What should be asked and never is

The last two modules taught you to read an earnings call — its fixed anatomy, and the difference between an answer that contains a number and one that contains a story. This module is about the part of the call that is silent. Because the most revealing thing about many companies is not in any answer given on the call; it is in the question that is never asked at all. Some questions would open a company up completely — and precisely because they would, they are the ones the analysts on the call systematically never raise. illustrative

There are three reasons a revealing question dies before it is spoken, and none of them is that the question is unimportant. The first is access. A sell-side analyst who wants management to take their calls, invite them to the plant visit, and keep the relationship warm cannot open the quarter by asking why ₹90 crore was lent to a company the promoter's brother owns. The pointed question is the relationship-ending question, and access is the analyst's livelihood, so the incentive is to stay friendly and ask about margins. The second is awkwardness. Some questions are socially costly to ask out loud — who succeeds the seventy-year-old promoter, what the family's unlisted companies actually do, whether the beloved brand is buying its own growth — and on a live call in front of peers most people will not pay that cost. The third, and quietest, is diffusion of responsibility: everyone assumes someone else checked. Surely the auditors looked at the related-party web; surely a bigger fund has been through the succession question; surely if the receivables were bad someone would have said. So no one does, and the question that everyone assumed was covered turns out to have been covered by no one.

The point of this module is not to lament the silence but to exploit it. You are not on the call, and you do not need to be. Every one of these un-asked questions has an answer that already sits, unglamorously, in a specific line of the filing — and you can go and read it whether or not anyone ever raised it. So the skill this module builds is a standing list: a fixed set of the questions that would most reveal a company, carried into every report, and answered from the document itself. In this market you read the controller before the business, and the controller is revealed most sharply by the questions the friendly call is engineered never to reach. And, as always in this book, there is an inversion threaded through it: the single most important un-asked question is not the same in every business — it moves by sector, and a reader who carries only one sector's version will miss the one that decides the others.

It helps to see the silence as a system rather than an accident, because a system is predictable and you can plan around it. The people with the microphone on an earnings call — mostly sell-side analysts, with a few large buy-side voices — are not neutral truth-seekers. They are operating under incentives, and the incentives select against exactly the questions that would reveal the most.

Consider what the sell-side analyst is optimising. Their research is read because they have access — to management, to guidance, to the colour that comes from a relationship. That access is granted at management's discretion and withdrawn at management's discretion. An analyst who asks the adversarial question in public spends relationship capital they need for the job, and does so in front of every competitor and client, for a payoff that mostly accrues to other people. The rational move, given those incentives, is to ask the questions that are useful and safe — the demand trajectory, the margin walk, the capex phasing — and to leave the dangerous ones alone. This is not cowardice; it is the structure working as designed. But it means the published set of questions is filtered for safety before it is filtered for insight, and the filter removes precisely the items that would most reveal the company.

Layer on the two human frictions and the silence deepens. The awkward question — succession, family entities, whether a national-treasure brand is quietly renting its growth — carries a social cost that most people will not pay on a live line. And the assumption that someone else checked is doing constant, invisible work: the audit, the rating agency, the marquee investor on the register are all treated as having already done the diligence, so each new reader relaxes, and the diffusion of responsibility means the hardest questions are everyone's job and therefore no one's. The net effect is a strange and exploitable gap — a company can be widely followed, heavily traded, and covered by a dozen analysts, and still have its single most revealing question sitting unasked and unanswered in plain sight.

Here is the core of the module: the questions that would most reveal a company, and, for each, the line in the filing where its answer already sits. The discipline is to run all six on every company, the dull ones and the exciting ones alike, because it is only by knowing what the honest version looks like that you recognise the year one answer goes strange. You will notice that not one of these needs the question to have been asked on the call.

What are the true unit economics? The headline is a blend; the truth is in the unit. Unit economics is the revenue, cost and profit of one repeatable unit of the business — one store, one order, one subscriber, one hospital bed, one branch — and it is where a business's real quality lives, because a company can grow its total profit for years by adding units whose individual economics are getting worse. No analyst on a friendly call asks the founder to admit that the newest cohort of stores earns half what the old ones do. But the segment note, the store and volume disclosures, and the contribution hints in the management commentary let you reconstruct it. Ask it of every company, and prefer the unit to the aggregate every time.

Where did the incremental capital actually go, and what did it earn? Over a decade, a business consumes an enormous amount of capital — retained profit, new debt, fresh equity — and the whole game is the return on the incremental rupee, not the average. The incremental return on capital — the extra operating profit earned divided by the extra capital deployed to earn it — is the truest test of capital allocation, and it is almost never asked because the answer is often embarrassing: the new capital went into a segment earning below its cost, or into capital work in progress that never commissioned, or into an acquisition quietly being written down. The investing section of the cash flow statement, the CWIP line, and the acquisitions note tell you where the money went; the segment returns tell you what it earned.

What is the real related-party web? A little related-party trade is normal; the question is whether the profit, the growth, or the cash is concentrated in dealings the promoter controls both sides of. This is the single most access-threatening question on any Indian call, so it is the one least likely to be asked and most likely to matter. The related-party note, the loans-and-advances line, the list of subsidiaries and the inter-corporate deposits together map the web — who the company buys from, sells to, lends to, and guarantees, and how much of the reported result runs through parties that are not at arm's length. A related-party siphon is caught here, by tying reported profit to where the cash actually went.

Why does cash lag profit? You have met this tie before, but it belongs on the standing list because it is the question a management most wants framed as "timing" and waved away. Reported profit is an accrual opinion; cash conversion — operating cash against reported profit, over three to five years — is the fact. When the two diverge, the working-capital schedule says where the profit went, and you age the gap rather than accept the reassurance. This is the difference between reported profit and the cash an owner could actually take out — the number that survives being tied to cash is the one worth valuing.

What do the promoter's other, unlisted companies do? This is the question almost no one asks and almost everyone should. The listed company is one entity in a family's constellation; the unlisted ones — the private trading firm, the holding company, the vendor, the landlord — are where value can quietly be routed to, and where the promoter's true incentives are revealed. The listed filing names many of these as related-party counterparties, and their own accounts are public at the registrar (the MGT-7 and financials filed with the Registrar of Companies), so you can see what the promoter builds when the minority is not watching, and compare it to what they build when the minority is.

Who runs this in ten years? Succession is awkward, personal, and almost never raised on a call, which is exactly why succession risk — the risk that a company built around one ageing founder has no credible answer to who leads it next — sits unpriced in so many Indian names. The board composition, the ages and tenure of the key managerial personnel, the roles held by the next generation of the family, and the depth of professional management below the founder are all disclosed. A company wholly dependent on a seventy-year-old promoter with no visible successor is carrying a risk no quarter's numbers will show.

The standing list is universal — you run all six everywhere — but if you had to pick the single most revealing un-asked question for a given business, it would not be the same one twice. The crux moves by sector, because the place where a business is most easily flattered and least willingly examined is different in each. A reader who carries only the consumer sector's un-asked question into a bank, or the bank's into a contractor, will interrogate the wrong silence and come away falsely reassured.

The inversion is sharpest, again, at the lender. Ask an ordinary reader for the most revealing question about a bank and, if they trained on consumer names, they will reach for something about demand or growth — and a bank will happily show them a briskly growing loan book. But the thing that decides a lender's fate is the quality of that growing book, and the most revealing question about quality is the one about the pool that has been restructured, rolled, or evergreened to keep it out of the non-performing column. That question has no analogue on a consumer company's call — an FMCG business has no restructured-advances note — so a reader who never learned to ask it will listen right past the one silence that matters. The contractor inverts differently: its most revealing question is not whether it can win work, which the order book already answers, but whether the money it is owed by the state will ever fully arrive, aged and disputed as it often is. The standing list you carry everywhere; the one question you press hardest you must relearn for each business. That is the whole reason a single silence can be reassuring in one sector and damning in another — the un-asked question was pointed at the wrong pool.

Take a composite mid-cap engineering company we will call Avira Industries, reporting a strong-looking year — revenue up 24%, profit up 29% — and a concall that ran ninety minutes and covered order inflow, margins, and the capex plan in detail. Twelve analysts asked questions; every one was answered. You listened to the whole thing, and you noticed what the busy call never reached: nobody asked where the last three years of capital had actually gone, or what it had earned. So you go and answer it yourself. illustrative

Start with the standing question on incremental capital. Over three years, Avira's capital employed rose from ₹1,200 crore to ₹2,100 crore — ₹900 crore of new capital deployed. Over the same three years, operating profit rose from ₹240 crore to ₹330 crore — ₹90 crore of extra operating profit. So the incremental return on capital was roughly ₹90 crore on ₹900 crore, about 10% pre-tax — well below the ~20% the average capital employed appears to earn, and barely at the cost of that capital. The headline return ratio, computed on the average, looked healthy; the return on the money actually put to work in the growth was mediocre. No analyst asked, and the blended ratio concealed it, but the cash flow statement and the balance sheet answered the question between them. illustrative

Now the second standing question, on where that ₹900 crore went. The investing section and the CWIP line show most of it flowing into a new subsidiary and a large capital-work-in-progress balance that has sat, uncommissioned, for two years. And the related-party note shows the twist: a meaningful share of the construction contracts for that CWIP were awarded to a private company the promoter's family controls, and ₹120 crore sits in loans and advances to the same group of related entities. Suddenly the mediocre incremental return has a possible explanation — capital is being deployed into projects that earn little for the listed company while the spending itself flows through parties the promoter owns. None of this is proof of wrongdoing; a family construction arm can be a legitimate, arm's-length vendor, and a two-year CWIP can be a genuine large project. But it is now the located question, and you did not need it asked on the call to find it.

The habit to build is exactly this: come to the report with the standing list, notice which of its questions the call left untouched — usually most of them — and answer each from the document. Where the answers close cleanly, you have earned more confidence than the headline alone could give. Where one goes strange, as Avira's incremental-return question did, you have found the thing to press before you believe the result — and you found it in a company a dozen analysts had already pronounced upon.

The standing list is a way of locating questions, not a machine for answering them, and mistaking the one for the other is its own trap. A related-party web, an uncommissioned CWIP, an incremental return below the average, a family construction vendor — each is a question the filing raises, not a verdict it delivers. The family vendor may be genuinely arm's-length and cheaper than the alternatives; the CWIP may be a real project on a long timeline; the low incremental return may be the early, loss-making phase of an investment that pays off later. The list tells you where to look and what to ask; the answer comes from the notes, the history, the segment detail, and the pattern across years. Treating a located question as a proven fault is the paranoid version of the discipline, and it burns out the reader who practises it, because most companies most of the time survive the questions honestly.

Nor can the list tell you that a company which answers all six cleanly is therefore safe. Some of the most careful frauds are built precisely to keep the visible answers clean — a related-party sale routed so that cash appears to flow both ways, a receivable financed near year-end so the cash conversion looks fine, a succession plan announced on paper that means nothing in practice. The six questions raise the cost of deception and catch the large majority of it, because keeping every answer consistent while the underlying reality is rotten is genuinely hard. But a clean set of answers lowers suspicion; it does not eliminate it. This is why the list has six items and not one — the more independent questions a management must keep consistent, the harder a lie is to sustain, and the more likely one of them betrays it.

And the list cannot substitute for judgement about which question matters most here. That is the point of the sector inversion: run all six everywhere, but know that for this business one of them is decisive and the others are hygiene. Pressing the consumer darling on succession while never asking what its growth cost, or interrogating the contractor's unit economics while ignoring the state receivables it may never collect, is applying the right list with the wrong emphasis. The discipline is to carry the whole list and, before each company, name the one silence that would reveal the most — and press there first and hardest.

The first way people get fooled is by treating a thorough-sounding call as a complete one. A ninety-minute call with a dozen answered questions produces a strong feeling of coverage, and the feeling is unearned, because the coverage was set by what the analysts had the incentive to ask, not by what would most reveal the company. A reader who leaves such a call reassured has confused the volume of questions with their reach, and the reach was filtered for safety before it was filtered for insight. The defence is to hold your own list and ask, at the end of every call, not "were my questions answered?" but "which of the six did no one even raise?" — because the un-raised question is invisible precisely when the call feels most complete.

The second way is the diffusion-of-responsibility trap, turned on yourself. It is tempting to assume that the auditors, the rating agency, and the large institutional holders on the register have already done the work — that the related-party web must be fine because a marquee fund owns 8%, that the succession question must be handled because the board is distinguished. But those parties are subject to their own incentives and blind spots, and the assumption that someone else checked is exactly the mechanism that leaves the hardest question checked by no one. The remedy is to treat every un-asked question as yours to answer, from documents that are public, regardless of who else you imagine has looked.

The third way is carrying one sector's decisive silence into another business. The reader who learned to press consumer companies on the cost of their growth listens to a bank and never asks about the restructured book; the reader who learned to press lenders on asset quality interrogates a contractor's provisioning while the stale government receivable that will actually decide the year sits unexamined. The standing list transfers across sectors; the single most revealing question does not. Before each company, name the one silence that would reveal it most — the restructured pool for the lender, the cost of growth for the consumer name, the government receivable for the contractor, the top-client concentration for the IT firm — and answer that one first, from the filing, whether or not a single analyst ever thought to ask.