Stress-Testing Your Cash Flow Before You Need To

Say your board wants to see what happens if revenue drops 30 percent for two consecutive quarters. That's a reasonable, concrete scenario to model rather than a vague instruction to be conservative, since a specific number forces specific decisions throughout the model instead of everyone quietly padding their own assumptions in different directions. Pick a scenario severe enough to be a genuine stress test, not just a slightly worse version of your base case, since the whole point is finding out where the plan actually breaks.

Revenue rarely drops in isolation. Collections typically slow as customers themselves come under pressure, which stretches your receivables days beyond their normal pattern even before you change any assumption about total revenue. Variable costs tied to revenue should decline with it, but fixed costs, payroll, rent, committed software contracts, don't move on their own, and that gap between falling revenue and sticky fixed costs is usually where the real cash pressure shows up first.

Run the scenario forward far enough to find the specific week or month where cash would go uncomfortably low or negative under your current spending plan. That breaking point is the actual output of the exercise, not the overall percentage decline you modeled. Knowing the specific point in time where action would be required, and how many weeks of warning you'd realistically have before reaching it, is what turns this from an abstract exercise into something you can actually plan around.

Once you know where the model breaks, build a specific, sequenced response plan tied to that breaking point: what gets cut first, what gets delayed, what credit gets drawn, and in what order. Having this plan written down in advance, reviewed by the leadership team while everyone is calm and the scenario is still hypothetical, produces far better decisions than trying to design the same plan in real time once revenue has actually started falling.

Rerun the stress test at least annually, and any time your cost structure or customer base changes meaningfully, since a plan built against last year's fixed costs and customer concentration may not reflect this year's actual exposure. Treat it the same way you'd treat any other standing risk review: not urgent most of the time, but valuable specifically because it's done before it's needed rather than after.

A simple sequence for running the test:

1. Pick a specific, severe scenario, such as a 30 percent revenue drop for two consecutive quarters, so every assumption in the model has to move.
2. Slow collections and let variable costs fall with revenue, while holding fixed costs like payroll, rent and committed software contracts in place.
3. Run the forecast forward until you find the week or month where cash goes uncomfortably low or negative under your current spending plan.
4. Write a sequenced response plan tied to that breaking point: what gets cut first, what gets delayed, and what credit gets drawn.
5. Review the plan with leadership while the scenario is still hypothetical, then rerun the test annually and after major changes.

A stress test is only useful if it's presented honestly, including the scenario where the current plan genuinely doesn't survive without a specific intervention. Framing the result as a completed planning exercise with a clear response already attached, rather than either hiding a bad outcome or presenting it without any plan for what happens next, is what turns this from an alarming disclosure into evidence that management is ahead of the risk.

Boards that see a well-built stress test once tend to ask for it again on their own, without prompting, which is a reasonable sign the exercise landed as intended. If a board never brings it up again after the first review, ask directly whether they found it useful, since the format or the scenario chosen may need adjusting rather than assuming silence means it wasn't valuable.