Why scenario modelling
Single-point forecasts are always wrong. Scenarios let you answer "what's the worst that could happen, and what do we do then?" before it happens.
The classic three
- Base — your best estimate
- Downside — realistic adverse case (e.g. top customer leaves, revenue down 15%)
- Stress — survival case (e.g. revenue down 35%, top customer doesn't pay final invoices)
For each scenario, model:
- Revenue and gross margin
- Headcount and overhead response
- Cash runway in weeks
- Break-even point in months
The variables to stress
- Top-customer concentration — what if your biggest 1–3 customers leave?
- Pricing — what if you have to cut prices 10% to hold volume?
- Cost inflation — payroll +5%, energy +20%
- Working capital — DSO worsens by 15 days
- Tax — corporation tax rises, R&D credit denied
- Funding — credit line withdrawn at renewal
Outputs that drive decisions
For each scenario document:
- Trigger — what would signal we're in this scenario
- Lead indicators — what we'd see 2–3 months ahead
- Response plan — specific actions, with owners and dates
- Cash impact — when does it bite, how much cash do we need
Tooling
- Spotlight Reporting and Fathom both support multi-scenario forecasts
- Float lets you layer scenarios over a base cash forecast
- For complex businesses, a custom Excel model still wins
Common mistakes
- Building "best case" scenarios — useless; never plan for them
- Not updating scenarios as the business evolves
- Modelling but not pre-deciding the response
Ernest & Co Premium clients get a quarterly scenario review built into their management pack.