Valuation

Pricing a company is an old problem with an old answer. A factory clearing a million dollars a year gets priced by multiplying that profit by some number. Buyer and seller argue over whether the multiplier is 8 or 10, but the million itself sits in a ledger both of them can open. The argument stays narrow because the starting point is shared.

A startup has no such starting point. A three-year-old software company usually posted a loss last year, and often the loss is deliberate: the money that could have been profit went into hiring and advertising instead. With nothing to multiply, the old calculation stops entirely. Yet a number still has to exist, because fundraising trades cash for a slice of the company, and the size of that slice can't be worked out until the whole is priced.

Without evidence, the negotiation turns into a contest of nerve. When Pagedeck went out for its first round, Grace opened with "we're thinking somewhere around fifteen million." The investor asked what that was based on, and the answer was one sentence about a strong team in a large market. The investor countered with six million, backed by nothing sturdier than "that's what early companies usually get." Two numbers two and a half times apart, neither one checkable, so the only remaining test was which side was in more of a hurry. The side with two months of cash left lost. Nobody in the room thought the outcome reflected the business, and everybody signed anyway.

Valuation is the practice that grew up to put checkable evidence in that seat. The idea is plain enough: when profit can't do the measuring, measure with something else. Look at what similar companies recently sold for, apply a multiplier to the revenue arriving every month, or project the cash the business will throw off later and discount it back to today. None of these produce a correct answer. What they produce is the ability to answer "why that number" with a table and an arithmetic step. They also produce a range rather than a point, and the spread between methods is treated not as a flaw but as the floor the negotiation stands on.

The reason companies care so much is that ownership is settled here. Raise three million dollars against a twelve-million post-money price and a quarter of the company goes out the door; raise the same three million against thirty million and a tenth does. One conversation moves the founders' and employees' combined share by fifteen percentage points.

The terms under valuation fall into three groups: which moment the number describes (basis), what arithmetic produces it (methods), and what label gets attached to the result (classification).

Basis explains why one company carries two prices at once. There are two.

1. Pre-money valuation — the price before the investment lands. It covers only what the company already has, and it is the figure that actually moves back and forth during negotiation.
2. Post-money valuation — the price once the money is in the account. The arithmetic is one addition: pre-money plus the amount raised.

The split matters because ownership percentages are calculated off the post-money number. A company priced at ten million pre-money that raises two and a half million has a post-money price of twelve and a half million, so the investor's share is two and a half divided by twelve and a half, or 20 percent. This is where conversations go sideways. If one side says "twelve and a half million" meaning pre-money, the post-money figure becomes fifteen million and the investor's share drops to 16.7 percent. Hearing the same sentence two ways costs more than three percentage points, which is why practitioners say "pre" or "post" out loud every time they say a number.

Methods cover the arithmetic that produces the figure. Three of them show up.

1. Comparable analysis — take recent deals for companies with a similar model at a similar stage as the reference. It dominates early-stage work, because the less usable data a company holds internally, the more the outside market becomes the only clue available. Its weakness is finding genuinely similar companies. Two collaboration tools priced differently depending on whether they sell to individuals or to enterprises, plus deal terms that often stay private, can leave a sample of five or six.
2. Multiple — apply a multiplier to revenue or profit. Subscription businesses usually run it off annual recurring revenue: three million in recurring revenue at a multiplier of 8 gives twenty-four million. The multiplier isn't fixed. It moves with growth rate and with how many customers leave, and a company tripling every year does not get the same number as one holding flat.
3. Discounted cash flow (DCF) — project the cash the company will generate over several years and pull those amounts back to a present value. The method assumes future money is worth less than the same sum today, because it subtracts both what the money could have earned elsewhere in the meantime and the risk that the projection is wrong. It looks like the most rigorous of the three and is the least dependable for startups. Miss the year-five revenue figure by a factor of two and the output moves by a factor of two. It earns its keep later, once a company has several years of steady cash behind it and the projection rests on a record rather than on hope.

Teams don't pick one. They run all three, build a range out of the results, and negotiate inside it, because a single figure invites the other side to ask where the rest of the evidence went. If comparables point to eighteen million, the multiple to twenty-four, and discounted cash flow to thirty, the discussion happens between eighteen and thirty. Walk in holding one method and there is nothing to say when the other side answers with a different one.

The sequence is fairly consistent. The company packages its financial statements and metrics and hands them over, the investor runs the three methods independently, and a number gets agreed where the two ranges overlap. That agreed figure goes into the term sheet as a single line, and then due diligence checks whether the evidence behind it holds. If diligence turns up revenue booked under the wrong rules, or customers whose contracts expire next quarter, the agreed price comes back down. On the company side the chief executive and the finance lead carry this; on the other side it is the investment associate.

Classification is the branch that names the result. A unicorn is a private company valued above one billion dollars. It labels an outcome rather than describing a calculation, which is why the branch holds a single term. The name exists because such companies were once about as common as imaginary animals; they are less rare now, and the label carries less weight than it did. One thing is worth holding onto: the billion is a post-money figure agreed in the most recent round, not a billion dollars the company earned, and whether that price still holds is unknown until the next deal.

The three branches run in sequence. A method produces a figure, the figure splits into pre and post depending on basis, and how large the post-money figure grew determines what label the company wears. So when the word valuation comes up, two things are worth confirming: which method produced the number, and whether it is pre or post.

"Is that pre or post? The ownership math comes out differently."

Checking that both sides are reading the same figure against the same moment.

"Eight times recurring revenue is high for this market. Comparable deals at this stage are going at five or six."

Answering a number produced by a multiple with evidence from comparables.

"Push the valuation too high now and the next round gets hard."

A warning that today's figure becomes the starting line for the next negotiation.

The quarter Pagedeck prepared its Series A, recurring revenue stood at two million dollars and was growing 2.4 times year over year. The investor opened with a multiplier of 6, putting pre-money at twelve million. Grace argued for 10 on the strength of the growth rate. Eight million dollars sat between the two numbers.

What closed the gap was a spreadsheet. The finance lead collected recurring revenue and deal terms for seven domestic collaboration-tool companies that had raised in the past year, worked out the implied multipliers, and found them scattered between 5 and 9 — with the three whose revenue had more than doubled sitting between 7 and 9. The meeting restarted around that table. The settlement was a multiplier of 7 and pre-money of fourteen million; with two and a half million raised, post-money came to sixteen and a half million and the investor took just over 15 percent. Grace didn't get her 10. What changed from the first round is that the argument moved from "our team is strong" to "three companies in our position got between 7 and 9." Disputing that requires producing different deals, and from there the winner is whoever brought more evidence rather than whoever can wait longer.

The most common mistake is reading a higher valuation as a better one. The press prints this number and treats it like a report card, so the assumption is natural. What actually rises alongside it is the starting line for the next conversation. Take twenty-five million this time and the next round is expected to clear that, which means revenue has to have grown into it. Accept a price the business can't grow into and a down round waits two years out — a round priced below the last one, where protective clauses for earlier investors trigger, the founder and employee share shrinks further than anyone expected, and stock options end up priced above the current share price and become paper. This is also the moment a candidate asks about equity and there is no good answer. Some companies deliberately accept less than they are offered for exactly this reason, giving up a few million now to buy room to climb honestly later.

The other frequent error is believing the number describes what the company is actually worth. Fifteen million sounds like a sum that exists somewhere, which makes the confusion easy. In reality it is the price one investor put on a fraction of the company at one moment, scaled up across every share. Somebody paid a hundred and fifty thousand dollars for one percent, that figure got multiplied by a hundred, and the result is fifteen million — not a buyer standing there with fifteen million for the whole thing. Miss that distinction and money that isn't in the bank starts appearing in plans. Companies hire against a valuation and sign a larger office lease every year, then miss payroll when the next round slips by six months. The valuation was never wrong about anything; it simply described a price, and a price is not a balance.