Decision rules: pre-deciding your way out of overwhelm

Overwhelm is usually not too much work; it is too many open decisions. Every unmade decision — should I take this call, learn this tool, lower this price, keep polishing this draft — stays open in the background and charges rent on your attention. Decision rules close those decisions in advance: you decide once, in a calm moment, how a whole category of situation will be handled, and the rule executes from then on.

Decision rules are the second component of what I call the personal operating system, and in my experience they are the component with the highest return per hour of setup. This piece shows real examples and the method for writing your own.

Why do open decisions cause overwhelm?

A task is bounded: it has a size, and doing it shrinks the pile. An open decision is unbounded: it can be revisited at any moment, and typically is — in the shower, mid-meeting, at 2am. Ten tasks feel like a list; ten open decisions feel like drowning, because each one re-presents itself every time the context shifts.

Worse, open decisions get made at the worst possible moments. The question "should I discount?" does not arise during calm planning; it arises live, on a call, in front of a client, when the pressure to say yes is at its peak. An improvised decision made at the moment of maximum temptation is not really a decision — it is a reflex with a rationale attached afterwards.

The mechanism of a decision rule is simple: when a rule exists, then the decision costs nothing at the moment of temptation. The thinking was done earlier, by a calmer version of you, and the moment itself requires only recall.

What do real decision rules look like?

These are the kinds of rules working operators actually run — short, absolute, and slightly uncomfortable:

• New idea → parked until quarterly review. Ideas go into a file, unexamined, and get one fair hearing per quarter. The rule does not kill ideas; it kills the mid-project swerve, which is where most momentum goes to die.
• New tool → only if a paying project requires it. Tools are researched at the moment of need, never speculatively. Speculative tool research is procrastination that produces invoices.
• Price question → quote the anchor, never improvise downward. The price list is decided quarterly, in cold blood. On a live call, the only permitted answer is the listed number. Any exception is written down and argued for at the next review — not granted in the moment.
• Stuck for 15 minutes → ship the current version. If a piece of work has been circling for a quarter of an hour without moving, the version that exists is the version that ships. Being stuck is data; the rule converts it into an ending.
• Task with no visible output this month → deleted. If I cannot say what a task will visibly produce within thirty days, it leaves the list. It can re-apply when it has an output.

Notice what these rules have in common: each one closes a category of decision that would otherwise reopen weekly, and each one is checkable in a second. "Be more focused" is a hope. "New idea → parked" is a rule.

How do I write my own set?

Do not copy anyone else's list wholesale — rules earn their authority by being answers to your actual failures. The method:

1. Harvest from your friction log. For two weeks, capture every moment of drag or regret in one line, as it happens. (The friction log is part of how I run my week; it takes seconds per entry.)
2. Look for repeats. A friction that appears once is weather. A friction that appears three times is a missing rule.
3. Write the rule as "when X, then Y". X must be observable ("stuck for 15 minutes"), Y must be a single action ("ship it"). If either side needs judgement to apply, the rule will collapse exactly when you need it.
4. Add one rule per week, at most. A rulebook installed overnight gets abandoned overnight. Rules bed in one at a time.
5. Review quarterly; retire freely. A rule that keeps getting overridden is either wrong or badly worded. Fix it or delete it — a rulebook you routinely ignore trains you to ignore rulebooks.

Keep the whole set under about ten rules. The aim is not to bureaucratise yourself; it is to close the specific decisions that were eating you.

What changes when the rules are in place?

The workload rarely shrinks, but the weight does. Days stop being a chain of negotiations and start being execution. And there is a compounding effect that matters beyond your own head: a founder who runs on rules becomes legible to their team and their business. Predictable pricing, predictable priorities, predictable endings.

That legibility is not cosmetic. Firms where every decision routes back through the founder's in-the-moment judgement exhibit a recognisable pattern of symptoms — I have listed them in five signs your growth depends on you — and pre-decided rules are the first step out, because a written rule is the one form of founder judgement that other people can execute.

Decide once. Let the rule pay for itself every week after that.

---

Keep the thinking coming: The Format — a fortnightly letter on business and personal systems. Subscribe by email.