Stress Scenario Simulator
Enter four quarters of actuals, then shock the next one. Each shock propagates the way it does in a real business — through fixed costs that do not flex, working capital that absorbs cash, and interest that rises when you draw more. The output is a projected quarter, a stress index, and the exact size of each shock that breaks you.
Companion to the Financial Stress Forecaster. That one asks where a business is. This one asks what breaks it.
Actuals & projection
Q1 oldest, Q4 most recent. The highlighted column is the projected quarter under your scenario.
| Metric | Q1 | Q2 | Q3 | Q4 | Q5 proj |
|---|
Scenario
Applied to the projected quarter only.
Projected quarter
How the shock lands, line by line.
| Line | Q4 actual | Q5 baseline | Q5 stressed |
|---|
Breaking points
Each shock applied alone, from today. The magnitude that tips this business into Serious (60) and Acute (80).
How a shock propagates
Most scenario tools apply a haircut to revenue and stop. That understates the damage, because a revenue shock does not arrive alone — it arrives through a cost structure that will not flex, a working-capital cycle that absorbs cash, and a facility that charges more when you lean on it harder. The engine here follows that chain.
The chain
Revenue falls. Fixed costs do not. With fixed costs at 52% of the base, a 20% revenue fall does not cut EBITDA by 20% — it cuts it by far more, because only the variable half shrinks with you. That is operating leverage doing its work, and it is the single biggest reason scenario models understate stress.
Receipts lag revenue. A receivable shock widens the gap between what you booked and what you banked: receipts = revenue × (1 − ΔDSO / 90). Stretching DSO by 20 days removes roughly a fifth of a quarter’s collections whatever the P&L says.
Working capital absorbs cash. Receivable and inventory build are charged against operating cash flow directly, so growth in a stressed quarter consumes rather than generates.
Interest rises twice. Once from the rate shock itself, and again because utilisation climbs to fund the shortfall. Utilisation is endogenous here — if the projected cash flow will not cover the quarter, the model draws the line down to cover it, which raises interest, which worsens the following quarter. That feedback is the point.
Cash absorbs everything. Whatever is left lands on the balance, and the runway recalculates.
Why the window rolls
The projected quarter does not just get scored on its own. It is appended to the history and the oldest quarter drops off, so the model scores Q2–Q5. Every trend feature — revenue slope, DSO trend, cash-to-revenue divergence, cash balance trend — recalculates on the new window.
This matters more than it sounds. A one-off shock in an otherwise improving business barely moves the index, because three of the four quarters still trend well. The same shock landing on a business already drifting compounds — it steepens a slope that was already negative. The same shock is not the same event. Context is carried in the trend, and that is exactly what a single-quarter haircut model cannot see.
Breaking points — the reverse stress test
The conventional question is “what happens if revenue falls 20%?” The more useful one runs the other way: how far can revenue fall before this business is in trouble?
The panel answers it by binary search. For each shock dimension independently, the tool finds the smallest magnitude that pushes the stress index to 60 (Serious) and to 80 (Acute), holding everything else at baseline. The result is a ranked list of what this specific business is most fragile to — which is rarely the thing management is watching.
A firm with heavy fixed costs breaks on demand. A firm financing a long receivable cycle breaks on DSO. A firm carrying floating-rate debt near its covenant breaks on rates. Same stress index today; three entirely different vulnerabilities, and three different hedges.
What this is not
It is a single-quarter projection, not a multi-period plan. It holds management passive — no cost actions, no facility renegotiation, no asset sales — which makes it a measure of exposure rather than a forecast of outcome. Real businesses respond, and the gap between this projection and what actually happens is the value of the management team.
The stress index comes from the same fitted model as the stress forecaster, so the same caveats apply: synthetic development data, a Gini around 0.52, and an ordering between trend and level features that proved sensitive to specification. Use the direction and relative size of the movements, not the absolute index value.
Educational model built on a synthetic quarterly panel. Not a production stress-testing framework, not financial or credit advice. Projections hold management action constant and are therefore a floor on the range of outcomes, not a forecast. Runs entirely in your browser — nothing is stored or transmitted.