how do i choose the right peer group for valuation

Relative valuation prices a company by comparing its multiples — P/E, EV/EBITDA, price to book — against a set of comparable listed companies, rather than against the cash the business itself is expected to generate. Almost all of its accuracy sits in the choice of peer set: change the comparables and the same stock can look cheap or expensive. It also cannot detect a whole category being mispriced, which is why an absolute method such as a reverse DCF belongs beside it.

the method of valuing a company by comparing it with similar listed companies is called

Relative valuation, also known as comparable company analysis or simply “comps”. It sets a value by applying the multiples at which similar businesses currently trade, in contrast with absolute methods such as a discounted cash flow, which value a company from its own projected cash flows without reference to what anyone else trades at.

how to choose the right peer group for a stock

Pick companies that share the same customer decision, similar economics and a similar growth stage — not simply the same sector classification code. Ask whether a buyer choosing this product would realistically consider the other one, then check that gross margins, capital intensity and the working capital cycle are broadly comparable. Build the set before you look at any multiples, because choosing peers after seeing the numbers reliably produces a set that flatters the stock you already like.

should a sector P/E use the mean or the median

Use the median, and state how many companies are in the set. A mean is easily wrecked by one loss-making company, whose negative or enormous P/E distorts the average, or by a single name trading on a takeover rumour. A “sector average” built from four companies is effectively one company’s opinion, so the sample size matters as much as the figure itself.

is a stock cheap if its P/E is below the sector average

Not by itself — a lower multiple is only a discount if the businesses are genuinely comparable. A company at 16× against a sector median of 24× that earns 11% on capital where the sector earns 24%, and grows at 6% where the sector grows at 14%, is cheaper because it is worse, which is exactly what a market is supposed to price. The narrower question worth asking is whether the discount is larger than the difference in economics justifies.

Peer group and relative valuation

Relative valuation is the most used and least examined technique in equity analysis. "It trades at 18× against a sector average of 26×" is a sentence that decides a great many purchases, and almost all of its work happens in the phrase "the sector".

Think of it like this

Everything in the restaurant category

A five-star hotel restaurant, a highway dhaba and a cloud kitchen are all in the restaurant category. Comparing their rent per square foot tells you nothing, because they are not doing the same thing — they only share a word.

In the market

A large diagnostics chain and a generic exporter are both "Healthcare". Their economics, capital needs, customers and risks have almost nothing in common. The label is administrative.

What makes a genuine peer

The same customer decision. Would a buyer choosing this product realistically consider the other one? That is a sharper test than any classification code.
The same economics. Similar gross margins, similar capital intensity, similar working capital cycle. A business turning inventory twelve times a year is not comparable to one turning it twice.
The same growth stage. A company adding 40 stores a year and one optimising 900 existing ones will trade differently and should.
The same regulatory exposure. A regulated utility and an unregulated one in the same sector face different worlds.
Comparable size, roughly. Very large and very small companies in the same industry carry different liquidity, index-inclusion and governance premia.

Reading the table properly

Company	P/E	ROCE	Sales growth (3y)	Debt/equity	What it suggests
Company A	34×	28%	19%	0.1	Expensive, and earning it
Company B	18×	14%	8%	0.6	Cheaper for identifiable reasons
Company C	31×	13%	9%	0.7	Priced like A, performing like B
Company D	9×	22%	15%	0.2	The one worth investigating

Illustrative figures. The point is the pattern: a multiple only means something next to the returns and growth that justify it.

Where relative valuation goes badly wrong

Two failure modes

The sector is uniformly expensive

"Cheapest in the sector" can still be expensive in absolute terms
In 2021 the cheapest Indian consumer names traded at 45× — a discount to peers and a demanding price
Relative valuation cannot tell you the whole category is mispriced
This is why a reverse DCF belongs alongside the peer table, not instead of it

The peer is the outlier

One company on a takeover rumour drags the "average" up
One loss-making peer makes the sector P/E meaningless
Use the median, not the mean, and say how many companies are in the set
A "sector average" built from four companies is one company's opinion

The comparison page in the live section lets you put up to four listed names on one rebased chart with their returns, volatility, worst falls and correlations side by side. It answers the market half of the question; the peer table above answers the business half. Both are needed — a stock can beat its peers on the chart while losing on every operating measure that matters.

Check yourself

Company X trades at 16× against a sector median of 24×. Its ROCE is 11% against the sector's 24%, and it grows at 6% against 14%. What is the most likely explanation?

Simple bhasha mein

Dhaba aur five-star, dono restaurant

Five-star ka restaurant, highway ka dhaba aur cloud kitchen — teeno "restaurant" hain. Inka kiraya compare karna bekaar hai. Sector ka label sarkari hai, dhandha alag hai. Peer wahi hai jise grahak aapki jagah chun sakta ho.

What to remember

Sector labels are administrative; peers share customers and economics.
Choose the peer set before looking at multiples, or you will pick flattering ones.
A multiple means nothing without the return on capital and growth beside it.
Use the median and disclose the sample size; a four-company average is noise.
Relative valuation cannot detect a whole category being mispriced — pair it with a reverse DCF.

You reached the endMark it done and keep your streak going.

Up nextJudging management by where the cash went Previous: Operating leverage: why small revenue moves become big profit moves

Finished this lesson?

Mark it done to track your progress through the curriculum.

Common questions

Short, direct answers to what people ask about this topic.

how do i choose the right peer group for valuation: Relative valuation prices a company by comparing its multiples — P/E, EV/EBITDA, price to book — against a set of comparable listed companies, rather than against the cash the business itself is expected to generate. Almost all of its accuracy sits in the choice of peer set: change the comparables and the same stock can look cheap or expensive. It also cannot detect a whole category being mispriced, which is why an absolute method such as a reverse DCF belongs beside it.
the method of valuing a company by comparing it with similar listed companies is called: Relative valuation, also known as comparable company analysis or simply “comps”. It sets a value by applying the multiples at which similar businesses currently trade, in contrast with absolute methods such as a discounted cash flow, which value a company from its own projected cash flows without reference to what anyone else trades at.
how to choose the right peer group for a stock: Pick companies that share the same customer decision, similar economics and a similar growth stage — not simply the same sector classification code. Ask whether a buyer choosing this product would realistically consider the other one, then check that gross margins, capital intensity and the working capital cycle are broadly comparable. Build the set before you look at any multiples, because choosing peers after seeing the numbers reliably produces a set that flatters the stock you already like.
should a sector P/E use the mean or the median: Use the median, and state how many companies are in the set. A mean is easily wrecked by one loss-making company, whose negative or enormous P/E distorts the average, or by a single name trading on a takeover rumour. A “sector average” built from four companies is effectively one company’s opinion, so the sample size matters as much as the figure itself.
is a stock cheap if its P/E is below the sector average: Not by itself — a lower multiple is only a discount if the businesses are genuinely comparable. A company at 16× against a sector median of 24× that earns 11% on capital where the sector earns 24%, and grows at 6% where the sector grows at 14%, is cheaper because it is worse, which is exactly what a market is supposed to price. The narrower question worth asking is whether the discount is larger than the difference in economics justifies.