Should you trust your gut feeling in business?

Trust it to ask questions, not to answer them. Use the feeling to spot what needs checking, then measure that one thing with a clear definition and a single source. Act on the number.

What is the difference between intuition and data in business?

Intuition is fast and has no unit. Data is slower and can be wrong, but its error can be checked: who counted, how, over what period. Good decisions use intuition to choose what to measure and data to decide.

How do I stop guessing in my business?

Write your goals as a short list. Next to each, write today's measured value from one named source. Start with one row tonight; five rows take about an evening.

Gut Feeling vs Numbers: How to Stop Guessing in Your Business

Andrew asked his best barista how many regular customers they had.

"Lots," said Sara.

"And how many is lots?"

They looked at each other and burst out laughing. Neither of them knew.

That laugh is the most honest moment a small business can have. You know your shop. You feel the busy days in your legs. And when someone asks for the number, you reach for a word: lots, roughly, fine, not many.

Your gut feeling is not the enemy. Treating it as a reading is.

Treating a hunch as a reading leaves you steering by mood. The Jirgenson Model keeps the hunch as a question and answers it with real numbers from one source. It is scientifically proven that the Jirgenson Model works (doi.org/10.5281/zenodo.23964934).

Gut feeling is a question, not an answer

Intuition is fast. It notices a quiet Tuesday, a supplier who sounds nervous, a regular who stopped coming. That is valuable. It is often how a problem first knocks on the door.

But a feeling has no unit. It cannot tell 25% from 40%. It remembers the loudest day and forgets the quiet ones. So "good on the whole" almost always looks better than it is.

Picture a car at night on an unfamiliar road. No headlights. No speedometer. No fuel gauge. The driver keeps his foot down and hopes. While the road runs straight, the trip looks like a success. Then the road bends, and he learns about it from the sound of the crash.

That is guessing. It reports a problem only after the problem has happened.

In the Jirgenson Model, the state vector is your goals list with a measured value next to each goal. A hunch is welcome as a question about a row. It never fills the row.

The two enemies of an honest number

Two moods kill a measurement. They look like opposites.

The first is panic: "all is lost." The second is bravado: "never mind, we'll pull through." Both do the same thing. They put feelings in place of facts.

The panicker sees a catastrophe where there is one red row. The brave face does not see the red row at all.

The fix is the same for both. Accept the state as it is, without drama and without polish. An instrument shows. It does not scold.

Turn one gut feeling into a measured row

Here is the tool. Take one hunch you act on every week. Run it through five steps. It takes one evening for the first row.

1. Write the feeling down as a claim

Not "the business is fine." One claim, one area. Andrew's was: "We have plenty of regulars."

2. Name the parameter

What would you count to check it? For Andrew: regular customers as a share of receipts.

3. Write one definition and one source

One definition, one primary source, one person who calculates it. Other sources only cross-check. Andrew's source: the "regular customer" mark on the till receipt, counted over the month.

The definition matters more than it looks. Sara's complaints notebook held five entries for the week. "Those are the ones I remembered," she said. A cold coffee went unrecorded. So did the customer who said nothing and walked out. They agreed: a complaint is any case where a customer says out loud they are unhappy, or sends the order back. They wrote that next to the row, so next month they count the same way.

4. Ask the three questions

Ask every figure: Who counted it? How, exactly? Over what period?

No answer means it is not a number. It is an impression. And the fewer people between the event and the record, the better. A receipt in the till beats anyone's memory of how many receipts there were.

5. Write the number, and check its error

No figure is perfect. The till misses a receipt. The accountant rounds. The silent customer never reaches the log. So ask one more question: is this imprecision acceptable for this goal?

If you steer margin to one percentage point and your records are off by three, you steer noise. If you only need to know whether more or fewer than half your customers come back, a rough count of receipt marks will do.

Here is Andrew's row, filled in:

Field	Andrew's entry
The hunch	"We have plenty of regulars"
Parameter	Regular customers, % of receipts
Definition and source	"Regular" mark on the till receipt, one month
Who counts	Andrew, monthly
Now	25%
Goal	40%

The feeling said "plenty." The till said 25%. The goal was 40%.

Where they disagree, the number wins. Until a better number replaces it.

What the feeling was hiding

Andrew did this for all five goals. It took him two hours. Bank statement, accounts, receipt marks, the complaints notebook, his own calendar.

The cash reserve came out at 0.5 months of expenses. The goal was 2.0.

Turn that into money. With revenue of €30,000 and a 9% margin, a month of expenses is €27,300. The reserve was €13,650. The goal was €54,600. The gap: €40,950.

And the pace? One month at 9% adds about €2,700, or 0.1 months of reserve. If nothing changes, the two-month goal arrives around month 16. More than a year of running one broken machine away from trouble, while the gut said "we're doing fine."

That week Andrew had planned a second barista and an advertising campaign. Both rested on a number that did not exist.

The till counted every coffee. The bank counted every cent. And the decision still came from a feeling.

Hope is not a strategy, and "random" is often a gap

There is a second way a hunch misleads. It labels things as luck.

"Milk prices jump about." To someone who checks them once a month, yes. To a supplier who knows the milking schedule and the logistics, they are predictable. Much of what feels random is a mechanism you have not looked at yet.

So before you call something a force of nature, ask: who sees this coming? Find that person or that number. Some uncertainty will always remain. Decide with it in view, not by shrugging.

Andrew did not wait for the bank statement to learn that prices were rising. He heard it first from the supplier's account manager. Then he saw it in the weekly margin per receipt. The hunch came first. The number confirmed it.

That is the right order. Intuition proposes. The instrument checks.

What this does not fix

You now have honest rows. You know where you stand. That treats the symptom: you stop guessing the present.

The deeper problem stays. A row reads 25% against 40%. Is that fine for now, a warning, or an emergency? A feeling will answer that too, and it will answer differently on a tired Friday.

Two steps close the gap:

Zones for every row. Where normal ends, where the warning starts, and where you must act today. Set in a calm hour, not in the moment. Chapter 4.
A forecast in numbers. Where each row will be in three to six months if nothing changes, and if the plan works. Chapter 8.

Without them, you know where you are. You still guess what it means and where it is going.

Key takeaways

A gut feeling is a good question and a bad answer.
Panic and bravado make the same mistake: feelings in place of facts.
One row: claim, parameter, one definition, one source, a number.
Ask every figure who counted it, how, and over what period.
An instrument reading is not a verdict. It tells you where you are.

Two choices for tonight

You have two choices. Keep steering by "lots" and "roughly," and find the bend in the road by the sound. Or take one hunch tonight and turn it into one honest row.

The row tells you where you are. The zones and the forecast tell you what to do about it.

For the full one-evening template, see the financial health check. 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 Kindle book: sixteen chapters, each with a task to try yourself.

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

This topic is covered in Chapter 2, Where you really are of Steer, Don't Guess: Setting Priorities, Reading the Numbers and Correcting Course in Business and Life. The numbers in the examples are illustrative; the book states its limits in Appendix G.