Volume, Delivery Volume and Turnover: What Each Measures

Trading volume counts shares that changed hands. Delivery volume counts the smaller number actually settled into a buyer account. Turnover is the same activity measured in rupees rather than shares. The three are different views of one day, and the gap between volume and delivery is usually the largest and the least discussed of them.

A matched trade is the single underlying event beneath all three. Quoted prices were established earlier, along with what one session of trading amounts to. The order in which market data must be read was settled separately, and in particular that no absolute figure means anything until it has been divided by something sensible. Trading volume, delivery volume and turnover all come off the daily report, and each of them has to meet that rule. The illustrative numbers below come from the case record for Sarvani Coatings Limited, an invented paint maker, and carry the stated date of 28 August 2026.

Trading Volume: what does a day's traded quantity actually count?

Trading volume is the number of shares that changed hands over a period, and the period is usually one day. Volume counts quantity across matched trades. If a thousand shares are bought and later the same day sold again, the exchange counts a thousand on the way in and a thousand on the way out, and both land in the day's volume. Nothing about that is a defect in the figure. A count of transactions is meant to do exactly that.

Trading volume counts transactions rather than shares that moved ownership permanently. A share bought and sold again inside the same session is counted each time. Most misreadings of market data begin right there. The word volume sounds like it should describe a quantity of stuff that moved from one place to another, and it does not. Volume describes how busy the counter was.

Think about a vegetable stall on a busy morning. Suppose the stall handles a hundred kilograms of tomatoes before noon. A hundred kilograms did not go home in a hundred different bags. A wholesaler standing beside the stall may have bought thirty kilograms at nine and sold the same thirty back at eleven when the price moved. Both were transactions the stall handled, and the stall's book records both. The number is a true record of activity at the counter and a poor record of how many tomatoes reached a kitchen. Trading volume is the stall's book.

Delivery Volume: how many of those shares really changed holder?

Delivery volume is the smaller count. The delivered count is the number of shares actually settled into a buyer account at the end of the process, meaning beneficial ownershipBeing the person who genuinely holds a share and takes its benefits, rather than merely having transacted in it during a session. Different from whose name appears on an intermediate record. genuinely moved from one holder to another. The shares leave one account and arrive in another at a depositoryThe institution that maintains share holdings in electronic form. Shares in India sit in an account there rather than as paper, and a delivered share moves between two such accounts., and after that a different person holds them.

Delivery volume is always lower than trading volume, and the difference is precisely the activity that opened and closed inside the period. That is not a rule someone imposed. The difference falls out of the arithmetic. A position taken and released in the same session never needs to arrive anywhere, so it lands in the first count and cannot land in the second. If everybody who bought during the day still held at the close, the two counts would meet. In practice they never do.

Back to the stall. Delivery volume is the tomatoes that actually went home in somebody's bag. The wholesaler's thirty in and thirty out show up in the day's handling and in nobody's kitchen. The gap between the stall's book and the kitchens is exactly the size of the round trip, and no arithmetic mistake was made by either count.

Why is the gap between the two the most misread number here?

The two counts taken as a ratio give the delivery percentage: 2,76,480 against 8,64,000, or 32 per cent. The delivery percentage is quoted more confidently than almost anything else on a market data screen, and it supports far less than people think.

The delivery percentage does establish a composition. A low delivery share means most of the day's activity was participants transacting with each other and closing out before the session ended. A high delivery share means more of the activity ended with somebody actually holding. Composition is a genuine and useful reading. The ratio says what kind of day it was.

The delivery percentage does not establish whether the buyers were informed, whether the price is right, or what happens next, and it is one of the most overinterpreted figures in this whole subject. A day dominated by participants who close positions before the close will produce a low delivery share whatever anyone believed about the company. A quiet day with a handful of long term buyers will produce a high one. The figure is describing who was at the counter, not what was in their heads.

The everyday version runs like this. Count the people who walk into a bank branch on a Tuesday and the people who leave with an account opened. The ratio says something real about the mix of visitors that day. The ratio says nothing whatsoever about whether the bank is well run. A person who reads poor conversion as doubt about the bank has attached a belief to a composition figure that never carried one.

What is turnover, and why is it not just volume in different clothes?

Turnover is the same day of activity measured in rupees instead of shares. Turnover is volume multiplied by the price at which the trades happened. On the illustrative day, 8,64,000 shares at about Rs 486/- gives Rs 41,99,04,000. The case record carries that product in rounded form as about Rs 42 crore.

Measuring the same day in rupees looks like a trivial restatement. It is not. The two figures move for different reasons. Volume changes only when the number of shares changing hands changes. Turnover changes when either the number of shares changes or the price changes, and it cannot say which of the two did it.

Turnover and volume diverge whenever the price moves, so a rising turnover against flat volume is a price effect and nothing more, and reading a turnover series without the price beside it is a standing error. Hold Sarvani Coatings' 8,64,000 shares completely fixed and move the price from the Rs 402/- of twelve months earlier to the Rs 486/- of the stated date. Turnover goes from Rs 34,73,28,000 to Rs 41,99,04,000, a rise of 20.90 per cent, and not one extra share changed hands. A person reading only the turnover line would report a fifth more activity, and there was none.

Which of the three answers the question actually being asked?

Now the practical part. The three figures are not ranked, and none of them is the good one. Each answers a different question, and the discipline is to notice which question is being asked before reaching for a number.

The question is asked in rupees, and only turnover is denominated in rupees. An investor sizing a position therefore wants turnover rather than volume. For an investor placing Rs 40 crore in Sarvani Coatings, knowing that 8,64,000 shares trade in a day settles nothing until it has been multiplied by a price. Knowing that about Rs 42 crore trades in a day settles it immediately: the intended position is very close to a full day of everything that trades. Whether the market can absorb a position that size is covered under market depth. Only the choice among the three numbers matters here.

What does each one have to be divided by before it means anything?

The rule that no absolute market figure means anything on its own was settled earlier. Applied here, each of the three turns out to have its own natural denominator, and they are not interchangeable.

Volume is a count of shares, so it is scaled against a count of shares. Against all 24,00,00,000 shares in issueEvery share the company has actually issued and which is outstanding, including the blocks nobody intends to sell. Corporate actions change this count. How they do is covered separately., a day of 8,64,000 shares is 0.36 per cent. Against the free floatThe portion of the share count that is genuinely available to trade, once the holdings nobody intends to sell are removed from the base. Where the free float line is drawn is covered separately. of 11,42,40,000 shares, which is the 47.6 per cent of the count taken as tradable, it is 0.76 per cent. Delivery is scaled against volume, giving the 32 per cent already discussed, or against the free float count, giving 0.24 per cent. Turnover is a rupee figure, so it is scaled against a rupee figure. Free float market capitalisation of about Rs 5,552 crore gives 0.76 per cent.

Notice what just happened. The share based scaling and the rupee based scaling landed on the same 0.76 per cent, and that was not a coincidence. Both are the same ratio. Volume over free float shares, and volume times price over free float shares times price, are identical fractions with the price cancelling top and bottom. Each of the three has a natural denominator, the share based and rupee based scalings must agree exactly when the same base is used, and using a denominator from the wrong column produces a figure that sounds informative and is not.

What does one illustrative day look like worked all the way to a year?

Put the whole thing together on Sarvani Coatings' illustrative day, with every step visible. Trading volume is about 8,64,000 shares. Delivery is about 32 per cent of that, or 2,76,480 shares. The remaining 5,87,520 shares, 68.0 per cent of the day, opened and closed without settlementThe step at which a trade is completed and the shares and money actually change places. How it happens, and on what timetable, is covered separately. ever moving a share to a new holder. Sit with that for a second. A volume figure sounds like a count of shares that found new owners, and roughly two thirds of this one did not.

Two notes on the table before anything is read into it. The turnover figure of Rs 41,99,04,000 is the exact product, and the case record carries it in rounded form as about Rs 42 crore. The rounded figure is the one used for the annualisedScaled up from a shorter period to a full year by multiplying, here by an assumed 250 trading days. The scaling is a stated assumption, not an observation. row, and the annual figure lands on a clean Rs 10,500 crore. The 250 trading day figure is a stated assumption and not a count of anything.

The final row says the tradable part of Sarvani Coatings changed hands nearly twice over in a year. The figure states how much activity occurred and nothing about anyone's conviction. The same annualisation run on the delivered part shifts the picture again: 2,76,480 shares a day over 250 days is 6,91,20,000 shares, which is 60.5 per cent of the free float count. So on the activity measure the tradable part turned over nearly twice, and on the settled measure it turned over rather less than once. Both are true. Each settles a separate question, and anyone quoting one while privately thinking about the other has confused a count of activity with a count of ownership.

What does none of the three establish?

All three measures are counts. Volume counts shares, delivery counts a subset of those shares, turnover counts rupees. A count is a record that something occurred. A count is complete on arrival and carries no cause with it.

All three are counts of activity, a count records that something happened and never why, and every explanation ever attached to a volume figure was brought in from outside the figure. Volume tripled because of a results announcement. Delivery rose because long term holders were accumulating. Turnover fell because interest faded. Each of those sentences has a factual first half and an invented second half, and the invented half is doing all the work.

The same discipline governs a footfall counter on a shop door. The counter says four hundred people came in. The counter does not say the sale was working, the weather was bad, or a bus stopped outside. Somebody has to go and find out. IntradayWithin a single trading session, from the open to the close. An intraday position is one taken and released before the session ends. activity in particular gets narrated constantly, and the narration is never in the data.

Which parts of this are arithmetic, and which parts are somebody's rule?

Some of what has been set out is arithmetic on Indian market data and some of it is somebody's published rule, and the two are worth separating plainly. The counting is arithmetic. A traded quantity, a delivered quantity and a traded value in rupees are recorded by the exchange that matched the trades, and the two Indian exchanges publish each of them in their daily market data. None of it needed a rule to produce.

India

Named, not quoted

The step from counting activity to classifying a company leaves arithmetic behind and enters somebody's published rule, and that rule's numbers are held by the body that publishes them. Where a company is placed in a large, mid or small capitalisation band, the Association of Mutual Funds in India is the body that fixes that boundary, publishes the classification list and revises it on a cycle of its own. Conduct expected of anyone writing trading activity into a research note is fixed instead by the market regulator, the Securities and Exchange Board of India. Traded quantity, delivered quantity and traded value are defined and published by the exchanges.

Each body is named here and no threshold, band boundary, ranking or reporting period is stated from memory. The current text of each rule sits with the body that issues it. These documents are amended, and an old copy reads exactly like a current one.

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