Stress-Testing a Small Business Before Reality Does It

Every small business has a shortlist of things that would genuinely hurt. Most owners could name them and have never worked out which one is closest, or how much warning they'd get.

Scenario planning is a two-hour exercise that converts a vague unease into a number of weeks — and a pre-decided response.

The four worth modelling

Skip elaborate scenario trees. Four cover almost everything that actually happens to small businesses.

1. Your largest client leaves

Not a hypothetical — client relationships end for reasons unconnected to your performance: budget cuts, a new decision-maker, an acquisition.

Model it: remove their revenue from next month onward, keep every cost, and look at the 13-week forecast.

The number you want is how many weeks until the balance goes negative.

| Weeks of runway | What it means | |---|---| | Under 6 | Structural exposure. Fix concentration now. | | 6–12 | Uncomfortable but survivable with fast action | | 12+ | Time to replace the revenue properly |

Concentration is the risk small businesses most reliably underprice. If one client is 40% of revenue, that's a structural exposure regardless of how good the relationship is. Track the percentage as deliberately as you track the revenue.

2. Revenue drops by a third

Not one client — a general downturn. Sector slowdown, seasonal collapse, a bad quarter for reasons outside your control.

Model a 33% drop sustained for two quarters against your survival floor. Two questions:

• How long can fixed costs be covered?
• What would have to be cut, and in what order?

Decide that order now. Under pressure, people cut marketing first because it's easiest, which lengthens the downturn they're cutting to survive.

3. A large invoice pays 60 days late

The most common and most underestimated. Not a bad debt — just late.

This is where profitable businesses fail. The P&L looks fine and the bank account doesn't.

Model your single largest expected payment arriving 60 days after its due date. If that breaks you, the fix isn't more revenue — it's deposits, shorter terms, or a credit facility arranged before you need it.

4. You are unavailable for a month

Illness, family emergency, anything. This one is qualitative and the most revealing.

• What stops entirely?
• What decisions queue up with nobody able to make them?
• Which clients would notice within a week?
• Who could access the systems and accounts?

Most owners discover the business tolerates a fortnight and struggles badly past that. The fixes are the ones already covered elsewhere — decision rules, documentation, and not being the only person who can do a category of work.

Run it against the forecast, not in your head

The exercise only produces useful output if the answer is in weeks and dates.

Take your 13-week cash flow forecast, copy it four times, and apply one scenario to each. Everything else stays the same — same costs, same timing, same commitments.

What you're looking for in each copy:

• The week the closing balance goes negative
• How much warning you'd have had before that week
• What single change buys the most time

That last one is usually surprising. Frequently it's not cutting costs at all — it's taking deposits, or invoicing weekly instead of monthly.

Decide the response now

The output isn't the scenario. It's the pre-decided response, written down while you're calm.

IF our largest client leaves
  Week 1:  Pause all discretionary spend (list: ______)
           Move [name] onto business development full time
  Week 2:  Contact the three warm prospects from Q2
           Approach [supplier] about extended terms
  Week 4:  If no replacement in pipeline, reduce my own pay to £____
  Week 8:  Draw on the credit facility rather than delay supplier payments

  Runway at current fixed costs: ___ weeks

Having chosen in advance is most of the value. Decisions made under financial pressure are reliably worse than the same decisions made calmly three months earlier, and the delay while you work out what to do is frequently more expensive than the choice itself.

Do it even when the answer is fine

If every scenario shows twelve weeks of runway and a clear response, that's not a wasted afternoon. You now know:

• You can decline bad work without existential worry
• You can commit to a hire with a clear sense of the downside
• You can price with confidence instead of fear

Knowing your downside is what lets you take reasonable risks. Owners who've never modelled it either overestimate the danger and stay too cautious, or underestimate it and find out the hard way.

When to re-run it

• Quarterly, as part of quarterly planning
• Before any decision that adds fixed cost — a hire, a lease, a commitment
• When client concentration shifts meaningfully
• When the cash conversion cycle changes

Fifteen minutes each time once the models exist. The first build is the only long one.

The mistakes

1. Modelling in your head. Produces adjectives, not weeks.
2. Ignoring concentration. The most underpriced risk in small business.
3. Only modelling revenue loss. Late payment breaks more businesses than lost revenue.
4. No pre-decided response. The delay costs more than the decision.
5. Cutting marketing first. Lengthens the downturn you're surviving.
6. Not re-running before adding fixed cost. That's the moment it matters most.

What to do next

Take your 13-week forecast, remove your largest client's revenue, and find the week the balance goes negative. That one number tells you more about your actual position than any amount of general worrying — and it takes about ten minutes.