DCF vs Comparable Company Analysis
A discounted cash flow values a company on its own projected cash flows, while comparable company analysis values it against what the market pays for similar firms. One is intrinsic, the other relative. Neither is the single right answer, and good analysts run both, then triangulate a range where the two methods overlap. Understanding when each is trusted is the heart of valuation judgment.
Intrinsic versus relative
A DCF projects unlevered free cash flow, discounts it at WACC, and adds a terminal value: Enterprise Value = Sum of PV of FCF + PV of Terminal Value. It is intrinsic because the value comes from the company's own forecast, not from the market.
Comparable company analysis applies a peer multiple to a metric: Enterprise Value = Metric * Peer Multiple, for example EV = EBITDA * EV/EBITDA. It is relative because it reflects how the market currently prices similar businesses. A DCF can disagree with comps when the market's mood differs from the forecast's fundamentals.
Worked example
A firm has next-year EBITDA of 200. Peers trade at 8x EV/EBITDA. Separately, a DCF on the same firm produces a present value of cash flows of 1500 and a present value of terminal value of 200.
- Comps enterprise value:
=200*8returns 1600. - DCF enterprise value:
=1500+200returns 1700. - Midpoint of the two methods:
=(1600+1700)/2returns 1650. - The range 1600 to 1700 is the triangulated value band.
| Method | Enterprise Value |
|---|---|
| Comparable companies | 1600 |
| DCF | 1700 |
| Triangulated midpoint | 1650 |
=(1600+1700)/2 returns 1650.
Laying it out in a model
Keep the two methods on separate tabs that feed a single summary, so the comparison is clean and each result is traceable.
- Run the DCF on its own tab with WACC and terminal value as visible inputs.
- Run comps on a peer tab with one multiple per peer and a chosen median.
- Pull both enterprise values into a summary that shows the range.
- Present the overlap, not a single point, since the band is more honest.
Pitfalls
Trusting a DCF blindly is dangerous because it is sensitive to WACC and the terminal growth rate; small changes there swing the value heavily, and the terminal value often dominates. Stress-test those two inputs before relying on the output.
On the comps side, a peer set that is not truly comparable, or stale and outlier multiples, produces a misleading benchmark; screen peers for size, growth, and margin, and use a median rather than a mean. The deeper error is forcing the two methods to match by tweaking assumptions until they agree. Let them disagree, then explain why, and present the resulting range honestly.
Formula Trace
Formula Trace follows each method's value back through its inputs so you can see exactly why the DCF and comps differ.
Get ModelMint See how it worksFAQ
When is a DCF more trusted than comps?
When the company's cash flows are forecastable and stable, or when no clean peer set exists. A DCF captures company-specific fundamentals the market may not yet price. It is less trusted for early-stage firms where forecasts are highly uncertain.
When are comps more trusted than a DCF?
When there is a deep set of genuinely comparable public companies and the business is hard to forecast far out. Comps reflect current market pricing and require fewer long-range assumptions, though they inherit whatever optimism or pessimism the market holds.
Should DCF and comps give the same number?
Not necessarily. They answer different questions, intrinsic versus relative, so a gap is informative. Rather than force them to agree, present the range and explain the difference. The overlap band is usually the most defensible conclusion.