Valuation & Deal Analysis

3 calculators on this page

Reach for this firstThe bridge from enterprise value to a share price Sum of the parts

Going deeperThe terminal value, two ways

This deskWhat it does, who is in the room, and how the transactions relate.

The desk at a glance

Valuation is not a desk in an org chart; it is the work every other desk does before it can say a number out loud. It is here as a shelf of its own because the techniques are shared, they are learnable in an evening each, and they are the part of investment banking a reader can actually check — unlike a negotiation, a discounted cash flow can be rebuilt on a page of paper.

The one idea that matters more than any technique: a valuation does not set a price; a negotiation does. Analysis sets the range within which an argument can be made. Where inside that range a deal lands comes from who else is bidding and what happens if nobody signs — which no model contains and no model should pretend to.

The second idea, which follows: there is no single value. The same business is worth different amounts to different owners, because a buyer that can remove duplicated costs or fund more cheaply can pay more without overpaying. That is why fairness opinions speak about ranges and about one specific offer, and never about what a company is worth.

Who does what

• The analyst builds the model, and owns the assumptions in it whether or not they were chosen. The first valuation of any target is built from public filings and is wrong in known directions.
• The client's management supplies the forecast. This is the single largest input and the one nobody outside the company can check.
• Diligence produces the adjustments that change the answer: real working capital, real capital expenditure, contracts that end. The gap between the model before and after is where renegotiations come from.
• The valuation committee inside the bank signs off before an opinion is delivered. It exists because an opinion is a liability, not a slide.
• The shareholders, eventually, who read the ranges in the takeover document — usually the most accessible worked valuation anybody will ever see, and it is free.

What decides whether this desk is busy

Six mechanisms, each with the direction it pushes in and the thing to watch. None of them is a forecast, and none is a number that goes out of date.

The calendar this business keeps

Every desk has a rhythm its regulars plan around and a newcomer discovers by being surprised by it.

How this desk reaches the rest of the site

The deals and the instruments are one subject. These are the corridors — each one a mechanism, not a resemblance.

How the work is actually doneWhat the client pays for, the documents in order, and the arithmetic.

What is actually being paid for

Almost none of this work is billed on its own. It is inside an advisory fee, which is why the incentives are worth stating plainly: the analysis is produced by somebody who is paid if the transaction happens. That does not make it dishonest and it is not a comment on anybody's conduct — but it is the reason the methods are conventional, disclosed and cross-checked against each other rather than left to judgement.

The one piece usually paid for separately is the fairness opinion, and often at a flat fee that does not depend on completion. The logic is exact: an opinion on whether a price is fair should not be delivered by somebody whose fee depends on the answer being yes.

What a fairness opinion says is narrower than almost every reader assumes. It says the consideration is fair, from a financial point of view, on a stated date, on stated assumptions. It does not say the price is the best obtainable, that the transaction is a good idea, or what the company is worth.

Which blocker decides across this desk

• Price6 of 10The two sides cannot agree what it is worth.Decides: Cost of capital, Discounted cash flow, Exchange ratio, Precedent transactions, Sum of the parts, Trading comparables. Matters on 4 more.
• Financing1 of 10The money is not there, or its cost moved under the deal.Decides: LBO analysis. Matters on 2 more.
• Approval1 of 10A regulator, a court, a board or a shareholder vote says no.Decides: Fairness opinion. Matters on 2 more.
• Diligence5 of 10What was found when the buyer actually looked.Decides: Accretion and dilution, Discounted cash flow, Precedent transactions, Sum of the parts, Synergies. Matters on 4 more.
• Execution3 of 10Documents, disclosure, timetable, and the people running it.Decides: Accretion and dilution, Fairness opinion, Synergies. Matters on 7 more.

The outputs, in the order they appear

• The comparable companies analysis — where the peer set is chosen, which is itself most of the valuation. See trading comparables.
• The precedent transactions analysis — what was actually paid for similar businesses, including the premium for control. See precedents.
• The discounted cash flow — the only method that values the business itself rather than by reference to others, and the one most sensitive to assumptions nobody can observe. See the DCF and the arithmetic underneath it.
• The buyout model — the same cash flows run backwards: not what is it worth, but what can be paid at a target return. See LBO analysis.
• The sum of the parts, where a group is worth more taken apart than together. See the sum of the parts.
• The accretion and dilution analysis — not a valuation at all, but the number a buyer's board will actually ask about. See accretion and dilution.

Run the numbers

Interactive: the bridge from enterprise value to a share priceEasy

Open this calculator on its own page →

Almost every argument about what a buyer paid is an argument about one line in this bridge rather than about the headline number.

Enterprise value ($m) Gross debt ($m) Cash ($m) Minority interests ($m) Pension deficit ($m) Associates and investments ($m) Shares outstanding (m)

Net debt

—

Equity value

—

Per share

—

Debt share of enterprise value

—

Claims ahead of the shares

—

Reading

—

Which items belong in the bridge, and at what value, is an accounting and negotiating question. This computes a bridge; it does not settle what goes on it. Information and education only. Not advice, not a valuation, and not a quote for anything.

Interactive: the terminal value, two waysHard

Open this calculator on its own page →

There are two ways to end a discounted cash flow and they rarely agree. Each one implies the other, and reading off what it implies is the only cross-check the method has.

Final-year free cash flow ($m) Discount rate % Perpetuity growth % Final-year EBITDA ($m) Exit multiple (×)

From perpetuity growth

—

From the exit multiple

—

Multiple the growth implies

—

Growth the multiple implies

—

Gap between the two

—

Reading

—

Both ends are undiscounted here — they are values at the end of the forecast, not today. A growth rate above the long-run growth of the economy the business sits in implies it eventually becomes all of it. Information and education only. Not advice, not a valuation, and not a quote for anything.

Interactive: sum of the partsMedium

Open this calculator on its own page →

Value three divisions on their own terms, then ask what the group does to them. The line that decides the argument is usually not a division at all: it is the central cost nobody reports as one.

Division 1 EBITDA ($m) Division 1 multiple (×) Division 2 EBITDA ($m) Division 2 multiple (×) Division 3 EBITDA ($m) Division 3 multiple (×) Central costs ($m/yr) Multiple on central costs (×) Net debt ($m) Shares outstanding (m)

Divisions, before central costs

—

Central costs capitalised

—

Enterprise value

—

Equity value

—

Per share

—

Blended multiple

—

Largest division's share

—

Reading

—

Which multiple belongs on which division is the whole of the disagreement, and this puts no view on it: the multiples are yours. Information and education only. Not advice, not a valuation, and not a quote for anything.

Go deeper, and practiseLonger pieces, what an interview asks here, and a page to print.

Go deeper

How the analysis goes wrong

• Comparing enterprise value with equity value. Debt, cash, pensions, leases and minorities sit between them, and this is the commonest error a newcomer makes.
• A terminal value doing all the work. In most discounted cash flows the terminal value is the majority of the answer, and it is a division by a small number.
• Peers chosen to reach a conclusion. Adding or dropping two names moves the median more than most analytical work does.
• Adjusted earnings that only adjust upwards. Each add-back may be defensible; the direction of all of them together is the tell.
• Synergies counted at announcement and never revisited. The buyer's shareholders pay for them up front whether or not they arrive.

Test yourself: valuation and deal analysis

Quick questions on this desk — four options each, checked entirely on your device, nothing stored or sent. Wrong answers come with explanations, and the deep dives above hold every one of them. For education only.

What an interview asks here

Valuation questions are not testing arithmetic. They are testing whether you know which assumption the answer depends on, and whether you will say so.

Try each one out loud before you open it. What is underneath is the shape of a complete answer, not a script — somebody who can produce those parts in their own words can also answer the four variations that follow, and somebody who has memorised a paragraph can answer one.

Q1Walk me through a DCF.

What it is checking. The most-asked question in corporate finance. The answer is standard; what is graded is whether you flag the terminal value.

A complete answer contains:

• Project unlevered free cash flow for an explicit period — operating profit after tax, plus depreciation, minus capex, minus the change in working capital.
• Discount those at the weighted average cost of capital to get their present value.
• Add a terminal value, either a perpetuity growth formula or an exit multiple, discounted back.
• That gives enterprise value. Subtract net debt and other claims to reach equity value, then divide by the share count.
• The part to say without being asked: the terminal value is usually most of the answer, and it depends on two numbers nobody can observe.

Read it properly: Discounted cash flow · WACC

Q2Which of your inputs matters most, and how would you show it?

What it is checking. The follow-up that separates people who have built models from people who have read about them.

A complete answer contains:

• The discount rate and the terminal growth rate, because the terminal value is the largest component and both feed it directly.
• The way to show it is a sensitivity table across plausible ranges of both, not a single number to two decimal places.
• A range with explicit assumptions says what is known; a point estimate says the opposite of what it appears to say.
• Cash flow definitions are the quieter risk — whether leases, capitalised development and working-capital swings are in or out changes the answer more than the discount rate does.
• So the honest deliverable is a range plus a statement of which assumption moves it.

Read it properly: Discounted cash flow · The valuation desk

Q3Why is enterprise value used rather than market capitalisation?

What it is checking. A definitional question that catches a very common error in practice.

A complete answer contains:

• Enterprise value is what the whole business costs, independent of how it is funded — equity plus net debt, plus minorities and other claims.
• Market capitalisation only prices the equity, so two identical businesses with different leverage are not comparable on it.
• Which is why an EV multiple must go over a pre-interest measure like EBIT or EBITDA, and an equity multiple over a post-interest one like net income.
• Mixing the two is the most common error in comparables work, and it always flatters the levered company.
• The bridge between them — debt, cash, minorities, pensions, associates — is where most arguments about what a buyer paid actually live.

Read it properly: The EV bridge · Trading comparables

Q4Two companies, same sector, very different multiples. Why?

What it is checking. Whether the candidate reaches for a story or works through the drivers.

A complete answer contains:

• Growth: a higher expected growth rate justifies a higher multiple on the same earnings.
• Returns: a business that grows without much capital is worth more than one that needs capital to grow.
• Risk: different cost of capital, different earnings volatility, different customer concentration.
• Accounting: the two may not define the metric the same way — capitalised costs, leases, one-off items.
• And the boring possibility that is right surprisingly often: the peer set is wrong, and they are not actually in the same business.

Read it properly: Trading comparables · Precedent transactions

Q5How would you choose the peer set?

What it is checking. Because the selection is the analysis, and almost nobody defends it in writing.

A complete answer contains:

• Start from the business model rather than the sector code — what the company sells, to whom, and how the money arrives.
• Then screen on size, growth, margin and geography, because a multiple is a claim that these companies are alike in the ways that price them.
• Say out loud which comparables you excluded and why. Adding or removing two moves a median more than any adjustment made afterwards.
• Use a median rather than a mean, because one outlier otherwise sets the answer.
• And check the direction of the bias your set creates — a set chosen after the answer is known is not evidence.

Read it properly: Trading comparables · Sum of the parts

Q6Precedent transactions give a higher multiple than trading comparables. Why?

What it is checking. The control premium, plus two effects most candidates miss.

A complete answer contains:

• Precedents are prices paid for control, which carries a premium over a minority stake in the market.
• They frequently include synergies the buyer paid for, so the multiple reflects the buyer's combined business rather than the target alone.
• They are historic, so they carry the rate environment and the competitive tension of their date.
• And there is a selection effect: transactions happen when a buyer is willing, which is not a random sample of the sector.
• So a precedent multiple is a fact about that transaction, in that year, not a valuation of this company today.

Read it properly: Precedent transactions · Trading comparables

Q7What does a fairness opinion actually say?

What it is checking. One of the most misread documents in finance, and the precise answer is short.

A complete answer contains:

• That the consideration is fair, from a financial point of view, to a specified group of holders, as of a specified date.
• It is addressed to the board, not to shareholders, and it is a decision aid for directors rather than a recommendation to anyone voting.
• It does not say the price is the highest obtainable, and it does not opine on the merits of doing the deal at all.
• Its assumptions are load-bearing and listed: management forecasts relied on without independent verification, no undisclosed liabilities, completion on the stated terms.
• The practice became near-universal after a 1985 Delaware decision holding a board liable for approving a sale without adequate information.

Read it properly: Fairness opinion · The valuation desk

Q8When is a sum-of-the-parts valuation useful, and when is it a trap?

What it is checking. The most flexible method there is, which is exactly the warning.

A complete answer contains:

• Useful when divisions are genuinely different businesses that the market prices on different measures, and the group discount is real.
• You value each division on its own peer set, add them, then subtract central costs, net debt and any conglomerate discount.
• The trap is that choosing a different peer set per division can make almost any total come out — it is the easiest analysis in finance to reverse-engineer.
• Central costs and stranded costs are routinely understated, because they are the least visible line.
• And a separation has its own cost and takes years, so a theoretical sum is not a price anybody can realise on Monday.

Read it properly: Sum of the parts · Spin-off

Do these against a clock → — one at a time, ninety seconds each, answer before you look.

Whose questions these are. Every question on this page was written for this publication. None is taken from anybody else’s question bank, and none is a claim about what any named firm asks — that is neither verifiable from here nor ours to assert. They are our own reading of which mechanism a question of this kind is testing. Information and education only.

How the analyses fit together

There are only two ways to put a number on a business, and everything here is one of them. A discounted cash flow values it from the inside: forecast the cash, discount it at a cost of capital, and accept that most of the answer usually sits in the terminal value. Trading comparables and precedent transactions value it from the outside, by what the market pays for similar businesses — the first at the price of a share, the second at the price of control, which is most of why the two disagree.

Three take a business apart before valuing it. Sum of the parts values each division on its own terms and then asks what the group adds or subtracts. LBO analysis asks a different question entirely — not what a business is worth but what a buyer could pay and still earn a return — which makes it a floor produced by a financing structure rather than a valuation. Synergies is the number that sits between the two sides of a deal and is paid for at announcement whether or not it arrives.

Two are outputs rather than methods. An accretion and dilution analysis says what a transaction does to earnings per share, which is arithmetic about a capital structure and not a statement about value — the distinction is the most commonly collapsed one on this desk. A fairness opinion is the formal statement that a price falls inside a range, addressed to a board, and it says what it says and nothing more.

Concepts to master

• The bridge from enterprise value to equity value, written out once and understood, prevents more errors than any other single thing on this shelf.
• A multiple is a statement about which companies this one belongs with, not a fact about it — see valuation as a mechanism.
• The discount rate is mostly unobservable, and anyone can move the answer by a quarter within a defensible range, which is the honest reason results are quoted as ranges. See the cost of capital.
• Accretion is arithmetic, not value. A deal can add to earnings per share and destroy value, and the mechanism is simple enough to see on one line.
• The same multiples are quoted every day on the listed side, where the market rather than a buyer sets them — see cash equities.

Analysis

⬇ The whole desk on one printable page

Who does this: Valuation & Deal Analysis is a sell-side seat in Investment Banking, alongside Mergers & Acquisitions, Equity Capital Markets, Debt Capital Markets, Leveraged Finance, Restructuring, Structured Finance. See the industry map.