Before an Irreversible Decision: What to Settle First

The state you are probably in

You have done the work. The spreadsheet exists, the lawyer has been consulted, the numbers hold up under the pessimistic case. You have talked to the people whose judgment you respect and every one of them said something sensible and none of them decided anything.

And you are still not moving. Not because you lack information — you have more than you can hold — but because at some point in the last few weeks the question stopped being analytical. The decision is large, part of it cannot be undone, and other people's lives are inside it. That kind of decision does not resolve by adding facts, which is why the last three weeks of research produced no movement at all.

This is the point at which some people go looking for an outside reading, and it is worth being honest about why. Not usually for information. Usually because a decision this size wants to be shared with something, and the available options — friends who will be too kind, advisors who are paid on the outcome, family who have a stake — all have a slant.

That instinct is not foolish. But there is work to do before any outside input helps, and it is work only you can do.

First: is it actually irreversible, or only expensive?

Almost everyone gets this wrong, and they get it wrong in both directions.

A genuinely irreversible decision has no recovery path or one so long it changes the shape of your life: selling a business you built and cannot rebuild, a decision about children, moving a family to a country where your professional credentials do not transfer, ending a marriage. These deserve the full apparatus — slow, deliberate, uncomfortable.

An expensive-but-reversible decision has a recovery path with a cost you can name. Taking the wrong job. Moving cities and moving back. A concentrated position you could exit at a loss. These feel identical from the inside, because the anxiety is the same, and they need a completely different process: decide faster, on less information, and pay the recovery cost if you are wrong.

The error runs both ways. People agonise for a year over something they could reverse in eighteen months, and they commit in a fortnight to something with no way back — usually because the second one arrived with urgency attached and the first did not.

So write the recovery path down. Not 'it would be hard' — the actual path, and how long, and what it costs. If you cannot write it, that is your answer about which category you are in, and everything below changes accordingly.

Second: what would change your mind?

This is the step that does the most work and the one that is almost never taken.

Before you hear anyone's opinion, write down the specific observation that would make you choose the other way. Not a feeling. Something you could observe and someone else could verify: a number below a threshold, a condition the other side will not accept, a person unwilling to commit in writing.

Two things happen when you do this. The first is that you find out whether you have a criterion at all. Many people discover at this point that they have been researching for months without any standard by which the research could conclude — which is exactly why it never concluded. Information cannot substitute for a criterion, and this is why capable people can study a decision for a year and end up precisely where they started.

The second is that it inoculates you against the next thing you hear. A criterion written down beforehand can be checked against. A criterion formed after hearing an answer is not a criterion; it is a rationalisation with better posture. This matters especially if you intend to consult anyone — including a reading — because the criterion is the difference between using the input and being used by it.

Twenty minutes. Before anything else.

Third: what are you actually buying?

Every consequential decision has a stated purpose and an actual one, and they are frequently different.

The stated reason for selling the company is the valuation. The actual reason may be exhaustion, or a desire to stop being the person everyone escalates to. Both are legitimate. But if the actual purchase is rest and you negotiate as though it were money, you will optimise the wrong variable and end up with an excellent price and a three-year earn-out that gives you no rest at all.

The stated reason for the move is the schools. The actual reason may be that one of you has wanted to leave for years and this is the first defensible argument. Also legitimate — and worth knowing, because if it is not said aloud it will surface eighteen months later in an argument about something else.

This step is uncomfortable in a way the first two are not, which is why it gets skipped. It is also where most of the regret in large decisions is generated. People do not usually regret the decision they analysed; they regret discovering afterwards that they were solving a different problem than the one they described.

So: what would have to be true, a year from now, for you to say this was the right call? Write that down too. Its distance from your stated reasons is the most informative thing on the page.

Where an outside reading legitimately fits

Now, and not before, outside input becomes useful — and it is worth being exact about what this kind of input can and cannot do.

What it can address is timing. Traditional practice is largely about periods rather than events: whether this is a stretch where pushing tends to cost more than it returns, where the same effort buys less. That is a claim about cost, not about outcome, and it is genuinely actionable — it argues for smaller steps, longer runways, and putting the irreversible part later rather than for abandoning the plan.

What it can also address, and this is the part sophisticated clients tend to find most useful, is your own pattern. Whether you characteristically overcommit and then defend the commitment. Whether you wait past the window and then move under pressure. Whether you keep entering arrangements where you carry more than your share. These are readable, they are checkable against your own history, and they are the variable most absent from your spreadsheet — because the spreadsheet assumes a decision-maker who is not you.

What it cannot do is tell you the outcome. Nothing can. Any reading that names what will happen has left what the method supports, and the more confidently it does so the less it knows. That should be a disqualifying signal at any price and particularly at a high one, where the confidence is part of what is being sold.

What to refuse to buy

At this level of decision, you will be offered certainty. Sometimes crudely, often with considerable polish.

Refuse it. Not on principle — on accuracy. The uncertainty in your situation is not a gap in someone's knowledge that a better advisor could close. It is a property of a decision that depends on other people's future choices. Anyone presenting it as closeable is either mistaken about their own method or aware and selling anyway.

The specific things worth walking away from: a named date arrived at without a stated basis. A judgment about a person who is not in the room and has not consented. Any statement about health, litigation, or a specific asset from someone whose method does not touch those things. And the oldest pattern of all — a vivid description of what could go wrong, followed closely by the thing you can buy to prevent it.

There is one question that separates most of the field, and it costs nothing to ask: which part of this did you read that from? Someone working within a method answers immediately and specifically — this position, that period, this relationship in the chart. The inability to answer, or a retreat into intuition, tells you more than everything said before it.

The document, and then the decision

What you should have, before you commit, is one page. Not a report — a page.

The decision in a sentence. Which category it is in, reversible or not, with the recovery path and its cost written out. The observation that would make you choose the other way. What you are actually buying, and what would have to be true in a year for this to have been right. Then, at the bottom, whatever outside input you gathered, marked for what it was: timing, or pattern, or opinion.

Date it. This matters more than it sounds. In six months you will not remember what you knew now versus what you learned later, and without a dated page you will either credit yourself with foresight you did not have or blame yourself for information you could not have had. Both distort the next decision.

Then decide, and stop consulting.

That last part is the discipline most people fail, and it is worth naming plainly. There is a point past which additional input stops adding information and starts adding noise — usually recognisable because the last two things you heard did not move you. Consulting beyond that point is not diligence. It is looking for someone to take the weight, and nobody can: not an advisor, not a friend, not a chart. The decision arrives with your name on it regardless.

What the preparation buys you is not certainty. It is that when the decision goes badly — and some proportion of well-made decisions do — you will be able to look at a dated page and see that you decided well on what you had. That is the only durable protection available, and it is worth considerably more than a confident prediction.

What Changes the Interpretation

Context changes the reading. Birth time, location, current stress, recent behaviour, and above all the quality of the question can each shift what matters. A page worth reading names those conditions instead of pretending a symbol speaks alone.

How to Use This Page Now

Take one real situation, write down one signal to watch, and set a date to look again. A reading becomes useful when it changes the next thing you record, not when it sounds impressive.

Important Limit

This method can inform timing, better questions, and what to watch next. It should not replace medical, legal, financial, psychological, safety, or professional advice, and it does not decide anything for you. Read it as one input with a stated limit, and let evidence you can check carry more weight than any single symbol.