# How do you tell a coachable manager from a damaging one? **A coachable manager** is not one who agrees instantly or performs remorse for senior leaders. They can examine a specific event, recognize their contribution, choose a different behavior, and produce a better result in the next few weeks. A damaging pattern persists after clarity: decisions still stall, dissent remains costly, people route around authority, and the team grows quieter. ## What does a useful response to feedback sound like? A coachable manager first makes sure they understand the event. They may ask when the rollback decision was needed, what information was missing, and what the delay caused. They can disagree with part of the interpretation while still owning that the decision had no clear deadline or delegate. They then name an action within their control. For the next release, they will decide by the readiness review or assign approval to the staff engineer. In planning, they will let the risk owner finish, summarize the concern, and only then challenge it. The commitment is specific enough for everyone to recognize. A polished apology without a changed process is not coaching success. Neither is agreement in a private meeting followed by defensiveness with the team. The relevant evidence appears when the same pressure returns. Watch the next real decision, not the manager's fluency in discussing leadership. ## What does a damaging response look like? A damaging manager searches for the speaker instead of examining the behavior. They argue that the team is too sensitive, explain every consequence as another person's failure, or demand a list of names before considering change. These responses increase the risk of honesty and shift attention away from the work. They may accept a small procedural adjustment while preserving the harmful pattern. A new approval form does not help if decisions still wait for the manager and arrive too late. A round for concerns does not help if criticism follows anyone who uses it. Inspect impact rather than visible activity. Most importantly, the team becomes quieter. Engineers stop raising design risks, bring polished answers only after private coordination, and allow the manager to believe meetings are smoother. Silence after feedback can look like improvement to a distant leader. Often it means people learned that disclosure has a cost. ## How long should you allow for change? Match the period to the frequency of the behavior. If release approvals occur weekly, several weeks provide multiple tests. If a quarterly process is involved, create smaller comparable opportunities rather than waiting a quarter with no signal. State the review date when coaching begins. Some results can change immediately. Retaliation, public ridicule, and demands for names must stop at once. Skills such as planning a calm incident role may improve through practice. Do not use the language of development to tolerate conduct that continues to expose people. Look for direction as well as completion. A manager may miss one new deadline but communicate early, delegate, and repair the result. That can show learning. Repeated excuses, hidden misses, or changes that occur only under observation show that the pattern remains. ## What should the skip level observe? Observe whether the manager can receive disagreement without making the speaker pay. During planning, does the design concern reach a complete statement? Does the manager test the idea or attack the person? Afterward, check whether assignments, access, or feedback changed in ways that could punish the contribution. Follow decisions through the work. A rollback plan should have an owner, deadline, and answer. If information is incomplete, the manager should say what is missing and when the choice will be revisited. Endless reconsideration is not rigor when the useful decision window closes. Watch the unofficial organization. Are engineers still asking a staff engineer for the real answer? Is the manager included only after the team has solved the problem privately? Reduced routing around the manager is evidence that trust and usefulness are returning. Continued workarounds show that the title and working authority remain separated. ## How do you distinguish skill gaps from refusal? A skill gap often improves with structure, rehearsal, and feedback. A new manager may not know how to run an escalation, yet can use a checklist for impact, owner, action, and next update. Their intent is not enough, but their ability to learn is visible in later incidents. Refusal appears when the manager understands the requested behavior and repeatedly chooses a different one. They may say the team needs pressure, that interruption saves time, or that only they can approve every choice, even after seeing the delays. The barrier is no longer lack of instruction. Capacity can also be the issue. An overloaded manager may be willing and skilled but unable to meet the role's volume. Reduce scope or redistribute authority and test again. Keeping impossible scope while blaming either manager or team avoids the organizational decision. ## Which signals separate coaching from damage? | After feedback | Coachable pattern | Damaging pattern | | --- | --- | --- | | Account | Examines own contribution | Blames the team | | Plan | Names a visible behavior | Promises to do better | | Pressure | Uses the new behavior | Returns to old control | | Team | Raises risks earlier | Becomes quieter | No single row should decide a career. Together across several comparable events, the rows show whether authority is becoming more useful or more dangerous. Include context, but require a result. ## What coaching test should you run? - Name the behavior and work consequence. - Hear the manager's account. - Choose one replacement behavior. - Identify upcoming comparable events. - Observe at least one event directly. - Ask several people what changed. - Decide whether to continue, narrow, or end the role. Tell the manager that the test informs a role decision. Hidden stakes are unfair and produce weak learning. Also tell the team what behavior should change without making them responsible for scoring the manager. Their experience is evidence, but the skip level owns judgment. ## When should you stop coaching? Stop when the manager has received clear examples, suitable support, and fair opportunities, yet the impact remains. Decisions still stall, people still route around the manager, criticism still falls on the team, and the manager still avoids inspecting their own part. More coaching then extends damage rather than offering development. Stop sooner when retaliation or severe misconduct makes continued authority unsafe. Use formal processes as required, but take interim protection seriously. Employees should not have to prove resilience while leaders complete a predictable sequence. A role change is not a declaration that the person has no value. They may succeed with narrower scope, different responsibilities, or an individual role. Stay direct about why current authority cannot continue. Warmth means preserving dignity while protecting the team, not softening a decision until nobody knows what changed. ## How can senior leaders avoid being misled by upward behavior? Some managers are highly responsive to their own leaders and unreliable with their teams. They arrive prepared for a skip level review, accept feedback gracefully, and produce a convincing plan. The team then experiences the same delay and defensiveness once observation ends. Senior leaders should treat upward fluency as one data point, not as proof of day to day change. Design the test around outcomes the manager cannot create through presentation alone. Review whether approval arrived before the release window, whether the recorded design concern received a complete response, and whether incident ownership remained stable. Ask several employees about recent comparable events without telling them what answer you expect. Work traces are harder to polish than a coaching conversation. Watch for selective improvement. The manager may become careful with senior engineers while continuing to interrupt newer people. They may respond promptly to visible projects and neglect routine decisions. A coachable manager generalizes the lesson because they understand the management principle. A manager focused only on avoiding consequences changes the smallest visible surface. Consider whether the manager repairs harm openly. They need not disclose private coaching, but they can tell the team that a decision process failed, state the new route, and invite a specific check. Repair shows ownership and gives employees evidence that raising concerns can improve work. Silence leaves people wondering whether any apparent change will survive. Keep your own incentives in view. You may want the coaching to succeed because replacing a manager is disruptive or because you selected them. That preference can turn minor movement into a passing result. Write the success measures before the test, review them with another leader, and decide from the observed pattern rather than the comfort of avoiding a transition. Notice whether improvement survives an error. A coachable manager may fall back once, recognize it without prompting, repair the effect, and adapt the process. A damaging manager treats the miss as proof that the standard was unreasonable or that the team caused the problem. Recovery reveals whether learning has become internal or remains a performance for the skip level. Do not confuse calmness with safety. A manager can speak gently while withholding decisions, dismissing risk, or changing opportunities after disagreement. Another manager may be direct and still welcome challenge, explain choices, and protect the speaker. Evaluate the authority pattern and work consequence. Tone matters, but it cannot replace evidence about what people are able to say and do. ## What are common questions? ### Can a manager disagree and still be coachable? Yes. They can question the interpretation while examining their part and testing a specific change. ### Is an apology evidence of improvement? Only when later behavior and work impact change in comparable situations. ### How quickly should behavior change? Immediate harmful conduct must stop at once, while practiced management skills should show movement over the next few weeks. ### What is the strongest warning signal? Retaliation or a quieter team after feedback shows that speaking honestly has become more costly. ### Can reducing scope be part of coaching? Yes. Narrower scope can reveal whether overload caused the problem and can protect the team during the test. Related: [a team afraid of their manager](https://isilta.com/blog/when-a-team-is-afraid-of-their-manager/), [evidence for manager coaching](https://isilta.com/blog/what-evidence-shows-a-manager-needs-coaching/), [when a manager will not change](https://isilta.com/blog/what-to-do-when-a-manager-will-not-change/).