McKinsey 7S
7S is a diagnostic for why a change did not take. You announced the new way of working, everyone agreed with it, and six months later people are doing what they did before. Nobody resisted, nobody argued, and nothing moved. That is the situation this framework exists for.
It was developed at McKinsey around 1980 by Robert Waterman, Tom Peters and Julien Phillips, and it is still the reference most consultants reach for when an organisation is not doing what it decided to do.
The seven#
The claim is that an organisation is made of seven elements, that they are all connected to each other, and that effectiveness comes from their alignment rather than from any one of them being strong.
The three hard elements are the ones you can change by deciding to:
- Strategy. What the organisation is trying to win at, and how.
- Structure. Who reports to whom, how work is grouped, where authority sits.
- Systems. The processes and tools that run the daily work: planning, budgeting, reviews, the tooling, how work gets tracked.
The four soft elements change only when people’s behaviour does:
- Shared values. What the organisation actually believes, as opposed to what is on the wall. Waterman and colleagues put this at the centre because everything else is judged against it.
- Skills. What the organisation is genuinely capable of, not what its job titles claim.
- Style. How leaders actually behave, especially under pressure, and what that teaches everyone else about what really matters.
- Staff. Who is here, how they are recruited, developed, and promoted.
The orchestra#
An orchestra makes the interdependence obvious, because you can hear it.
The score is the strategy. The seating plan is the structure. The rehearsal schedule and the way parts are distributed are the systems. What the players can actually play is skills, who is in the chairs is staff, how the conductor behaves in a difficult rehearsal is style, and the shared sense of what this piece is supposed to be is the shared values.
Now hand them a new score and change nothing else. The seating is wrong for the new balance, the rehearsal schedule assumes a piece with different demands, half the players have never played this repertoire, and the conductor is still running rehearsals the way the old programme needed. Everyone is trying. It sounds worse than what they were playing before.
That is what a reorganisation feels like from inside, and the framework’s whole contribution is telling you to go and look at the other six.
Using it as a diagnostic#
The value is in the walk, not the diagram. When something has been decided and is not happening, take the change and ask each element one question.
| Element | The question |
|---|---|
| Strategy | Is the change actually consistent with what we say we are trying to win at, or does it quietly contradict it? |
| Structure | Does anyone own this, and do they have the authority to make it happen? |
| Systems | Does our tooling, planning and reporting make the new behaviour easy, or does the old path still exist and remain faster? |
| Shared values | Does this ask people to act against something they genuinely believe here? |
| Skills | Can people actually do the new thing, or have we asked for a capability nobody has? |
| Style | Do leaders do it themselves, especially in the week everything is on fire? |
| Staff | Are the people we hire and promote the ones who do this well? |
How to tell which one is fighting you. Whichever question produces an uncomfortable pause is the answer, and it is usually not the one that was announced. The most common culprit is systems, because the old path is still there and still faster, and people are not defying the change, they are taking the route that works. The second most common is style, because one senior person visibly skipping the new process teaches everyone more than any amount of documentation.
When it goes wrong. The failure mode of 7S is doing the walk, finding six misalignments, and launching six workstreams. That is the framework being used as a to-do list, which it is not. Fix the one that is actually load-bearing, ship it, then walk it again.
Where the framework stops#
Three limits worth saying out loud, because it is usually taught as if it had none.
It has no causality. It tells you which elements are misaligned. It does not tell you which one is causing the others, and it will happily present seven symptoms as seven problems.
It has no prioritisation. All seven are drawn the same size. In practice one is load-bearing and the rest are consequences, and the framework offers no help in telling those apart. That is what an issue tree is for.
It is a diagnostic, not a design. If you are designing an organisation rather than debugging one, Galbraith’s Star Model is the better instrument, because it is built around choices rather than around alignment. Reach for 7S when something has already been decided and is not happening.
How we use it here#
Two places, and they are narrow on purpose.
When a working agreement keeps sliding back. A standard we adopted, a review that keeps getting skipped, a cadence that quietly stopped. Rather than restating the rule louder, walk the seven and find the one element making the old behaviour cheaper. Usually it is systems, and usually the fix is removing the old path rather than adding a reminder.
Before announcing a change, not after. The walk is more valuable run in advance. If we are about to change how something works, going through the seven questions first tells us what else has to move on the same day, which is the difference between a change and an announcement.
What we do not use it for is designing anything, or as a slide. A 7S diagram in a document is almost always decoration; the output of using this framework is a sentence naming one element and what we are doing about it.
Related#
- Structured Problem Solving is where this sits among the frameworks we use, and which ones to reach for instead.
- Diagnosis is the same instinct applied to a metric rather than to an organisation.
- DRI is the structure question made concrete: does anyone own this.