The coaching shift now centres on performance, change and proof
- Coaching is being pulled into leadership, performance and change work, not kept separate as a luxury add-on.
- The coaching market is still expanding, but buyers now want clearer outcomes and better evidence.
- UK organisations are blending internal coaching, external specialists and AI support, depending on the problem.
- Coaching works best when the issue is behaviour, confidence or decision-making, not a broken process or a missing skill.
- The strongest programmes define success early and measure more than satisfaction at the end.
Why the coaching conversation feels different now
I read this as a reset, not a fad. At a major UK people-profession event in 2026, leadership, skills, wellbeing and AI kept appearing in the same conversations, which says a lot about where leadership development is heading. The message was simple: leaders are expected to perform while navigating change that rarely pauses.
The ICF’s 2025 Global Coaching Study points in the same direction, with record growth, rising revenue and strong confidence in the profession. For me, the practical takeaway is that coaching is no longer being sold only as personal development. It is being positioned as part of how organisations adapt, retain talent and improve execution.
- AI is changing the shape of work, so leaders need more than advice; they need a way to think, test and adjust.
- Change fatigue is making direct, human support more valuable.
- Boards want evidence, which pushes coaching toward clearer outcomes.
That shift changes the buying decision as much as the coaching itself, and it leads straight into the question of who should actually deliver the support.
How UK organisations are blending coaching models
In practice, I see organisations using three models more often now: internal coaching, external executive coaching and AI-supported coaching. The right choice depends on the stakes, the level of trust needed and whether the issue sits inside the manager’s control.
Internal coaching is usually the most scalable option. It works well for first-time managers, routine performance conversations and everyday capability building. External coaching is still the better fit for senior leaders, sensitive transitions and situations where the coach needs to be independent from the organisation. AI support is useful, but only as a layer around the coaching relationship, not a replacement for it.
| Model | Best for | Main strength | Main limitation |
|---|---|---|---|
| Internal coaching | Day-to-day development, new managers, regular feedback habits | Fast, familiar and closely tied to the real work | Can blur the line between support and accountability |
| External executive coaching | Senior leaders, transitions, confidential or high-stakes issues | Neutrality, challenge and fresh perspective | Cost and the need for a clear sponsor brief |
| AI-supported coaching | Reflection, rehearsal, action prompts and session follow-up | Consistency and speed | Weak on judgement, trust and context if used too broadly |
The CIPD has also been clear that organisations often use coaching internally through line managers, while external coaches are common in senior roles. I think that mix is sensible, but only if the intervention matches the problem. Once the delivery model is clear, the next question is where coaching works best and where it is the wrong fix.
Where coaching creates value and where it falls short
Coaching is strongest when the issue is behaviour, confidence or decision-making. A new manager who needs to delegate better, a senior leader navigating a merger, or a director who needs to lead more calmly through uncertainty can all benefit. I also see strong value when coaching supports change management, because people rarely resist change for only one reason; they resist the speed, ambiguity or loss of control attached to it.- Good fit: transition into a bigger role, stakeholder influence, feedback habits, executive presence, resilience under pressure.
- Poor fit: unclear role expectations, a skills gap that needs training, a process failure, conduct issues or urgent performance correction.
If the root problem is structural, coaching alone will disappoint. In those cases, it should sit beside job redesign, training or direct management action rather than replace them. That distinction matters, because it keeps coaching honest.
What strong coaching programmes do differently
The best programmes I see are specific from day one. They start with one business issue, not a vague wish for someone to be “better at leadership”. The conversation should name the behaviour to change, the context in which it matters and the measure that will show progress.
- They set a clear contract. Everyone should know the goal, the timeline and what success looks like.
- They protect confidentiality without making the work invisible. Sponsors need enough visibility to support progress, but not so much that the leader stops being open.
- They use a baseline. A short 360 review, manager observation, pulse feedback or delivery metric gives the coaching something to move against.
- They create action between sessions. Coaching without practice becomes a pleasant conversation instead of a behaviour change process.
- They use AI carefully. It can help with prompts, reflection and follow-up, but not with judgement, trust or confidential decision-making.
One simple framework still earns its place here: GROW, meaning goal, reality, options and way forward. I like it because it keeps the conversation structured without making it robotic. The point is not to sound clever; it is to keep the leader moving. From there, the most common question is not how to coach, but which mistakes waste the investment.
The mistakes that quietly waste budget
Most coaching failures are not dramatic. They are slow, expensive and easy to miss because the conversations feel productive while the behaviour barely shifts.
- Using coaching to soften a difficult management conversation instead of facing it directly.
- Buying coaching before the real problem has been defined.
- Matching by availability rather than by issue, style and experience.
- Expecting coaching to fix broken processes, unclear roles or poor line management.
- Measuring satisfaction only, then assuming that means impact.
- Letting AI blur confidentiality or dilute the quality of the human relationship.
The pattern I see most often is simple: organisations buy coaching before they have named the problem accurately. Once that happens, the programme feels busy but changes very little. That leads naturally to what I would watch next if coaching is part of your leadership strategy.
What I would watch next if coaching is part of your leadership strategy
AI is moving from experimentation to everyday infrastructure, and leadership development will have to keep up. Recent UK research on AI adoption suggests that confidence is strongest where leaders already have the skills to redesign work and weakest where they must rethink jobs, workforce planning and reskilling. That matters for coaching because the next demand will be less about motivation and more about helping leaders work through ambiguity, redesign decisions and trade-offs.
- More internal coaching capability inside organisations, especially for first-line managers.
- Stronger evidence standards for outcomes and behaviour change.
- More coaching attached to change programmes, not isolated talent pools.
- Clearer rules for confidentiality, data use and AI support.
The organisations that get the most from coaching will be the ones that treat it as a management capability, not a ceremonial perk. That is the real story behind the current market: better questions, tighter goals and a more honest match between the problem and the intervention.
