Building a Diligence Workplan Around the Price Mechanism

A transaction due-diligence workplan ordered around valuation lists every term in the documents that can move money, names the figure each term needs, names who produces that figure and as at which date, and states what happens if it cannot be produced. The plan is ordered by which findings move the price rather than by subject, and that ordering decides what gets finished.

What is a diligence workplan ordered around valuation?

Most diligence plans are built the same way. Someone lists the subject areas, gives each one a scope and a deadline, and lets each set of specialists decide what to look at inside their own lane. The arrangement feels orderly. It produces a great deal of paper. And it has no mechanism at all for making sure the figures the transaction actually needs are among the things produced.

A diligence workplanThe working document that says what will be found out before a transaction completes, who will find it out, and by when. It is a plan for producing information, not a report on it. ordered around valuation starts at the other end. Such a plan starts with the documents that have already been drafted, reads them for the terms that can change the money that changes hands, and builds every line of the plan backwards from one of those terms. The subject areas still get covered. The subject areas simply stop being the organising idea.

Packing for a week away has the same shape. One way is to walk through the house opening drawers and putting in whatever looks useful. The bag ends up full, with no certainty that the item actually needed is inside it. The other way is to write down what will be done there, then pack the item each of those things requires. The second list is shorter, it is checkable, and its completion is visible. A workplan built from the price mechanism is the second list, and a workplan built by subject area is the first one.

The word doing the work here is price-moving termA clause in the transaction documents whose outcome changes the amount of money that passes between buyer and seller. A term that cannot change that amount, whatever the facts turn out to be, is not one.. Not an important term, not a heavily negotiated term, not a term the lawyers spent a fortnight on. A term whose outcome changes the amount of money that passes. The plan is built around price-moving terms, and the test for whether something belongs on the list is simply whether a different finding would produce a different number on the completion statement.

What is the procedure, in order?

Eight steps. Seven of them build the plan and the eighth states its limit. The order is not a preference. Steps one to four produce the lines, step five decides which lines get attention, step six decides what happens when a line fails, and step seven is the only honest test of whether the work is over.

1. Inventory every term that can move moneyRead the documents, not the subject areas. Write down each term whose outcome changes the amount that passes between the two sides.
   Checking: would a different finding under this term produce a different number on the completion statement?
2. Name the figure each term needsOne term, one quantity to be measured. Write the quantity down in the words the document uses, not in the words the accounts use.
   Checking: what would be written in the box, and in what unit?
3. Name the producer of each figureA person, a team or a named set of accounts. Record it even when the producer sits on the other side of the transaction.
   Checking: if this figure were asked for tomorrow, who exactly would be asked?
4. Name the date each figure is struck as atThe date the quantity is measured on, rarely the date the document is signed and never the date the figure arrives.
   Checking: is the date written on the line, and does the document agree with it?
5. Order the plan by how much money each finding can moveLargest first, using gross amounts. A small item that moves the price outranks a large review that does not.
   Checking: is the top line of the plan the item with the largest amount at stake?
6. State a fallback for every lineWhat happens if the figure cannot be produced by the date it is needed. Three exist: proceed on an agreed assumption, move the item into the price mechanism to be settled later, or stop.
   Checking: is the third option written down as a live choice rather than left unsaid?
7. Apply the finish testEvery price-moving term has a figure, a producer, a date and a fallback. Count them. When the count is complete the plan is complete.
   Checking: is any line missing one of its five fields?
8. State the limit of the planThe plan produces figures. The plan decides nothing. Write that into the front of the document so nobody reads it as a view.
   Checking: does any line in the plan contain a recommendation rather than a quantity?

Five fields per line is the whole discipline. The term comes from the documents, the figure comes from reading the term carefully, and the last three are the ones that get skipped. A line carrying a term and a figure but no producer, no date and no fallback is not a plan for finding something out, it is a wish written in a table. The four lines above are the entire plan for the purchase this sequence works with, and what follows builds them one field at a time.

How is every term that can move money found?

Step one is a reading exercise, and it is done before anybody is asked to scope anything. Harivansh Packaging Limited is buying the whole of Sundarban Polymers Private Limited. The documents have been drafted. Devyani Kulkarni, chief financial officer of Harivansh Packaging Limited, sits down with Ashwin Rege, who leads the transaction team, and they read the agreement with one question in front of them: which of these terms, if the facts came out differently, would change the amount that leaves the account?

Four terms answer yes. The working capital peg: actual working capital at completion is compared against a normalised level, and the difference adjusts the price. The net debt definition: what counts as debt-like decides which items land inside the completion figure and which do not. The earn-out threshold: a further payment is due only if the acquired business reaches a stated level of earnings before interest, tax, depreciation and amortisation (EBITDA) in its first year. And the completion date itself, the date on which two of the other three are measured.

Everything else in the agreement matters and none of it belongs on this list. The test is not importance, it is whether a different finding produces a different number, and applying that test honestly is what keeps the plan short enough to finish.

What figure does each of those terms need?

Step two turns each term into one quantity. The exercise sounds mechanical. In practice the step exposes more trouble than any other in the sequence, and a term that cannot be turned into a quantity is itself telling.

The working capital peg needs a normalised working capitalThe level of working capital the two sides agree a business would carry in ordinary trading, stripped of seasonal peaks, one-off build-ups and anything done specially in the run-up to a sale. level for Sundarban Polymers Private Limited. The agreement sets it at Rs 96 crore. Notice that this is not a figure sitting in anybody's accounts waiting to be copied out. The peg is a judgement about what ordinary trading looks like for this business, and somebody has to produce it by making that judgement on stated grounds. Actual working capital at completion came in at Rs 108 crore, so the price moved up by Rs 12 crore.

The net debt definition needs a list, not a number: the schedule of what counts as a debt-like itemSomething that is not borrowing in name but behaves like it, in the sense that the buyer will have to fund it after completion. Whether an item sits on the list is agreed in the documents, item by item.. The schedule decides whether a given balance lands inside the Rs 195 crore of actual net debt at completion or outside it. The transaction had assumed net debt of Rs 180 crore for Sundarban Polymers, so the price moved down by Rs 15 crore.

The earn-out threshold needs a definition of EBITDA that will still work when Sundarban Polymers is sitting inside a larger group. The definition is a harder object than it looks. The Rs 145 crore threshold is 9.8 per cent above the Rs 132 crore the business earned on its own, and after completion the business will be carrying group charges, shared costs and management time that did not exist before. The figure the earn-out needs is not a number at all until somebody has written down which costs may be charged against it.

Who produces each figure, and how firmly is that recorded?

Step three attaches a name. Not a subject area, not a stream, not a phrase like the finance workstream. A producerThe named person, team or set of accounts that will actually make a figure exist. A line that names a subject area rather than a producer has named a place, not a source.: the person or the set of accounts from which this quantity will come into existence.

Run it on the four lines. Normalised working capital and the debt-like schedule are both produced from completion accounts prepared as at the completion date. The earn-out EBITDA comes from accounts prepared for the first full year after completion. The completion date itself is produced jointly, once the conditions in the agreement have been met.

Two of those producers are on the seller's side of the table at the time the plan is written, and one of them sits inside the buyer's own group afterwards. Neither of those is a problem to solve at this stage. Both are facts to record on the line. A producer outside the buyer's control is a different kind of risk from a producer inside it, and the plan is more useful when it says which is which. A figure with no named producer will not exist, and the commonest way a diligence line quietly fails is that everybody assumed somebody else was making it.

The household version is familiar enough. Three people share a flat and agree that the electricity bill will be split. Nobody is named as the one who reads the meter. The bill still arrives, the split still has to happen, and the argument that follows is not about the arithmetic.

As at which date is each figure struck?

Step four is the shortest field on the line and the one that wastes the most work when it goes wrong. The as-at dateThe date a quantity is measured on. It is different from the date the work is done and different again from the date the figure is delivered, and only the first of the three belongs on the line. is the date the quantity is measured on. The as-at date is not the date the analysis was done and not the date the file landed.

Working capital of Rs 108 crore is a figure as at the completion date. Net debt of Rs 195 crore is a figure as at the completion date. Neither of them means anything as at any other date. Working capital swings with the trading month and net debt swings with whatever was paid out that week. The earn-out EBITDA is a figure for the twelve months after completion. The quantity is not yet in being, and no amount of effort can produce it early.

A figure struck at the wrong date is not a partial answer that can be adjusted into place, it is a different quantity that happens to carry the same name. Working capital measured a month before completion describes a month before completion. The peg does not compare against that month. The peg compares against completion. The two numbers look alike, they carry the same label, and putting one where the other belongs produces a completion statement that is wrong by an amount nobody can later reconstruct.

How is the workplan ordered, and why does the ordering matter?

Step five is where this plan stops resembling a conventional one. The four lines are ranked by how much money each finding can move, largest first, and the ranking is done on gross amounts.

The earn-out definition goes first. Rs 60 crore turns on it, or 5.28 per cent of the Rs 1,137 crore actually paid at completion. The net debt definition goes second on minus Rs 15 crore, or 1.32 per cent of the Rs 1,140 crore headline equity value it adjusted. The working capital peg goes third on plus Rs 12 crore, or 1.05 per cent of that same Rs 1,140 crore headline. The completion date sits alongside them because it is what makes two of the three measurable at all.

The three percentages are struck on two different bases: the earn-out is measured against what was paid, and the two adjustments against the headline they adjusted. Placed on the single Rs 1,137 crore base, all three come out at 5.28, 1.32 and 1.06 per cent, so the ranking does not change. The check is worth doing rather than assuming. A ranking that flips when the base changes is not a ranking.

Read the two columns against each other and the point lands without help. The item that moves the most money is the one measured last and settled a full year after everybody has gone home, and a workplan ordered by subject area would have reached it after the two smaller ones. The mismatch is not a scheduling inconvenience. It is the reason the ordering rule exists.

One thing in that ranking looks like a mistake and has to be defended. The two completion adjustments are ranked on their gross amounts, plus Rs 12 crore and minus Rs 15 crore, and not on the minus Rs 3 crore they netted to. Ranking on the net would put both at the bottom of the plan, and burying them is exactly the error purchase price mechanics warns against. The net is where the two adjustments happened to land after both were computed. The net is not what either of them could have moved, and a buyer who checked only one of the two would have been wrong by Rs 12 crore or Rs 15 crore rather than by Rs 3 crore.

What happens to a figure that cannot be produced in time?

Step six is the field most plans leave implicit, which is another way of saying most plans leave it to whoever is in the room on the day. The fallbackThe decision written down in advance for what happens if a figure cannot be produced by the date it is needed. Written before the deadline, it is a choice; written after, it is whatever was convenient. is the course the line takes when its figure does not arrive.

Three fallbacks exist and there is no fourth. Proceed on an agreed assumption: both sides write down what is being assumed and sign it, so the figure is replaced rather than skipped. Move the item into the price mechanism: the question is then settled after completion by a route the documents state rather than before completion by a figure nobody has. Or stop.

The fallback is a decision, not a hope, and the third option is a real one: a workplan that does not carry stopping as a live choice has no fallback at all, only two ways of continuing.

Work it on the hardest line. Suppose the EBITDA definition for the earn-out cannot be agreed before completion. The first choice is an agreed assumption about what may be charged against the acquired business once it sits inside the larger group, written into the documents so that the Rs 145 crore threshold means one thing to both sides. The second is to move the dispute into a stated resolution route, so the definition is settled afterwards by an agreed process rather than argued about from scratch. The third is to decline the earn-out structure altogether and settle the whole price at completion.

Notice what the third choice does not mean here. Stopping does not mean walking away from the purchase. It means removing the term that cannot be made to work and pricing without it. Stopping is available at the level of a term as well as at the level of a transaction, and the finer version is the one that actually gets used.

How is a workplan known to be finished?

Step seven is a count. The finish testThe condition that says the plan is complete. It has to be something countable, so that two people looking at the same plan on the same day give the same answer about whether it is done. for this plan is that every price-moving term has a figure, a producer, a date and a fallback. Four terms, four figures, four producers, four dates, four fallbacks. Count them and the plan is either done or it is not, and two people counting will agree.

The other finish line, the one most plans are written against, is that every subject area has been reviewed. The review finish line cannot be reached. There is always another contract to read, another year to go back, another supplier to check. Reviewing a subject area is an activity, not a condition, and an activity finishes when the deadline arrives rather than when the work is complete.

Two finish lines exist and only one of them is a line: the countable test ends, and the review-everything test only ever runs out of time. The difference is the whole reason the finish test is written into the plan at the start rather than assessed at the end.

What does the workplan never do?

Step eight is one sentence written into the front of the document. The plan produces figures. The plan decides nothing.

The distinction sounds like modesty and it is a working rule. The moment a line of the plan says that the earn-out looks expensive, or that the peg was set too low, or that the transaction should proceed, the line has stopped producing the quantity it was written to produce. Somebody downstream now has an opinion where they needed a number, and the opinion cannot be checked the way a number can.

It also matters for the same reason that runs through everything in this sequence: the arithmetic is knowable and the merit is not. Whether Harivansh Packaging Limited was right to buy Sundarban Polymers Private Limited depends on what the money would otherwise have done and on what the combined business goes on to achieve, and no completion statement contains either. A workplan that starts recommending is doing somebody else's job, and it has stopped doing its own.