Gary Crittenden was CFO of American Express and Citigroup before he started buying companies, and the first thing he'll tell you about underwriting one is that EBITDA can't be read straight. Two businesses can post the identical number and be worth wildly different amounts, and the figure alone won't say which is which. What he trusts instead is return on capital measured against the cost of capital, and whether that spread is widening as the business grows.
The one spread that drives value
Strip away the complexity and durable value comes down to one relationship. "Return on capital above the cost of capital, executed well over time and growing, is the key factor that kind of drives profitability," Crittenden says.
The spread is what matters, not the raw return. A business earning 12% on capital when its capital costs 10% is creating value slowly. A business earning 30% on capital that costs the same 10% is creating it fast, and if that spread widens as the company grows, the compounding is what produces extraordinary outcomes. EBITDA tells you what the business earned. Return on capital above cost tells you whether earning it was worth the capital it consumed, which is the question an underwriter actually needs answered.
Why a dollar of EBITDA isn't a dollar
This is the part most analysis skips. "Not all EBITDA is created equal," Crittenden says, and the reason is that two dollars of EBITDA can sit on very different amounts of invested capital.
A dollar of EBITDA generated by a capital-hungry, low-return business is worth less than a dollar generated by a high-return one, because the first dollar required far more capital to produce and will require more to grow. Fee income is the clearest example of the high-quality kind. "Businesses that have the ability to drive fee income at a relatively modest capital investment have the ability to drive up that margin and tend to be the leaders in terms of value creation," Crittenden says. The lesson for a deal team is to stop reading EBITDA as a single number and start asking what it cost to produce.
Three businesses, one lesson
Crittenden's examples make the abstraction concrete. At American Express, the charge card and the credit card look similar from the outside and behave completely differently underneath. "The return on capital on a credit card is, say, something like 14 or 15%. The return on capital on a charge card is more like 38%," he says. That’s the same company and same customer base, but more than double the return on capital. The charge business simply consumes far less of it.
The Platinum card's annual fee, $895, is the purest version of this idea: fee income that arrives with almost no incremental capital behind it. Costco runs the same playbook in retail, where the membership fee, not the thin margin on the groceries, is the high-return engine. These different businesses share the same underwriting lesson: find the part of the model that earns a high return on modest capital, because that’s where the value compounds.
The hard-to-grow paradox
Here is the counterintuitive corollary. High-return businesses are usually hard to grow, and that difficulty is exactly why they’re valuable. If a business could pour capital in and reliably scale its high returns, every competitor would do the same and the returns would compete away. The fact that high-ROC models resist effortless scaling is what protects the spread. So when a deal team finds a target with genuinely high return on capital, the right reaction is recognizing that the moat and the difficulty are the same thing rather than disappointment that it’s hard to grow quickly.
Scoring a target
This lens turns into a practical checklist. Start with the spread: what is the target's return on invested capital, and how does it compare to its weighted average cost of capital? A narrow or negative spread is a warning regardless of headline EBITDA. Then test the quality of the EBITDA: how much invested capital sits behind each dollar of it, and how much fee or recurring income is in the mix versus capital-intensive volume?
Next, ask about the trajectory. Is the spread widening or compressing as the company grows? And finally, run the paradox check: if the returns are high, is the difficulty of scaling them a sign of a real moat, or of a business that has simply topped out? A target that scores well across those four questions has the kind of EBITDA worth paying up for.
How a relationship intelligence CRM fits the diligence
None of this lives in a spreadsheet alone. Scoring EBITDA quality and the durability of a return profile depends on what a deal team learns from operators, former executives, customers, and co-investors who know the business from the inside. Affinity is the AI-first private capital CRM, and it keeps the relationship and diligence signals behind a target organized and current, so the team can see who at the firm has talked to people close to the company and what they learned. The return-on-capital lens tells you what to look for; the relationships are how you confirm whether the quality is real.
See how Affinity keeps your diligence relationships and signals organized through a deal. Request a demo.
Read the guide to private equity due diligence.
FAQ
What is return on invested capital, and why does it matter in private equity?
Return on invested capital (ROIC) measures how much profit a business generates relative to the capital deployed to run it. It matters in private equity because value is created when that return sits above the cost of capital and grows over time. A high headline EBITDA on a low return on capital is far less valuable than a smaller EBITDA on a high one, so ROIC is a sharper underwriting lens than EBITDA alone.
Why isn't all EBITDA created equal?
Because two dollars of EBITDA can require very different amounts of invested capital to produce. A dollar earned by a capital-light, high-return business (fee income, for example) is worth more than a dollar earned by a capital-hungry, low-return one, since the second required more capital to generate and will need more to grow. Quality of EBITDA, not just the amount, determines the value.
What is a high-quality EBITDA business?
One that earns a high return on modest invested capital, often through fee or recurring income rather than capital-intensive volume. Former Citi and American Express CFO Gary Crittenden points to the AmEx charge card (around 38% return on capital versus 14 to 15% on the credit card), the Platinum card's annual fee, and Costco's membership model as examples where the high-return engine is fee income carrying little incremental capital.
Why are high-return businesses often hard to grow?
Because if high returns could be scaled easily by adding capital, competitors would do the same and compete the returns away. The difficulty of scaling is what protects the spread between return on capital and cost of capital. For an underwriter, a high-ROC business that resists easy growth often signals a real moat rather than a weakness.
How should a deal team score a target's EBITDA quality?
Compare return on invested capital to the cost of capital and check whether the spread is positive and widening; measure how much invested capital sits behind each dollar of EBITDA; weigh fee or recurring income against capital-intensive volume; and assess whether any difficulty in scaling reflects a moat or a topped-out business. A target that scores well across those checks has durable, high-quality EBITDA.
.png){kind=logo}
.webp)
.webp)
.webp)