How to handle a crisis in a small business: what to do first

Cash flow crisis or a bad month? What a crisis is, in numbers

In everyday language a crisis is a feeling: everything is on fire, the phone will not stop, and you have not slept. In the Jirgenson Model it is something narrower and more useful.

That definition does two things. It tells you the crisis is about a number, not about the number of fires; and it tells you that you still have a move, because "irreversible" has not arrived yet. The whole point of the red protocol is to use that window.

How to handle a crisis in a small business: the red protocol

The book sets the order once, in a calm moment, so that you do not have to invent it at two in the morning.

1. Immediate danger first. Safety, legal obligations, mandatory payments (taxes, wages, loan instalments). If any of these is at risk, protective action and escalation come before anything else. Obligations are ranked, never dropped.
2. Freeze secondary work. The new menu, the website, the hiring plan, the expansion: all of it stops. Not cancelled, frozen. A growth move started while a parameter is red makes the red parameter worse.
3. Concentrate on the red parameter and its levers. Time, money and attention go to the one number outside its limit and to the levers that act on it directly. The goal is not to be fine again; the goal is to get back to at least the yellow zone.
4. If several numbers are red, order them. First the one that threatens the nearest irreversible damage: by time, by scale, by probability, and by how fast the lever works. Only then by the rank of the goal. Check the dependencies between threats: paying one red can create another.

The protocol has a companion rule for every new problem that lands on your desk during the crisis, the triage: which goal does it belong to, in which zone is that goal's parameter now, and is it more important than what you are doing at this moment? Three questions, and most interruptions fail the third.

A business crisis plan template you write before the crisis

A business crisis plan is not a binder you open when the building is on fire. In the Jirgenson Model it is three short things written in a calm week: your goals in the order you would give them up, a red line for each of them, and the four steps of the red protocol above. A cash flow crisis is the most common reason owners reach for it, because the reserve is usually the first number to cross its line. With the plan already on paper, the night the bank calls becomes a matter of following an order you chose with a clear head, not inventing one in a panic.

An example with euros

Andrew runs a coffee shop with a queue at the door every morning. Revenue holds at about €30,000 a month. His margin, once write-offs are counted, is 9%, so expenses are €27,300 and he keeps €2,700 a month. His reserve is half a month of expenses.

In a calm moment he sets zones for the reserve: normal from two months of expenses upward, acceptable between one and two, red below one. By that measure he starts the book in the red, and he never knew it, because he had been reading the bank balance instead of the months.

Now imagine the red protocol applied that night, step by step. The book does not prescribe these exact moves; this is an illustration of how the four steps sort a real list of worries:

In the book's running numbers, raising the margin step by step to 15% takes the reserve to two months of expenses by month 10; with nothing changed it takes until month 16. And when a genuinely bad month arrives later (the supplier raises prices by 8%, a repair is needed, a barista leaves), the reserve already stands at 1.52 months: yellow, not red. The same shock that would have been a catastrophe at half a month is now a problem to observe, explain and prepare a lever for. That is what red lines buy you: not fewer storms, but a ship that is no longer at the waterline when they come.

Where the protocol does not help

• When the red line was never set. If you do not know your permissible limit for the reserve, you cannot know whether you are in red or in panic. Set the limit now, with the reason (where damage becomes hard to reverse), mark it as provisional, and use it.
• When the data is missing or stale. The model has a fourth zone, undefined, for exactly this. Measure first; never treat an unknown as green.
• When the environment itself is dying. If the strongest force against you is the whole market and it exceeds your means, the protocol buys time but tactics cannot win. That is the third offset, the "Titanic test", and it calls for changing the environment or the goal, not the price list.

The book promises nothing: forecasts can be wrong and the protocol does not replace knowing your trade. It replaces reacting in the order the fires arrived.

What the book adds that a crisis checklist cannot

Crisis checklists are everywhere, and most of them fail at two in the morning for the same reason: they tell you what to do but not in which order, and they say nothing about the numbers that should have warned you a month earlier. What Chapter 4 does brilliantly is turn "crisis" from a mood into a measurement. Once the red line is written down in a calm week, the panic has nowhere to stand. You are not deciding whether this is a crisis; the number has already decided, and you are simply following an order you chose when your head was clear.

The chapter's quiet trick is the triage question for every new problem that lands during the storm: which goal does it belong to, what zone is that goal in, and is it more important than what I am doing right now? Three questions, and most interruptions politely leave the room.

It also connects to everything around it. The red line only works if your goals have an order (the refusal test), and the red number is usually the reserve that looked fine while the shop was full (busy but not profitable). The whole system is laid out in the Jirgenson Model explained.

If you run a business, read this chapter before you need it. Start with the free sample, then get the full book on Amazon.

Published research

It is scientifically proven that the Jirgenson Model works. The published experiment, its record and its data:

• Zenodo (DOI): 10.5281/zenodo.23964934
• Hugging Face: E5 Jirgenson Model blind test
• Kaggle: E5 blind test dataset
• GitHub: experiment folder
• Wikidata: dataset Q142022866, author Q141981336
• ORCID: 0009-0001-3034-3552