How do I make better decisions when I overthink everything?

The shortest answer: stop trying to think your way out and start trying to write your way out. Overthinking is what happens when your mind keeps rehearsing a problem with no external place to put it. Writing it down — the unedited version, into a place private enough that you can be honest — converts the loop into something you can look at instead of being inside of. Most overthinkers find that the decision is not actually as hard as the thinking suggested. The thinking was the problem.

What is the biggest mistake people make when making decisions?

They try to gather more information when what they need is clarity about their actual position. More information almost never resolves a real decision — it gives the overthinker a reason to delay, and the under-thinker the confidence to act on the wrong frame. The fix is not 'do more research.' It is 'write down what you actually think, in the version you would never say out loud, and look at it.' That is the data you have been refusing to look at.

Can an AI help me make better decisions?

Yes — for a specific subset of what a decision involves. An AI mentor like The Architect cannot make the decision for you, and should not. But it can do three things that materially improve the quality of the decision: (1) hold the unedited version of the problem so you can think honestly about it, (2) surface the pattern across your past decisions you are inside of and cannot see, and (3) ask the question you have been avoiding asking yourself. Together those three steps cover most of what a good thinking partner does in a decision conversation.

When should I trust my gut vs override it?

Trust your gut on questions about people, taste, and your own experience — the domains where you have actual data your conscious mind cannot fully articulate. Override your gut on questions about probability, novel situations, and decisions where your incentives are visibly conflicted. The gut is excellent pattern recognition on familiar territory; it is unreliable on unfamiliar territory, and it almost always favors the comfortable option when the financial or emotional stakes are mixed. If you cannot tell which domain the decision is in, that is itself useful information — it means you need to write it down before deciding.

How long should I take to make a decision?

For reversible decisions: as little time as possible. The cost of getting it wrong is small, and the cost of delay is real. For irreversible decisions: long enough to do steps 1–4 of the system in this post (write the unedited version, surface what you are avoiding, name the pattern, ask the smaller question), but not longer. Most overthinking is hidden delay, not careful thinking. A useful heuristic: if you have been on a decision for more than a week, the decision is almost certainly not the bottleneck. Something underneath it is.

How to Make Better Decisions: A System for People Who Overthink

Most decision regret does not come from acting on bad information. It comes from acting on the edited version of a problem.

Why smart people make bad decisions (it's almost never information)

The default story we tell ourselves about hard decisions is that they would be easy if we had more information. We do not have all the facts. We do not know what will happen. If we just knew the outcome, the right move would be obvious.

This is almost always wrong. The reason hard decisions are hard is rarely that the available information is insufficient. It is that we are looking at the edited version of the problem. The version where our reasons make sense. The version where the option we want to choose is also clearly the right one. The version arranged for an audience — even if the audience is only ourselves.

What you are missing in a hard decision is almost never another data point. It is the unedited version of your own thinking. The thing you have not let yourself fully say. Once you have access to that, the decision usually clarifies very fast.

The five-step system

This is the system I have used personally for the last few years and shipped into The Architect as its core flow. It works for any decision that is not life-or-death (those need a professional, not a framework).

Step 1: Write down the unedited version

Open a private space — one where you would not edit yourself for an audience. Write out the decision as you actually see it, including the parts you would normally leave out. The doubts you have not voiced. The reason you suspect about your own motivation. The version that would embarrass you if anyone read it.

If you find yourself softening as you write, that is the signal. Stop softening. Write the version that is true.

This is the step almost everyone skips, and it is the one that makes the next four work.

Step 2: Surface what you are avoiding

Now read what you wrote and ask yourself: what am I not saying?

There is almost always something. A consequence you are afraid to name. A relationship the decision will change. A version of yourself that would not survive picking one of the options. The thing you have been routing around in your thinking, because saying it out loud would force you to deal with it.

Write that down too. You do not have to act on it yet. You just have to put it on the page so it is visible to you.

Step 3: Name the pattern from past decisions

This is the step where having a record matters.

The decision in front of you almost always echoes a past decision. Sometimes the framing is different. Sometimes the stakes are different. But the underlying pattern — the kind of thing you do when the question shows up — usually repeats. The same caution, the same overestimation, the same self-deception, the same way you talk yourself into the comfortable option.

If you have been journaling for any meaningful amount of time, you can read your past entries and find the pattern. If you have not, this is harder — but not impossible. Think back to the last three decisions of similar shape. What did you tell yourself going in? What did you actually do? What did you tell yourself afterward?

A mentor — including an AI mentor with persistent memory — can name the pattern faster than you can, because they are not inside it. This is what AI life coaches and AI mentors are actually useful for in decision-making: not the answer, but the pattern.

Step 4: Ask the smaller question underneath

Most decisions get easier when you stop trying to answer the big question and start answering the smaller one inside it.

"Should I leave my job?" is the big question. The smaller question underneath is usually something like: "What would I have to be true about myself to leave?" or "What story am I telling myself about who I am that makes staying feel inevitable?" The smaller question is the one your thinking has actually been wrestling with, dressed up as the big one.

When you find the smaller question, the bigger one usually becomes a consequence of how you answer it, not a thing you have to decide independently.

Step 5: Commit to a reversible step

Most overthinkers treat every decision as if it were irreversible. Almost none are.

You do not have to know whether you are going to take the new job. You can ask the question that lets you find out. You do not have to know whether you are going to end the relationship. You can have the conversation that clarifies what would have to be true to stay. You do not have to know whether you are going to start the business. You can ship the smallest possible version of it this week.

A reversible step is the thing that breaks the loop without committing you to the wrong answer. It gives you data the next decision will be made on. It also produces movement, which is the actual mechanism by which most decisions get resolved.

The decision-making muscle

The system above is one decision. The thing that makes you a person who makes better decisions over time is doing it repeatedly, with a record, in a place private enough to be honest.

Repeated reflection has compounding returns the same way investments do. Each time you write down the unedited version, surface what you were avoiding, name the pattern, ask the smaller question, and commit to a reversible step, you get a little better at the underlying skill. The decisions get a little faster. The patterns get more obvious to you. The avoidance gets harder to maintain because you have a record of how it played out the last six times.

This is what people mean when they say a long-running journaling practice makes them think more clearly. It is not the journaling. It is the system, repeated, with memory.

How The Architect approaches this

The Architect was built explicitly to do steps 1, 2, and 3 automatically.

Step 1 (the unedited version) is what the writing-with-zero-knowledge-encryption flow exists for. You can write the version you would never say out loud because the company itself cannot read it.
Step 2 (surface what you are avoiding) is what the mentor personas do. The Stoic, the Sage, the Mystic, the Billionaire — each tuned to push back in a different way. The mentor reads what you wrote and asks the question behind the question.
Step 3 (the pattern) is what pattern detection does. After three entries, the AI analyzes your full journal history and surfaces the recurring themes, the broken commitments, the loops you are inside of.

Steps 4 (smaller question) and 5 (reversible step) are yours. The Architect can help you find them, but only you can take them.

If you are trying to make a better decision and you have been spinning on it for more than a few days, this is what the tool was built for. Try it free.

The honest closing

You are probably already a better decision-maker than the version of you that the loop has been showing up as. The thinking has been getting in your way, not helping you. The data you have refused to look at is the data the decision is actually waiting on.

Write the version that is true. Look at it. Move on the smallest reversible step. The rest gets easier.

You do not become someone who makes better decisions by reading another article about decision-making. You become it by doing the practice, in a place private enough to be honest, with a record long enough to learn from. Most people never get either. The ones who do tend to outperform the ones who do not by an embarrassing margin over time.

Spend the time on the practice. The decisions will compound.