Immunity to Change for Conflict Avoidance
Nour
Founder, Coachful
Using Immunity to Change for conflict avoidance means examining a specific contradiction: an executive wants to raise a concern, yet repeatedly removes the point, delays the discussion or repairs the work without speaking to the person involved.
The aim is not to make someone enjoy disagreement. It is to understand what the avoidance may protect, investigate a prediction and prepare an appropriate conversation. A real threat, missing skill or unclear responsibility needs attention in its own right.
This guide follows a complete fictional executive case. The first conversation produces irritation and a useful correction. A later scheduling refusal does not become proof of relationship collapse, and a return to avoidance remains in the final review.
Check whether reflective mapping fits the problem
Robert Kegan and Lisa Laskow Lahey’s Immunity to Change approach examines protective commitments that can work against a genuinely desired improvement. In their 2010 Harvard interview, the authors describe how behavior that frustrates one goal can serve another self-protective purpose.
This is the developers’ account of the model, not an independent finding that every delayed conversation has the same cause. The immunity metaphor is not a diagnosis, a brain scan or proof of an executive’s developmental stage.
Separate three situations before mapping:
- A person who does not know how to give specific feedback may need instruction and rehearsal.
- A person who lacks relevant authority or evidence may need clarification before raising the issue.
- A person who has workable conditions and repeatedly edits away a concern they want to raise may find the map useful.
A credible risk of retaliation or a formal grievance is not a convenient experiment in courage. Discuss suitable organizational or qualified support. Do not use a reflective worksheet to persuade the client that a dangerous setting is safe.
Agree the client-owned purpose. “Raise the evidence concern with my peer before we finalize the brief” identifies a work behavior. “Become fearless” does not specify what should change or acknowledge that some concern may remain reasonable.
The Immunity to Change process guide explains the broader model. This article focuses on the difficult interpersonal moment, the partner’s actual response and the cooperation that follows.
Distinguish worry, protection and the assumption
The four-column map separates an improvement goal, behavior working against it, a competing protective commitment and a big assumption that makes the protection feel necessary. Kegan and Lahey distinguish the worry itself from the commitment to prevent the feared outcome.
For example, “I worry my colleague will be angry” names a concern. “I am committed to never being seen as a difficult partner” identifies a possible protective position. “If I disagree once, they will withdraw every future offer of help” states an assumption that can be examined.
These are working descriptions chosen by the client. A coach should not declare a hidden motive from a delayed email. The person may reject the proposed wording or discover that several concerns matter.
A value such as cooperation is not the enemy. The practical question is whether avoiding every explicit disagreement is the only way to protect that value. Cooperation can include accurate information, a clear refusal, a difficult exchange and limits on what either person can provide.
Do not turn the map into a moral ranking. The client is not automatically mature when they challenge someone or deficient when they pause. Evaluate the actual situation and the action they choose.
Worked case: a peer disagrees and still cooperates
All people, dates, documents, figures and responses in this case are fictional. Damon is a research director. Keira is a partnership director. They jointly prepare an internal board brief and both have authority to review its wording. Sima is Damon’s sole coach.
They agree three sixty-minute sessions on 4, 11 and 25 November. Damon wants to raise specific evidence concerns with Keira rather than silently changing a footnote or complaining privately afterward.
Keira sent him the draft and asked for review on 2 November. Damon already has the relevant response summary. He sees an overstatement but waits, hoping the issue will disappear in editing. Their work deadline for a finalized brief is 9 November.
1. Complete the map in Damon’s language
| Map element | Damon’s completed working entry |
|---|---|
| Improvement goal | State an evidence concern explicitly with Keira before we finalize a joint brief |
| Doing or not doing | I remove the direct request from my email, make quiet edits and wait until the final review |
| Worry | She may become angry and decide working with me is more trouble than it is worth |
| Competing commitment | I am also committed to never being seen as a difficult partner |
| Big assumption | If I explicitly disagree with her conclusion, she will withdraw all further cooperation with me |
The worry row is shown separately for clarity; it is not a fifth column in the authors’ model. Damon says the protective wording fits. He wants collaboration, but he notices that protecting his image as uncomplicated can make the shared document less accurate.
Sima asks about actual previous threats. Damon recalls irritated exchanges, but identifies no threat to end collaboration or punish him for using the ordinary joint-review process. This does not guarantee the discussion will be easy. It gives them a specific, routine context in which to consider a modest test.
2. Verify the issue and prepare a bounded test
On 5 November, Damon checks the response summary against the source record he is authorized to use. Twelve partner organizations were invited to comment. Four replied: two support the proposed model, one supports it only if training is provided and one opposes it. Eight have not responded.
The draft’s conclusion, “Our partners support the proposed model,” goes beyond those responses. Damon is not testing Keira by inventing an error; he is raising a genuine issue in their actual work.
He proposes a private twenty-minute conversation on 6 November, and Keira accepts through their existing work email thread. They will examine the paragraph before finalizing the brief. The board is not present, and this is not a performance hearing.
The test record separates the prediction from the desired work result:
| Part | Completed plan |
|---|---|
| Action | State the response limitation and request revised wording privately |
| Prediction | Keira will stop cooperating after the disagreement |
| Immediate observations | Her actual response, whether the paragraph can be discussed and what is agreed |
| Observation period | 6–13 November, including ordinary joint-work interactions that actually arise |
| Limits | Do not provoke, manufacture requests or infer permanent relationship change from one reply |
| Work result sought | A brief that accurately describes the four responses and the eight nonresponses |
In the authors’ published SMART test guidance, safe and modest scope, near-term action and a research stance matter. The test gathers information about the assumption; it is not a demand to perform a dramatic confrontation. The specific plan above is an original application.
3. Say the concern and hear the criticism
On 6 November, Damon raises the point directly:
Damon: We have four replies from twelve invitations, with different positions. I do not think we can describe that as our partners supporting the model. Can we state the actual responses and the limit?
Keira: You have had this since Monday. Why are you bringing it up now? We have very little time to change the brief.
Damon: You are right that I delayed. I saw the issue and avoided saying it clearly. I am sorry for adding pressure. The response limit still needs to be visible. I have a draft paragraph we can examine.
Damon’s apology concerns his actual delay. It does not withdraw the evidence concern or require Keira to stop being annoyed before they can continue. Keira’s criticism is not automatically retaliation.
They examine the four replies. Keira asks Damon to make the conditional response clear rather than simply group it with opposition. Damon agrees; their correction needs to represent the respondents accurately, not merely weaken the conclusion.
4. Complete the shared output
They agree this replacement paragraph:
“Four of the twelve invited partner organizations responded. Two supported the proposed model, one supported it subject to training being provided, and one opposed it. The eight nonresponses do not establish support or opposition. These comments inform the proposal but do not demonstrate a common partner position.”
Damon saves the agreed wording in the joint draft after the conversation. Keira opens it and confirms through the work email thread on 6 November that it reflects their agreement. The final brief on 9 November retains that paragraph.
The completed work output is a corrected account of consultation, not a decision that the model should proceed. Any later board decision remains outside this case. Damon has also not established that all conflict is harmless or that an apology removed Keira’s frustration.
Interpret the next interaction without mind-reading
On 9 November, Damon asks Keira to review an additional source note that afternoon. She declines that timing. His first private interpretation is, “She has stopped helping because I challenged her.”
He checks rather than treats the interpretation as a fact. Keira says she is attending a previously arranged offsite and offers 10 November at 11:00 for a short review. Damon accepts. She completes that review at the agreed time.
This is one declined timing request followed by an accepted and completed alternative. It is not a refusal of all cooperation. It also does not prove that the earlier disagreement had no influence on her feelings.
On 12 November, an actual further work need arises: Damon asks for a source-summary review and a full annex review by 13 November. Keira replies on 13 November that she can review the short source summary, but not the whole annex in that window.
Damon accepts the limited review and confirms that the annex is not covered. Keira sends her summary comments that day. The record shows bounded help and a real limit, not unlimited availability. The annex remains unreviewed at the end of the test period.
| Observation | What it supports | What remains unknown |
|---|---|---|
| Keira was irritated on 6 November | The conversation involved discomfort and criticism | Her full private judgment or future relationship choices |
| She co-approved corrected wording | They completed a substantive disagreement and shared output | Every future disagreement will be workable |
| 9 November timing declined; 10 November review completed | A scheduling refusal was not total withdrawal | Every reason behind the original refusal |
| 13 November limited review completed; annex declined | Cooperation continued with a genuine boundary | Whether or when the full annex will be reviewed |
The response counts also stay separate: twelve invited organizations, four replies, two supportive, one conditional, one opposed and eight nonresponses. None is a measure of Damon’s conflict competence. The colleague’s three work interactions are a different set of observations.
Keep a return to avoidance in the review
At the 11 November coaching session, Damon can already see that the first conversation did not produce immediate total withdrawal. He still finds the criticism difficult. Sima does not ask him to replace the assumption with “everyone welcomes feedback.” They agree to review the remaining ordinary interactions and the next joint-editing occasion.
On 18 November, Damon has another concern about wording in a new joint draft. Instead of naming it to Keira, he rewrites the paragraph quietly. On 19 November, she asks why the conclusion has changed and says the unannounced edit made her review harder.
Damon acknowledges that he avoided the conversation again. Through their existing work email thread, he supplies the old and proposed versions, identifies the actual point of disagreement and asks Keira to examine it. He does not call the silent edit a successful test.
On 20 November, Keira agrees to one part of the proposed change and rejects another. They save the accepted version and leave the disputed sentence out pending the responsible project owner’s review. Damon accepts that the owner’s decision is still needed. He does not report that the whole difference has been settled.
At the final 25 November session, Damon’s revised assumption is conditional: “A specific private disagreement with this peer can produce irritation without ending cooperation. My delay and silent edits can also damage the work. I still need to raise the actual point and hear her limits.”
This statement preserves both the learning and the remaining practice. The case ends with a corrected first brief, bounded later help, a return to avoidance and one later sentence still awaiting an authorized decision. It is not a permanent change claim.
Prepare the conversation as well as the experiment
An assumption test still needs a usable interpersonal action. Use this original preparation sequence:
- Check the issue. Identify the specific information, behavior or decision you want to discuss. Correct any mistake in your own account first.
- Choose an appropriate setting. Agree the conversation with the relevant person. A private routine review differs from a public challenge or formal process.
- State the concern directly. Name the point you usually edit away. Avoid personal accusations or a long introduction that never reaches the issue.
- Invite their account. Ask what they see differently and listen for a factual correction, constraint or legitimate reciprocal concern.
- Own your contribution. Acknowledge an actual delay, overstatement or abrupt delivery without erasing the original work issue.
- Close the work agreement. Record what is accepted, who will do it and what remains undecided. Check the actual saved output or follow-up.
A reusable sentence can be: “The point I need to discuss is [specific issue]. I see [evidence or effect]. What am I missing, and can we agree [bounded next step]?” It should sound like the client and fit the relationship.
If an initial delivery is sharp, a repair can be simple: “I said that abruptly. I want to discuss the issue accurately.” Do not use a formula to make the other person accept the client’s view. They retain their own judgment and right to decline a request.
The safe behavioral test guide develops the experiment-design task. Conversation preparation and evidence review complement it; they are not proof that a chosen interaction is risk-free.
Use a completed review rather than a courage score
Keep the map, the interpersonal action and the work result distinct. Damon’s final review can be recorded in a short original template:
| Field | Completed case entry |
|---|---|
| Assumption examined | Explicit disagreement would end all cooperation with Keira |
| Actual difference voiced | The brief overstated four mixed replies as partner support |
| Response retained | Keira criticized Damon’s delay and examined the wording |
| Completed output | Agreed paragraph saved, checked and retained in 9 November brief |
| Later cooperation | 10 November review completed; 13 November source-summary comments completed |
| Actual limits | 9 November timing declined; full annex review declined |
| Setback | 18 November silent edit, acknowledged and reopened explicitly on 19 November |
| Open work | Later disputed sentence awaits responsible owner; annex review not established |
| Proportionate learning | This private disagreement did not lead to total withdrawal; irritation and boundaries still occurred |
A useful review allows the assumption to be supported, weakened or narrowed. If the concern proves realistic, change the plan and seek appropriate support rather than insisting that a better attitude will fix the setting.
Do not ask an employer to rate the executive’s vulnerability. A sponsor-facing update can describe agreed work behavior and completed outputs without disclosing the private map, fears or peer interpretations.
Notice problems the map cannot resolve alone
The client still removes the actual disagreement. Review whether the planned action occurred. A friendly conversation that avoids the point supplies little evidence about the predicted response to disagreement.
The client interprets every refusal as punishment. Separate the exact request, timing and response. Ask for clarification where appropriate, while accepting that motives may remain unknown.
The peer identifies a real problem in the client’s conduct. Address it. Damon’s delay and quiet rewriting matter independently of his fear.
The work responsibilities conflict. Bring the relevant owner into the work decision rather than turning the conflict into a contest of confidence. Neither client nor coach should promise another person’s acceptance.
The inquiry becomes distressing or exceeds scope. Pause and discuss appropriate support. Completing a map is not a condition of receiving coaching or proving readiness to lead.
The current ICF Code addresses engagement agreements, confidentiality, appropriate boundaries and competence. A client may choose what to share, while the coach remains responsible for staying within the actual professional role.
Keep the map and agreed follow-through in Coachful
Coachful is our #1 recommendation for organizing an original private worksheet, preparation and review of the chosen work action. It does not identify hidden commitments automatically or turn a task check into evidence that a difficult conversation happened.
Sima assigns Damon’s worksheet through Clients > Damon > Worksheets & resources > Assign. The shared version uses Damon’s chosen wording. Confidential research responses and the authoritative board brief stay in the organization’s approved systems.
They agree one brief clarification through their individual Messages conversation between 11 and 20 November, submitted by 12:00, Europe/London, with a reply by 17:00 that working day. On 13 November at 10:00, Damon asks whether Keira’s refusal to review the full annex means the test failed. Sima replies at 14:00 that the actual accepted scope, completed work and declined portion all belong in the record. Damon remains responsible for the work arrangement.
The final coaching review remains 25 November. Ordinary tasks can hold preparation and the review point, but they do not send the work correction or approve a document. The coach has no authority over Keira’s workload.
Unshared journal and takeaway pages remain private to the client. Where enabled, Share with coach lets the client deliberately share an individual page with the coach. Keep sensitive personal words unshared. Damon chooses what to submit. Sima checks automatic recap settings, Approve before sending, session-type overrides and held-content review before sharing a summary. A sponsor receives no automatic access to the map.
Explore Coachful’s features for the actual resource and client-plan workflow. Keep the client’s learning alongside the next choice without converting the private inquiry into a workplace assessment.
Frequently asked questions
Does avoiding a conflict always reveal a hidden commitment?
No. Missing skills, uncertain facts, unclear authority and credible consequences can explain hesitation. Explore those conditions before assuming the map fits.
Does a good test need a friendly response?
No. The purpose is learning about the assumption. Keira’s irritation, criticism, continued cooperation and limits all matter; friendliness is not the outcome measure.
Should I share the whole map with my colleague or sponsor?
Not automatically. Agree an appropriate explanation of the work behavior and any reporting. Private fears and personal reflection do not need to become team material.
What if the colleague refuses a request?
Record which request was refused and what happened next. One scheduling or workload limit does not establish total withdrawal, but do not invent a reassuring explanation when the reason remains unknown.
What if I avoid the next conversation again?
Review the actual event, acknowledge any work consequence and decide what to repair. Damon’s later quiet edit stays in the case rather than being erased to preserve a success story.
