What needs to be finished first
A dynamic situation exists with time pressure, incomplete information, and a need for action, where OODA is stronger than structured decision methods.
What needs to be ready before start
Situation board with four quadrants (Observe, Orient, Decide, Act); access to data sources (dashboards, logs, status channel); timer for loop timeboxing; shared document for decision log; communication channel with all participants.
One lead who runs the loop and makes decisions; one observer with data access who continuously feeds information; two to four additional contributors bringing options and risks; one scribe documenting every loop.
Current status of the situation in two to three sentences; available data sources; decision authority of the lead; known constraints (time window, resources, external dependencies); escalation path when thresholds are crossed.
5-30 min per loop, multiple loops in sequence
Make the four quadrants visible as working space. Feed first observations. Set loop timer to 5-10 min. Rule: after every Act return to Observe; do not remain in one loop. Keep the decision log as a running document with a timestamp per loop.
The one question this method answers
Which observation, orientation, decision, and action is one step ahead in the current loop, and which follow-up loop comes next?
Marker: Minute
| Step | Duration | Action | Hint |
|---|---|---|---|
10-2 min Observe | 2 min | Capture current situation: data, signals, actor behavior, own position. Record observations without interpretation in the Observe quadrant. | If Observe already contains conclusions ("the competitor will acquire them"), this belongs in Orient. Keep the separation clean, otherwise bias contaminates the data. |
22-4 min Orient | 2 min | Interpret: what the observations mean against experience, models, and hypotheses. Make options visible and test mental models. | Orient is the most important phase and is often underutilized. Place at least two interpretations of the situation side by side before deciding. |
34-5 min Decide | 1 min | Choose one option. The lead decides, while others contribute risks and alternatives. Document the decision with rationale in the log. | If nobody can decide, OODA stalls. The lead must decide, even under uncertainty. "We wait" is a valid decision but it must be made explicitly. |
45-10 min Act | 5 min | Execute the chosen action. Make the effect observable (telemetry, status update, customer feedback). Prepare for the next loop. | If Act takes longer than 5 min, the action is too large for the loop. Choose a smaller action or reduce loop frequency, but not both. |
510+ min Restart | continuous | After every Act immediately return to Observe. The situation has changed through your action and external movement. A new loop starts. | The longer the team analyzes after Act, the more speed advantage is lost. Loop discipline is core to the method, not speed for its own sake. |
What comes out at the end
Decision log with timestamp per loop, including observations, orientation, decision, action, and observed effect. One log per situation with lead and date, archived for lessons learned.
One new log per situation, loops numbered chronologically. In retrospective review the log is used as source. Attach corrections or clarifications at the end, do not overwrite.
Compact working template for OODA Loop with context, input, output artifacts, and next step.
Iterative decision cycle from Observe, Orient, Decide, and Act for dynamic situations.
When and for what do we use this method?
Which data, observations, decisions, or materials are available?
Short notes along the runsheet.
Owner, date, and success signal.
Concrete filled scenario, fictional example
Concrete filled scenario, fictional example
Lead: @sabine. Scribe: @tobias.
Recognize symptoms and steer against them
Observations contain conclusions ("there is an attack") instead of data ("Login attempts from IP range X at 500/min").
Keep the Observe quadrant strictly data-only. Interpretation belongs in Orient. The coach interrupts if words like "clear" or "obvious" appear in Observe.
Team jumps directly from Observe to Decide and the first interpretation becomes the decision, with no alternatives.
Make Orient a mandatory phase with at least two interpretations. Ask: "Which other explanation fits the same data?" If only one interpretation exists, pause the loop.
Team analyzes for 20 min and does not move to Decide or Act. The situation develops further without own intervention.
Enforce the timer strictly. Better a decision with 60% certainty than 100% certainty too late. Escalate to the next level if the lead cannot decide.
Next loop starts without data on the previous action; effect is inferred rather than observed.
Define before Act which metric makes effect measurable. Keep telemetry visible in Observe. If the action is not measurable, redefine the action size.
Decide requires escalation and repeated approvals, so the loop stalls and the advantage is lost.
Clarify decision authority before OODA and make it explicit in writing. If the lead is not decision-authorized, appoint a higher-level lead or pre-approve escalation.
Done signals checkable in under a minute
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.