How to Read Equity Market Data Without Overreading It

Read market data in a fixed order: establish the date and the unit, scale every absolute figure against something, separate what is observed from what is inferred, and write down the questions the data raises. Read the figures last and the date first. Almost every absolute market figure means nothing until it is scaled.

Underneath that order sits one awkward fact about market numbers. Almost every one of them is a level, and a level is a number with its divisor left off. A price is rupees for one share with the share count missing. A day of trading is rupees with the size of the company missing. The order below exists so that the missing half gets supplied before anything is said out loud.

Everything below runs on Sarvani Coatings Limited, an invented maker of decorative paint and industrial coating whose every market figure carries the date 28 August 2026. The order is what transfers from one company to the next.

In what order should market data on a share be read?

Consider the screen as it actually arrives. A price, a change for the day, a volume, a value traded, a capitalisation, a holding split, a couple of ratios, a chart. About a dozen numbers, all the same size in the same font, none of them announcing which ones can carry weight and which ones cannot. Left alone, most readers start at the biggest number and work outwards, and that is exactly backwards.

The household version of the same mistake runs like this. A shop took Rs 40,000/- yesterday. Nothing can be done with that until it is known whether the shop is a tea stall or a jeweller, and whether yesterday was a Tuesday or the day before a festival. The takings are real. The sentence is empty. Market data offers a hundred versions of that sentence and dresses each one as a finding.

Step one: what is the date on every number on the screen?

Start with the vintageHow old a number is. A figure carries the moment it was true, and for market data that moment is often a single day or a single minute rather than a quarter. of everything on the screen. Market data ages faster than anything else a reader handles. A set of accounts stays true for the year it describes. A price stays true for as long as the next trade takes to arrive. The date comes first because it is the step that makes every step after it mean something, and a figure with no date attached cannot be used at all.

So the discipline is small and mechanical. The as-of date goes at the top of whatever is being built, in words, before a single figure goes in. For Sarvani Coatings the price of Rs 486/- belongs to 28 August 2026 and to nothing else. The shareholding patternThe published split of who holds a company's shares, given by category rather than by name. Listed companies file it on a fixed rhythm. The split is usually older than the price standing beside it. beside it will almost certainly be older than the price, because the two are published on completely different rhythms, and a reader who does not notice that is already mixing vintages inside one sentence.

Step two: shares or rupees, one day or an average, whole base or tradable part?

Step two settles the units. Four questions, all four answered before a single quantity is looked at. Is this counted in shares or in rupees? Is it one day or an average of many? Does it sit against every share in issue or only against the free floatShares that can actually change hands, after the blocks nobody intends to sell are taken out of the count. Where that line gets drawn is covered separately.? Is it one exchange or several added together?

The same underlying activity produces wildly different numbers under different answers to those four questions. Most arguments about market data turn out to be arguments about basis rather than about the data. Two people will look at Sarvani Coatings on the same Tuesday and quote figures that differ by a factor of two, and both will be right, because one of them counted against 24,00,00,000 shares and the other counted against the 11,42,40,000 shares that make up the tradable part.

Watch it happen. On 28 August 2026 Sarvani Coatings traded about Rs 42 crore. Against the full market capitalisationOne share price scaled up by every share the company has issued. Market capitalisation is covered separately, including the limits of that total. of Rs 11,664 crore that is 0.36 per cent of the company changing hands in a day. Against the tradable part, Rs 5,552 crore, it is 0.76 per cent. Neither number is wrong. The two percentages answer two different questions, and nothing on the screen says which question was asked.

The basis question is also why comparing Sarvani Coatings with Nandivarman Paints Limited, Kesaria Surface Solutions Limited or Thottam Chemicals Limited needs the basis question asked once per company rather than once for the comparison. Each has a different split between the shares that sit still and the shares that move, so a comparison built without checking each basis is comparing two different measurements wearing the same word.

Step three: which figures say almost nothing on their own?

Only now do the figures themselves come into it, and the first thing to do with them is sort them. Some quantities on the screen carry information by themselves. Most do not. A traded value alone, a price level alone, a capitalisation with nothing beside it and a movement measure with no period attached are the four that get quoted alone most often, and quoting them alone is the commonest overreading in this whole subject.

Take them one at a time. What can be said about Rs 42,00,00,000 traded depends entirely on the divisor chosen for it. The quantity is real and supports no statement on its own. Rs 486/- a share is a price for one unit and says nothing about the size of the thing until the number of units is known. Rs 11,664 crore is a total that needs a companion before it means anything. And a volatilityHow far a price wandered, always across some stated stretch of time. Covered separately, including how a daily figure gets stretched to a yearly one. number without its period attached is not a number at all, because the same price history produces different figures over a week and over a year.

Step four: what exactly is the figure divided by?

Now do the arithmetic, and do it deliberately. Scaling is what turns a number into a statement. Different divisors answer different questions, and the choice between them is a decision rather than a technicality.

Here is the whole of step four on one figure. Sarvani Coatings traded about Rs 42 crore on 28 August 2026. Divide it three ways and watch what comes out.

Four readings, all correct, all from one unchanged observation. The year figures are annualisedStretched up to a full year from a shorter period, using a stated count of trading days rather than calendar days. The count is an assumption and belongs beside the figure. at 250 trading days, which is an assumption stated here rather than a fact, and Rs 42 crore times 250 gives Rs 10,500 crore. The same daily figure has just produced four very different sounding sentences, so a reader who does not name the divisor has not actually said anything.

Notice which pairs go together. The 90.0 per cent figure asks how much of the whole company changed hands over a year. The 189.1 per cent figure asks how much of the part that can actually move changed hands over the same year, and it goes above one hundred because the tradable shares turn over more than once. Neither is the right answer. The two figures answer two questions, and choosing between them means deciding which question is being asked.

Step five: where exactly does observation cross into conclusion?

Step five is the one nobody does, and it takes ten seconds. A line runs down the middle of the sheet. On the left goes what the data literally shows. On the right goes what was concluded from it. Market data invites the crossing so smoothly that readers routinely fail to notice the moment they made it, and writing the two columns separately is the only reliable way to catch yourself.

An example with no ambiguity in it. Sarvani Coatings traded 8,64,000 shares on 28 August 2026, and the trade record holds that quantity where anybody can point at it. Calling the share liquid is on no record anywhere. Liquidity is a conclusion, and the conclusion depends entirely on the size somebody wants to trade. For a holding of 1,000 shares, that is about a thousandth of the day and the question barely arises. For a holding of 50,00,000 shares, it is about 5.8 whole days of every share that traded, and the same day of data now reads very differently. Same observation. Different conclusion. The observation did not choose.

Step six: what should the reading actually produce?

Here is the part that feels wrong the first few times. Market data can show that something changed and almost never why. The correct output of a pass is therefore a written list of questions, not a set of conclusions.

Consider what the numbers physically are. The numbers record that trades happened, at prices, in sizes, on dates. The numbers carry no reason for any of it. So at the end of the pass on Sarvani Coatings, what is legitimately in hand is a set of things worth asking somebody: whether the 32 per cent share taken to deliveryTrading that finishes with shares landing in an account somewhere, as against positions opened and closed before the session ends. Covered separately. is usual for this company or unusual, what a quiet day looks like as against the average day, and which divisor the last note about this company used. Each of those is a good question. None of the three is a conclusion, and the absence of conclusions is the point.

A conclusion written at this stage is worth stopping over. Market data contains no reasons, and no conclusion can have come out of the figures. The conclusion came from somewhere else, and that somewhere else deserves naming rather than smuggling in wearing a number.

Step seven: when does the reading stop?

Market data is effectively infinite and there is always another chart. Stopping therefore needs a rule. The reading stops when four conditions are met, and not when the figures run out.

1. The date and the basis are written downNot remembered. Written, at the top, in words, so that anybody reading the work later can see which vintage and which base were used.
2. Every level has been divided by somethingEvery absolute figure carries a divisor, and the divisor is named beside the reading rather than assumed.
3. What was seen and what was concluded sit apartTwo separate lists, in writing, so a reader can check the conclusions against the observations rather than taking them together.
4. The questions are listedThe open questions are written out. An empty list means either that everything has been understood, which is unlikely, or that noticing has stopped.

What is never a step in this order?

Two things, and both of them get done constantly. Calling a figure high or low without naming what it is being compared with is not a step, and neither is treating a market figure as evidence about how the business underneath is performing.

The first is a missing divisor wearing an adjective. High compared with what? Last month, the other three makers, this company at another time, some figure carried in the head? Any of those might be a defensible comparison, and none of them is defensible unnamed. The second is a category error. Trading records what holders did with their shares. Trading does not record what the paint business did with its factories, and the two are joined only by a chain of reasoning that has to be spelled out rather than assumed.

What does the whole order look like on one company, start to finish?

Here is the full pass on Sarvani Coatings, run once, in order, on the illustrative data as at 28 August 2026. The pass takes about four minutes and produces no verdict whatsoever. No verdict is the correct result.

Notice the cross-check in the shares row. A year of volume at that pace is 21,60,00,000 shares against 24,00,00,000 in issue. The share arithmetic gives 90.0 per cent, exactly the reading the rupee arithmetic gave, and agreement between the two is worth a few seconds. When the shares version and the rupee version disagree, the basis question at step two was answered wrongly somewhere.

How does somebody actually use this at a desk?

Meghna Iyer covers the coatings makers for a research desk and reads market data on four companies most mornings. She does not read it to form a view. She reads it to build the question list she takes into a management call later, and the order above is what stops that list from filling up with things she merely assumed.

Her working habit is the two-column sheet. Left column, what the screen said, with the date at the top. Right column, what she thinks it might mean, each line tagged with the divisor she used. When she gets to the call, only the left column is presented as fact. The right column becomes questions. The whole discipline reduces to never letting a right column line escape into a note dressed as a left column line.

A household investor uses the same order with less of it. For a holding worth a few thousand rupees of a listed company, the holding is a rounding error against any day of trading. The size question at step five collapses, and most of what the market depthHow many shares sit on offer close to the current price at a given moment. Covered separately, and a very different quantity from a whole day of trading. screens report is irrelevant at that size. Step one and step two do not collapse. A price seen last week is not a price, and a figure compared against the wrong base is not a comparison. The date and the basis are worth the ninety seconds they cost, at any size of holding.

Meghna Iyer, Sarvani Coatings Limited, Nandivarman Paints Limited, Kesaria Surface Solutions Limited and Thottam Chemicals Limited are invented.
Educational material. Not advice on any investment, tax, budget or market position.