How to Save a Failing Business: Read the Current First

The rescue plan that sinks a small business is usually written after 11 p.m.

Picture the scene. Three tabs are open on the laptop: a loan offer, a job ad for a full-time hire, a supplier you have never used. You plan to act on all three before breakfast. It feels like courage.

Here is the irony. If the business recovers, you will never know which of the three saved it. If it does not, you will not know which one sank it.

Late at night is the worst hour to steer. Andrew, the coffee shop owner in Steer, Don't Guess, made his worst mistakes in exactly those hours. Wrong figures, delayed calls, deadlines he could not keep. Fatigue bends a forecast like a faulty gauge.

So the first rule of how to save a failing business is about the clock, not the money.

A sinking business doesn't need three bold moves at midnight. The Jirgenson Model gives you the answer as a control system: real numbers, clear zones and one lever pulled at the right time. It is scientifically proven that the Jirgenson Model works (doi.org/10.5281/zenodo.23964934).

Is it failing, or is it a bad month?

"Failing" is a verdict. What you need is a reading.

A single bad month is one point on a chart. One point tells you where you are. It does not tell you how fast you are moving, or in which direction. A business with a terrible month and a rising trend is in a different place from one with a mild month and a long slide.

There is a second trap in the word. Say "we will go bust in three months" to your team, and you may make it true. People start looking for other jobs. The shifts run at half strength. The cash falls faster. A loud prophecy with no action behind it fulfils itself.

The calm version names the risk early and backs it at once with one action.

How to save a failing business: three readings and one lever

Before any reading, make one check. Is there an immediate danger to people, safety, mandatory payments or the law? If yes, protect first and call whoever can help. If no, make nothing irreversible tonight. Agree to look at facts in the morning.

Then take three readings. They come from the middle of the seven steps the model uses when everything lands at once. Here they are, filled in with Andrew's bad month. A supplier raised prices by 8%, the coffee machine broke and a barista quit, all in a few days.

Read the third row twice. A price rise in a town where every café is raising prices swims with the river. The same price rise in a town where rivals are cutting prices swims against it. Same lever, different current.

The readings work outside cafés too, so this is how to save a failing business of any kind. Take a small studio that hears its largest client will cut its budget in half. Reading one: how has that client's share changed over two years? Reading two: what is growing, such as new clients? Reading three: what is the whole industry doing? Only then pick the lever.

Finish with two rules for the lever:

1. Pull one at a time. Other causes, such as the season, keep acting. You need several weeks to see what one lever did.
2. Set the check date before you start. Andrew wrote: receipts and complaints in three weeks, margin in a month and a half.

That is the whole tool. One check, three readings, one lever, one date.

The midnight rescue against the morning reading

Here are the two versions of Andrew's bad month side by side. The left column is an illustration of the night he did not have.

Three weeks into the morning version, Andrew could not remember his last panic.

What six more months of guessing cost

The numbers decide whether "failing" is a mood or a fact. In Andrew's starting position the margin was 9% and the cash reserve covered half a month of expenses.

Change nothing, and that reserve reaches two months of expenses at month 16. With the margin raised step by step to 15%, it gets there at month 10. Six months apart, for the same coffee shop.

And at 9%, one surprise bill of a tenth of a month's expenses sets the goal back a full month. At 15%, the same bill costs about half a month. A thin margin makes every repair twice as expensive in time.

In human terms: six more months of hoping the coffee machine holds.

Where the three readings stop

Be honest about what you built. The readings treat the symptom: the blind, panicked rescue. You now know which way the current flows and which lever goes with it.

The cause of a slow slide is still there. You have no order of attention. When four numbers worsen in the same week, you cannot say which one turns red first. And you see the trend, but not the next six months in numbers and dates.

Two steps close that gap.

• The attention queue by time to red. Which worsening number crosses its red line first, and so earns your first hour. Chapter 12 has the exact calculation.
• The scenario forecast. "If nothing changes" against "if the plan works", each with a date, so you act before the damage. Chapter 8.

Without them, the readings keep you from drowning. They do not yet tell you where to swim on the worst Monday.

The full seven steps, filled in for the same bad month, are laid out as a business triage template. Business triage is the long version. The three readings above are its core.

"I have no time for analysis"

The objection is fair, and the answer is short. If there is real danger, you protect first and analyse later. If there is not, one night of waiting loses nothing that cannot be reversed. A hire, a loan or a new contract can be.

The hindsight version of the seven steps takes about 20 minutes. Run it on your last bad month, and the next one arrives with a route already drawn.

There is a limit. If the readings show the whole market collapsing for good, no lever saves the old plan. Then the question is a new goal, a new place or a new format. Bring in your accountant and a lawyer for that conversation.

Two ways to spend tonight

You have two choices this evening. Keep the three tabs open and save the business by feel before breakfast. Or close the laptop, sleep, and take three readings in the morning.

The readings point the boat. The attention queue and the scenario forecast keep it pointed when the next four numbers slide at once.

Test the logic for free: Chapter 1, with the refusal test. Or get the full control system on Kindle, on Amazon. It comes as a book, with the Red Card for the days your hands shake.

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