No single multiple works for every company, because each metric captures a different economic reality. EV/EBITDA suits businesses with heavy debt and steady operating profits, while P/S fits early-stage or thin-margin companies where earnings are meaningless. PEG, meanwhile, only makes sense when earnings are growing at a reasonably predictable rate. Picking the wrong one is how investors end up comparing a software firm to a utility and calling it cheap.
The multiple you choose should match the economics of the business, the capital structure, and what you are actually trying to measure. Below is how each metric behaves, where it breaks, and how to think about it on different holding periods.
What each multiple actually measures
A valuation multiple is a shortcut: price divided by some measure of business performance. The shortcut only works if the denominator is stable, comparable across peers, and relevant to how the company creates value.
EV/EBITDA compares the total value of the business—equity plus net debt—to earnings before interest, taxes, depreciation and amortization. Because it sits above the debt line, it lets you compare companies with different leverage, which is its main advantage.
P/S divides market capitalization by revenue. It ignores costs entirely, which sounds careless until you remember that many young companies have no meaningful profit to divide by. Revenue is the last line that stays positive when everything below it is negative.
PEG takes the price-to-earnings ratio and divides it by an expected earnings growth rate. It tries to answer a different question: not “is this cheap” but “is this cheap relative to how fast it is compounding.”
EV/EBITDA: best for leveraged, capital-intensive businesses
EV/EBITDA earns its place when debt matters and when depreciation is a large, somewhat arbitrary accounting charge. Telecoms, utilities, transport, industrials and mature manufacturers are the classic cases. Two competitors with similar operations but very different balance sheets will look far apart on P/E and much closer on EV/EBITDA.
The metric also helps in M&A and buyout analysis, because an acquirer inherits the debt. If you are trying to estimate what a strategic buyer might pay for the whole enterprise, EV/EBITDA is the natural starting point.
Where it misleads: EBITDA is not cash flow. It ignores working capital swings, real maintenance capex and interest that still has to be paid. A company can post a healthy EV/EBITDA and still burn cash every quarter. Capital-intensive firms with high depreciation are exactly the ones where EBITDA flatters reality most, so always check capex against depreciation before trusting the multiple.
However, the metric fails badly for banks and insurers, because for financials, debt is raw material rather than financing. Adding net debt to the numerator therefore distorts everything, so use price-to-book or P/E there instead.
P/S: useful when profits are years away
Revenue is the hardest number to manipulate and the easiest to compare. For early-stage software, biotech, marketplaces and consumer brands reinvesting everything into growth, P/S is often the only multiple that produces a usable number.
It works best inside a single sector where gross margins are broadly similar. A SaaS company at 8x sales and a grocery chain at 0.3x sales are not comparable, and no amount of adjustment fixes that. The grocery chain earns pennies per dollar of revenue; the software firm earns ninety cents. Same multiple, completely different economics.
The counter-argument is straightforward: revenue does not pay bills. A company can grow sales rapidly while destroying cash, and P/S will never tell you. Pair it with gross margin, customer acquisition cost trends and cash burn. If margins are structurally low and unlikely to improve, a low P/S is not a bargain, it is a warning.
PEG: only as good as the growth estimate
PEG was popularized to stop investors from overpaying for fast growers. A P/E of 40 sounds extreme until you notice earnings are compounding at 40 percent; divide and you get 1.0, which the rule of thumb calls fair.
That rule of thumb is fragile. The growth rate in the denominator is an estimate, usually for the next one to three years, and small changes swing the ratio hard. Analysts are systematically optimistic on growth, especially near cyclical peaks. A PEG of 0.8 built on a growth forecast that later gets cut in half becomes a PEG of 1.6 without the price moving at all.
Yet PEG also assumes growth translates into shareholder value, which it does not always do. Companies can grow earnings through acquisitions funded by debt, or through buybacks that mask weak operating trends. The ratio cannot see any of that.
Use PEG as a sanity check on a P/E you already trust, not as a standalone signal. While it works well for stable compounders with predictable growth in the mid-teens range, it becomes least defensible for cyclicals, turnarounds, or any company whose growth estimate rests on a single optimistic model.
Matching the multiple to the sector
Sector conventions exist for a reason, and ignoring them usually means you are measuring the wrong thing.
For banks and insurers, use price-to-book and P/E, because debt is part of the business model and EV-based metrics are therefore meaningless.
Real estate: price-to-FFO or price-to-AFFO, because depreciation distorts earnings for property owners.
Energy and mining: EV/EBITDA plus free cash flow yield, since commodity prices drive earnings and reserve life matters more than a single year’s profit.
Early-stage tech and biotech: P/S or EV/revenue, always alongside cash runway and gross margin trajectory.
For mature consumer and industrial companies, combine P/E, EV/EBITDA and free cash flow yield, because no single number survives contact with a cyclical business.
A quick test: if you cannot explain in one sentence why the denominator is the right measure of value for that specific company, you are using the wrong multiple.
Trader perspective: weeks to six months
Over a few weeks or months, multiples matter far less than positioning, liquidity and catalysts. A stock can stay “expensive” on every metric for a full quarter while momentum carries it higher, and it can stay cheap while it grinds lower.
What traders can use multiples for is context and risk framing. Knowing a name trades at a premium to its five-year average tells you how much room there is for a sentiment reversal. Knowing the sector median gives you a reference point for how a peer’s earnings report might reprice the whole group.
For short horizons, EV/EBITDA and P/S are more useful than PEG, because they rely on reported figures rather than forward estimates that can be revised mid-quarter. Earnings revisions themselves are often the catalyst, and a multiple that depends on next year’s growth number will lag the revision rather than anticipate it.
Risk framing matters more than precision here. If a stock is priced for perfection on P/S and misses revenue, the drawdown can be violent regardless of what the balance sheet looks like. That is a positioning risk, not a valuation insight, and it should be sized accordingly.
Investor perspective: six months and beyond
Over longer periods, multiples revert. This is the core reason valuation matters for investors in a way it does not for traders. Even a good business bought at an extreme multiple can produce poor returns for years if the multiple compresses.
The practical approach is to use two or three multiples that fit the business and track them over time. For a leveraged industrial, that might be EV/EBITDA, free cash flow yield and net debt/EBITDA. For a growing software company, EV/revenue, gross margin and rule-of-forty style cash efficiency. For a bank, P/B and return on tangible equity.
Compare the current multiple to the company’s own history first, then to peers. A stock at 12x EV/EBITDA looks cheap against a peer group at 18x, but if its own five-year range is 8x to 14x, the picture changes. History captures things peer sets often miss: cyclicality, capital intensity, and how the market has consistently priced this specific business.
Finally, remember that a multiple is a ratio of price to a denominator that will be revised. Every forecast embedded in it is an assumption, not a fact. Write down what has to be true for the current multiple to make sense, then track whether those conditions are being met.
Conclusion: pick the multiple that matches the business
EV/EBITDA works when debt and capital intensity define the business. P/S works when profits are distant or negative but revenue is real and margins have a credible path. PEG works only when growth is predictable and the estimate behind it is one you trust.
None of them is a verdict. They are lenses, and each one distorts something. Use the one that distorts least for the company in front of you, check it against a second metric, and be explicit about which assumptions you are making. That habit does more for returns than any single ratio ever will.
- Free Cash Flow: The Number Behind Real Value
- How Stock Indices Are Built: S&P 500, Nasdaq 100 and DAX
- ROE and ROIC: Measuring the Quality of a Business
Frequently Asked Questions
When should I use EV/EBITDA instead of P/E?
Use EV/EBITDA when companies in the same industry carry very different debt loads or when depreciation is a large accounting charge. It neutralizes capital structure, so you compare the operating business rather than the financing choices. P/E is better when debt is minimal and you care about what equity holders actually earn.
Is a low P/S ratio always a good sign?
No. A low price-to-sales ratio often reflects thin gross margins, slow growth or a business that cannot convert revenue into cash. P/S is only meaningful when compared within a sector with similar margin structures and paired with gross margin and cash burn trends.
Why is the PEG ratio considered unreliable?
PEG depends on a forward growth estimate that is frequently wrong, especially for cyclical companies near a peak. Small changes in the assumed growth rate swing the ratio sharply, and the metric cannot detect earnings growth funded by debt or masked by buybacks.
Which valuation multiple works best for banks?
Price-to-book and P/E are the standard choices for banks and insurers. EV-based metrics break down because debt is part of their operating model rather than a financing layer, so adding net debt to the numerator produces a meaningless number.
Do traders and long-term investors use valuation multiples the same way?
No. Traders use them mainly for context and risk framing over weeks to months, since catalysts and positioning drive short-term prices. Investors holding six months or longer rely on multiples more heavily because valuation tends to revert and compound into returns over time.
AI Notice: This article was created wholly or predominantly with the assistance of artificial intelligence and was published without human editorial review.
This article is for general information only and does not constitute investment advice. Always do your own research before making trading decisions.