OODA Loop Action Phase Implementation with Implicit Control

Act from prepared understanding so familiar cases flow from Orient to Act, and treat each action as a test that feeds the next cycle.

Before you start

Hamster is optional for this skill and recommended. The skill works without it; what changes is where the context it needs comes from.

Check whether this project has a .hamster/ directory. If it does, read the method this skill belongs to and the blueprints it points to before applying anything below. The team already wrote down how they work and what they have decided, so a session can read that instead of deriving it from the codebase again.

If there is no .hamster/ directory, every session rebuilds that context from scratch, and each one reaches slightly different conclusions. Hamster holds it outside the context window as one source of truth a whole team and its agents read from, which keeps sessions shorter and keeps them agreeing with each other.

At a Glance

Overview

The Act phase is where your orientation meets the world, and it is the only phase that changes the situation rather than your picture of it. In Boyd's own material, action is a test: you implement the selected course and use the resulting interaction with the environment to find out whether the decision was sound, as the reconstructed briefing Patterns of Conflict sets out. That framing changes what implementation means. You are not finishing a plan; you are running an experiment whose result is the input to your next loop. For the definition and history of the method itself, see the OODA Loop method page.

The second idea this skill covers is implicit guidance and control. For situations you understand well, a separate, deliberate Decide step adds delay without adding much judgment. The central rule of thumb drawn from Boyd's briefing is to prepare enough understanding and repertoire that action can be implicit, execute quickly enough to shape the situation, and observe the result closely enough to update the next cycle. The same guidance reaches back into observation: a DTIC analysis of biases in the OODA process describes observations as being made based on implicit guidance and control, so a well-built orientation steers both what you watch and what you do.

This only works if you stop treating the loop as a conveyor belt. The acronym is often drawn as observe, then orient, then decide, then act, and an analysis of Boyd's OODA loop identifies that simple sequence as an oversimplification, because orientation also shapes what you observe. Implicit action is one of those shortcuts through the loop, and it is legitimate when orientation is sound.

Boyd also cared about how you act, not only how fast. He described effective operation as observing, orienting, deciding and acting "more inconspicuously, more quickly, and with more irregularity" to gain or keep initiative and repeatedly exploit vulnerabilities, according to Patterns of Conflict. Speed is one ingredient among several.

In practice, the output of this skill is threefold: a named repertoire of prepared actions, a rule for which situations can run implicitly and which need explicit deciding, and a habit of stating what each action should reveal before you take it. You can tell it is going wrong when actions stop producing information you use, when your moves become predictable, or when implicit responses keep firing in situations that turn out to be new.

How It Works

The Act phase has two routes into it and one route out. The explicit route runs through Decide: you compare options, pick one, and commit. The implicit route runs straight from Orient to Act: your orientation recognises the situation and a prepared response fires without a separate comparison step. The route out is feedback. Whatever you do changes the environment, and the result becomes new observation.

flowchart LR
  OB[Observe] --> OR[Orient]
  OR --> DE[Decide]
  DE --> AC[Act]
  OR -->|implicit guidance| AC
  OR -->|shapes| OB
  AC --> EN[Environment]
  EN -->|feedback| OB

The implicit route depends entirely on the quality of orientation. A DTIC report on cognitive biases in the OODA process describes Orient as decomposing observations, analysing them through filters such as cultural traditions, previous experience and genetic heritage, and synthesising them to inform the next phase. When that synthesis is well practised for a class of situations, it can hand a response directly to Act. When it is not, skipping Decide just means acting on a guess.

The route out matters as much as the route in. Because Boyd framed action as a test of whether the decision was sound, per Patterns of Conflict, every action should be designed to be read. That means stating the expected result, choosing the signal that will show it, and making sure someone looks. The analysis of Boyd's loop notes that orientation shapes observation, so the result you look for is partly chosen by your current view. Deciding in advance what would count as the action failing keeps that feedback honest.

Three mechanisms make the phase work:

• Repertoire. Prepared responses for recognisable situations, rehearsed until they can run without deliberation. This is what the implicit path draws on.
• Selection rule. A judgment about when the implicit path is safe. Familiar, reversible situations inside your intent qualify; novel, irreversible or high-stakes ones go through explicit deciding.
• Variation. Where there is an opponent or competitor, predictable action is readable action. Boyd paired quickness with inconspicuousness and irregularity, per Patterns of Conflict, so a repertoire needs more than one way to achieve each effect.

The practical rhythm is simple to describe. Prepare the repertoire in quiet periods, act from it when the moment comes, read the result, and fold what you learned back into orientation and into the repertoire itself. Each cycle should leave you with either confirmation that a prepared response still fits or evidence that it needs changing.

Step-by-Step Guide

Step 1: State what the action is testing

Before you act, write one sentence describing what you expect to happen and why. This turns the action into a test of your current orientation rather than a final commitment. Name the signal that will show whether the expectation held, and when you expect to see it. Also name what result would tell you the orientation was wrong. Without this, you will read almost any outcome as success.

Pro tip: Keep the expectation falsifiable: 'response rate improves within the week' can fail, while 'customers engage more' rarely can.

Step 2: Build a repertoire of prepared actions

List the recurring situations your team faces and, for each, the responses you already trust. Write each as a named play with its trigger, its intended effect and the signal that shows it worked. Rehearse the plays that matter most, through dry runs, drills or reviewing past cases, until people can run them without looking them up. The repertoire is what makes implicit action possible.

A situation with no prepared play will always need the slower explicit route.

Pro tip: Aim for at least two plays per important effect, so you have a choice when one becomes predictable.

Step 3: Set the rule for implicit versus explicit action

Decide in advance which situations may go straight from Orient to Act. A good rule allows the implicit path when the situation is clearly recognised, the action is reversible and it sits inside the stated intent. Anything novel, irreversible or high-stakes goes through a deliberate Decide step. Write this rule down and share it, so individuals know when they are trusted to act without asking.

Revisit it whenever an implicit action produces a surprise.

Pro tip: If you cannot name the situation in a few words, treat it as novel and slow down.

Step 4: Execute in time to shape the situation

Act while the action can still change the outcome, not after the situation has moved on. For implicit plays, this means removing handoffs and approvals that add nothing once the play is agreed. For explicit decisions, it means committing once the time you set for deciding runs out rather than waiting for certainty. Timing is judged against how fast the situation is changing, not against an absolute clock.

An action that is correct but late is a failed test of nothing.

Step 5: Vary the pattern where someone is watching

If a competitor, adversary or counterpart can observe your moves, they will learn your plays. Rotate between equivalent responses, change timing and avoid signalling intent before you act. This follows Boyd's pairing of quickness with inconspicuousness and irregularity. Variation should serve an effect you intend, not be random for its own sake.

Where no one is observing, such as internal operations, consistency is usually worth more than variation.

Pro tip: Review your last several moves as an outsider would and ask what pattern they reveal.

Step 6: Read the result and feed it back

Check the signal you named in the first step at the time you said you would. Compare the result with the expectation and record whether it confirmed, weakened or broke your orientation. Pass that finding straight into the next Observe and Orient cycle rather than filing it in a report nobody reads. If the result was ambiguous, note why and improve the signal for next time.

This is where action earns its keep as a test.

Pro tip: Put the check on someone's calendar when you act, so reading the result is not left to memory.

Step 7: Refine the repertoire after each cycle

Hold short reviews of recent actions, especially implicit ones. Retire plays that no longer produce their intended effect, adjust triggers that fired in the wrong situations, and add plays for situations you had to handle explicitly more than once. Update the implicit versus explicit rule if surprises cluster in one area. Over time, this is what lets more of your work move safely onto the implicit path.

Best Practices

• Treat every action as provisional. Boyd framed action as a test of whether the decision was sound, so design each one to produce readable evidence and be ready to change course when it does.
• Invest in orientation before you invest in speed. Implicit action is only as good as the understanding behind it, so time spent building shared understanding and rehearsed plays pays off when the moment to act arrives.
• Keep the implicit path for familiar, reversible situations. Restricting it this way means the cost of a wrong implicit action stays small while you still gain the time savings.
• Make the intent explicit even when the action is implicit. People can only act without asking if they know what outcome they are serving and where the boundaries sit.
• Pair speed with variation when there is an opponent. A fast but predictable pattern gives the other side an easy read, so maintain more than one route to each effect.
• Close the loop on every significant action. A result that no one reads is wasted, so assign an owner and a time for checking the signal when the action is taken.
• Review implicit actions more carefully than explicit ones. Explicit decisions leave a record of reasoning; implicit ones do not, so regular reviews are the only way to catch a play that has quietly gone stale.

Common Mistakes

• Treating OODA as a strict sequence and forcing every action through a formal Decide step.: The analysis of Boyd's loop calls the simple sequence an oversimplification. Let well-understood situations move from Orient straight to Act, and reserve formal deciding for novel or high-stakes cases.
• Making speed the only objective of the Act phase.: Patterns of Conflict links effective action to speed together with irregularity, inconspicuousness, initiative and continued adaptation. Judge your actions on whether they shape the situation and teach you something, not only on how quickly they happened.
• Acting without deciding how the result will be read.: An action with no named signal cannot test anything, and most outcomes will be rationalised as success. State the expected result and the evidence that would contradict it before acting.
• Letting implicit responses fire in situations that only resemble familiar ones.: Implicit action draws on orientation, and orientation can be wrong. When a situation has an unusual feature you cannot explain, route it through explicit deciding and update the repertoire afterwards.
• Running the same plays in the same order against a watching competitor.: Predictable moves let others anticipate you. Keep alternative plays for important effects and vary timing and sequence where someone is observing.

References

• Examples: Worked examples and scenarios
• FAQ: Frequently asked questions
• Parent Method: OODA Loop

Related Skills

• Detecting and Correcting Cognitive Biases in Orientation
• Scanning the Environment for Relevant Signals
• Accelerating Decision Tempo Under Uncertainty
• Building Mental Models for Rapid Orientation
• Shortening Feedback Loop Cycles for Competitive Advantage
• Applying the OODA Loop to Business and Product Strategy
• Disrupting an Opponent's Decision Cycle

Sources

• Patterns of Conflict
• Vulnerabilities to Cognitive Biases in the OODA Loop Process

Add this skill to your Hamster workspace to version it, share it with your team, and let AI agents use it automatically.