Immunity to Change Collective Team Maps: A Practical Guide

Immunity to Change collective team maps help a group examine why an improvement it wants keeps losing to its everyday behavior. The useful question is specific: what are people doing, what might that protect, and what would help them examine the prediction behind it? A team that reports risks late may need to investigate fear of blame. It may also need a clearer reporting threshold and a leader who actually responds differently.

The map becomes harmful when the facilitator announces what the team secretly believes. Members can share a behavior without sharing a motive. A finance colleague waiting for a credible range and a product colleague avoiding criticism need different responses, even if both delay an update.

This guide explains the method's boundaries, then follows an original fictional team through preparation, one experiment, a missed deadline and a corrected account. Its working records are original examples, not reproductions of the authors' worksheets. The result is a narrower agreement about risk reporting, with disagreement and a refusal to circulate personal material still intact.

Understand what belongs to Kegan and Lahey's method

Robert Kegan and Lisa Laskow Lahey developed Immunity to Change. The four-column structure connects an improvement commitment, contrary behavior, competing commitments and big assumptions. A possible protective purpose is investigated rather than treated as a character defect. Their Harvard interview explains the move toward inquiry and testing. That originator account describes a way of working; it does not establish that every stalled team has this explanation.

The 2009 book distinguishes collective work from individual work. Its HarvardX excerpt introduces that transition, while the full collective chapter requires separate reading. The selected testing passages emphasize manageable risk and learning about an assumption. This article does not reproduce the book's complete process, diagnostic sheets or cases, and it cannot replace suitable facilitator preparation.

The current originator provider, Minds at Work, distinguishes individual maps, collective patterns and shared experiments in its ITC with Teams program. It also recognizes that the wider organization may matter. This is provider guidance about its methodology, not an independent effectiveness study or evidence that a workshop changes a team's culture.

For a shared discussion, use provisional language: “One explanation we want to examine is...” Invite someone to say that it does not fit. Do not add together confidential individual maps and present the total as a discovery about the group. Our guide to examining individual assumptions addresses the different one-person inquiry.

Make six preparation decisions before inviting disclosure

First, name the business purpose. Choose an issue the team recognizes, such as giving decision-makers usable risk information before a delivery commitment becomes irreversible. “Make everyone less defensive” is a judgment with no agreed finish line. Identify who has authority over reporting rules and who can approve changes.

Second, agree the audience. Decide who may see the working discussion, individual preparation, experiment record and eventual sponsor account. These can be different audiences. A sponsor's purchase of coaching does not settle permission to receive private reflections. Explain retention and access arrangements in terms participants can understand before they write.

Third, offer genuine participation choices. Someone may speak, prepare a written contribution, keep a personal note private, request a different format or decline the particular exercise. Explain any real limits to confidentiality and organizational obligations. Do not promise that declining has no career consequences unless the people with that authority can establish it.

Fourth, include the leader's conduct. Ask what the leader does when information is incomplete. A request for more candor is weak if the same manager interrupts each early warning. The coach should be able to question that response without the leader using the exercise to identify “resistant” staff.

Fifth, distinguish observation from explanation. Record the actual message, date or decision before interpreting it. Missing data, contradictory priorities, unclear ownership and insufficient resources may require operational changes. Do not force a competing-commitment explanation onto a problem that the evidence does not support.

Sixth, reserve the work on both sides. A speaker needs preparation time, but recipients also need time to read, respond, correct and review. Agree the invitation, acceptance, working window and end point. A request for “quick feedback” does not establish another person's availability.

Keep serious concerns on their appropriate route. Acas guidance, in its UK workplace context, describes written options, records and informal or formal reporting. A collective map must not recast alleged retaliation, discrimination or misconduct as an employee's mistaken fear. Coaching is not a substitute for investigating a complaint or obtaining relevant support.

Start with one episode and preserve different explanations

In this fictional case, Tamsin coaches a three-person planning team: Bea, its leader; Ivo, product lead; and Lina, finance lead. The bounded work runs from 1 to 14 February 2061. Their existing engagement and ordinary project work sit outside the account below, rather than being assumed free. No customer release, safety decision or employment investigation is used as an experiment.

On 1 February Tamsin sends the purpose, proposed time allocations and information arrangements. On 2 February each member reads and replies within ten minutes. They accept the listed work and windows through 14 February. Lina explicitly declines sharing a personal map or allowing personal interpretations into a wider management report. Tamsin reads all three replies within her twenty-minute invitation-and-agreement block and honors that limit.

Each person separately prepares for the 3 February conversation. Bea spends fifteen minutes selecting and sending two risk-reporting episodes. Ivo and Lina each spend ten minutes selecting and sending information they choose to discuss. Tamsin spends ten minutes reading their selected contributions. They use the agreed internal email route for these messages and the later plan, records and summary replies. Private writing remains optional and uncollected; selected preparation is not permission to share everything someone considered.

During their forty-five-minute meeting, they choose an operational goal: make one accurate, bounded supplier-risk flag available by 6 February at 17:00. The earlier pattern is that potentially useful warnings arrive only after the weekly decision. Bea initially summarizes this as reluctance to look uncertain. Lina objects that the previous reporting rule did not say whether a range without a final quote was acceptable.

Bea: We all wait until we can sound certain.

Lina: That is not my account. I waited because two different thresholds were being used.

Ivo: I did worry you would call an early warning poor preparation.

Tamsin: Can both accounts stay in the record without requiring Lina to adopt Ivo's explanation?

Bea: Yes. I need to clarify the threshold, and my first response is part of what we examine.

Lina: I can review one operational flag. I am not agreeing that my private motives belong in a management summary.

They retain Ivo's prediction as his contribution: an incomplete but clearly labeled warning will cause Bea to dismiss the sender as unprepared. They do not call it the team's unanimous assumption. Lina's threshold concern remains a separate process question. Bea's wish to avoid unnecessary escalation is acknowledged without deciding that every challenge she raises is defensive.

Their collective working note links the agreed reporting improvement to the observed delay. Alongside it, Ivo names a possible commitment to protecting his reputation and the prediction about Bea's response. Bea records her concern about unnecessary escalation. Lina leaves the personal-interpretation space unfilled and records the threshold problem instead. The map is therefore partial and disputed. It is useful because those differences are visible, not because the facilitator has filled every space with one team-level explanation.

A useful shared account can therefore contain an agreed improvement and behavior alongside multiple explanations. An empty or disputed interpretation is better than a forced consensus. If only the leader recognizes the goal, return to the contract rather than polishing a map that participants do not own.

Before moving on, ask each member to check three distinctions: what actually happened, which explanation they recognize, and which action they are willing to try. A person can accept the proposed action while rejecting an explanation. Record that combination rather than treating the action as retrospective agreement with the map. Also ask what would improve simply through a clearer rule or better information. Comparing those possibilities helps prevent an emotional account from crowding out a practical fix. Neither discussion nor a single trial separates every cause; the purpose is to make the next decision more informed.

Turn the discussion into one bounded experiment

The team designs an operational trial rather than testing anyone's worth. Six design questions matter: what prediction is being examined, which situation is included, what action changes, what safeguards apply, what evidence will be collected, and when it will be reviewed? The following seven-field record adds ownership to those questions. It is this case's coordination record, not the published ITC instrument.

On 4 February Bea uses fifteen minutes to prepare and send the record and explicit requests. Ivo and Lina each use ten minutes on 4–5 February to read it and send acceptance of the fifteen-minute trial and their later ten-minute handling slots. Tamsin reads the plan and sends a short design comment within ten minutes. Bea uses five minutes on 5 February to read those two acceptances and Tamsin's comment. The experiment begins with actual receiving agreements.

Do not ask recipients secretly to perform a special reaction so the speaker can discover that a fear was irrational. Here Bea's agreed clarification-first response is a deliberate change being examined. That limits the conclusion: it can show what happens under these negotiated conditions, not how all managers or even Bea would respond without them.

For a team preparing a different trial, retain evidence that would support the concern as well as evidence that would challenge it. Specify what would cause a stop, whose approval is necessary and whether the proposed change is reversible. A modest coaching experiment cannot authorize a risky commercial decision.

Keep the failed response and the missed finish line

The fifteen-minute trial happens on 6 February with Bea, Ivo and Lina. Bea's first response is, “Why bring this without the number we need?” Ivo says this is the dismissal he had predicted. Tamsin is not present. Lina points to the agreed clarification-first approach. Within the same meeting Bea acknowledges that her opening was judgmental and asks which figure is missing. Ivo explains that he has only a single supplier estimate, without its uncertainty range.

Lina still requires a usable range before describing the flag as ready. That is a substantive evidence requirement, not automatically a failure of openness. They separate Bea's opening from the quality of the information. Ivo leaves with the action to correct the range, while Bea agrees that a preliminary flag may be recorded as incomplete rather than suppressed or treated as approval.

Ivo uses his ten-minute 6–7 February recording block to document the exchange, prepare the corrected range and send it at 10:00 on 7 February. The accurate flag is seventeen hours late against the original 6 February 17:00 target. A useful conversation does not erase that miss. Lina spends her ten-minute block reading the record, checking the range and returning her limited confirmation by noon. Bea spends ten minutes reading both returns, recording the planning decision and sending the selected operational record to Tamsin; no additional acknowledgment is assumed.

Tamsin reads the selected operational record for ten minutes on 8 February. At the thirty-minute 9 February review, all four examine the evidence, including Bea's actual words and the late flag. Ivo says he is still uneasy. Lina accepts the clearer threshold but declines a wider exercise about personal assumptions. They close this trial without authorizing another. A future experiment would need its own agreement and time.

Deliver a reviewed account without manufacturing consensus

On 10 February Tamsin spends twenty minutes drafting and sending an internal summary to all three. It wrongly describes “our shared fear of appearing incompetent” and suggests the exercise cleared the reporting obstacle. Each member spends ten minutes reading and replying on 11 February. Lina rejects the shared motive and wider circulation. Ivo rejects the claim that his concern has gone. Bea asks for the missed deadline and her opening response to remain visible.

Tamsin uses twenty minutes on 12 February to read all three corrections, revise the account and send this replacement:

I described one shared fear and overstated the result. That was inaccurate. Ivo described concern about being judged; Lina identified inconsistent reporting thresholds and has not authorized a personal interpretation. Bea's first response repeated the problem, then changed after challenge. The corrected flag arrived seventeen hours late. We have one usable operational record, not proof that the team's immunity has been overcome. This version replaces my earlier account for the three of you. It is not cleared for wider circulation. No further experiment is agreed.

On 13 February each member spends five minutes reading and sending a limited reply: this is an accurate internal account of the trial, without authorization for wider circulation or an endorsement of the method. Tamsin spends ten minutes on 14 February reading those replies and recording closure. She sends no further acknowledgment. Her initial summary remains in the correction history, rather than disappearing behind the replacement.

This is a practical test of the facilitator too. If a participant corrects an interpretation, the response should change the record. A sponsor's wish for a success story cannot override it. For ongoing individual support, peer coaching triads require their own information boundaries; they are not a route for extracting private material into a team report.

Account for the work, including receiving and repair

The following account covers the described preparation, trial coordination, meetings, selected records, report correction and closure. The allocations were accepted at the outset. It excludes ordinary project execution and prior coaching setup, without pricing them at zero. No external manager reads a summary because wider circulation remains withheld.

Total described work is 625 person-minutes against 715 allocated, leaving 90 within these allocations: 25 for Tamsin, 25 for Bea and 20 each for Ivo and Lina. The mapping and review involve all four; the trial involves three. Together those live activities total 345 person-minutes, already included in 625. The other 280 cover preparation and written handling. None of the difference authorizes new work or represents measured business savings.

Use Coachful for agreed follow-through and chosen records

Coachful is our #1 recommendation for organizing this coaching workflow. That is an editorial recommendation about the described preparation and follow-through, not a measured market ranking, ITC certification or guarantee of team change. Review the Coachful features against the actual audience and service you agreed.

A form or worksheet can structure selected preparation. Agree what is submitted and who should receive it before asking people to write. Private Coach review is an internal review facility; using it does not deliver feedback to a participant. In this case the coach must actually send the selected account and each recipient's reading and reply must occur through the agreed route.

Tamsin could use the client conversation to send individually agreed material and keep a manual record of the operational trial's state. The My Coaching overview can organize preparation and next actions, but a visible task does not prove the recipient agreed, understood or completed it. Do not infer a shared team workspace audience merely from putting content into a coaching platform.

The demonstration images show product organization, not this fictional team's data or a product trial performed for this article. The ICF Code of Ethics provides professional context for clear roles, agreements and information handling. Neither a platform setting nor an agreement removes the need to check the appropriate audience for the actual message.

Frequently asked questions

Must everyone agree on the same competing commitment?

No. Keep a proposed explanation attributed to those who recognize it. The team can agree to change a reporting procedure without agreeing that every person has the same protective purpose. Forced agreement makes the map less trustworthy.

Can people keep individual preparation private?

Yes, when that is the agreed arrangement. Explain collection and audience choices before writing. Someone may contribute an operational observation without submitting a personal map. Do not treat the coach's access to one contribution as permission to circulate it.

What if the experiment confirms a concern?

Retain that evidence. Examine the recipient's conduct and the operating conditions, then decide whether to repair, redesign or stop. In the example, Bea's opening mattered even though she subsequently changed it. The coach does not get to declare the fear disproved.

How long should collective mapping take?

There is no guaranteed completion time. Agree a bounded piece of work and include preparation, recipients' handling and review. A completed worksheet is not proof of changed behavior. Continue only when participants accept the purpose, conditions and additional effort.