What Is a Decision About Decisions? Meta-Decisions, Explained

A decision about decisions — a meta-decision — is a choice about how you choose: how fast you decide, how big you're allowed to bet, which calls deserve deliberation and which deserve a default. It's the highest-leverage class of decision there is, because a single meta-decision governs every future decision beneath it. It's also the title of Paper II of the DaD Papers.

Here is the strange thing about your decision process: you almost certainly never chose it. It accreted — from your parents' anxieties, your first boss's habits, the culture of whatever industry raised you. Most people run inherited decision software their whole lives and never once open the source. The meta-decision is the act of opening it.

What does a meta-decision actually look like?

A normal decision resolves one situation. A meta-decision sets policy for a class of situations. Compare:

The left column pays once. The right column pays on every decision it governs, for as long as the rule stands. That asymmetry is the entire argument for working at the meta-level: it's the difference between making a good trade and writing a good trading system.

Why does the meta-level compound?

Because decision quality times decision velocity is a compounding function, and the meta-level is where velocity gets set.

Picture the founder who stalls — asks twelve people, builds a pros-and-cons list that grows until both columns are equally useless. Meanwhile the person next to them made the same call in eleven seconds, got two of the next four wrong, adjusted by dinner, and is now three months ahead on a compounding curve the staller cannot catch. The staller didn't lose on decision quality. Averaged over the year, their individual calls may have been better. They lost at the meta-level — they never decided how to decide, so every decision paid the full toll of an unexamined process.

This is the insight behind Colonel John Boyd's OODA loop (Observe, Orient, Decide, Act), developed from fighter combat: the pilot who cycles through the loop faster gets inside the opponent's loop, and the opponent's beautifully-reasoned maneuvers start answering a situation that no longer exists. Tempo is not a nice-to-have on top of good judgment. Past a threshold, tempo is judgment — because a fast decider gets more feedback per quarter, and feedback is what judgment is made of.

What are the core meta-decisions?

Paper II builds the full doctrine; the load-bearing ones are four:

1. Reversibility triage. Jeff Bezos's language for this is one-way doors versus two-way doors: irreversible decisions deserve deliberation, reversible ones deserve speed, and treating the second kind like the first is how organizations calcify. The meta-decision is committing to sort every incoming choice into one of those two bins before engaging with its content.
2. Sizing. The Kelly Criterion — the formula gamblers and quantitative investors use for bet sizing — carries one lesson that survives translation out of the math: there is an optimal fraction to bet given your edge, overbetting it ruins you even when you're right about the edge, and the first rule is that no bet may be allowed to remove you from the game. Deciding your sizing rule once beats re-negotiating it with yourself inside every exciting opportunity.
3. Stopping rules. Herbert Simon won the 1978 Nobel in economics partly for the observation that real minds can't optimize — they satisfice, searching until an option clears a threshold, then stopping. The meta-decision is choosing your thresholds deliberately: what does "good enough to act" mean for this class of choice? Undefined thresholds are how twelve-advisor paralysis happens.
4. Signal admissibility. Antonio Damasio's Iowa Gambling Task research showed subjects' bodies reacting to bad card decks before their conscious minds could say why — and patients whose emotional circuitry was damaged, though their logic stayed intact, became pathologically indecisive. The meta-decision: does bodily signal count as data in your process, and how much? Pretending you decide on spreadsheets alone isn't rigor; it's a refusal to audit one of your inputs. That question gets its own treatment in frameworks versus gut feel.

How do you make your first deliberate meta-decision?

Small and enforceable beats grand and vague. Pick one class of recurring decision that currently costs you drag — say, "requests for my time" — and write a one-sentence policy you could detect yourself violating. Then let it run for a month before touching it. The step-by-step version of this is in how to build a personal decision framework, and the pruning discipline that pairs with it is in how to make fewer, better decisions.

One caution, in the spirit of the trilogy: the meta-level is upstream of individual choices, but it is not the top of the stack. What your decision process even gets to choose between is set by your distinctions, and which options deserve collapsing is signaled by delight. Decision is one movement of a three-part loop — the movement that makes the other two consequential.

Distinction reveals possibility. Delight signals direction. Decision collapses reality.

FAQ

Is a decision about decisions just decision theory?

Decision theory is the academic field — expected value, utility functions, probability. A decision about decisions is the personal, architectural act: choosing which process you will actually run when choices arrive. Decision theory is one input to that choice. Most people study neither and inherit their process by accident.

How is a meta-decision different from a normal decision?

A normal decision resolves one situation: hire this person, take this deal. A meta-decision sets policy for a whole class of situations: how fast we decide reversible things, how big we're allowed to bet, who decides below a threshold. One normal decision pays once. One meta-decision pays on every future decision it governs.

Isn't thinking about how you decide just sophisticated procrastination?

It can be, if it never terminates in policy. The test is simple: a real meta-decision produces a rule you can state in one sentence and violate detectably — "reversible calls get decided same-day." If your reflection on decision-making hasn't changed the speed or shape of your next ten decisions, it was rumination wearing a suit.

Where does the phrase "decision about decisions" come from?

It's the title of Paper II of the DaD Papers by Brad Hart — The Decision About Decisions: A Theory of Velocity, Intuition, and the Economics of Choice. The Paper works through Kelly Criterion bet sizing, Boyd's OODA loop, Damasio's somatic marker research, and a five-level taxonomy of decision failure with the math and the cures.