so the asking price was not the problem. The roof was: Rs 2,00,00,000 of work that any owner would have to do and that neither the description nor the seller's valuation had allowed for. Taken off the asking price, the agreed figure is Rs 32,00,00,000. The agreed Rs 32,00,00,000 sits just below the lower valuation, exactly as it should: the valuers priced a warehouse with a sound roof, and this warehouse does not have one. Add back the Rs 2,00,00,000 Aravalli Agro Foods will now spend on repair and the all-in cost is Rs 34,00,00,000, the market price for a warehouse in the condition the seller described. Diligence did not get the buyer a bargain. Diligence got the buyer the price that was fair for what actually stood on the plot. Nothing more is asked of it.

How do lenders, investors and households actually use due diligence?

A lender does diligence every time it lends, and the loan agreement is the fossil record of what it checked. Before the invented bank advanced its Rs 1,20,00,00,000 term loan to Aravalli Agro Foods, it verified title to the plants it took as security, tested the cash flows that would service the debt, and wrote the two covenants, net debt under three times earnings before interest, tax, depreciation and amortisation (EBITDA) and interest cover above three times, around the numbers its own analysts had confirmed rather than the numbers in the borrower's presentation. A lender's diligence is proportionate too. The same bank lends Rs 5,00,000 to a distributor on a bank statement and a visit: small stake, short exposure.

Practitioners do not ask whether they trust the other side; they ask which of the other side's claims their decision depends on, and then check exactly those from a source the other side does not control. An investment committee deciding whether to hold Aravalli Agro Foods shares runs the same discipline on management's claims: the 2,200 distributors, the 18 per cent of sales to one retail chain, the leverage of 2.3 times, each traced to a filing or a disclosure rather than a slide. And Farida Shaikh, the invented retail investor who bought the shares on a distributor's tip, learned the household version the hard way. Her diligence, had she done it, would have been an evening with the annual report checking whether the tip's story about a big new contract appeared anywhere the company itself had signed; it did not, and the check would have cost her nothing but the evening. The same instinct that makes a household check the encumbrance certificate before buying a flat is the whole of the subject, scaled up.

Aravalli Agro Foods Limited, Devika Rathore, Suresh Menon, Farida Shaikh, the selling company, the cousin who controls it, the lending bank and the warehouse are invented.
Educational material. Not advice on any investment, tax, budget or market position.