Alex Hormozi’s Decision Framework, Applied Backwards

You searched for the Alex Hormozi decision framework. Here it is, accurately, and then the part that matters more than the framework itself.

It is not his. He says so.

Where the Alex Hormozi Decision Framework Comes From

In his 2015 letter to Amazon shareholders, Jeff Bezos split decisions into two types.

Type 1 decisions are consequential and irreversible, or close to it. He describes them as a one-way door: walk through, dislike what you find, and you cannot get back to where you were. Those deserve slow, careful, deliberate treatment.

Type 2 decisions are changeable. A two-way door. You can walk back through. Bezos argues these should be made quickly, by individuals or small groups, and that a bad Type 2 decision costs far less than the delay spent avoiding it.

Then he names the failure mode, and this is the sentence that made the framework famous. As organisations grow, they start applying the heavy Type 1 process to everything, including all the Type 2 decisions. The result is slowness, unthoughtful risk aversion, too little experimentation, and less invention.

Hormozi’s contribution is a threshold and a translation. Get to roughly 80% confidence, decide, and keep enough agility to correct the remaining 20%. His framing of speed is blunt and useful: the speed of the boss sets the speed of the team. Naval Ravikant’s version circulates as a one-line heuristic about how to break a tie when you cannot decide.

Three names, one idea, one origin. If you came here for the Alex Hormozi decision framework, that is the framework.

The Mistake Almost Everyone Makes With It

Knowing the Alex Hormozi decision framework and applying it correctly turn out to be very different skills.

Watch what actually happens in a small business. The reversible decisions get agonised over. Which CRM. Whether to change the logo. What to call the offer. Which platform to post on. These are two-way doors, all of them, and they absorb weeks.

Meanwhile the one-way doors get made in an afternoon, usually under pressure and usually because someone else was waiting. Signing a client who does not fit. Dropping the price to close a deal. Taking on a partner. Committing to a technology everything else will be built on.

The Alex Hormozi decision framework did not fail. It was applied precisely backwards, and Bezos predicted the pattern, though he framed it as something that happens to large organisations. It happens faster in small ones, because there is nobody to slow you down on the decisions that deserve slowing down and nobody to hurry you on the ones that do not.

Reversibility Is a Cost Question, Not a Possibility Question

Here is where I think the popular version of the framework is too crude to use, and where it needs one correction.

Almost every business decision is technically reversible. You can fire the hire. You can raise the price back. You can migrate off the platform. Ask “can I undo this?” and the answer is nearly always yes, which makes the question useless.

The right question is not whether you can go back. It is what going back costs, and who pays it.

Three costs, and they behave very differently.

Money is the cheap one. A wrong software choice costs a migration. Annoying, priced, finite. Most decisions people treat as one-way doors are only expensive in money, which makes them two-way doors with a bill attached.

Time is worse, because you cannot recover it. Eight months building the wrong product is eight months, whatever you do next. This is the cost that makes people wrongly cautious about small decisions and wrongly casual about direction.

Reputation and trust are the genuine one-way doors. You can un-hire someone. You cannot un-tell the team what firing them said about you. You can raise a price back up. You cannot un-teach a client that your price moves when they push. You can leave a market. You cannot re-enter it as though you were never there.

That third category is the one that matters, and it is almost never on the list when people apply the framework, because it does not appear on any invoice.

The Test

Before any decision that feels significant, one question:

If this turns out wrong and I reverse it, what specifically cannot be put back?

If the answer is money, decide fast. Price the mistake, check you can absorb it, move.

If the answer is months, slow down enough to write the reason on one page. Not a plan. Just the reason, so that in six months you can tell the difference between a decision that failed and a decision that was wrong.

If the answer names a person, a relationship or a reputation, that is a one-way door regardless of how small the decision looks on the invoice. Those are the ones to sit with overnight.

The Decisions That Look Reversible and Are Not

Four that catch small businesses repeatedly, all of them wearing two-way-door clothing.

Discounting to close. The revenue arrives, and it looks like a one-off. What actually happened is that you taught one buyer, and everyone they talk to, what your price does under pressure. Which is why what you charge at the start sets more than your margin.

The first hire. Reversible on paper. Except the first hire defines what the second one expects, and how you behave when it goes wrong is the culture, permanently, in a way that no later handbook edits.

Taking the client outside your niche. One project, easy money. Then your case studies say something different about who you are, and your positioning is now an average of the work you accepted rather than the work you chose.

Building on someone else’s platform. Every individual step is reversible. The accumulated dependency is not, and you find out on the day the terms change.

Notice what those have in common. Each is cheap to undo in money and expensive to undo in trust or in position. That is exactly the class the crude version of the Alex Hormozi decision framework misses.

For the genuine two-way doors, the case for moving fast, and how to put a deadline on a choice that keeps coming back, is in our piece on rapid decision making.

What to Take From This

The framework is right. Speed on reversible decisions is close to free, and most people are far too slow on them.

What it needs is one added step before you classify anything. Do not ask whether you can go back. Ask what you would be unable to put back if you did, and check whether the answer is a number or a person.

Numbers you can price. People remember.

The same reasoning applies at the level of the whole business, not just single decisions, which is why staying in long enough for things to compound depends far more on avoiding the unrecoverable losses than on winning the recoverable ones.

A vision does the same job one level up, which is why a company without one keeps relitigating the same call.