Scenario Analysis and Sensitivity Testing

Valuation Methods — Lesson 7 of 10

Every intangible asset valuation is built on assumptions. The royalty rate in an RFR analysis. The attrition rate in an MPEEM. The discount rate in every income approach method. The growth rate, the useful life, the tax rate, the competitive landscape. Each assumption carries uncertainty, and that uncertainty compounds across the model.

A single-point valuation — "the brand is worth $45 million" — is a fiction of precision. No asset has a single correct value. It has a range, and the width of that range depends on how sensitive the valuation is to its key assumptions. Professional valuers present ranges, not points. They test assumptions systematically, identify which inputs drive the most variation, and communicate uncertainty transparently.

This lesson covers the three principal techniques for analysing uncertainty in intangible asset valuations: sensitivity analysis, scenario analysis, and Monte Carlo simulation.

★ Key Takeaway

A single-point valuation is the beginning of the analysis, not the end. Professional practice requires sensitivity testing (varying one input at a time), scenario analysis (varying multiple inputs simultaneously), and ideally Monte Carlo simulation (probabilistic modelling). These techniques identify the key value drivers, quantify the range of plausible outcomes, and enable informed decision-making.


Sensitivity Analysis: One Variable at a Time

Sensitivity analysis tests how much the valuation changes when a single input is varied while all other inputs are held constant. It answers the question: which assumptions matter most?

The Tornado Chart

The most effective way to present sensitivity results is a tornado chart — a horizontal bar chart showing, for each input, the valuation range when that input is varied from its low to high estimate.

For a brand valued using the Relief from Royalty method, a typical tornado analysis might test:

Sensitivity Analysis — Brand Valuation ($M)

Input Low Estimate Base Case High Estimate Value at Low Value at High Swing
Royalty rate 4.0% 5.5% 7.0% 32.8 45.0 57.2
Revenue growth 2.0% 5.0% 8.0% 38.5 45.0 52.3
Discount rate 10.0% 12.0% 14.0% 50.2 45.0 40.6
Useful life 10 yrs 15 yrs Indefinite 36.1 45.0 52.8
Tax rate 20% 25% 30% 46.8 45.0 43.2
$24.4M swing from royalty rate variation alone
$32.8M - $57.2M full valuation range from all sensitivities
54% of total swing driven by royalty rate

This analysis reveals that the royalty rate drives more than half the total valuation uncertainty. It tells the valuer — and the reader — exactly where to focus diligence efforts. If the royalty rate can be pinned down more precisely (through additional comparable licensing agreements or profit-split analysis), the overall uncertainty narrows significantly.

ℹ Note

Sensitivity analysis tests one variable at a time, which means it does not capture interactions between variables. In reality, a higher revenue growth rate might justify a higher royalty rate (because the brand is more valuable in a growing market). Scenario analysis addresses this limitation.


Scenario Analysis: Coordinated Assumptions

Scenario analysis tests the valuation under multiple coordinated sets of assumptions. Unlike sensitivity analysis, it varies several inputs simultaneously to create internally consistent alternative futures.

Constructing Scenarios

Professional practice typically employs three scenarios, though five can be appropriate for complex valuations.

Three-Scenario Framework

  • Base case: Management's best estimate — the most likely outcome
  • Downside: Conservative assumptions — weaker growth, higher churn, competitive pressure
  • Upside: Optimistic but plausible — stronger-than-expected performance

Five-Scenario Framework

  • Severe downside: Recession, competitive disruption, regulatory change
  • Moderate downside: Below-plan performance
  • Base case: Management's best estimate
  • Moderate upside: Above-plan performance
  • Strong upside: Market expansion, breakthrough adoption

Application: Customer Relationships (MPEEM)

Consider the DataCo example from Lesson 3. The base case valued customer relationships at $58.1 million using a 10% attrition rate and 5% revenue growth.

Scenario Analysis — DataCo Customer Relationships

Assumption Downside Base Upside
Annual attrition rate 15% 10% 7%
Revenue growth (year 1) 2% 5% 8%
Operating margin 13% 15% 17%
Discount rate 14% 13% 12%
Customer relationships value ($M) 38.2 58.1 76.5
As % of purchase price 25% 39% 51%

The scenario analysis tells a richer story than the base case alone. It reveals that customer relationships could be worth between $38 million and $77 million — a range of nearly 2x — depending on customer retention performance and market conditions.

✔ Example

In a PE acquisition, the downside scenario is particularly important because it informs the acquirer's understanding of risk. If the downside value of customer relationships ($38.2M) still supports the deal thesis at the agreed purchase price, the transaction is robust. If the deal only works in the base and upside cases, the buyer is making a bet on customer retention that may not be adequately compensated by the purchase price.


Monte Carlo Simulation: Probabilistic Modelling

Monte Carlo simulation is the most rigorous approach to quantifying valuation uncertainty. Instead of testing discrete scenarios, it models each uncertain input as a probability distribution and runs thousands of simulations to generate a probability distribution of the output value.

How It Works

1. Define input distributions

For each uncertain input, specify a probability distribution. Common choices: normal (symmetric uncertainty), triangular (with a most likely value), uniform (equal probability across a range), or lognormal (for skewed inputs like royalty rates).

2. Run simulations

The model randomly draws values from each distribution and calculates the asset value. This is repeated 10,000+ times, producing a distribution of possible values.

3. Analyse the output distribution

The output provides: the mean (expected value), median, standard deviation, and percentile ranges (e.g., 10th to 90th percentile). It also reveals which inputs contribute most to output variance.

When to Use Monte Carlo

Monte Carlo simulation is justified when:

  • The asset value is material to the transaction (typically >$10 million)
  • Multiple inputs carry significant uncertainty
  • The relationship between inputs is non-linear
  • Stakeholders require a probabilistic assessment (common in litigation and regulatory contexts)

It is rarely necessary for routine valuations of smaller assets, where sensitivity analysis and three-scenario analysis provide sufficient insight.

Monte Carlo Input Specification Example

Input Distribution Parameters Rationale
Royalty rate Triangular Min 4%, Mode 5.5%, Max 8% Comparable data suggests this range
Revenue growth Normal Mean 5%, SD 2% Symmetric uncertainty around plan
Discount rate Uniform 10%-14% Genuine uncertainty about risk premium
Useful life Triangular Min 8, Mode 15, Max 25 years Brand longevity is uncertain

Interpreting Monte Carlo Output

A Monte Carlo simulation might produce: Mean value $47.2M, Median $45.8M, 10th percentile $31.5M, 90th percentile $65.3M. This tells you that there is an 80% probability the true value lies between $31.5M and $65.3M. The mean ($47.2M) is the expected value — the probability-weighted average across all simulations. For accounting purposes (IFRS 13 fair value), the base case single-point estimate is typically used, but the Monte Carlo range informs the reasonableness assessment and the disclosure of measurement uncertainty.


Presenting Ranges to Stakeholders

How you communicate uncertainty is as important as how you measure it. Different stakeholders need different presentations.

Stakeholder Preferred Presentation Why
Board / acquirer Three scenarios with probability weights Supports strategic decision-making
Auditor Sensitivity table + WARA reconciliation Demonstrates assumption reasonableness
Tax authority Single value with documented support for each input Transfer pricing requires a specific arm's-length price
Litigation Full Monte Carlo with confidence intervals Courts require explicit quantification of uncertainty
Investor presentation Range with base case highlighted Communicates conviction with appropriate caveats
⚠ Warning

Never present a single number without disclosing the sensitivity to key assumptions. A valuation of "$45 million" without context implies a precision that does not exist. "$45 million (range $32-58 million, primarily sensitive to the royalty rate assumption)" is a honest and professional presentation.


What Comes Next

In Lesson 8: Selecting the Right Method, we synthesise the lessons so far into a practical decision framework — matching asset types, data availability, and valuation purposes to the most appropriate method.


Tony Hillier is an advisor to Opagio with over 30 years of experience in structured finance, valuation, and due diligence across private equity and corporate transactions. Meet the team.