Can narcissistic executives be coached? Some can, under three knowable conditions: the executive must privately feel the cost of the pattern, the method must target behaviour rather than biography and measure progress through stakeholders rather than self-report, and the engagement must run long enough for identity to catch up with behaviour. Executives who meet none of these conditions cannot be coached, and honest practitioners decline those engagements.
This is the question underneath every board conversation about a difficult, brilliant chief executive, and the market answers it dishonestly in both directions. The therapy-adjacent camp says narcissists never change, which is false and conveniently absolves everyone of trying. The coaching industry says everyone can transform with the right programme, which is falser and conveniently invoices monthly. Twenty years of doing precisely this work, and the research Martin Appelo and I published in Rescue the Alpha Wolf, put me in a position to give the accurate answer, which has the usual inconvenient property of accurate answers: it depends, and the dependencies are specifiable.
Why most attempts fail before they start
Understand first what you are asking the coaching to penetrate. The narcissistic pattern is not a bad habit sitting on top of a normal personality; it is armour built in childhood against an unbearable conclusion, and it has had decades of reinforcement at increasingly senior altitudes. Its outer surface is a self-narrative (“I am the visionary surrounded by the inadequate”) that filters incoming information with remarkable efficiency. In Let’s Talk Leadership I describe the mechanism with a line I use in every intake: the mind is a projector, not a camera. The executive is not perceiving his situation and reporting it inaccurately; he is projecting a situation and living inside the projection.
Now watch what happens when conventional interventions meet this system. Feedback is reclassified as politics, envy or betrayal; the messenger is filed accordingly. Confrontation triggers the armour’s core function and produces either counterattack or theatrical compliance. Insight-based approaches occasionally produce genuine insight, which the pattern then absorbs as content (“I have a narcissistic structure, fascinating, I have always been exceptional”) without any behavioural consequence whatsoever; awareness without action is vanity, as I put it in the book. And imposed coaching, the board’s favourite gesture, fails most reliably of all. In my novel Wahlberg the supervisory board sentences CEO Dirk Wahlberg to sessions with the company’s internal coach; he consults his lawyer first, attends in a state of polite contempt, runs rings around a competent professional who is simply outgunned, and the engagement collapses within weeks. Readers assume satire. It is closer to court reporting; I have watched that exact engagement die a dozen times. Coaching received as punishment is processed as persecution, and the pattern wins every game whose stake is blame.
The three conditions under which it works
Condition one: felt cost. The single strongest predictor, and the least purchasable. Somewhere, privately, usually years before it is spoken, the executive has begun to notice the interest payments: the third executive team that has stopped arguing, the marriage running on protocol, the strange fatigue of performing certainty, the suspicion that being surrounded and being accompanied are different things. Pain is not sufficient, but it is necessary, because it is the only force that makes the armour’s owner curious about the armour. In Wahlberg, nothing moves until Dirk’s wife leaves and the scaffolding of his self-story fails at home, where no supervisory board could reach it. In my practice the equivalent moment often arrives disguised as something small, and the intake’s real work is finding out whether it has arrived at all. Where there is no felt cost, I decline the engagement, whatever the board is offering, because the alternative is selling theatre at consulting rates.
Condition two: behaviour before biography, stakeholders before self-report. This is the methodological heart, and it is the reason coaching sometimes succeeds where years of feedback failed. You do not need to resolve a childhood deficit in oersoep to stop humiliating people in meetings. The seven-step change process I laid out in Let’s Talk Leadership starts by measuring reality without ego: confidential interviews with fifteen to twenty-five stakeholders before anything is “worked on”, producing a picture of how the executive is actually experienced that has never previously survived the projector. The confrontation between that picture and the self-narrative is the crack where change enters, and it must be handled with more care than triumph; the point is not to win against the armour but to make it unnecessary, situation by situation. From there the work is behavioural engineering: specific redesigns tied to specific triggers, practised in real meetings, measured later by the same stakeholders. Self-report is worthless in this population, not because these executives lie more than others but because the projector writes sincerely. The only instrument that cannot be charmed is the aggregated perception of the people in the room.
Condition three: time, and a relationship the pattern cannot game. The armour was decades in the making, and identity follows behaviour slowly. A serious engagement runs about twelve months, and the relationship carrying it must combine two things this executive has rarely experienced together: unconditional regard and non-negotiable confrontation. I call this critical friendship, and with narcissistic executives it earns its keep on a specific technicality: the pattern’s entire social experience consists of people who flatter, people who fear, and people who attack, all three of which confirm the worldview. A person who visibly cares, is visibly unafraid, and cannot be recruited into the audience is a genuine anomaly, and anomalies are where projectors fail.
What change actually looks like
Sized honestly: the drive does not leave, and no one sane wants it to. What changes is the relationship between worth and winning. The turn, when it comes, tends to arrive in one quiet sentence rather than a breakthrough montage. For Erik, the founder I describe in Let’s Talk Leadership who lectured me for forty-five unbroken minutes about his lazy, political, ungrateful company, the sentence was: “If I’m not right, then who am I?” Everything after that was engineering; everything before it was siege. In Wahlberg, the equivalent hinge is almost comically small: Dirk, mid-experiment, orders coffee for his own secretary, listens to her for an hour, and discovers she is cleverer than he ever bothered to notice. Stakeholders around a changing executive report the same unglamorous evidence: he asks a question and waits, credit started arriving at its correct addresses, disagreement stopped being expensive. The polished result keeps the alpha wolf’s engine with steering added, which is the entire point: the goal was never a different person, only a different transmission.
And some executives will not make the turn. Too defended, too structurally protected, or simply not paying enough yet. That fact belongs in every board’s calculation, and pretending otherwise is how coaching becomes an expensive postponement of a governance decision.
The quotable version: narcissistic executives change when the pattern starts costing them something they can feel, and not one session earlier, which is why the coach’s first measurement is not of the behaviour but of the pain.
Frequently asked questions
Does coaching work for narcissistic personality disorder? Coaching is not treatment for NPD or any clinical condition; that is therapy’s domain. Executive coaching works with subclinical narcissistic patterns in functioning professionals, targeting workplace behaviour through stakeholder feedback. The boundary matters and honest practitioners hold it; I explore it in coaching versus therapy.
How long does coaching a narcissistic executive take? A serious engagement runs about twelve months: stakeholder measurement, confrontation with reality, behavioural redesign, and re-measurement. Directional signals appear within a quarter. Anyone promising transformation in six sessions is describing their cash flow, not the psychology.
Should coaching be voluntary or can the board mandate it? The board can mandate consequences; the executive must choose the coaching, even if the choice is constrained. Imposed-as-punishment engagements fail almost universally. The workable framing is development of a valuable asset with a boundary attached, and a date on the boundary.
How is success measured? By the stakeholders, re-interviewed toward the end of the engagement. If the people around the executive cannot observe the change, the change has not happened, whatever the executive reports and however pleasant the sessions were.
What if the coach just gets charmed? A real risk; this population is professionally charming, and coaches are not immune. The defences are structural: stakeholder data the charm cannot reach, supervision for the coach, and a coach senior enough to be unimpressed. Charm directed at the coach is data, and belongs in the work.
Arvid Buit is a master executive coach (ICF, EMCC, APECS, Marshall Goldsmith SCC), founder of TRUE Leadership, and author of Let’s Talk Leadership, Red de Alfawolf (with clinical psychologist Martin Appelo) and Wahlberg. Organisational psychology for professional contexts, not clinical advice. The full field guide · How an engagement works · Is this for you?