Creating Sensitivity Tables and Heat Maps
A one-at-a-time sensitivity table ("What if growth is 5% vs. 8% vs. 12%?") tells you which input matters. A two-way sensitivity table shows you how two assumptions interact—revealing where your valuation lives under realistic combinations of scenarios and making trade-offs visceral.
Quick Definition
A sensitivity table (or heat map) is a matrix where rows represent values of one assumption (e.g., WACC), columns represent values of another (e.g., terminal growth rate), and cells display the resulting intrinsic value. Shading (gradient from red to green) instantly visualizes which zones are overvalued, fairly valued, and undervalued relative to current price—transforming abstract assumptions into actionable conviction zones.
Key Takeaways
- Two-way tables surface critical trade-offs: "If growth is disappointing, margin better not compress"
- Heat maps let you identify the minimum assumptions needed for your thesis to work
- Color coding (red/yellow/green) communicates valuation likelihood to non-technical stakeholders
- Sensitivity grids show you which assumptions have the biggest valuation impact and deserve most analysis effort
- Anchoring current stock price on the heat map reveals how much downside you're buying and where base case sits
Building a Two-Way Sensitivity Table
Start with your completed DCF model. Choose two high-impact assumptions to vary; typical pairs:
-
Terminal Growth Rate (rows) × Discount Rate / WACC (columns)
- Captures perpetual value driver × cost of capital
- Shows how valuation compresses if rates rise or risk increases
-
Long-Term Revenue Growth (rows) × Operating Margin (columns)
- Core P&L drivers; shows whether revenue growth requires margin sacrifice
- Answers: "Can we grow and keep margin?"
-
EBITDA Multiple (rows) × Next 5-Year Revenue Growth (columns)
- For terminal-value-multiple approaches
- Shows premium justified by growth trajectory
Worked Example: Mid-Cap Software Company
Base case assumptions:
- 5-year revenue CAGR: 20%
- Terminal FCF growth: 3%
- WACC: 8%
- FCF margin (year 5): 28%
- Current stock price: $85/share
Build a table: Rows = Terminal Growth Rate (1%, 2%, 3%, 4%, 5%), Columns = WACC (6%, 7%, 8%, 9%, 10%)
| Terminal Growth | 6% WACC | 7% WACC | 8% WACC | 9% WACC | 10% WACC |
|---|---|---|---|---|---|
| 1% | $92 | $78 | $68 | $60 | $53 |
| 2% | $105 | $87 | $74 | $64 | $56 |
| 3% | $122 | $99 | $82 | $70 | $61 |
| 4% | $145 | $115 | $93 | $78 | $67 |
| 5% | $178 | $138 | $108 | $89 | $76 |
Current price: $85. Base case (8% WACC, 3% growth): $82. Stock is fairly valued.
Interpretation:
What does this tell you?
- Fair value is robust across reasonable assumptions (stays $74–$115)
- Downside scenario (rates spike, growth stalls): $53–$64
- Bull case (rates fall, growth sustains): $99–$178
- Current stock price is near the edge of the base case—limited margin of safety in a normalized rate environment
Heat Maps: Visual Sensitivity
Translate the table into color:
- Red: Stock is expensive (intrinsic value << current price)
- Yellow: Stock is fairly valued
- Green: Stock is cheap (intrinsic value >> current price)
Same example, color-coded:
| Terminal Growth | 6% WACC | 7% WACC | 8% WACC | 9% WACC | 10% WACC |
|---|---|---|---|---|---|
| 1% | 🟡 $92 | 🟡 $78 | 🟢 $68 | 🟢 $60 | 🔴 $53 |
| 2% | 🟢 $105 | 🟡 $87 | 🟡 $74 | 🟢 $64 | 🟢 $56 |
| 3% | 🟢 $122 | 🟢 $99 | 🟡 $82 | 🟡 $70 | 🟢 $61 |
| 4% | 🟢 $145 | 🟢 $115 | 🟢 $93 | 🟡 $78 | 🟢 $67 |
| 5% | 🟢 $178 | 🟢 $138 | 🟢 $108 | 🟢 $89 | 🟡 $76 |
Cutoff: Green if intrinsic value ≥ $85; Yellow if $70–$84; Red if < $70.
The heat map immediately shows: Your base case needs either low rates OR sustained growth. If both disappoint, you're in trouble. That's the opposite of reassuring if you're buying at $85 in a rising-rate environment.
Three-Way Sensitivity (Advanced)
For more nuance, create separate matrices for a third dimension:
Example: Revenue Growth (rows) × Margin (columns), with separate sheets for different WACC scenarios
Bull case (WACC 7%):
| Revenue CAGR | 20% Margin | 25% Margin | 30% Margin |
|---|---|---|---|
| 15% | $72 | $85 | $101 |
| 20% | $89 | $110 | $135 |
| 25% | $108 | $138 | $172 |
Base case (WACC 8%):
| Revenue CAGR | 20% Margin | 25% Margin | 30% Margin |
|---|---|---|---|
| 15% | $61 | $71 | $84 |
| 20% | $74 | $90 | $112 |
| 25% | $90 | $114 | $144 |
Bear case (WACC 10%):
| Revenue CAGR | 20% Margin | 25% Margin | 30% Margin |
|---|---|---|---|
| 15% | $48 | $55 | $63 |
| 20% | $58 | $70 | $84 |
| 25% | $70 | $88 | $110 |
Now you see the interaction: in the bear case, high growth alone doesn't save you—you also need margin expansion. In the bull case, you have flexibility.
Common Mistakes
Cherry-Picking Ranges If you set WACC from 5%–11% and terminal growth from 0%–8%, you're not exploring realistic scenarios—you're just widening the range to encompass any outcome. Use peer data, historical ranges, and management guidance to justify your bounds.
Ignoring Correlation High revenue growth rarely coexists with margin compression in healthy industries. Yet most sensitivity tables treat them as independent. Consider noting on the table: "Correlation: High growth + margin expansion expected" or create separate matrices for different competitive scenarios.
Overloading Too Many Dimensions Five rows and twenty columns becomes unreadable. Stick to 5×5 or 6×6 matrices. If you need more granularity, create multiple two-way tables for different scenario pairs.
Setting Current Price as the Baseline Instead, anchor to your base-case intrinsic value. Color code relative to fair value, not stock price. This keeps the table grounded in fundamentals, not market sentiment.
Forgetting to Stress-Test Against Reality If your "bearish" scenario (WACC +150 bps, growth -5 points) never actually occurs in history or is impossible given industry dynamics, it's not useful. Validate your grid bounds against actual precedent.
Practical Applications
Investment Committee Presentation Show the base-case matrix with current price overlaid. The visual immediately tells board members: "In 60% of realistic scenarios, we're within 10% of fair value. In 20%, we're 20%+ cheap. In 20%, we're overvalued."
Thesis Pressure Testing Ask: "What assumptions must be true for this to be a good buy?" Find that intersection on the heat map. Is it 1 cell, or a whole quadrant? One cell = high conviction needed. One quadrant = resilient thesis.
Scenario Communication Instead of saying "bear case is $60," use the table: "If rates go to 10% and growth drops to 2%, fair value is $64—but that scenario is unlikely (only 8% probability in our Monte Carlo)." You're anchoring the bad outcome while contextualizing its likelihood.
Position Sizing Heat maps reveal volatility of outcomes. If intrinsic value swings from $50 to $180 across realistic assumptions, size your position smaller (higher uncertainty). If fair value stays $85–$95, size it larger (high conviction).
FAQ
Q: Should I show the heat map to the company or only keep it internally? A: Internally. External sensitivity tables can be misused by management to justify favorable scenarios or by shorts to attack weak-case assumptions. Keep the disciplined version to yourself; communicate your conviction range (e.g., "Fair value $80–$110") publicly.
Q: What if I can't agree on reasonable ranges for an assumption? A: That's valuable information. It means the assumption is uncertain and deserves more work. Zoom in: run 10 micro-scenarios instead of 5 broad ones. Or escalate to management for clarity.
Q: How often should I update the heat map? A: After each quarterly earnings (new growth trajectory), major market moves (rate changes), or when you're reconsidering the thesis. A six-month-old heat map in a 20% market rally is dangerously stale.
Q: Can I use a heat map for sell decisions? A: Absolutely. If the stock rises and pushes into the red zone of your map (overvalued in 70%+ of realistic scenarios), that's a sell signal. The table makes that discipline objective.
Q: What if the stock price falls between two cells? A: Interpolate or note: "Fair value most likely $82–$88 based on matrix; stock at $85 is fairly valued with modest upside." Avoid false precision, but do establish a range.
Related Concepts
- Scenario Analysis: Discrete Bull/Base/Bear cases; heat maps are the continuous version
- Monte Carlo Simulation: Probabilistic outcomes; sensitivity tables show the mechanics
- Margin of Safety: Discount to intrinsic value; heat maps reveal where your true margin sits
- Break-Even Analysis: What assumption level makes the stock fairly valued? Find on the heat map
- Stress Testing: Similar logic applied to liquidity, solvency, competitive scenarios
Summary
Sensitivity tables and heat maps translate abstract assumptions into visual, actionable conviction zones. They answer the core investor questions: "How wrong can I be and still have a winning thesis? Where does my conviction actually sit? What happens in realistic bad outcomes?" By anchoring valuation to a matrix of plausible assumptions rather than a single point estimate, you build resilience into your process and defend against the inevitable moments when assumptions prove wrong.
The discipline of building a robust sensitivity table forces you to think deeply about correlations, trade-offs, and the true range of outcomes—work that often reveals why a stock looks cheap (it's not; you're assuming unrealistic margin expansion) or why a seemingly expensive name has hidden appeal (two-thirds of realistic scenarios support it).
Next Steps
Sensitivity tables show you what the market is implicitly assuming about a company's future. What Growth is Priced In? teaches you to reverse-engineer those assumptions—to ask, "What must investors believe for the stock to be at this price?"