Profit is an opinion, cash is a fact

A construction company is building a large bridge. Part-way through the year its finance team revises how far along it judges the project to be, from 60% complete to 70%. On the strength of that revision alone, reported profit goes up. No new section of the bridge was billed to the customer. Not one extra rupee of cash came in. illustrative

Did the company earn more money? In its profit-and-loss account, yes. In its bank account, nothing happened. The profit rose because a judgement about the bridge changed, not because any money moved - and that is the uncomfortable truth at the centre of reading accounts. Profit is not a fact you can weigh or count. It is the sum of many judgements: when a sale counts, how complete a project is, how long a machine lasts, how much to set aside for losses that have not happened yet. Change the judgements and the profit changes, even when the business has not. Cash is different. Cash either arrived or it did not. This module is about the gap between the two, and about reading profit against cash so the opinion is always checked against the fact.

That profit is an opinion is not an accusation of dishonesty; it is built into how accounting works. To turn a messy, ongoing business into a single yearly figure, someone has to make dozens of judgements, each a defensible range rather than a single number. The profit that comes out depends on where in each range management chose to sit. Once you see this, a class of puzzles resolves: two similar businesses report very different profits because one judged more optimistically; a company grows profit for years while its cash stagnates because the profit is being lifted by judgements the cash never confirms.

Profit is an opinion because judgement enters the P&L at a few specific, identifiable points. Learn where they are and you know which lines to check first.

The first is revenue recognition - when a sale counts as earned. For a shopkeeper this is trivial: the customer pays and takes the goods. For a company building a bridge over three years it is a genuine judgement, resolved by percentage-of-completion accounting, which books revenue in step with how complete the project is judged to be. The completion estimate is a lever: nudge it and profit moves. The second is inventory valuation. Stock is meant to be carried at the lower of cost and what it can now be sold for, but judging the latter is a choice - carry slow-moving stock at full cost and profit is higher; write it down and profit falls. The third is depreciation, where the judgement is an asset's assumed useful life: assume a machine lasts fifteen years rather than ten and its yearly cost falls, lifting profit with no change in the machine at all. The fourth is capitalisation - whether a spend is an expense that hits this year's profit or an asset charged slowly over future years; treat a cost as an asset and profit rises now, at the expense of later. The fifth is provisioning. A provision is an amount set aside for a likely-but-uncertain future loss - bad debts, warranty claims, a legal dispute - and how much to set aside directly moves reported profit.

Put numbers on the bridge and you see how much profit one estimate can conjure. Say the contract is worth ₹1,000 crore and is expected to cost ₹750 crore, so the whole project should earn ₹250 crore of profit over its life. Under percentage-of-completion, profit recognised so far is simply the completion fraction times that ₹250 crore. At a judged 60% complete, that is ₹150 crore; revise to 70% and it becomes ₹175 crore. The revision alone adds ₹25 crore - the whole project's profit multiplied by the size of the nudge - and no bridge was billed, no cash arrived. On a large, long project with a fat margin, a ten-point revision well within what an auditor would accept can swing reported profit by tens of crores, entirely on judgement.

Against all of this stands one figure that is not a judgement: the cash. Where reported profit and cash agree, the profit is solid. Where they part company, believe the cash.

Profit is an opinion in every business, but the single line where the opinion does the most work moves from one sector to the next. Knowing which lever matters most tells you where to look first.

The contractor is the instructive case, and the inversion worth carrying. In most businesses revenue feels like the one hard fact in the P&L - a sale happened, or it did not. But under percentage-of-completion a contractor's revenue is itself an estimate of project progress, so the line that feels most factual becomes one of the most judged. The lesson is not that one of these businesses is more honest than another. It is that "profit is an opinion" holds everywhere, and the first place to check is wherever that particular business concentrates its judgement.

Take a composite EPC contractor with a large bridge project on its books, judged 60% complete at the start of the year and revised to 70% during it. illustrative Revenue and profit are recognised in proportion to completeness, so the revision pulls another slice of the project's value into this year's P&L and reported profit rises. But look at what did not change. No new milestone was billed, so the amount the company can actually demand payment for is the same, and the cash collected is the same. What rose instead is unbilled revenue - work now claimed in profit but not yet invoiced, sitting as an asset on the balance sheet.

So profit went up and cash did not. Whether that is fine depends entirely on whether the bridge really is 70% done. If it is, the billing and cash catch up over the next quarter or two and the profit was simply recognised a little ahead of the invoice. If it is not - if the 70% was an optimistic estimate to lift a soft year - then unbilled revenue keeps growing, the cash keeps failing to arrive, and the profit was an opinion the fact never confirmed. What tells the two apart is the cash flow statement and the unbilled-revenue trend, read together over several quarters. Rising profit alongside rising unbilled revenue and flat operating cash is the pattern to distrust. The profit figure alone will never show it to you; only holding the opinion against the fact will.

When profit outruns cash, a good analyst goes after the judgement that lifted it. For a contractor the question sounds like this: "Unbilled revenue rose faster than billings again and operating cash was soft. How much of this year's profit is from completion estimates versus billed work?" That asks management to separate the opinion from the fact.

A candid answer is specific about how much converts, and when:

"Fair to probe. Unbilled revenue rose about ₹200 crore; roughly ₹120 crore is on two projects where we crossed engineering milestones the client certifies at the next billing date in Q2, so it converts then. The remaining ₹80 crore is genuinely slower client sign-off, and we've flagged it. Our completion methodology is unchanged this year - I can share the milestone certificates. Cash should catch up as Q2 billings go out."

The evasive version reassures without reconciling:

"Unbilled revenue is a normal feature of the project cycle and always normalises as billings catch up. Execution has been strong, the order book is at a record, and we're very confident in the portfolio's profitability. I wouldn't read too much into one quarter's working capital."

It is fluent and plausible - the kind a real team gives. What makes it evasive is that it never says how much of the unbilled revenue converts, or when, never addresses whether completion estimates changed, and redirects a question about profit quality into reassurance about the order book. The follow-up that settles it - "of the increase in unbilled revenue, how much bills out in the next two quarters, and did any completion estimates get revised upward this year?" - splits the profit into the part the cash will soon confirm and the part resting purely on a revised judgement. When "it always normalises" is allowed to stand while unbilled revenue climbs and cash stays flat, that silence is the signal.

Knowing that profit is an opinion tells you to be sceptical; it does not, by itself, tell you whether a particular judgement was honest or merely optimistic. A completion estimate, a useful-life assumption, a provision - each sits in a defensible range, and being within that range is not the same as being conservative. The cash flow statement is the best single check, because cash is the fact judgement cannot bend for long, but even cash can be flattered for a few quarters, and a determined fraud can move faster than the cash catches up.

There is a limit in the other direction too, and it is the trap most readers fall into. A lower reported profit is not automatically the more honest one, and a higher one is not automatically suspect. A company that writes down stock or takes a large provision reports less profit and may be the more trustworthy; a rival reporting more may simply be recognising less pain, storing it up for a later year. But sometimes the higher profit is just a better business. The point is not to assume the worst of every judgement - it is to know exactly where the judgements live, and to check them against the cash rather than take the profit on trust. Which judgements are honest, and how to catch the ones that are not, is the work of Part Three.