Value is not a fact you discover. It is an estimate produced by a method, and every method embeds a claim about where value comes from. A candidate who knows this can handle almost any valuation question, including the ones designed to be unanswerable, because the honest answer is usually about which method is least wrong for this company on this day.
Two families of answer
Every valuation approach belongs to one of two families. Intrinsic methods say a business is worth the cash it will generate, discounted for time and risk. Relative methods say a business is worth what the market pays for similar businesses.
They answer different questions. Intrinsic valuation asks what the asset is worth. Relative valuation asks what it would fetch. In a rational market these converge, and the interesting cases are the ones where they do not, because the gap is where an investment thesis or a deal rationale lives.
The four methods a banker uses
| Method | The claim | Strongest when | Breaks when |
|---|---|---|---|
| Discounted cash flow | Value is future free cash flow, discounted | Cash flows are forecastable and the business is stable | Growth is uncertain; most of the value sits in the terminal value |
| Comparable companies | Value is what the market pays for similar listed firms | There are genuine peers with public prices | Peers are not comparable, or the whole sector is mispriced |
| Precedent transactions | Value is what buyers have paid to own similar firms | Recent, relevant deals exist | Deals are stale, or each had idiosyncratic reasons |
| Leveraged buyout | Value is the most a financial buyer could pay for a target return | Setting a floor on what a sponsor would bid | Read as fair value rather than as a floor |
Why the answers differ, and what the differences mean
Precedent transactions typically produce the highest values, because an acquirer pays a control premium and expects synergies that a passive shareholder does not get. Trading comparables sit below that, because a share price is the price of a minority stake in a company nobody is currently trying to buy. An LBO usually produces the lowest value of the three, because a sponsor is solving for a return rather than paying for a strategic fit.
A DCF can land anywhere, which is both its strength and the reason people distrust it. It is the only method that does not import a market view, so it is the only one that can tell you the market is wrong. It is also the only one whose answer you can move by two hundred basis points of assumption.
The football field, and what it is for
The output of valuation work in a bank is rarely a number. It is a chart of ranges, one bar per method, with the current share price or the offer price drawn across it. The purpose is not to average the bars. It is to show where the methods agree, where they do not, and which of them a board should weight.
When you present one in an interview, say what the overlap implies and then say which range you trust least and why. A candidate who says "the DCF range is wide because seventy per cent of the value is terminal, so I would lean on the comparables here" has demonstrated the judgement the entire exercise exists to test.
Choosing a method for a company that resists all of them
Interviews love the awkward cases, because they separate the memorised from the understood.
- A pre-revenue biotech: no cash flows and no meaningful earnings multiple. Risk-adjusted net present value per asset, plus precedent transactions at similar trial phases.
- A bank or insurer: enterprise value is meaningless because debt is raw material rather than financing. Use equity value directly, with price to book, price to tangible book and return on equity, plus a dividend discount model.
- A loss-making high-growth software company: no earnings to multiply. Revenue multiples against growth and gross margin, a cohort view of retention, and a DCF that is honest about how much sits in the terminal value.
- A property company or an infrastructure asset: value the assets. Net asset value, capitalisation rates, and cash flows that are long and contracted rather than forecast.
- A cyclical industrial at the bottom of a cycle: normalise. Mid-cycle margins in the DCF, and multiples on a normalised earnings figure rather than on a trough one.
The questions this frame lets you answer
Once the map is in your head, most valuation interview questions become the same question in different clothes: which method, and why this one here.
- "Which method gives the highest value?" Usually precedents, because of control and synergies. Say usually, then say why it might not.
- "You can only use one method. Which?" Name the company type first, then choose. There is no answer in the abstract, and the interviewer is testing whether you know that.
- "How would you value a company with negative cash flows?" Ask why they are negative. Early growth, a cyclical trough and structural decline point to three different approaches.
- "Two companies are identical but one trades at a higher multiple. Why?" Growth, margins, capital intensity, risk, liquidity, accounting, or the market being wrong. Work down the list rather than guessing.
Frequently asked questions
Which valuation method gives the highest value?
Precedent transactions, usually, because an acquirer pays a control premium and prices in synergies that a minority shareholder never receives. Trading comparables sit below, and an LBO analysis is typically lowest because it solves for a sponsor return rather than a strategic price.
What is a football field in valuation?
A chart with one horizontal bar per valuation method, each showing the range that method produces, against the current price or an offer price. It exists to show where methods agree and disagree, not to be averaged into a single number.
Why can you not use enterprise value for a bank?
Because debt is the raw material of a bank rather than a financing choice, so subtracting it to reach an operating value is meaningless. Banks are valued on equity metrics: price to book, price to tangible book against return on equity, and dividend discount models.
Is a DCF better than comparables?
Neither dominates. A DCF is the only method that does not import the market view, so it is the only one that can say the market is wrong, and it is also the most sensitive to assumptions. Comparables are anchored in observable prices and inherit any mispricing in the sector.
Related guides
Build the three methods properly
The modelling course works through a DCF, a comparables set and a merger model from a blank sheet, with the checks bankers actually run.