Multiples are the quickest way to turn messy company details into something investors and boards can debate. They are also the fastest way to fool yourself.
I learned that the hard way on an early deal when the spreadsheet looked tidy and the math was clean, yet the conclusion felt wrong in the room. The peer set was “close enough” on industry and headline revenue, but the margins told a different story. The multiple was doing all the work, while the story behind the multiple was quietly missing. The valuation was not just off, it was off for the same reason people often get burned in finance: they trust the numerator and denominator without interrogating what they actually represent in practice.
Below is a practical guide to using valuation multiples wisely, with enough realism to handle the edge cases that show up once you get past the first pass.
What a multiple is really buying you
A valuation multiple is not just a ratio. It is a bundle of assumptions about growth, risk, margins, capital intensity, and accounting choices.
When you see something like EV/EBITDA, you are effectively saying: “The market pays X times operating cash earning power for companies with characteristics similar to this one.” But that “operating earning power” can vary wildly. Two companies can both report EBITDA, and still have different:
- working capital dynamics (one funds customers through receivables, another collects fast), maintenance versus growth spend embedded in “cost of revenue,” stock based compensation policies, lease accounting effects that show up differently in leverage and capital structure.
Even within the same sector, you can end up valuing different economic realities with the same multiple headline.
That is why the first discipline is to understand what the multiple is trying to normalize. If you treat the multiple as a universal constant, you will miss the conditions that make it meaningful.
The most common multiples, and where they break
Most valuation work starts with a small menu of multiples. Here are the ones you will see again and again, and the main reason each can fail.
- EV/EBITDA: Popular for capital-structure neutral comparisons, but sensitive to lease capitalization, one-time adjustments, and how “EBITDA” is derived. P/E: Intuitive for equity value, but heavily affected by capital structure, tax rates, and non-operating items that distort earnings. EV/Sales: Useful when margins are unreliable or earnings are temporarily suppressed, but it can reward growth that never turns into durable cash flow. Price-to-book (P/B): Sometimes relevant for financials, but for asset-light businesses it can be disconnected from economic value. EV/Gross profit or EV/FCF: Often more grounded, but requires quality of forecasting and careful treatment of working capital and reinvestment.
That list is enough to remind you of the trade-offs. The next step is where “using multiples wisely” starts: you pick the right multiple for the business you’re valuing and then you adjust the inputs until they represent comparable economics.
Start with the business quality, not the peer list
A peer group is rarely “wrong” because of geography or SIC codes. It is wrong because the economics do not match.
In practice, the peer selection becomes a series of questions you can answer with diligence, not vibes:
Is the company in a transitional phase where margins are compressing or expanding? Is growth funded by retention of cash in the business, or by external capital? Does it have customer concentration that makes forecasting brittle? Are earnings distorted by unusual litigation, restructuring, or acquisitions? Are there recurring revenue elements and what is the churn profile?
I once reviewed a software company where two public peers looked similar on ARR and headline growth, but one had a much higher proportion of expansion revenue. On paper both had “high growth,” but one was compounding and the other was buying growth with aggressive discounting. The multiple anchored to the first company’s compounding profile, not the second company’s reality. The valuation range widened once we separated sustainable growth from short-term commercial intensity.
Multiples are most defensible when the peer set shares the same drivers: margin structure, growth durability, and capital intensity.
Normalize the inputs before you trust the outputs
Most multiple-based valuations fail at input normalization. The market multiple is based on a specific definition of EBITDA, sales, or earnings. Your target company must align to that definition.
EBITDA and “real” operating profitability
EBITDA sounds objective, but it is often a negotiated number. Even if every peer reports EBITDA, you still need to check whether:
- stock-based compensation is excluded consistently, restructuring or acquisition related costs are handled the same way, leasing is treated similarly (or at least the differences are understood), revenue recognition practices differ materially, “adjusted EBITDA” removes items that are actually recurring in disguise.
If you are adjusting EBITDA, you need a rule for what is legitimate. In deal teams, I’ve seen two approaches that work better than others:
Use consistent policy adjustments across peers and the target, anchored to what management excludes but with reasoned scrutiny. Prefer cash-based proxies (where available) if adjusted metrics cannot be made comparable without inventing a new company.The danger is “adjustment drift,” where your target becomes a best version of itself and the peers remain audited. That gives you a multiple that flatters your work, not the business.
Sales and the illusion of growth
EV/Sales is tempting because it sidesteps profitability. That is also why it can be dangerous. Revenue can grow while economics deteriorate, especially in businesses with:
- rising discounting, churn that is masked by reactivation, higher fulfillment costs, expanding incentives or channel payments.
If you use a sales multiple, you should connect it to margin trajectory and cash conversion, even if you do not calculate full discounted cash flow. At minimum, verify that the implied margin and reinvestment needs do not contradict the operating plan.
A helpful mental check is to ask: “If the company reaches the peer-like margin profile, does it require unrealistic improvements in working capital or capex?” If the answer is yes, the sales multiple is likely pricing hype rather than deliverable fundamentals.
Earnings and the context behind P/E
P/E can be straightforward when earnings are stable and tax rates are comparable. It becomes tricky when:
- earnings include large one-time items, amortization policies differ across acquisitions, capital structure differences change interest expense and tax, the business is experiencing a temporary disruption.
You can still use P/E, but you have to normalize the earnings base and account for the company’s tax and capital structure differences. Otherwise you are comparing equity outcomes driven by financing choices rather than operations.
Don’t ignore control, liquidity, and the “deal premium” reality
A public trading multiple is not the same thing as an observed transaction multiple, and even within public markets, liquidity and governance can matter.
In private company valuation, buyers often pay for control, certainty, and governance. Those factors can justify a premium to what you see in thinly traded stocks. On the other hand, a private company may face discounts due to lack of liquidity, less transparency, and sometimes higher risk.
If you are comparing multiple-based valuations across public and private comps, be careful with your framing. It is not that one is correct and the other is wrong. It is that they embed different risk pricing.
A practical approach is to triangulate: use public comps to understand market pricing of fundamentals, then evaluate how a control holder would translate those fundamentals into a deal process. Even without “perfect” adjustments, you can set a range that reflects the direction of the differences.
Watch the denominator: trailing, forward, and “what the market already knows”
Most practitioners have a preference between trailing and forward multiples, and it usually shows up as a quiet bias.
Trailing multiples are anchored in history. They can be stable, but history can be stale. If your company is in a turnaround, you might be valuing yesterday’s economics.
Forward multiples use forecasted earnings, EBITDA, or sales. They can align better with current expectations, but forecasts are where optimism lives. Market pricing already reflects consensus expectations to some degree, so your private forecast needs to be disciplined and grounded.
One technique I trust is to model a “bridge” from trailing results to the forward denominator. If the forecast requires margin expansion that contradicts past cost structure, or growth that depends on customer acquisition economics that do not cash out, you should either adjust the denominator or recognize that the multiple you are applying might not be realistically supported.
Cyclicals, business model changes, and the danger of averaging
In cyclicals, a single-year multiple can be misleading. During peaks, EBITDA looks strong and the multiple compresses into “good times pricing.” During troughs, EBITDA looks weak and the multiple can expand, not because the company is improving, but because earnings are depressed.
The right method is not “always use a five-year average.” Sometimes average hides regime changes. If the business model changed, or if the company shifted from one customer segment to another, then averaging across incompatible eras will distort the denominator.
Instead, focus on normalization tied to operating reality:
- If the company’s margins are seasonal, use an approach that reflects average seasonality. If costs are restructuring, separate structural changes from cyclical noise. If the company recently improved pricing power or distribution, do not average it away with older margin data.
The goal is to avoid anchoring your valuation to a temporary earnings level, while also avoiding the opposite mistake, ignoring that the business has permanently changed.
A disciplined way to apply multiples (without pretending it’s mechanical)
You can get a useful valuation quickly by layering judgment on top of the math. Here’s the process I’ve seen work reliably when teams are under time pressure but still want credibility.
First, choose the multiple that matches your business economics. Second, build comparables that share those economics. Third, normalize the denominator so you are applying a valuation concept, not an accounting artifact. Fourth, sanity-check the implied outcome against what the market would need to believe.
A short checklist can help teams avoid the most common failures:
- Confirm the multiple’s denominator definition is consistent across peers and the target (especially EBITDA adjustments). Check whether growth and margin are comparable, not just revenue growth. Look for earnings quality issues: one-time items, stock comp differences, and working capital distortions. Test trailing versus forward assumptions, and quantify the range rather than betting on one year. Verify the implied economics through a “reasonableness” lens, such as cash conversion and reinvestment needs.
That is the difference between using multiples and letting multiples talk you into a number.
From implied value to decision-grade insight
Even when the multiples are “right,” you rarely stop at a single valuation figure. You typically translate the multiple outputs into a view on what matters for the deal or decision you’re supporting.
For example, suppose EV/EBITDA suggests a valuation range that is higher than what your cash flow model implies. That doesn’t automatically mean your DCF is wrong. It might mean the market is pricing:
- growth durability beyond your base case, margin expansion, a competitive advantage not captured in your cash flow assumptions, or simply different reinvestment expectations.
Conversely, if multiples imply a much lower value than your cash flow model, you should scrutinize:
- whether the business carries hidden liabilities (customer churn risk, contract term risk, regulatory exposure), whether maintenance capex is undercounted in EBITDA, whether working capital needs are larger than forecast.
I remember a situation where EBITDA margins looked healthy, and the EV/EBITDA multiple supported a strong value. Then we traced working capital and found that revenue growth required a disproportionate build of receivables. The cash conversion was the limiting factor. The multiples had effectively ignored that cash reality because the denominator was anchored to EBITDA, not cash earnings.
Multiples and cash flows are not competitors. They are different lenses. Good judgment is knowing when the lens is missing key details.
Deal context changes the interpretation of the same multiple
Valuation is not only about the company. It is about the buyer and the transaction.
A strategic buyer may pay more for synergies, talent retention, or faster go-to-market. A financial buyer may pay less due to required returns and financing structure. Both can be “reasonable” depending on the expected synergies and risk allocation.
If you apply a multiple from a one-off acquisition to a company without synergies, you may overstate value. If you apply a public trading multiple to a private sale with meaningful control premiums and governance improvements, you may understate value.
The most useful outcome is a range with drivers. Instead of saying “the multiple is 8.5x,” you translate it to: “At a market-like growth and margin profile, value lands here; if the margin profile is lower, value shifts down by this amount.” That framing helps stakeholders make trade-offs.
Common mistakes that keep coming back
Multiples-based valuation can become a ritual: pull peer data, compute an average, multiply it by your company’s denominator, and call it done. That is the path to fragile conclusions.
Here are the mistakes I’ve seen repeatedly, along with why they matter:
Using peers because they look close on the surface, ignoring margin and capital structure differences. Applying adjusted multiples without understanding whether adjustments are recurring or one-time. Anchoring to a single multiple without checking cyclicality, customer concentration, or forecasting credibility. Mixing denominators, for example using forward EBITDA but trailing enterprise value based on a different period’s market conditions. Overconfidence in averages. Outliers exist for a reason, and those reasons can be information.Some of those mistakes are avoidable with better process. Others require better judgment about what the business actually is.
When multiples are the wrong tool, and what to do instead
Multiples are powerful when you are comparing similar businesses with stable, measurable economics. They are less reliable when:
- the company is truly unique with no comparable economics, accounting differences cannot be normalized without inventing adjustments, the business is early stage and financial metrics are not meaningful, profitability is too volatile and forecasts are essentially scenario fiction.
In those cases, you can still use multiples as a reference point, but you should lean harder on other methods: unit economics analysis, cohort retention modeling, and cash flow forecasts tied to operational assumptions.
The point is not to abandon multiples. The point is to stop pretending the ratio is the valuation. It is a pricing signal, not the entire decision.
Bringing it together: what “using multiples wisely” looks like in practice
When you use multiples wisely, you end up doing three things that look boring on paper but matter in the outcome.
First, you make the denominator comparable, not just available. Second, you select peers based on the drivers behind the multiple, not just the label. Third, you treat valuation as a range governed by assumptions, not a single precise number.
If you do that, the work becomes defensible in a room. People can disagree with your assumptions, but they cannot dismiss the process as a finance tools and calculators spreadsheet exercise. And that is the real win with valuation in finance: you provide something stakeholders can challenge productively, instead of something that collapses the moment someone asks what sits inside “EBITDA” or “growth.”
Multiples will always be imperfect. Used thoughtfully, they can still give you an honest sense of where the market is pricing risk and opportunity, and where your specific company story changes the answer.