Limited Review: Why Quarterly Numbers Are Not Audited

A limited review is a much lighter engagement than an audit. The reviewer mostly asks questions and studies movements in the numbers, and does little of the testing and verification an audit performs. The review exists because quarters come round too often for an audit to fit. Its conclusion is written in the negative, and that single grammatical difference carries the whole difference in what a reader may rely on.

The point is easier to see in a household than in a set of accounts. Somebody hands over a shopping bill for the month, and it gets a glance. Nothing looks odd. The milk is roughly what milk costs, the electricity is close to last month, and nobody bought anything strange. Not a single receipt has been opened, the shop has not been rung, and nothing in the cupboard has been counted. The honest verdict afterwards is that nothing jumped out. The one thing nobody can say is that the bill is right.

The gap between nothing jumped out and this is right is where a limited review lives. A statutory audit closes that gap once a year, working to a threshold no reader gets to see. Between one annual report and the next sits a second, much lighter engagement over the numbers a business puts out, and it produces a sentence that looks reassuring and claims far less than a reader assumes.

What is a limited review, and what does it leave out?

A limited reviewA short independent engagement over a set of interim figures. The reviewer asks questions and studies movements rather than testing transactions, and concludes in the negative rather than the positive. is an independent engagement over a set of interim figures, and it is defined far more usefully by what it omits than by what it includes. Two things happen in it. The reviewer asks the finance team questions about the figures, and the reviewer performs analytical proceduresComparing a number against what it ought to be, from last period, from the rest of the business, or from a ratio, and asking about anything that does not fit. Analytical procedures study the shape of a number rather than the evidence behind it., which means comparing each number against what it ought to look like and asking about anything that does not fit.

Now the omissions, and there are more of them than most readers expect. There is no substantive testingGoing to the underlying evidence behind a figure, such as invoices, contracts, delivery notes or bank statements, and checking that the figure is supported by it. of individual transactions and balances. Nobody goes to the invoices behind a revenue figure. There is no direct confirmation from banks, customers or lenders, so nobody writes to the counterparty and asks whether the balance the business shows is the balance the counterparty agrees. There is no physical verification, so nobody stands in the warehouse counting notebooks. And there is no work on the controls that produce the figures in the first place.

A limited review is not a small audit, and every practical mistake a reader makes with quarterly figures starts by imagining that it is one. A small audit would be the same activity done in less depth: fewer invoices opened, a smaller sample counted, the same kinds of evidence gathered in smaller quantities. A limited review does none of that. The review is a different engagement with a different objective, and whole categories of evidence that an audit treats as essential are simply never gathered at all. Sampling less is a difference of degree. Not going to the evidence is a difference of kind.

The household analogy holds precisely. Opening three of the twenty receipts would have been a smaller version of a proper check. The actual check was a look at the shape of the total and a comparison against memory of last month. The second activity never touches the evidence, so no number of extra glances turns it into the first.

Why does a quarter get a review rather than an audit?

Because of the calendar, and it is worth being plain that the reason is practical rather than principled. A statutory audit is a long piece of work. Evidence has to be gathered from outside the business, counts have to be attended, estimates have to be examined and challenged, and the whole of it has to be finished, documented and signed. The elapsed time is measured in months, not days.

Now put four of those inside one year. A quarter ends, and the work to audit it begins. Before that work could reasonably be finished, the next quarter has ended and its own audit is due. An audit of every quarter would not be a more assured version of quarterly reporting, it would be a permanently unfinished one, and the trade that produces a review instead is deliberate rather than a shortcut anybody is concealing. The engagement is disclosed. The heading on a set of quarterly figures says what has been done to them, and the conclusion at the end says it again.

There is a cost limb too, and it is smaller than people assume but real. Four audits a year cost roughly four times one, and every rupee of that lands on the business and therefore on its shareholders, for information that the annual audit will cover anyway a few months later. The arrangement buys speed instead. Numbers arrive within weeks of a quarter ending, with a light independent engagement over them, rather than assured numbers that could not be had until long after they would have been any use.

How is the wording different, and why does that difference matter?

The point is the grammar rather than the exact text of any report, so the two sentence shapes below repay slow reading.

An audit concludes positively. The statements are the subject of the sentence, and something is asserted about them: that they give a true and fair viewThe standard the annual statements are judged against. A true and fair view means the accounts present the position and the result faithfully, within the accounting rules, not that every figure is exact. of the position and the result. One direction, one claim, the accounts on the receiving end of it.

A review concludes negatively. The sentence turns inside out. Its subject is nothing. Its verb belongs to the reviewer, and reports what did not reach the reviewer during the engagement. A second negative sits inside the claim. Nothing came to attention causing a belief that the statements do not give a true and fair view. Two negatives wrapped around the same phrase, with the reviewer rather than the accounts at the centre of the sentence.

The second sentence reports the absence of a finding rather than the presence of assurance, so it is a far weaker claim than it looks, and a reader who treats the two as equivalent has misread the most important sentence in the document. This is what negative assuranceA conclusion phrased as the absence of a contrary finding rather than as a positive statement. Negative assurance says nothing came to attention. The statement is about the engagement rather than about the figures. means, and the name is exact. The claim is about the engagement, not about the figures.

The test that makes the difference concrete is what happens to each sentence when the work behind it shrinks. The positive sentence gets harder and harder to say. With no basis left for asserting anything about the accounts, at some point it becomes impossible to say honestly at all. The negative sentence gets easier. The less the reviewer looks, the less can come to the reviewer's attention, and the more comfortably the reviewer can say that nothing did. As the work behind them thins, the two sentences move in opposite directions. The review therefore carries the negative form and the audit the positive one.

None of that makes the review dishonest or the reviewer casual. The engagement is real, independent and disclosed, and the sentence at the end of it is a scrupulously accurate description of what was and was not done. The failure is entirely on the reading side, and it happens because the negative sentence contains the same comforting words as the positive one and a reader skims past the frame around them.

What does that mean once the numbers actually land?

Two consequences follow, and both are ordinary rather than sinister. The first is that reviewed figures move more often than audited ones. A number that nobody tested is a number more likely to be adjusted once somebody finally does test it. Nothing in that is a criticism of anybody. The movement describes what the two engagements were designed to do.

The second consequence is arithmetic, and it is where most readers trip. When a business reports four quarters and then an audited year, the fourth quarter is very often a balancing figureA number that is not measured on its own but derived by subtracting what is already known from a total. Whatever the total absorbs, the balancing figure absorbs too.: the audited annual result with the first nine reviewed months taken out of it. There is nowhere else for it to go, so whatever the year-end work changed lands inside that quarter. The landing point is why the fourth quarter is the least comparable of the four, and why the sum of the quarters and the audited year can sit apart from each other without anything having gone wrong.

Anjani Stationers Private Limited, an invented stationery business, files no quarterly results at all, so the split below is a hypothetical laid over figures already familiar from its annual accounts. The audited annual profit before tax of Rs 38,00,000 is fixed, and every quarter below was split out of that one figure.

Suppose the first three quarters had been reviewed and had reported profit before tax of Rs 11,90,000, Rs 10,40,000 and Rs 9,50,000. The three come to Rs 31,80,000 for nine months. Suppose the fourth quarter then traded exactly as well as the average of the first three, at Rs 10,60,000. Nine months plus that fourth quarter is Rs 42,40,000. But the audited annual figure is Rs 38,00,000, and the difference of Rs 4,40,000 is not trading at all. The difference is two year-end adjustments that no review would ever have reached.

The fourth quarter as published would therefore be Rs 6,20,000 rather than the Rs 10,60,000 it traded, and that Rs 4,40,000 difference is a reporting fact about where the adjustments landed rather than evidence of anything about the business or about anybody's work. Look at what the two adjustments actually are. The provision judgement of Rs 3,77,000 is the part of the Rs 6,00,000 charge that no ageing calculation produces, so a quarterly review running the same mechanical ageing would never have found it. The depreciation catch-up of Rs 63,000 comes from revisiting useful lives at the year end. Neither is an error. Both are estimates settling.

Worth noticing what is not on that list. The Rs 2,40,000 disputed claim from the Sunrise Public School group is the third of the recurring judgement areas. The claim was assessed as not probable, so it is disclosed rather than recognised and produced no charge at all. A judgement can be difficult, be examined carefully, and still change nothing in the profit figure.

What should be done differently with a reviewed figure?

Four changes, and none of them is dramatic. The four are adjustments of weight rather than of attitude.

First, treat a reviewed quarter as provisionalGood enough to work with now, and openly subject to change later. A provisional figure is not a doubtful one, it is one that has not yet been through the process that settles it.. Provisional is doing precise work here. Provisional does not mean doubtful, and it certainly does not mean anybody is being careless with it. The word means the figure has not yet been through the process that settles it, and settling can move it. The figure belongs in whatever is being built as a number that can move.

Second, let the annual audited figures govern. Where a quarterly figure and an annual audited figure disagree about the same thing, the audited one is the assured figure and the reviewed one is not. The ordering never reverses, whatever the two numbers happen to say.

Third, the fourth quarter should be expected to look odd, and read rather than reacted to. A large fourth quarter movement shows where the year-end work landed, so it is a question worth asking. The movement is not a conclusion. Absorbing the year-end work is exactly what a balancing figure is built to do.

Fourth, and most practically, building a conclusion on a single reviewed quarter that the annual audit has not yet confirmed rests a decision on the one class of figure that was designed not to carry one. Ask a question of the business instead. A question costs nothing and can be withdrawn. A conclusion, once said out loud about somebody, cannot.

Does Anjani Stationers have a limited review at all?

No, and the reason is worth holding on to because it applies to most Indian companies rather than to some unusual corner of them. Anjani Stationers Private Limited is unlisted. The business publishes no quarterly results, so there is nothing between one annual report and the next for a reviewer to look at, and no limited review of any kind exists on it.

The remaining position is clean and worth stating plainly. Its only assured figures are the annual audited ones: revenue of Rs 2,70,00,000, profit before tax of Rs 38,00,000, profit after tax of Rs 30,00,000, total assets of Rs 1,80,00,000 against liabilities of Rs 38,00,000, and everything else inside the annual statements. One engagement a year, one opinion, and nothing in between.

The quarterly review regime that many readers take for granted belongs to listed companies, and for the large majority of companies in India the position is the one on this business: audited once a year and unassured the rest of the time. To a reader trained on listed company reporting, the absence of quarterly figures on an unlisted business can feel like something missing. Nothing is missing. The obligation was never there.

The shape of that gap is worth noticing. Between one annual report and the next, a reader of Anjani Stationers has nothing independent at all. Not a lighter engagement, not a negative conclusion, not a set of provisional numbers. Whatever the business says about itself during that period, it says on its own. Vaidehi Rao, as finance controller, may produce perfectly good monthly figures for the board, and there is no reason to doubt them, but nobody outside the business has looked at them and nobody has concluded anything about them in any form.

Where does the assurance actually sit, level by level?

One framing makes the panel below worth working through. Assurance is not a single dial that goes up as effort goes up. Two separate things decide what reaches an outside reader: how much work was done, and whether whoever did it is independent of the business and reports to readers rather than to management. A rung can be heavy on the first and give nothing on the second.

Move the control through all four levels rather than looking at the default. The panel redraws the procedures, the assurance a reader receives, and the shape of the sentence that comes out at each level.

The four levels behave like this. At level zero nothing independent has happened, so there is no sentence at all and no bar to draw, and that is the position Anjani Stationers Private Limited is in for eleven months of every twelve. At level one, internal audit work, three procedures are performed including the controls testing that a review skips, and yet nothing reaches an outside reader, because the work is not independent of the business and its report goes to management and the board. Level one is the rung that breaks the intuition: more procedures than a review, and nothing at all for an outside reader. At level two, the limited review, only two procedures are performed and both of them are enquiry and analytics, but the work is independent and a conclusion is addressed to readers, so a sentence arrives, in the negative. At level three, the statutory audit, all six procedures are performed and the sentence turns positive. No engagement at any level produces certainty about a set of accounts, so at every one of the four the bar stops short of the certainty marker.

Who changes their behaviour because a figure is reviewed rather than audited?

Three people, in three different ways, and watching them is more useful than any restatement of the definition.

A lender times its covenant tests to audited figures, an analyst widens the error bar around a reviewed quarter, and a finance controller such as Vaidehi Rao knows which numbers she will be asked to defend twice. Take the lender first. A bank that tests a covenant against a reviewed quarterly figure has built a trigger on a number that can move, and it will eventually find itself in a conversation about a breach that reversed at the year end. The sensible design is to test against audited figures, or to leave enough headroom that a provisional figure moving does not by itself trip anything.

The analyst is where the difference bites hardest. Somebody modelling a business from quarterly results is working with figures that arrived fast precisely because nobody tested them, and the correct response is not to distrust them but to hold them loosely. Holding them loosely means widening the range around a forecast built on reviewed data, and expecting the fourth quarter to carry adjustments rather than reading them as new information about trading.

And the finance controller uses it in reverse. Vaidehi Rao, preparing figures that would go out during the year if this business were listed, would know which estimates the reviewer will ask about and which ones only the annual audit will genuinely examine. The judgement layer inside the provision, the useful lives behind the depreciation charge, the assessment that the Sunrise Public School group claim is not probable: these are the three that get looked at properly once a year rather than four times. For anybody who prepares accounts rather than reads them, knowing which of those numbers are only ever tested annually is the most practical point here.

There is a household version of all three. A relative who says the car is fine because they drove it round the block and nothing rattled has given something a weekend can be planned around. A thousand kilometre journey cannot be planned around it, and that waits for somebody to put the car on a ramp. Nobody in that story was careless, and nobody lied. The two statements simply carried different weight, and the whole skill is in noticing which one was given.

Where can I read the underlying material myself?

Three bodies between them decide who reviews a quarter, who audits a year and what each one is allowed to conclude. The current wording of any requirement belongs at its own source.

Anjani Stationers Private Limited, Chitra Binding Works, the Sunrise Public School group and Vaidehi Rao are invented.
Educational material. Not advice on any investment, tax, budget or market position.