UNDERSTAND THE METHOD
Select comparable companies before choosing the valuation
A comparable company should help explain the target’s economics: cash generation, growth and risk. An industry label is a starting filter, not proof of comparability. Decide the selection criteria before seeing which set gives a preferred valuation.
| Set | Multiples | Median | EV | Equity Value |
|---|---|---|---|---|
| All candidates | 6 / 7 / 8 / 14 / 20× | 8× | 160 | 130 |
| Documented A–C set | 6 / 7 / 8× | 7× | 140 | 110 |
On this page
Write a target profile and a consistent measurement basis
The synthetic target sells mainly project-based B2B services in one region, with modest growth and a 10% EBITDA margin. Target EBITDA is EUR 20 on the stated 2027 forecast basis; net debt is EUR 30. Candidate values are a supplied snapshot at 2026-12-31. A–D use LTM earnings; E supplies NTM earnings. The raw native table is explicitly labeled “Supplied sample” and E’s name records NTM, so its mixed period is visible before exclusion. This forward-target/LTM-peer convention is explicit, not a claim that the denominators are perfectly matched.
For a real analysis, normalize accounting, leases, exceptional items and financial periods before treating differences as pricing. Separate an economic exclusion from a missing-data exclusion. Keep a record of both, including the source date.
Use a matrix with documented reasons
All names and numbers below are synthetic. A–C share the target’s project model, region, margin neighborhood and LTM convention. D has a materially higher recurring mix and growth; E uses a different earnings period. Those differences are the reasons for exclusion, not their high multiples.
| Peer | Mix / growth / margin | Scale / geography / period | EV/EBITDA | Decision |
|---|---|---|---|---|
| A | 80% project / 4% / 10% | 5× target revenue / same region / LTM | 6× | Include; scale remains a limitation |
| B | 75% project / 5% / 11% | 4.55× / same region / LTM | 7× | Include; comparable delivery economics |
| C | 85% project / 3% / 9% | 5.56× / same region / LTM | 8× | Include; inspect margin difference |
| D | 90% recurring / 20% / 25% | 2× / global / LTM | 14× | Exclude: different business economics |
| E | 80% project / 4% / 10% | 5× / same region / next-year earnings | 20× | Exclude until period alignment |
Reconcile the effect without cherry-picking
The native all-candidate median is 8×: 8 × 20 = EV 160, less net debt 30 = Equity Value 130. With A–C included, the median is 7×: EV 140 and Equity Value 110. The EUR 20 equity difference comes entirely from the selected multiple here; the target metric and bridge did not change.
A smaller selected set can be more coherent but less statistically stable. A–C still differ in scale, margin and growth; selection does not remove every comparability issue. Report the original candidates, exclusions and sensitivity to plausible alternatives rather than deleting unfavorable evidence.
Download the synthetic example, explicitly import it under Model, select Advanced and open Value → Trading Comparables. Import creates a separate model with inspectable inputs and sources.
Carry the evidence into the native peer table
In Trading Comparables, enter your own peer values and financial metrics, document the period and use the inclusion control. Check the selected statistic and the EV-to-equity bridge separately. A percentile range does not repair an inconsistent metric basis.
BalanceCheat does not supply a market-data feed or automatically screen real companies. This page demonstrates selection discipline with synthetic evidence; it does not identify investable securities or certify fair value.
Related decisions and model reference
For adjacent questions: Trading Comparables & Comparable Company Analysis; Mean vs Median Valuation Multiples: Trading Comps Example; Company Valuation — Build a Defensible Value Range; Valuation Methods — Choose DCF, Comps or Precedents.
For the exact controls and source behavior, use Trading Comparables and EV-to-equity bridge and value per share.
Sources and conventions
Synthetic examples. Model-based results were reconciled with the current BalanceCheat calculations; assumptions, units and boundaries are specified in each example.