When is executive coaching the wrong intervention? It is the wrong intervention when the evidence points to a problem that coaching cannot change: unclear authority, an impossible workload, a broken incentive, a missing capability pathway, a team relationship that needs a direct decision, or a role that no longer fits the organisation. Coaching can support the work, but I would not use it to avoid naming the work.
Coaching is attractive because it is private, respectful and easier to approve than a structural change. That can make it a useful first step when a leader needs reflection and practice, provided the organisation knows how to choose an executive coach who will name the system problem rather than just build the person's resilience to it. It can also become a polite delay that keeps the person accountable for a system the organisation has not designed to work.
Is executive coaching the wrong intervention for your leadership problem?
It is if any of five tests comes back with a clear no that you cannot explain away. Run them in order and stop at the first no. The table is the whole method in one view, and the sections that follow explain how to run each test.
| Test | The question to answer | If the honest answer is no | What coaching can still do |
|---|---|---|---|
| Capability | Is this a skill or behaviour the leader can learn and practise in live work? | The gap may be too large for the role, so move to the appointment test | Support practice only if the gap is learnable and tied to a real assignment |
| Authority | Can the leader act on the expected behaviour within the decision rights they actually hold? | Redesign the role or decision rights first | Help the leader communicate and cope while the boundary is repaired |
| System | Do workload, incentives, routines and relationships support the behaviour? | Repair the operating model, because the system will overpower the intention | Help the leader handle the change, never stand in for it |
| Consequence | Has the organisation responded clearly when the problem appeared before? | Take a leadership action or a direct conversation first | Support the leader inside a response people can understand |
| Appointment | Is the role and person combination still right for the next stage? | Make the appointment decision with the evidence you already have | Support a fair transition, not a delay |
If all five come back yes, coaching is a strong candidate and I would pair it with a live assignment and a review point. If any one comes back no, coaching alone is the wrong intervention, and the response belongs to whoever owns that part of the problem, usually the chief executive or the CHRO.
Start with the problem, not the intervention
Describe the observed problem in work terms. Decisions are slow. A team does not trust the leader's priorities. A role cannot carry its remit. A successor is not developing. A leader avoids a conflict that the executive team needs to resolve. These descriptions are more useful than saying that someone lacks executive presence or needs to be more strategic.
Then ask who experiences the problem, what consequence it creates and what evidence would show improvement. The answers help separate a person gap from a design gap. A leader can be part of the problem and still be working inside a system that makes the desired behaviour difficult.
Use more than one source of evidence. A complaint from a senior stakeholder may describe a real impact, but it may also reflect a conflict over authority or style. A team survey may show frustration without revealing the decision that created it. Combine the account of the leader, the experience of the team, peer observations and the operating record before choosing an intervention.
This is not a case for endless diagnosis. It is a case for a better first decision. A short evidence review can prevent months of coaching that produces insight without a change in the conditions that created the problem.
| Evidence pattern | Likely intervention | Why coaching alone is insufficient |
|---|---|---|
| The leader lacks a specific skill and can practise it | Coaching plus live practice | Reflection needs an assignment and feedback |
| Authority is unclear between leaders | Role or decision-rights redesign | Insight cannot grant missing authority |
| Targets reward the wrong behaviour | Incentive or operating-model repair | The consequence system will overpower intention |
| The team avoids a necessary conflict | Guided decision or leadership action | A private conversation cannot replace a public call |
| The role has outgrown its design | Role redesign or appointment decision | Personal development cannot shrink an impossible remit |
What does the evidence say about when coaching works?
The research supports coaching on average and warns against treating it as a universal answer. A 2023 meta-analysis by Janis Cannon-Bowers and colleagues in Frontiers in Psychology concluded that workplace coaching is effective in achieving positive organisational outcomes, consistent with earlier meta-analyses. The same paper describes the field as still relatively immature, with limited controlled studies and little shared understanding of exactly what works. That is a sensible reason to match the intervention to the evidence instead of reaching for coaching by default.
The CIPD's 2025 factsheet on coaching and mentoring is more direct. It says coaching is one of a range of interventions and is not the solution for all developmental needs, that it must only be used when it is clearly the best way of helping an individual learn and develop, and that organisations should assess an individual's readiness and motivation before coaching starts. It also warns that coaching is sometimes used without a great deal of thought or reflection. Among the situations it lists as a good fit are helping competent technical experts develop better interpersonal skills, developing a more strategic perspective after promotion to a senior role, and dealing with the impact of change on an individual's role.
Two Harvard Business Review pieces point at the edges. Ben Dattner and Elizabeth Wood write that coaching sometimes reveals that the person being coached is in the wrong role, and that bosses too often leap to the conclusion that they must fire the person. A second piece by Matt Brubaker and Chris Mitchell sets out four signs that an executive isn't ready for coaching. Between them they describe the two cases my five tests are designed to catch: a problem that sits in the appointment and a problem that sits in the person's readiness.
I use these sources to mark out the boundary, not to claim a rate. I know of no study that tells you the share of leadership problems that coaching alone will solve, and I would distrust anyone who quoted one.
The five intervention tests
Coaching is one response to a leadership problem among several. These five tests help you decide whether it fits, or whether the problem sits in authority, the system, the consequences or the appointment itself.
Choose the right response
- Capability test: Does the leader know what to do and need practice, feedback or confidence?
- Authority test: Can the leader act on the expected behaviour within the decision rights they actually hold?
- System test: Do incentives, workload, routines and relationships support the behaviour?
- Consequence test: Has the organisation responded clearly when the problem appeared before?
- Appointment test: Is the role and person combination still right for the next stage?
Capability test
Coaching is often useful when the leader can learn a specific behaviour and apply it in live work. The goal might be better challenge, clearer delegation, more disciplined prioritisation or stronger stakeholder listening. Define the behaviour, the situation where it matters and the evidence a team should notice.
A capability intervention needs practice. If the leader leaves each session with insight but no decision, conversation or assignment to test, the work may feel productive without changing the system. Coaching should connect to a real commitment and a review point.
Authority test
If a leader is expected to own an outcome but another person can overturn every decision, the issue is authority. Coaching may help the leader communicate better, but it cannot resolve a boundary the executive team refuses to define.
Map the decision. Who owns it, who contributes, who can veto it and who carries the consequence? If the map is confusing, repair the decision rights before judging the leader's confidence or presence.
System test
Look at the conditions around the person. Is the workload possible? Does the team have the capability needed? Do measures reward the stated priority? Are meetings and hand-offs creating clarity or noise? A leader may be using a sensible behaviour that the system punishes because speed, certainty or local targets receive the visible reward. This is the same diagnostic question I use when I fix uneven leadership quality across an organisation: the gap is rarely one person, it is the conditions around several of them.
System repair might involve a role boundary, a routine, an incentive, a staffing decision or a clearer escalation path. Coaching can help the leader handle the change, but it should not be the organisation's substitute for making the change.
Consequence test
Ask what happened the last time the problem was raised. Was the leader given clear feedback? Did peers support the decision? Did the organisation keep rewarding, without saying so, the behaviour it says it wants to stop? If the consequence is ambiguous, another coaching referral may teach the person that nothing substantive will happen.
A consequence can be developmental, structural or formal. The important point is that the organisation responds in a way people can understand. Coaching belongs inside that response only when it matches the evidence and the leader has a real chance to act.
Appointment test
Sometimes the question is not how to help a leader succeed in the role. It is whether the role and person still fit the next stage. The role may have expanded beyond the leader's current capability, or the organisation may need a different kind of judgement. Coaching can support a transition, but it should not be used to postpone an appointment decision that the evidence has already made visible.
Make the appointment test humane and direct. Explain the work the next stage requires, the evidence that has been gathered and the options available. A leader deserves a real development opportunity when the gap is learnable, and the organisation deserves a clear decision when the role needs a different fit.
Directness is kinder than an indefinite intervention that leaves everyone guessing.
The intervention should make the next conversation easier because everyone can see what the organisation is asking the leader to change and what the organisation must change around them.
- Describe the work problem: State the behaviour, decision or outcome that is creating difficulty.
- Collect multiple views: Use evidence from the leader, team, peers and operating record rather than a single complaint.
- Run the five tests: Check capability, authority, system, consequence and appointment before selecting an intervention.
- Choose a matched response: Select coaching, practice, role redesign, a structured intervention, system repair or an appointment decision.
- Set a visible review: Agree what should change, who will notice it and when the intervention will be reconsidered.
When coaching should be part of a wider intervention
Coaching and structural action are not opposites. A leader may need coaching while the CEO clarifies decision rights, the people team changes the development pathway or the executive team repairs a relationship. The coaching goal should be connected to the wider action so the leader can practise in the environment that is being redesigned.
- A live decision that gives the leader a safe place to practise
- Feedback from people who experience the behaviour
- A clear boundary that prevents the system from cancelling the change
- A sponsor who can remove structural obstacles
- A review point where the intervention can be changed
This makes coaching more rigorous, not less humane. The leader knows what the work requires, the organisation owns its part and the coach can focus on reflection, challenge and practice rather than carrying an unspoken performance process.
A wider intervention also improves the quality of the coaching conversation. The leader can bring a real decision, a real relationship and a real constraint into the room. The coach can challenge the leader's choices while the sponsor works on the conditions around those choices. Both parts remain visible, so neither becomes an excuse for the other.
Keep confidentiality clear. Coaching can create a private space for reflection, but the organisation still needs a fair way to know whether the agreed work is changing. Agree in advance what themes may be shared, what evidence the sponsor will review and what remains between coach and client. Ambiguity here can turn trust into suspicion.
Be especially careful when the leader is carrying a role transition. A person may be learning a larger remit while the organisation is still deciding whether the remit is sustainable. Coaching can help them explore the identity shift, but the CEO must still make the design decision. Reflection should inform the decision, not replace it.
How to talk about the decision
Use plain language with the leader. Explain what has been observed, what consequence it creates and what response is being chosen. Do not hide a role or appointment question inside a development label. People can work with difficult information when the organisation is direct about the decision and fair about the evidence.
Also explain what support is available and what remains the leader's responsibility. Clarity protects the relationship. It prevents coaching from becoming either a secret warning or a promise that insight alone will change the conditions around the work.
Use the same language with the team where appropriate. If a role boundary is changing, say what decisions will move. If a relationship needs repair, say who will convene the conversation. If coaching is one part of the response, explain the work the organisation is doing alongside it. This stops colleagues from treating coaching as a hidden explanation for every difficult interaction.
A direct conversation can feel slower than a referral, but it creates a better contract. The leader knows the standard, the sponsor knows the responsibility and the coach knows the purpose. That clarity is what allows development to be both supportive and accountable.
Review the intervention and the person
At the review point, ask whether the original problem changed, whether the evidence was strong enough and whether the intervention matched the problem. If the leader improved but the system did not, change the system. If the system improved but the behaviour did not, revisit capability or appointment. If the evidence remains mixed, define the next test rather than repeating the same referral.
Look for changes in the work that others can describe. Are decisions clearer? Are disagreements raised earlier? Does the leader make more space for the team to own its part? Has a routine become less dependent on personal rescue? These signs show whether the development is travelling beyond the private coaching relationship.
Also notice what has not changed. A leader may become more articulate about the problem while the same decision remains stuck. A team may report that conversations feel better while the operating outcome remains fragile. Do not punish that evidence. Use it to select the next intervention with greater precision.
Intervention choice is a leadership responsibility. It asks the organisation to be clear about the problem, fair about the evidence and willing to change the system when the system is part of the cause. Coaching is valuable when it meets those conditions, and it is the wrong answer when it is being used to avoid them. Choosing it for that reason is one of the more common causes behind why leadership development programmes fail: the intervention was chosen to feel supportive, not because it fit the evidence.
How does this compare with other ways of investing in leadership?
The same logic applies one level up. If several leaders show the same pattern, the cause is probably the conditions around them and not any one of them, which is the case I make in executive coaching or leadership systems: which should you invest in and in how to fix uneven leadership quality across an organisation. If the question is which kind of outside help fits, the distinctions in executive coaching versus leadership consulting and executive coaching versus leadership coaching are worth reading before you sign. If coaching passes the five tests, choosing well still matters, and how to choose an executive coach covers it.
To choose the right intervention, begin with the Leadership Capability Diagnostic, review the Leadership Capability Architecture, or explore Leadership Assessments.
Select the intervention the evidence supports
A coaching referral should follow a diagnosis, not replace one. The related pieces on capability-based executive coaching, the leadership capability architect role and leadership succession planning help separate capability, authority, role and succession questions before money and attention are committed. If coaching is still the leading option, read what corporate coaching is and how it works first, so that the referral follows the diagnosis.
Three questions for the sponsor before a coaching referral is approved
Before a coaching referral is signed, put three questions to the sponsor. What is the problem in work terms, such as slow decisions or an unclear remit? What happened the last time it was raised? What would have to change for it to improve, and can coaching change that directly? A sponsor who cannot answer the second is likely to be arranging the polite delay this article warns about.
Use more than one source for the answers. How to measure leadership capability across an organisation helps gather evidence beyond a single complaint, what causes leadership development programmes to fail shows what follows when an intervention is chosen to feel supportive, and the team collaboration and effectiveness framework is worth reading if the problem turns out to be a team relationship.
My own work starts from the same place. The CapabilityAI page and the Work With Me page set out how I separate capability, authority and system questions, and the Architecture Accelerator is for the cases where the answer turns out to be the structure and not the person.
Coaching can change how a leader operates inside a system. It cannot change the system. The best intervention leaves the organisation with more clarity: people know what good work looks like, who owns the decision and what will happen when the standard is not met.