When a number carries
more weight than a story
Valuation is where financial analysis gets honest. These guides cover the methods analysts actually use — not simplified for a classroom, but structured for real decisions.
Picking the wrong model for a situation, or applying the right model without understanding its assumptions, is one of the most common and costly errors in financial practice. Each guide here addresses a specific point of failure.
Six areas where valuation gets complicated
Discounted Cash Flow — assumptions that distort
DCF produces a precise figure from imprecise inputs. Terminal value often accounts for 60–80% of the output, yet analysts frequently use a perpetuity growth rate that does not match the sector or economic cycle. The guide examines how to set defensible WACC, why mid-year discounting changes results by 3–6%, and how to anchor terminal value assumptions with observable data.
Comparable company selection without confirmation bias
Selecting peers is judgement-based, which means it is easy to weight the sample toward a preferred outcome. The guide outlines a structured screening process across 4 quantitative filters — revenue band, margin profile, geographic exposure, and capital intensity — before applying any multiple. Includes a worked example across 11 candidate companies narrowed to 5.
Precedent transactions — reading the deal context
Transaction multiples embed strategic premiums that have nothing to do with standalone value. A deal closed during a credit cycle peak at 12× EBITDA tells a different story than one at 8× in a tighter market. The guide covers how to adjust for deal timing, buyer type (strategic vs. financial), and target size before including any transaction in a reference set.
Sensitivity analysis that informs rather than decorates
Most sensitivity tables appear in presentations as reassurance — the central case is always centre. The guide builds analysis around the inputs that actually move the output: revenue growth rate and operating margin interact differently than WACC and capex intensity. Covers tornado diagrams, scenario-pair tables, and Monte Carlo framing without requiring specialist software.
Choosing between methods — a structured decision framework
No single valuation method fits all situations. Early-stage companies with negative EBITDA cannot use EV/EBITDA multiples. Asset-heavy industrials rarely yield useful results from a pure DCF without a separate asset appraisal. The guide provides a decision tree that maps company lifecycle stage, data availability, and purpose of valuation to the most appropriate primary and cross-check method.
- DCF — stable cash flows, predictable capital structure
- Comps — mature sectors with active public peers
- Precedent transactions — acquisition or exit context
- NAV — real estate, holding companies, asset-heavy
- Revenue multiples — high-growth, pre-profit businesses
- LBO model — private equity acquisition analysis
Running a full valuation from context to conclusion
Each step builds on the previous. Skipping context in step 1 produces model outputs that look precise but cannot withstand scrutiny from a buyer, regulator, or investment committee.
Before opening any model, document what the valuation is for — a minority investment, a full acquisition, a restructuring, or management planning. Purpose changes which methods are appropriate and which metrics carry weight with the decision-maker.
Identify the company's stage, sector, and capital structure. A 4-year-old SaaS business and a 30-year-old industrial manufacturer require entirely different starting frameworks.
Allow 2–4 hours for context documentation on a new companyUse the decision framework from Guide 05 to select your primary method. Then choose 1–2 cross-check methods that draw on different data sources — ideally one market-based and one intrinsic-value approach.
Document why you rejected alternatives. In a professional setting, the reasoning behind method selection is as important as the output itself.
Minimum 2 independent methods for any formal valuationConstruct a 5-year financial forecast with explicit revenue drivers, cost assumptions, and capital expenditure rationale. Avoid extrapolating historical growth rates without testing whether the underlying conditions still hold.
Run sensitivity analysis on the 3 inputs your model is most responsive to. A well-structured sensitivity table communicates more honestly than a point estimate.
Test at least 3 sensitivity variables, not just WACCCompare the implied value ranges from each method. If DCF returns a range of $38M–$52M and the comps analysis returns $44M–$58M, the overlap zone ($44M–$52M) is your defensible reference band.
Large divergence between methods — say 30% or more — usually signals a broken assumption somewhere, not that one method is simply wrong. Investigate before presenting.
Divergence above 30% warrants assumption reviewThe final output is not a single number — it is a range with a clear rationale. Document every material assumption, the source of comparable data, and what would cause the estimate to shift significantly.
A valuation that cannot explain its own inputs is not analysis — it is arithmetic. Stakeholders should be able to disagree with your assumptions and re-run the logic, not just accept or reject the conclusion.
Present a value range, not a single point estimate