← Workplace culture

Culture · diagnosis & change

Schein’s three levels of culture

Move from visible routines to the beliefs beneath them, and investigate why stated values and daily decisions can diverge.

Edgar H. Schein · 1985 · Original editorial explanation
Schein’s three levels of culture — simplified model sketchArtefacts · what is visible · Espoused values · what is said · Assumptions · what is taken for granted. Editorial interpretation, after Edgar H. Schein (1985).Artefacts · what is visibleEspoused values · what is saidAssumptions · what is taken for grantedADM HR · ORIGINAL INTERPRETATIONEdgar H. Schein · 1985
Original simplified sketch after Edgar H. Schein (1985). Read the explanation for assumptions and limitations. Download SVG ↗

The core idea

Schein separates visible artefacts, espoused beliefs and values, and underlying assumptions. Artefacts include language, spaces, documents and repeated behaviour. Espoused values describe what people say matters. Assumptions are deeply learned understandings that can become so familiar they are rarely questioned. An artefact is observable; its meaning needs investigation. A silent meeting could reflect agreement, fear, fatigue or a norm of thinking before speaking. The model concerns depth of interpretation rather than a ranking of good and bad cultures.

Source and attribution [1]

Using it in practice

Start with a bounded question such as why colleagues hesitate to report staffing problems. Observe a routine, inspect relevant decisions and invite accounts from people with different roles and lengths of service. Separate what happened from your explanation. Ask what a newcomer would learn from that incident. Compare what leaders announce with what gets rewarded when costs rise. Return tentative interpretations to participants and actively seek cases that do not fit. HR can then test changes to the routines that reproduce the problem.

An example, not a reported case

Worked example · illustrative

A fictional guarding business says safety comes first. At shift handover, supervisors praise teams that fill every post but barely discuss fatigue concerns. Interviewees describe learning that escalation signals personal failure. HR treats this as a possible assumption, not a proven fact about every supervisor. One site introduces a short staffing-risk discussion and recognises early escalation. The pilot tracks the response to reports, overtime pressure and staff accounts of the next difficult shift. A rise in reporting could mean greater trust rather than more underlying danger.

What to watch for

Do not diagnose unconscious beliefs from office decor or a single survey. People may disagree within the same team. Your position as a manager or researcher affects what they feel able to disclose. Values gaps can reflect competing constraints as well as hypocrisy. A workshop can generate explanations, but repeated observation and follow-up are needed before claiming that deeper assumptions have changed.

Reading the three levels

At the artefact level, keep a factual record: who spoke, what was decided and what followed. At the values level, collect both formal commitments and employees’ own descriptions of good work. At the assumption level, formulate a tentative sentence that might explain several events, such as “a competent person solves difficulties alone”. Label it as an interpretation. Ask participants when that sentence fits, when it fails and what history might explain it.

A practical evidence notebook

Use four columns: observed incident, stated expectation, possible explanation and disconfirming evidence. Include normal routines as well as crises. A disciplinary decision, promotion conversation or response to a customer complaint may reveal different priorities. Remove unnecessary identifying details from working notes. The aim is to understand recurring organisational learning, not to compile a secret list of supposedly difficult employees.

Changing what teaches the culture

If staff learn that reporting creates extra work and no support, a new slogan will have little effect. Redesign the response: acknowledge the issue, agree a decision owner, explain the outcome and check back with the person who raised it. Give supervisors time to do this. Review whether promotion, resourcing and recognition support the same expectation. Ask new starters what they have actually learned after several weeks.

Use alongside other models

Use Schein to deepen an explanation suggested by a culture profile. Then use the Cultural Web to identify stories, routines and controls that reinforce it. This sequence is an editorial working method, not a standardised Schein assessment. Keep the original observations visible so that an appealing interpretation does not replace the evidence.

Investigate a contradiction between words and practice

Choose a concrete tension, such as 'we encourage challenge' alongside meetings where junior staff rarely disagree. Collect examples of what happens before, during and after someone raises a concern. An artefact is evidence to explore, not a direct reading of an underlying assumption. Silence could reflect fear, lack of preparation, meeting design or something else. Ask people how they interpret the situation and compare accounts across roles before deciding what the pattern means.

Use a critical-incident conversation

Invite participants to describe a recent decision under pressure: what happened, what options were available, who influenced the outcome and what was rewarded. Ask what a newcomer would learn from the incident. Keep the discussion specific and avoid asking people to reveal sensitive cases in a public group. Several incidents may reveal a recurring expectation, such as protecting the client relationship even when internal concerns are raised. Treat that as a hypothesis to test, not a hidden truth uncovered by one workshop.

Design an experience that challenges the assumption

If staff believe bad news damages careers, a new value statement is unlikely to be sufficient. Leaders can invite an early risk review, respond constructively and show what changed because someone spoke. Repeat the behaviour when the news is inconvenient. Make supporting changes to escalation and reward practices. The aim is to provide credible experience that a different behaviour is possible; asking people to believe a slogan while the old consequences persist will weaken trust.

Look for change below the visible surface

Observe how decisions are made when time or money is tight. Do employees use the new escalation route, and do leaders respond consistently? Ask newcomers what behaviour seems to be expected. Visible changes such as a new office layout may support a shift, but they do not prove one. Schein's lens is especially useful for examining meaning and taken-for-granted beliefs; combine it with operational evidence and avoid claiming to measure unconscious assumptions precisely through a short survey.

Investigate an assumption without declaring it a fact

Begin with a repeated puzzle: employees say safety matters, yet supervisors praise people who work through checks. Gather several examples across shifts, ask what people expected would happen if they acted differently, and compare the answers with promotion, scheduling and incident-review decisions. A possible assumption is that keeping the client happy matters more than stopping work. Treat that as a hypothesis to discuss, not a psychological diagnosis. Look for exceptions: a team that routinely stops unsafe work may reveal conditions that make a different assumption credible.

Connect depth with observable change

Schein helps explain why a practice feels natural; Nabeel’s behaviour-and-outcome distinction helps specify what to change. In this example, define the observable action as supervisors pausing a job when a check fails, and the intended outcome as safer delivery. Track the response to a real concern, not merely completion of a values workshop. If employees still expect punishment, investigate the incentive and authority structure. A visible ritual changes little when the consequences of speaking up remain unchanged.

Take it into your next conversation

Three useful questions.

  1. What observable evidence supports this interpretation?
  2. Whose experience differs from the dominant story?
  3. What small change could we test, and how would we know it helped?

Related terms

Go to the evidence

Sources & attribution

Further reading: Hani Nabeel (2025), Culture Analytics: An evidence-based approach to company culture, First edition, Chapters 3 and 13; PDF pp. 38–39, 190–191 and 214. Edition & reading notes ↗

[1] MIT Sloan: Edgar Schein’s enduring management ideas ↗

The core idea is an original summary of the cited work. Application notes, examples and sketches are our interpretations, not quotations or reproductions of the authors’ figures. Publisher records may require access to read the full original work.

Compare all eight culture models → · Read the practical playbook →

Expanded on 20 September 2026 with practical diagnosis, application and limitations; original URL retained. Application guidance expanded on 26 September 2026. These practical suggestions and invented examples are editorial explanations; further-reading sources are not the original authors of the historical models unless expressly identified. Original overview review dates are retained.

Published 2026-09-20 · Reviewed 2026-09-22. Editorial approach

Keep exploring

A little more context.

Culture · diagnosis & change01

Deal and Kennedy’s culture model

Explore how the stakes of decisions and the speed of learning about their outcomes shape working habits.

Terrence E. Deal7 min
Culture · diagnosis & change02

Handy’s four organisational cultures

Understand where authority comes from: personal influence, defined roles, specialist expertise or professional autonomy.

Charles Handy7 min