Guides
Part of Strategic planning: steps, examples and decisions for 2027
Strategic planning case study: findings and lessons
Strategic planning case studies where decision quality and outcome quality come apart: variance, intent added afterward, and what actually transfers to you.
A strategy story is a bet plus a result, and the result is the least informative part of it. Nothing on this page describes a real organization, a real person or a real outcome, because the point is the reading method rather than another example to admire.
The general problem with published write-ups is covered in operations case studies. Strategy adds a specific difficulty on top: strategic choices are bets with real variance, so a sound decision can lose and a reckless one can win, and the story is always sorted by which happened. Reading only the accounts that got written and inferring the odds from them is survivorship bias in its purest form.
What to take away
- Judge the decision by what was knowable when it was made, not by how it ended. These come apart often enough that using the outcome as the grade teaches you the wrong lesson roughly as often as the right one.
- The intent in the story was mostly written afterward. Much of what reads as a plan was a series of adjustments that later acquired a narrative.
- Take the question and the trade from a strategy case. Never take the move.
Decision quality and outcome quality come apart
Four combinations exist, and stories are only written about two of them.
| Good outcome | Bad outcome | |
|---|---|---|
| Sound decision | Written up as genius. Was partly luck. | Never written up. This is where most of the learning is. |
| Unsound decision | Written up as bold. The most dangerous category to imitate. | Written up only as a warning, and usually by a competitor. |
The top-left and bottom-right cells produce almost all published material, which makes the two look like the same axis. They are not. A decision made with a reasonable read of the information available, a stated way to find out it was wrong, and a loss the organization could absorb, is a good decision whichever way it goes.
The practical use of this distinction is in your own reviews. When a bet pays off, ask what would have happened had the uncertain thing gone the other way, and whether you had a plan for that. If the answer is no, you got away with something and should not repeat the method.
Bets have variance, and the winners write the books
Where a large number of organizations make bold, uncertain bets, some will do well for no reason other than that outcomes vary. Those are the ones with a story to tell, an audience for it, and the credibility to be believed.
This means the association you observe between a strategy and success can exist even if the strategy is neutral or harmful on average. You are not looking at a biased estimate of the effect. You are looking at a sample selected by the outcome, from which no effect can be estimated at all.
The honest use of a strategy case is therefore narrow. It shows that an approach is possible, and it may describe a mechanism worth thinking about. It cannot tell you the odds, and the odds are exactly what you would want to know before making the same bet.
Ask, when you are shown one, how many organizations you would expect to have tried something similar. If the answer is "a lot," the story tells you nothing beyond that the good outcome was attainable.
Intent gets added afterward
Strategy write-ups describe a coherent direction pursued over years. Coherence is usually a property of the telling.
What often happened was smaller and messier: a set of local responses to pressure, a customer who pulled the business somewhere unexpected, a capability that existed for historical reasons and turned out to be worth something, a market that moved underneath a company that stayed still. Once the direction is visible, it becomes the plan, and the early adjustments are retold as the first steps of it.
Two tells are worth watching for. Dates that arrive without the deliberation that would have preceded a real decision, so a shift appears fully formed. And an absence of abandoned directions, when any organization operating over that period will have started several things that went nowhere.
The reason this matters practically: if you read an emergent path as an executed plan, you conclude that the lesson is to plan more precisely, when the actual lesson may be about how quickly they noticed something and reallocated. Those imply completely different habits.
What is never in the story
Five things are systematically missing, and each one changes how the case should be read.
- The option they killed. Every commitment forecloses others. The alternative that was rejected, and the argument for it, is the most useful part of any strategy decision and is almost never recorded.
- What made the loss survivable. A bet is only bold if you can afford to lose it. Reserves, a profitable existing line, a patient owner: these determine whether the same move is courage or recklessness, and they are rarely stated.
- The timing they did not choose. Some moves work because of when they happened, and the timing was frequently not a decision. It was when the money arrived, or when a key person left.
- The people who were right and lost the argument. Their reasoning would tell you what the genuine uncertainties were at the time. It is the first thing edited out.
- The scale of the commitment relative to everything else. A move that reads as a transformation may have been a small share of attention and money. Without that proportion, you cannot tell whether they bet the company or ran an experiment.
What actually transfers
Not the move. Three other things do.
The question they asked. If they worked out that their real limit was distribution rather than product, the transferable item is the diagnostic, not their answer. Your limit is elsewhere and the same conclusion would be wrong for you.
The trade they accepted. Any real strategy gives something up. Finding the sentence that names what got worse, or noticing that no such sentence exists, tells you whether you are reading a strategy or a list of ambitions. The same test applies to your own draft, and is the subject of strategic planning examples.
The thing that would have told them earlier. In hindsight there is almost always a signal that preceded the outcome by months. Ask what it was and whether you would currently see the equivalent. That question converts somebody else's story into a change to your own reporting, which is the only kind of change a case study can honestly produce.
Reading your own past plans
The same biases apply to your own history, more strongly, because you were there and remember being right.
Do it in a specific order. Read the plan and the decision record first, without looking at what happened. Ask whether the decision was reasonable given what is on the page, whether the assumptions were named, and whether anybody had agreed what would count as it not working. Then look at the outcome.
Reversing that order makes the exercise worthless, because once you know the ending, the earlier document reads as either prescient or naive and nothing in between. That flattening is hindsight bias, and it is not something you can decide not to do. Where no decision record exists, the honest conclusion is that you cannot evaluate the decision at all, only the result. Keeping a personal record so this becomes possible is covered in management foundations case study.
Two questions worth asking every time. Which of the assumptions were never checked, and would checking have been cheap? And which parts of the outcome would have happened anyway, without the plan, because the market moved or a large customer did something? The second question is uncomfortable and is where most inflated attributions live.
For what a plan can contain that does not depend on being right about the future, see strategic planning trends; for the underlying structure it should have, the strategic planning pillar.
Common questions
Is it not overly cautious to discount every success story?
The recommendation is not to discount them. It is to extract the mechanism and the question, and to decline the odds. A story is good evidence that something is possible and poor evidence about how often it works, and treating those as the same claim is the error.
What about a case where the organization failed? Are those more honest?
More candid, and subject to the reverse bias. Failure invites a search for the fatal mistake, so an account gets built in which one decision explains everything. Most failures are several ordinary decisions plus conditions, and the tidy postmortem is as constructed as the tidy success.
How do I discuss this without sounding like I am against every idea?
Ask about the trade rather than about the evidence. "What would we give up to do this?" is a question about their proposal rather than about their reasoning, and it produces the same information without putting anybody on the defensive.
