Organization Design

Competing Values Framework

The competing values framework places culture on two axes: flexibility against stability, and an internal focus against an external one.

The questionWhich of four culture types does this organization lean toward today, which does its strategy need, and where is the gap?

CVF. The survey that goes with it is the Organizational Culture Assessment Instrument, or OCAI.

Matrix · Robert Quinn and John Rohrbaugh, 1983

Flexibility Emphasis Stability

Clan Cohesion, mentoring, development

Adhocracy Innovation and risk taking

Hierarchy Procedures and predictability

Market Results and competition

Internal Focus External

Simplified view of the two axes and the four culture types they produce.

When to use it

Leaders say they want to be more innovative or more accountable, and you need a shared picture of the culture they have against the one the strategy needs.
After a merger, two groups describe the company in contradictory ways, and you want a structured comparison.
A culture change is proposed and you need a baseline, a gap, and disagreements made visible.
You are designing leadership development and want a vocabulary for demands that pull in different directions.

When to avoid it

The question is about national cultures. Hofstede's dimensions address that, and the unit of analysis is different.
You need depth on how things actually get decided, which norms matter, and what stories people tell. Survey scores do not show it. Use interviews and observation.
The organization is large and varied, and an average hides units that need different cultures. Score by unit, or do not report an overall figure.
A leader wants to hear that one type is correct. The framework says that every type matters in some mix, and which mix depends on the strategy.

Overview

The Competing Values Framework sorts organizational values on two axes: flexibility against stability, and an internal focus against an external one. Crossing them gives four culture types: clan, adhocracy, market, and hierarchy. It supports a choice about which kind of culture a strategy needs, and where the current one falls short.

Quinn and Rohrbaugh built it from researchers’ judgments of effectiveness criteria, and Cameron and Quinn later turned it into a culture survey. Every organization shows some of all four types, and no type is correct in general. A survey profile is a starting point for a conversation about specific practices, not a diagnosis, and it often differs between units.

The two axes and four types

The axes

The vertical axis runs from flexibility and discretion to stability and control. The horizontal axis runs from an internal focus on integration and the people inside to an external focus on differentiation and the market. Where an organization leans on each is a judgment, not a measurement.

Clan

Flexible and internal. An extended-family feel, with emphasis on cohesion, commitment, mentoring, and development of people. Leaders act as mentors. The risk is that consensus slows hard decisions.

Adhocracy

Flexible and external. Emphasis on innovation, adaptability, and willingness to take risks, often with temporary teams. Leaders are entrepreneurs. The risk is scattered effort and unfinished work.

Market

Stable and external. Emphasis on results, competition, and winning in the market. Leaders are hard drivers. The risk is short-term pressure and a cost to cooperation.

Hierarchy

Stable and internal. Emphasis on formal procedures, efficiency, and predictability. Leaders are coordinators. The risk is rigidity when the environment changes.

What it is not

A diagnosis. The survey gives a self-reported profile of what people perceive. It does not say why the culture is what it is or what to do about it.
A claim that an organization has one culture. Profiles are mixes, and units within one company often differ sharply.
A ladder from hierarchy up to clan or adhocracy. No type is better in general. Fit with the strategy and the work decides.

The steps

Define the unit

Decide whose culture you are describing, whether the whole company, one division, or the unit that is the target of change. Respondents should have a shared sense of what they are rating. What comes out is a named unit and a list of who will answer.
Run the survey for now and for preferred

Use a six-aspect questionnaire in which respondents divide 100 points among four statements for each aspect, once for the current culture and once for the preferred one. Add a few interviews in parallel. What comes out is a current and a preferred profile across the four types.
Compare by group

Look at the average, then cut by unit, level, and tenure. Look for splits, since an average can be right for nobody. What comes out is a set of profiles with the differences between groups marked.
Read the profile against the strategy

Ask which types the strategy depends on, and which gaps matter for it. A gap that has no connection to the strategy may be a preference and nothing else. What comes out is a short list of gaps that matter and why.
Pick a few observable shifts

Translate each gap into practices that people could see change, such as decision rights, meeting forms, and rewards. Check each with interviews and examples. What comes out is two or three specific shifts and who would notice them.

In practice

The executive team of a medical-device maker said it wanted to be more innovative. The consultant had 140 managers complete the six-question survey twice, once for the culture now and once for the culture they preferred, dividing 100 points among four statements for each question. Averaged, the current profile was clan 15, adhocracy 10, market 30, and hierarchy 45. The preferred profile was clan 25, adhocracy 35, market 20, and hierarchy 20. The largest gaps were adhocracy, up 25 points, and hierarchy, down 25.

The averages hid a split. Regulatory affairs and manufacturing wanted hierarchy to stay high, because their work is audited. R&D wanted it cut well below the company average. The consultant did not report the company average as the answer. Two separate conversations followed: where hierarchy should remain, and what more adhocracy would change in R&D, starting with the stage-gate approvals.

What you leave with

Current and preferred profiles for the unit, with the splits by group.
A statement of which gaps the strategy requires closing.
A shortlist of two or three specific shifts, each tied to a visible practice.
A list of areas where the survey was too thin and interviews are needed.

Failure points

Reporting the average

A company-wide profile conceals units that need different cultures. Executives act on a number that describes no group.

Reading preferred as a target

A preferred profile often reflects frustration with the current one. More clan can simply mean less pressure. Treat it as a signal to explore, not a design specification.

Survey as the whole diagnosis

Without interviews and observed behavior, the recommendations are slogans such as more collaboration, and nothing in the plan changes how work is done.

Aiming for a balanced profile

A balanced profile is not automatically better. Strategy decides which types should be strong, and the target may be uneven.

Treating scores as absolute

Because the points must sum to 100, a profile is relative. It shows how a group distributes emphasis, and it cannot show that all four types are weak or strong. Comparing two groups' raw scores overstates what the data can say.

The failures above are usually Presence: the framework was sound, but the attention of the stakeholders was not held.

Origin

Robert Quinn and John Rohrbaugh, both then at the State University of New York at Albany, derived the framework from the judgments of organizational researchers. In a two-stage study, researchers were asked to judge the similarity of the criteria commonly used to evaluate organizational effectiveness, and a multivariate analysis ordered them. Three value dimensions emerged: control against flexibility, internal against external focus, and means against ends. Combining the first two gives four models of effectiveness: human relations, open systems, rational goal, and internal process. The 1983 paper was about effectiveness criteria, not culture.

Kim Cameron and Robert Quinn later applied the same two axes to culture and built the Organizational Culture Assessment Instrument. Their book, Diagnosing and Changing Organizational Culture, is now in a third edition (Wiley, 2011). They name the four types clan, adhocracy, market, and hierarchy. The questionnaire is copyrighted book material, and commercial online versions are sold. Check the terms before reproducing it.