Developed by John R. Boyd in his briefings (Patterns of Conflict, 1986) and carried into business by Chet Richards (Certain to Win, 2004).
Runs decisions as a continuous cycle of observe, orient, decide, and act, and treats cycling faster and with better orientation than a rival as the source of advantage.
Two loops in a price war, and the faster one gets two turns in
The startup's observe, orient, decide, and act run down the left; the incumbent's slower loop on the right. At each choice, pick how the startup moves: the reply shows what it learned and how much time it gave away. Illustrative.
Startup
Incumbent
Startup
Incumbent
Fast, well-read moves: 0 of 4Fast, well-read moves: 1 of 4Fast, well-read moves: 2 of 4Fast, well-read moves: 3 of 4Fast, well-read moves: 4 of 4
- The incumbent acts (Incumbent). Monday: the incumbent payroll provider cuts its mid-tier price 20 percent for new customers in the Northeast, where the startup sells mostly to restaurants.
- Observe (Startup). The cut was announced Monday. Pick what the startup does first.
- Wait for the quarter's win and loss numbers before reacting. Six weeks pass. By then the incumbent has signed 300 restaurants. Slow: waiting for full information hands the rival a free loop.
- Pull this week's lost deals, call five prospects, and read the offer's fine print. By Wednesday: the cut applies only to annual contracts, and three of the five prospects pay monthly. Fast and specific: new information in two days.
- Orient (Startup). Orienting is reading the facts against what the team knows about its customers and the rival. Pick the reading.
- It is a price war: match the 20 percent everywhere. Margins halve in every region, including the ones the incumbent never touched. An old reading: every price cut treated as a fight over price.
- It targets annual buyers, and our restaurants pay monthly and close for the off season. The team sees that the cut misses most of its market. A sound reading: the same facts, read against who the customers are.
- Decide (Startup). The reading points to restaurants that want to pay only in the months they are open. Pick the decision.
- Commission a pricing study before changing anything. The study is due in five weeks. A decision put off runs at the rival's pace.
- Launch a seasonal plan in the Northeast on Friday and watch sign-ups for two weeks. A test with a stop date: small enough to reverse, fast enough to learn from. Decided on good enough information, with a way back.
- Act (Startup). Friday: the seasonal plan goes live in one region. Sign-ups, churn calls, and the incumbent's response are now the next things to observe.
- The incumbent orients (Incumbent). Two weeks in, the incumbent's pricing committee is reviewing the first month of its cut. Its monthly meeting has not yet discussed the seasonal plan.
- Observe again (Startup). Two weeks of data: sign-ups in the test region are up 40 percent, and almost all the new customers are restaurants that close in winter.
- Orient again (Startup). The test worked. Pick what the startup does with the result.
- It worked: roll it out to every region at once. Two regions have no off season, and the plan loses money there. Speed without the check: the loop stops learning.
- Check where it worked: regions where restaurants close for a season. The plan works in the four regions with seasonal closures and does nothing in the other two. The loop corrects its own reading before the next move.
- Decide and act (Startup). The plan rolls out to the four regions where it worked. Second loop done, five weeks after the cut.
- The incumbent decides (Incumbent). The incumbent's committee approves a seasonal plan of its own. By then the startup has run two full loops and holds the regions where the plan matters.
An example
A payroll startup that sold mostly to restaurants woke up on a Monday to a 20 percent price cut from the largest incumbent in its main region. The sales head wanted to match it that afternoon.
The founder asked for two days. By Wednesday the team had pulled every deal lost that week, called five prospects, and read the offer's fine print: the cut applied only to annual contracts, and most restaurants paid monthly and closed for part of the year. Matching the cut everywhere would have halved margins to fight for customers the incumbent was not even chasing. On Friday the startup launched a seasonal plan, billed only in the months a restaurant was open, in one region. Two weeks later sign-ups there were up 40 percent, but almost all of them were restaurants that closed in winter, so the second rollout went to the four regions with an off season. The incumbent's pricing committee, meeting monthly, approved a seasonal plan of its own six weeks after its cut.
Why it works
Every competitor runs some version of the loop: it notices what changed, makes sense of it, chooses, and acts. The side that completes the loop faster, with a sound reading, acts on a situation the other side has not yet understood, so the slower side is always answering a move that has already been overtaken.
Speed counts only with a sound reading behind it. Orientation is where the loop is won or lost, because the same facts mean different things depending on what a team knows about its customers and its rival; reading the cut as a price war would have led the startup to act fast in the wrong direction. Treating each action as the next observation keeps the reading honest, since the test's results corrected the startup's own view before the second rollout.
The thinking behind it
John R. Boyd John R. Boyd, a US Air Force fighter pilot and strategist, set out his ideas in briefings instead of books, chiefly Patterns of Conflict (final version 1986) and The Essence of Winning and Losing (1995). His observe, orient, decide, act loop describes competition as a contest of decision cycles, in which the side that orients and acts faster than its opponent can disrupt the opponent's picture of the situation. His last briefing gives the loop's fullest drawing.
Chet Richards Chet Richards, a writer on Boyd's strategy, carried the loop into business in Certain to Win: The Strategy of John Boyd, Applied to Business (2004). The book takes observe, orient, decide, act from Boyd's military strategy and applies it to competition between companies, where the rivals are firms and the contest is for customers and markets. It is a strategist's application of Boyd's ideas, with no research of its own behind it.
How to run it
- Observe what changed. Gather fresh, specific signals within days, such as lost deals, customer calls, and the rival's actual terms.
- Orient deliberately. Read the facts against what you know about customers and the rival, and ask how the reading could be wrong.
- Decide on good enough information. Choose a move that can be reversed, with a date to judge it, and skip the wait for certainty.
- Act small and fast. Ship the move where its effect can be measured, such as one region or one segment.
- Treat the result as the next observation. Read what happened as new information about the market, and keep blame out of the room.
- Review the orientation. After each loop, ask whether the reading was right, and change it before the next move.
Where it goes wrong
The loop gets mistaken for a call to hurry. A team that acts fast on an old reading of the market, matching every price cut on reflex, is simply making mistakes sooner. Watch for moves made before anyone has asked what the change means.
A second failure is a loop that never closes. Teams observe and orient for weeks, waiting for complete data, while the rival acts twice. If a decision keeps waiting on one more report, set a date and decide with what exists.
The third is skipping the review. Speed comes from repeating loops, and each one is only as good as the reading behind it; a team that never checks its orientation keeps running the same wrong loop faster. End each cycle by asking what the results say about the reading.
When to use something else
- Recognition-Primed Decision. When one experienced person must decide fast under pressure, with no rival reacting to the choice.
- PDCA Cycle. When the loop is improving a process through measured trials, without a competitor setting the pace.
- Second-Order Thinking. When the risk is a quick move whose later effects, such as a rival's retaliation, outweigh the first result.
Where it sits in SOLVE eX
Go to the source
- Patterns of Conflict; The Essence of Winning and Losing, briefings, final edition of Patterns of Conflict December 1986, 1986, 1995
- Certain to Win: The Strategy of John Boyd, Applied to Business, Xlibris, 2004