The main barriers are usually trust, clarity, accountability, and context
- Coaching loses momentum fastest when the goal is broad, borrowed, or impossible to measure.
- Trust matters more than technique because people only change honestly when the room feels safe.
- The best fix is often a smaller, clearer next step rather than a bigger framework.
- Coaching works better when it sits inside a wider learning and development strategy, not as a side project.
- Remote and AI-enabled coaching can improve access, but they do not replace human judgement in complex situations.
Why coaching breaks down in real life
Coaching is meant to be a non-directive process that helps people think, decide, and act more effectively. That sounds straightforward until the work meets reality: a busy leader wants advice, a sponsor wants faster results, and the client may not feel safe enough to be fully honest. Once those expectations collide, the session can drift into polite conversation instead of real development.
The CIPD’s guidance is useful here because it treats coaching as part of a broader learning and development strategy, not an isolated benefit. That matters in practice. When coaching is disconnected from the organisation’s real priorities, it can feel useful in the room but weak outside it. That is why the first question I ask is rarely, “Is coaching working?” It is usually, “What is this coaching actually supposed to change?”
When that question is clear, the next step is to look at the specific failure modes rather than blaming the whole process.
The most common obstacles and what they look like
Most breakdowns in coaching follow a small number of patterns. Once you can name the pattern, the response becomes much more obvious.
| Pattern | What it looks like | What I would do |
|---|---|---|
| Unclear goal | The sessions are thoughtful but they do not lead to any visible change. | Reframe the goal as one specific behaviour, outcome, or decision the client can control. |
| Low trust | The client answers safely, avoids hard topics, or performs insight without real honesty. | Reset boundaries, revisit confidentiality, and spend more time on the relationship before pushing for action. |
| Passive ownership | The client waits for the coach to solve the problem or provide the answer. | Shift from advice to questions, and make the client choose the next move. |
| Weak accountability | Good intentions are discussed, but nothing changes between sessions. | Agree one visible action, one deadline, and one check-in point. |
| Organisational mismatch | The sponsor wants one thing, the client wants another, and the manager is quietly pulling in a third direction. | Clarify the contract early and keep returning to it when the agenda starts to split. |
| Context overload | The client understands the issue but their workload, politics, or team dynamics keep crushing the follow-through. | Treat the environment as part of the problem, not as background noise. |
This table matters because not every stalled engagement needs a different style of coaching. Sometimes the intervention is not “better coaching” at all; it is clearer contracting, a safer relationship, or less friction in the working environment. Once the pattern is visible, the next question becomes: how do you diagnose it quickly without overcomplicating the process?
How I diagnose the real issue before changing the method
When coaching feels stuck, I work through four checks. They are simple, but they usually expose the real problem faster than a long theory about motivation.
- Contract check. Who asked for the coaching, who owns the outcome, and what exactly stays confidential? If that is unclear, trust and honesty usually suffer.
- Goal check. Is the goal behavioural, specific, and within the client’s control? “Be more strategic” is a vague aspiration; “lead one cross-functional meeting differently” is workable.
- Relationship check. Does the client feel safe enough to tell the truth? Psychological safety simply means they can speak without fearing ridicule, punishment, or retaliation.
- Environment check. Is the person trying to change inside a system that keeps rewarding the old behaviour? A client cannot coach their way out of every structural problem.
I also ask one practical question: if this coaching worked, what would be different in the diary, in the behaviour, and in the business outcome? That question cuts through a lot of fog. It forces the conversation out of abstract development language and into observable change. With that diagnosis in place, the intervention becomes much easier to design.
What actually moves coaching forward
When coaching is stuck, I rarely try to make the process more complicated. The fix is usually to make it smaller, clearer, and more honest.
- Re-contract the work. If the goal has drifted, say so. A clean reset is often better than pretending the original agreement still fits.
- Shrink the next step. One meaningful action over the next seven days is better than a grand plan nobody can execute.
- Use evidence, not impressions. Ask what the client will do differently, who will notice, and what proof will show that the change happened.
- Make accountability visible but light. Accountability should support ownership, not turn the coach into a supervisor.
- Work with emotion, not just behaviour. Avoidance, fear, frustration, and embarrassment often sit underneath the visible issue.
- Change the format when needed. Some conversations work better live, some by video, and some only after the client has had time to reflect alone.
One line I often use is this: progress does not need to be dramatic to be real. A leader who finally has the difficult conversation, delegates one task properly, or stops hiding behind busyness has already moved the work forward. That said, coaching is not always the right tool, and pretending otherwise wastes time.
When coaching is the wrong fit
Not every problem should be handled through coaching. In fact, one of the most useful things a good coach can do is recognise when the work needs something else.
- Performance management. If someone is failing to meet basic standards, management action may be required before coaching can help.
- Technical advice. If the person needs expertise, instruction, or a decision, reflection alone may not solve it.
- Serious wellbeing concerns. Coaching is not therapy or crisis support, and it should not pretend to be.
- No real commitment. If the client is there only because a sponsor insisted, progress is usually limited until ownership exists.
- Conflicting agendas. If the organisation wants compliance and the client wants exploration, the gap has to be named early.
The CIPD is explicit that organisations need clarity on when coaching is, and is not, the right intervention for poor performance. I think that is the right standard. Coaching should support growth, not replace management, leadership, or specialist support. Once that boundary is respected, organisations can build a much stronger coaching culture around it.
How UK organisations can support better coaching
In UK workplaces, coaching works best when it is treated as a serious development practice rather than a perk for high potentials. The organisations that get better results usually do a few things well.
- They align coaching with the wider learning strategy. Coaching should support leadership, capability, and business priorities, not float on its own.
- They define the sponsor role. A sponsor can support the process, but they should not quietly rewrite the goals midway through.
- They train internal managers carefully. Manager-as-coach can work, but only if people understand the limits of that role and where line management begins.
- They protect confidentiality. Clients will not explore difficult behaviour if they think every comment will be repeated upward.
- They measure the right things. Better measures are behaviour change, decisions made, conversations completed, or capability gains, not just attendance.
- They invest in supervision and reflection. Coaches need space to examine their own blind spots, especially in complex or political situations.
That last point is easy to ignore, but it matters. In the UK I often see coaching programmes lose quality because they scale faster than they mature. A thoughtful structure beats a large one every time. With that foundation in place, the final question is how 2026 changes the picture.
What changes in 2026 and what to keep in mind next
In 2026, coaching is more hybrid, more digital, and more influenced by AI tools than it was a few years ago. That creates real opportunities. Remote sessions can make coaching more accessible, easier to schedule, and cheaper to deliver across a wider population. AI can help with reflection prompts, session preparation, administrative follow-up, and personalised learning paths.But the limits are just as important. EMCC Global has been clear that AI can improve scalability and consistency, yet it still lacks empathy, emotional intelligence, and contextual understanding. That means it is useful for preparation and support, but not for replacing human judgement in conversations where trust, nuance, politics, or emotion matter.
- Use AI for notes, prompts, scheduling, and lightweight practice.
- Keep sensitive feedback, ambiguity, and behaviour change conversations human-led.
- Check privacy and data handling before using any digital tool in coaching.
- Do not assume a video call gives you the same non-verbal insight as an in-person session.
The simplest rule I use is this: if the issue is emotional, keep a human in the room; if it is structural, change the system; if it is administrative, let the tools do the heavy lifting. That keeps coaching practical, honest, and useful instead of merely busy.
