Series B: Scaling Capital and Tightening Terms

A Series B is a larger priced round struck at a higher price per share. Aravalli Learning Systems Private Limited, an invented company, agreed a pre-money of Rs 3,00,00,00,000 across 1,25,00,000 shares at Aravalli Year 5. The price is Rs 240.00. The Rs 75,00,00,000 raise created 31,25,000 shares, carrying the total to 1,56,25,000. Nilgiri Venture Fund I, invented, took none of its pro-rata, so it holds 9.6 per cent.

Almost nothing structural changes between one priced round and the next. The two familiar divisions run again, on a larger pre-money and a larger share count, and they behave exactly as they did before. The change is not in the arithmetic but in the situation: who is at the table, what the people already on the register now have to decide, and how much money it takes simply to stand still. The cost of standing still is the sharpest of those three. A holder who does nothing at a Series B has not opted out of anything. The holder has made a decision with an exact price and an exact consequence, and both are worked out below to the rupee and to the share.

What is a Series B doing that a Series A did not?

The shape is easier to see away from finance. Four cousins run a sweet shop between them. Two years ago they let in a neighbour who put money into a second counter, and everyone's slice of the shop got a little smaller while the shop got bigger. Now a second outsider wants in, with more money, for a bigger kitchen and two delivery vans. The cousins are not being asked whether they like the arithmetic. Each cousin is being asked one question: put in more of their own money alongside this new person, or hold the shares they already hold and accept a smaller slice of a larger shop. Nobody is taking anything from them. There is simply more shop.

A Series BA later priced round, usually larger than the one before it and at a higher price per share. is that second outsider at institutional scale. A Series B is a later-stage roundA round raised once a company has a record the last round's money paid for.. The company arrives at it carrying something the Series A round did not have: a record of what the Series A money actually bought. The seed round was funded on a plan. The Series A was funded on early evidence. A Series B is negotiated against eighteen months or more of results that the previous investor's cash paid for. So the conversation is longer, the document is longer, and the list of things written down about what happens if the plan does not hold is longer too.

The mechanism does not change. There is a pre-money figure the parties agree. There is a raise. The price per share is the first divided by the shares already in issue, the new shares are the second divided by that price, and the post-money is the two figures added together. That is it. A larger round does not have larger arithmetic; it has the same two divisions with more zeroes in them, and every surprise a reader gets at a Series B comes from misreading a denominator rather than from anything new in the mechanics.

Aravalli Learning Systems Private Limited is an unlisted company held by Nilgiri Venture Fund I, one of six vehicles managed by Nilgiri Alternatives Advisors Private Limited. Aravalli runs on its own clock, counted from its own beginning, and this round falls at Aravalli Year 5. Every rupee, share count and percentage below belongs to that one company on that one day.

What was on the register the day before the round?

Five holders and 1,25,00,000 shares. The register is the output of two earlier rounds, both covered separately, and it stands as settled rather than rebuilt. Every percentage below is that same set of share counts divided by a moving denominator, so the table is worth reading slowly.

The shape of it matters. Two of those five holders are people who founded the thing, one is a pool of shares set aside for employees who may not have been hired yet, and two are outside investors who paid cash at a stated price. The angel syndicate paid Rs 30.00 a share at Aravalli Year 1. Nilgiri Venture Fund I paid Rs 4,50,00,000 for its 15,00,000 shares at the same round and the same price, and it has not put in one rupee since. The fund's decision never to put in another rupee turns out to be the sharpest fact on this register.

Where does the price of Rs 240.00 come from?

From one division, and it is the same division that produced every price in this company's history. The parties agreed a pre-money of Rs 3,00,00,00,000 for Aravalli Learning Systems Private Limited at Aravalli Year 5. There were 1,25,00,000 shares already in issue. The price per shareThe pre-money divided by the number of shares already in issue. is therefore Rs 3,00,00,00,000 divided by 1,25,00,000, or Rs 240.00.

Readers slip on two things about that division, so both are worth saying out loud. The first is the denominator. The denominator is the shares that exist before the new money arrives, not the shares that will exist afterwards. Dividing by the post-round count gives a smaller number, and that number is not the price anybody paid. The second is that the pre-money is negotiated and the price is derived. Nobody in the room argues about Rs 240.00. The argument is about Rs 3,00,00,00,000, and the price falls out of it. A price per share in a private round is an arithmetic consequence of a negotiated valuation and a known share count. So the price can move by a different factor from the valuation itself.

How do 31,25,000 new shares appear, and what happens to the total?

The second division. The raise is Rs 75,00,00,000 and the price is Rs 240.00, so the company issues Rs 75,00,00,000 divided by Rs 240.00, being 31,25,000 shares. The new shares did not exist a moment earlier and they were not taken from anybody. The company created them and the company sold them, and the Rs 75,00,00,000 went into the company's own bank account rather than to any existing shareholder.

Added to the 1,25,00,000 already in issue, the total becomes 1,56,25,000. Multiplied by Rs 240.00 the new total gives Rs 3,75,00,00,000, and Rs 3,00,00,00,000 plus Rs 75,00,00,000 gives the same figure. The check ties both ways. A post-money that does not equal the pre-money plus the raise means one of the three inputs is wrong, so the check is worth running every time.

Who is on the register once the round has closed?

Six holders instead of five, and every one of the original five still holds exactly the share count it held the day before. Only the denominator moved. The founders still have 70,00,000 shares, the option pool still has 10,00,000, Nilgiri Venture Fund I still has 15,00,000, the angel syndicate still has 5,00,000, and the Series A investor still has 25,00,000. Dividing each of those by 1,56,25,000 instead of by 1,25,00,000 gives the register after.

Read the two percentage columns against each other and a pattern jumps out. Every single one of the five original holders fell by exactly the same proportion: 56.0 to 44.8, 8.0 to 6.4, 12.0 to 9.6, 4.0 to 3.2, 20.0 to 16.0. The uniformity is not a coincidence and it is not a negotiation. One number has been applied five times, and where that number comes from is worked out below.

Why did the pre-money multiply by 2.5 and the price by only 2.0?

The gap between the two multiples is the sharpest idea here and it is pure arithmetic, so it is worth deriving rather than asserting. The pre-money multipleOne round's pre-money divided by an earlier round's, which is not the same as the price multiple. is Rs 3,00,00,00,000 divided by Rs 1,20,00,00,000, or 2.5 times. The price multiple is Rs 240.00 divided by Rs 120.00, or 2.0 times. Both statements are true of Aravalli on the same day, and neither is a mistake.

Written out, each price is a division. The Series A price was the Series A pre-money divided by the shares in issue before that round: Rs 1,20,00,00,000 over 1,00,00,000. The Series B price is the Series B pre-money divided by the shares in issue before this round: Rs 3,00,00,00,000 over 1,25,00,000. Divide the second by the first. The pre-money figures give 2.5 and the share counts give 1,00,00,000 over 1,25,00,000, or 0.8. So the price multiple is 2.5 multiplied by 0.8, or 2.0. Turned the other way round, 2.5 divided by 1.25 is 2.0, where 1.25 is 1,25,00,000 over 1,00,00,000. The whole of the difference between the two multiples is the shares the Series A round issued, and there is nothing else in it.

The same result shows in rupees, and rupees convince more than the algebra does. Of the Rs 3,00,00,00,000 pre-money, the 1,00,00,000 shares that already existed at the Series A carry 1,00,00,000 times Rs 240.00, being Rs 2,40,00,00,000. The 25,00,000 shares that the Series A round created carry 25,00,000 times Rs 240.00, being Rs 60,00,00,000. The two blocks add back to Rs 3,00,00,00,000 exactly. Set each against the Series A pre-money of Rs 1,20,00,00,000. The old block is 2.0 times it, matching the price multiple to the decimal place. The new block is 0.5 times it, and 2.0 plus 0.5 is 2.5, the pre-money multiple. Half a turn of the pre-money multiple is not price at all: it is 25,00,000 shares that did not exist when the earlier pre-money was struck, and the company was paid Rs 30,00,00,000 for them at the time.

The general rule holds anywhere there is a share register. The price multiple equals the pre-money multiple divided by the share count multiple. If no shares were issued in between, the share count multiple is 1.00 and the two figures agree. If shares were issued in between, they cannot agree, and the size of the disagreement gives exactly how many shares went out. A reader who does not have this will look at 2.5 and 2.0 and assume somebody has made an error somewhere, when in fact both figures reconcile to the share.

Why is every existing holder multiplied by 0.8?

Because this round sold 20.0 per cent of the company. The Series B investor ended with 31,25,000 shares out of 1,56,25,000, exactly one fifth. Whatever fraction of the company the new shares take, the rest of the register keeps the remainder, and here the remainder is four fifths. In share counts it is simpler still: the old total of 1,25,00,000 divided by the new total of 1,56,25,000 is 0.8. The 0.8 is this round's dilutionThe fall in a holder's percentage caused by the total share count rising. factor, and multiplying any existing holder's percentage by it gives that holder's percentage afterwards, with no exceptions and no negotiation.

Run across the register, it checks. Founders 56.0 times 0.8 is 44.8. The option pool 8.0 times 0.8 is 6.4. Nilgiri Venture Fund I 12.0 times 0.8 is 9.6. The angel syndicate 4.0 times 0.8 is 3.2. The Series A investor 20.0 times 0.8 is 16.0. Adding 44.8, 6.4, 9.6, 3.2, 16.0 and the new investor's 20.0 gives exactly 100.0. Nobody sold anything to anybody. The denominator grew by a quarter and every share of it shrank by a fifth. Both sentences say one thing twice.

And here is the half of dilution that a percentage column alone never shows. The founders' 70,00,000 shares were 56.0 per cent of Aravalli before this round and are 44.8 per cent after it. Put that way it sounds like a loss. The same 70,00,000 shares were priced at Rs 120.00 each at the Series A, being Rs 84,00,00,000, and are priced at Rs 240.00 each at this round, being Rs 1,68,00,00,000. Both sentences describe the same unchanged 70,00,000 shares on the same register. Both rupee figures are a share count multiplied by the price of a round, and no buyer has offered the founders anything. Neither figure is a valuation of the founders' position. The difference between a price somebody paid and a price nobody offered does most of the damage in this subject, so it gets a block of its own below.

What is a pro-rata right, and what does it not do?

Go back to the sweet shop. When the second outsider is let in, the four cousins might have written down, two years earlier, that any of them may put money into a future expansion in proportion to what they already hold, before the money is offered to anybody outside. The whole idea is a place in the queue, at the price everybody else is paying.

A pro-rata rightA right to buy part of a new issue so as to hold the same percentage afterwards. in a share subscription arrangement is the institutional version: a contractual right for an existing holder to subscribe for a stated share of a new issue, at the round's price, before the balance goes to the incoming investor. The right is written into the documents at an earlier round and survives into later ones. Nothing about it is automatic. A pro-rata right is a right to buy, not a right to keep, and the entire follow-on decision lives inside that distinction.

Say it the blunt way. If a holder with a pro-rata right writes the cheque, its percentage is preserved. If it does not write the cheque, its percentage falls by exactly the same 0.8 that everybody else's falls by, and the right it held made no difference at all. The right changes who is allowed to buy. The right does not change what happens to somebody who does not buy.

What does it cost to stand still?

The holder's own percentage of the round. The identity is worth carrying away even if every other number here is forgotten. Nilgiri Venture Fund I held 12.0 per cent of Aravalli before this round. To hold 12.0 per cent afterwards the fund needs 12.0 per cent of 1,56,25,000, or 18,75,000 shares. The fund has 15,00,000, so it must buy 3,75,000 more. At Rs 240.00 those cost Rs 9,00,00,000. And Rs 9,00,00,000 is exactly 12.0 per cent of the Rs 75,00,00,000 being raised. Participating pro rata costs a holder its own percentage of the round, whatever the round is, and that identity holds at any size and any price.

A lot of readers quietly change their model of what is happening at exactly this point. Standing still is not the default. Standing still is the expensive branch. The cheap branch, the one that requires no meeting, no committee paper and no cash, is the one that ends at 9.6 per cent. Doing nothing is not the absence of a decision; it is a decision whose price happens to be nil and whose consequence happens to be 2.4 percentage points.

What happened to the fund that did not follow on?

Nilgiri Venture Fund I paid Rs 4,50,00,000 for 15,00,000 shares at Aravalli Year 1 and has not put in a rupee since. The fund did not take its pro-rata at the Series A and did not take it here. So its holding fell from 15.0 per cent after the seed to 12.0 per cent after the Series A, and from 12.0 per cent to 9.6 per cent after this round. Each of those steps is the same single multiplication by that round's own factor of 0.8, applied twice, once per round.

The language people use about a fund in this position is careless, so being exact is worth the effort. Nothing was taken from the fund. The fund was not squeezed, forced out or written down. The fund holds the same 15,00,000 shares it has held since Aravalli Year 1 and can still count them on the register at the Ministry of Corporate Affairs. The fund declined twice to buy any of the new shares pushing the denominator up, so what fell was a ratio. A percentage is not a possession. A percentage is a fraction whose bottom half belongs to everybody.

Why would a fund decline? The record does not say. The record does say that Nilgiri Venture Fund I is a venture capital fund with Rs 1,50,00,00,000 of commitments and eighteen investments, of which Aravalli is one. A follow-on investmentMore money put into a company a holder already has a stake in. at Aravalli would have to be found from the same commitments that every other position is competing for, and any fund making one is choosing it over something else. The competition for the same commitments is structural. Anything beyond it would be a story about people this record does not describe.

Why do the two follow-on questions have to be kept apart?

Because they start from different places, and a figure taken from one of them and used in the other is true of nothing. Both are counterfactuals about Nilgiri Venture Fund I, and neither happened. Both assume the follow-on shares are bought inside the round rather than added on top of it, so the total after the round stays at 1,56,25,000 and the incoming Series B investor takes correspondingly less.

Question one asks what it would have cost this fund to hold 15.0 per cent of Aravalli all the way through, and the answer is a chain in which each leg depends on the one before it. To hold 15.0 per cent at the Series A the fund needed 15.0 per cent of 1,25,00,000, being 18,75,000 shares, so it had to buy 3,75,000 at Rs 120.00 for Rs 4,50,00,000. To hold 15.0 per cent here it needed 15.0 per cent of 1,56,25,000, being 23,43,750 shares, so it had to buy a further 4,68,750 at Rs 240.00 for Rs 11,25,00,000. The two legs come to Rs 15,75,00,000 of further capital on top of the original Rs 4,50,00,000, being Rs 20,25,00,000 all in for 23,43,750 shares. The Rs 11,25,00,000 is correct only because the Series A leg was paid first. Change that assumption and the second figure changes with it.

Question two asks something narrower and its answer is a single step: from where the fund actually stood, at 15,00,000 shares and 12.0 per cent, what would it cost to hold 12.0 per cent through this round alone? It needed 12.0 per cent of 1,56,25,000, being 18,75,000 shares, so it had to buy 3,75,000 at Rs 240.00 for Rs 9,00,00,000. The single leg is Rs 13,50,00,000 all in with the original seed cost, for 18,75,000 shares. The record states the Rs 9,00,00,000 nowhere: the figure follows from the share counts, which makes it a derivation rather than a reported number.

Now look at what happens if somebody blends them. The Rs 11,25,00,000 and the Rs 9,00,00,000 are both cheques at Rs 240.00 into the same round, so they look comparable and they are not. One buys 4,68,750 shares on the way to 15.0 per cent and presumes a Series A cheque that was never written. The other buys 3,75,000 shares on the way to 12.0 per cent and presumes nothing. Adding them, subtracting one from the other, or presenting the difference as a saving all produce numbers that describe no version of this company's history.

What is Rs 36,00,00,000, and who offered it?

Nilgiri Venture Fund I holds 15,00,000 shares of Aravalli. The Series B price is Rs 240.00. Multiplied, that is Rs 36,00,00,000. Every part of that sentence is arithmetic and none of it is in dispute. The trouble starts at the very next sentence a reader writes in their head: the fund's stake is now worth Rs 36,00,00,000.

What tightens in a later term sheet?

A term sheetA short document setting out the shape of a proposed investment, read rather than drafted here. at a Series B is longer than the one at the seed, and the reason is not that lawyers have got busier. A later round has a record to argue about. At the seed there was a plan and two people, and almost everything in the document was about what happens if the plan works. By a Series B there are results, employees, customers, a prior investor with rights already written down, and a much longer list of things that could go differently from the way everyone is hoping. The document grows where the uncertainty is, in the part after the good case.

Such a document can be read at a structural level. The structural reading Feld and Mendelson set out in Venture Deals, 2011, means asking what category each block of text belongs to rather than what any particular clause should say. Four categories carry almost all of it, and a document sorted into them can be read at any stage without knowing a single piece of drafting.

Two of those four are already fully worked above. The first category is the arithmetic just done: the pre-money of Rs 3,00,00,00,000, the raise of Rs 75,00,00,000, the price of Rs 240.00 and the 1,25,00,000 shares the pre-money was divided by. The second category is where a pro-rata right lives, alongside everything else attaching to the instrument the incoming investor subscribes for. The third and fourth are where a later document does most of its growing. Working them would mean stating terms Aravalli's record does not fix, so they stand as categories rather than worked examples.

How does an analyst actually use any of this on a Monday morning?

Three ways, and none of them involves doing a deal. The first is reading somebody else's number. A note lands saying a company has raised at two and a half times its last valuation. The first question is not whether that is good; it is whether the price per share moved by two and a half as well, and the answer is almost always no. The share count multiple between the two rounds, divided out, gives the price multiple. If the pre-money multiple has been quoted while the talk was about price, or the other way round, that gap is visible rather than a vague sense that the numbers do not fit.

The second is reading a fund's own report. A fund reports a holding at a carrying value, and any competent report says alongside it what that figure is: usually a share count multiplied by the price of the most recent round, sometimes adjusted, sometimes not. The single most useful habit available here is asking, of any figure attached to an unsold private holding, who was on the other side of the transaction that produced the price. If the answer is that somebody bought newly issued shares from the company, then the figure is a mark on the holder's position rather than a bid for it, and the report should say so.

The third is the operations and reporting side, quietly doing the hardest part of all this. Somebody has to keep the share register right after the round: the new shares recorded, the old counts unchanged, the percentages recomputed on the new denominator, and the whole thing agreeing with what is filed about the company. The people who do this work meet this arithmetic more often than anybody else does, and they meet it as a reconciliation rather than as a concept. Their check is the one run above. Does the post-money equal the count times the price, and does it also equal the pre-money plus the raise? If those two readings disagree, something in the file is wrong and it has to be found before anything else can be trusted.

The everyday version sticks, so take it one last time. Four cousins let a fifth person into the shop. Nobody took anything from the cousins. The shop got bigger, and each cousin's slice of it got thinner, and any cousin who wanted to keep the same slice had to reach into their own pocket for a share of the new money. The cousin who kept their hands in their pockets still has exactly the same number of shares in exactly the same shop. The size of the shop those shares are a fraction of is what changed, and that is the whole of dilution in one sentence.