Quick answer
DCF sensitivity is the check that tells you whether your valuation conclusion still holds when discount rate and terminal growth move within a defensible range. If a small change in assumptions creates a huge swing in value, the model is fragile and the single headline number should not be treated as truth.
Why sensitivity matters
Use sensitivity analysis when the decision matters more than the headline valuation. A DCF can look precise while the answer is still fragile, so the real question is how valuation changes when WACC and terminal growth move inside a defensible range.
How to run DCF sensitivity (step-by-step)
- Start with your base-case DCF (FCF forecast, discount rate, terminal growth).
- Pick a range for discount rate (r) and terminal growth (g).
- Recalculate value for each r/g pair to build a grid.
- Compare how far valuations move from the base case.
How to tell if the DCF is robust
- If the valuation story stays directionally similar across reasonable WACC and terminal-growth pairs, the conclusion is more robust.
- If a 1-2 point move in WACC or terminal growth changes the decision, treat the DCF as fragile and widen the decision range.
- If terminal value dominates most of enterprise value, challenge the forecast length, reinvestment logic, and terminal assumptions before defending the headline output.
How to pick ranges
- Discount rate: start with WACC as a base, then test +/-1-3%.
- Terminal growth: test conservative long-run rates (often 0-4% depending on context).
- If terminal dominates EV, consider extending the forecast or making assumptions more conservative.
DCF sensitivity example
If your base case uses a 12% discount rate and 3% terminal growth, test 10-14% for discount rate and 2-4% for terminal growth. A stable valuation across the grid signals robustness.
How to read the grid
- Look for stability: if value swings wildly, the conclusion is fragile.
- Use the base case as a reference point, not as the only outcome.
- Check whether the same story holds across reasonable rate and growth pairs.
Common traps
- Terminal growth >= discount rate (invalid in perpetuity model).
- Picking a range that's too narrow and creating false confidence.
- Using accounting earnings instead of cash flow for valuation.
Modeling checklist
- Keep units consistent (annual cash flows with annual discount rates).
- Reconcile terminal value to a reasonable share of total EV.
- Run sensitivity on margin and reinvestment if they drive FCF.