The six axes every business decision should survive

Every recommendation Kauzio makes is checked against six axes. Not three. Not eight. Six. We tested every other count. Six is what stuck.

This post is the why. Why six, why these six, and how they catch the decisions that look fine on the day and quietly cost you six months later.

How we landed on six

We started with one axis: revenue. Will this move grow the top line? That was the question every dashboard in retail and hospitality was already trying to answer. So we built that first, and then we ran it on a few hundred real decisions from operators we trust.

It missed everything.

It missed the price change that boosted revenue 4 percent and tanked margin. It missed the new hire that lifted output and crushed culture. It missed the supplier switch that saved 8 percent and exposed the business to a single delivery route through one port.

So we added a second axis. Then a third. We kept adding until each new axis was catching genuinely new failures — and stopped the moment a new axis was just rephrasing one that was already there. We landed at six. Anything fewer and you miss a class of mistake. Anything more and you are double-counting.

The six

1. Revenue risk. How much top-line could this decision put on the line. Not the upside, the downside. The most common mistake in retail and hospitality is sizing a decision on its best case and forgetting to model the worst.

2. Reversal cost. What does it cost you to undo this if it goes wrong. A markdown can be reversed in a day. A new lease cannot be reversed for a decade. Decisions with cheap reversals deserve fast yeses. Decisions with expensive reversals deserve slow ones — even when the upside looks the same.

3. Time to impact. When will you know if this worked. A reorder gives you signal in a week. A brand campaign gives you signal in a year. We tag every decision with its honest feedback window so the team knows when to stop waiting and start judging.

4. Blast radius. Who else is affected. A till change touches one register. A supplier change touches every shop and every customer. Decisions with large blast radius need more upfront scrutiny than the size of the decision alone would suggest.

5. Uncertainty. How confident is the recommendation, and where are the gaps. This is the axis most software refuses to expose because it makes the system look weaker. We expose it deliberately. A 60 percent confidence answer with the gaps shown is more useful than a 90 percent answer with the gaps hidden.

6. Opportunity cost. What is the next-best use of this money, this time, this attention. The decision in front of you is almost never being weighed against doing nothing. It is being weighed against the other thing you could do instead. Most decision tools forget this. We make it the sixth axis on purpose.

A worked example

A small chain runs four pharmacies. The owner asks Kauzio whether to mark down a slow-moving range of vitamins by 30 percent.

A normal dashboard would tell them how much stock is sitting and how long it has been there. Then leave them to it.

Here is the six-axis view:

• Revenue risk: Low. The current stock is dead weight; clearing it at 30 percent off recovers cash that is otherwise frozen.
• Reversal cost: Cheap. Markdown lasts two weeks; price restores after.
• Time to impact: Fast — 14 days to read whether the markdown moved units.
• Blast radius: Small. Four shops, one shelf each. No customer cohort affected long-term.
• Uncertainty: Medium. We do not know whether the slow movement is the product or the placement. A markdown will not distinguish those — but the next decision after this one will benefit from the data this one creates.
• Opportunity cost: Higher than it looks. The cash freed up should fund the reorder of the top-selling cold remedy line, which is about to stock out. Flagged.

That last point is the one a dashboard would miss. The right call here is not just to run the markdown — it is to run the markdown specifically so the cash funds the reorder that is about to bite. The six-axis view surfaces that. A one-axis view never would.

Why this is the work

The pitch for "AI" in retail is that it will make better decisions for you. That is wrong. It cannot. It does not know your business as well as you do, and it never will.

What it can do is make sure no decision goes through without surviving every axis it should have to clear. That is a different and more useful job. The owner is still the decider. The system is the second pair of eyes that refuses to sign off until each axis has been argued.

Six axes. Same scrutiny on every call. That is the standard.