AI Coaching Works Best When Managers Lead the Way

AI Coaching Can Extend What Great Managers Do, but It Can't Stand in for Them

AI in Leadership Development Is Still Early

AI in leadership development is still new territory for most HR teams, but that's starting to change. Twenty-eight percent have AI-based learning experiences, such as AI coaching, chatbots, or personalized recommendations, in their leadership development programs. Another 44% are considering it. AI coaching specifically is rarer still: only 4% of HR professionals say their organizations use it frequently, and 81% rarely or never do.

Leaders are ready to try it. Nearly 7 in 10 are at least somewhat enthusiastic about using AI in their own development, and 32% are very enthusiastic.

Early results are mixed. Among organizations already using AI in development, 37% say they're seeing strong results, a sign that access to AI alone doesn't make the difference. What matters more is how AI fits into the rest of a leader's development, including the people who help them grow.

That question will soon be in front of many HR teams. Many organizations considering AI will evaluate AI coaches over the next year or two, often pitched as a way to deliver coaching at scale without developing managers. But that shortcut leaves out what makes coaching work: the relationship between a leader and their manager.

Manager Coaching Is the Most Powerful Coaching Leaders Get

Managers are uniquely positioned to coach. They know a team member’s work, see how they handle day-to-day challenges, and influence their assignments and career path. Leaders are also accountable to their managers, which gives a manager’s coaching a weight that advice from other sources rarely carries. Acting on it is part of the job. When managers coach well, they can give feedback in the moment and create a safe space to try new things, work through difficult conversations, and learn from mistakes. Not every manager does this well, but the leaders whose managers do stand out in the data.

The data reflects that. Compared with leaders who get no coaching, leaders whose managers are effective coaches are:

• 2.3X more likely to say they consistently bring out the best in their team (43% vs. 19%).
• 3.2X more likely to feel very prepared to handle high-stakes conversations (41% vs. 13%).
• 5.6X more likely to rate their development programs as high quality (45% vs. 8%).

They are also able to apply 15% more of what they learn back on the job.

Effective coaching and trust go hand in hand. Among leaders who strongly agree their manager is an effective coach, 86% strongly trust that manager to do what is right. Effective coaching also shapes engagement. Leaders who never wanted the role but have an effective manager coach are more likely to be fully engaged than leaders who wanted it but lack one. Leaders' ratings of their manager's coaching also rise modestly the longer they've worked together, a reminder that coaching relationships build over time.

Few Leaders Get Coaching That Works

The problem is that most leaders aren't getting effective coaching. Only 1 in 4 leaders say their manager is an effective coach. At the levels where most leaders sit, the number is even lower: only 17% of individual contributors and 25% of frontline managers strongly agree their manager coaches them effectively.

Even leaders with a great manager coach need more than their one-on-ones. Building a skill takes practice, and leaders may not want every rough first attempt to happen in front of the person who evaluates their performance. They need somewhere to rehearse a hard conversation, test an idea, or think out loud: a space that’s private, available day to day, and calibrated to their organization’s leadership model and culture. That frees their time with their manager for the conversations that matter most.

Only 1 in 4 leaders say their manager is an effective coach.

AI Coaching Gives Leaders a Place to Practice

An AI coach can be that space, without taking a manager’s place.AI coaching delivers real benefits on its own. Compared with leaders who get no coaching, leaders who receive only AI coaching are:

• 1.9X more likely to say they consistently bring out the best in their team (36% vs. 19%).
• 2.8X more likely to feel very prepared to handle high-stakes conversations (36% vs. 13%).
• 4.9X more likely to rate their development programs as high quality (39% vs. 8%).

But AI coaching delivers the strongest results when leaders also get effective manager coaching. Leaders with manager and AI coaching are 8X more likely to rate their development as high quality than leaders with no coaching (64% vs. 8%), and well ahead of leaders with manager coaching alone (45%). They’re also the most likely to say they consistently bring out the best in their team: nearly 6 in 10, compared with about 4 in 10 of those with an effective manager coach alone.

Leaders Are Finding AI Coaching on Their Own

Leaders are moving faster than their organizations. Eleven percent say they already use AI to personalize their development, such as through AI coaching, compared with just 4% of HR teams that use AI coaching frequently. That shouldn’t be surprising. Leaders who aren’t getting effective coaching still need guidance, so they pull from whatever they can: their own experience, colleagues they admire, a quick search, and increasingly, general-purpose AI tools.

For HR, that changes the question. It’s no longer only whether to offer AI coaching. If leaders are already turning to AI for development help, what’s guiding that experience?

In many cases, it’s a tool that wasn’t built for leadership development. General-purpose generative AI predicts the next likely word. It doesn't inherently know what helps people grow lasting leadership skills. It pulls from anywhere, can give two leaders different advice for the same situation, and has no connection to an organization's culture, values, or leadership expectations.

For HR, the question is no longer whether to offer AI coaching. If leaders are already turning to AI for development help, then HR needs to know: what’s guiding that experience?

The stakes rise quickly when the question involves people. A leader who asks a general-purpose tool how to put someone on a performance improvement plan, or how to respond to a disability accommodation request, may get an answer that sounds confident but conflicts with company policy or employment law. In situations like these, bad advice can create legal risk for the organization, not just a harder conversation for the leader.

That makes the decision more urgent for HR teams still weighing AI coaching. Organizations that provide AI coaching built for leadership development, grounded in proven methods, and connected to the coaching leaders get from their managers can shape what leaders learn from AI, instead of leaving it to whatever tool they find.

External Coaching Works Best When It’s Targeted

Across all leaders, external coaching on its own shows the smallest benefit of the three. But that average blends very different situations, and the value of external coaching depends on who receives it. High-potential leaders who understand their future career path are the most likely to rate their external coaching as high quality. Senior leaders see more value from it, too. C-level leaders are 2.8X more likely than individual contributors to rate their external coaching as high quality, and C-level leaders with quality external coaching apply about 12% more of what they learn on the job. At the top of an organization, where leaders have fewer people above them and face decisions with broad business consequences, an outside perspective can fill a gap that’s hard to fill internally.

Even then, external coaching delivers the most alongside strong manager coaching, not in place of it. Leaders who get all three forms of coaching are the most likely of any group to rate their development as high quality (86%).

Every form of coaching has a role, but manager coaching matters most. It’s also the one organizations can strengthen the most.

AI coaching delivers the most value as part of a broader development approach, one anchored by managers who coach well and grounded in proven leadership methods. Organizations don't have to choose between investing in managers and investing in AI. The data suggests they work best together.

1. Develop managers as coaches and build a coaching culture

With so few leaders getting effective coaching from their manager, the biggest opportunity is helping more managers coach well. Coaching is a skill, and it doesn’t come with the management title. Build it from a leader’s first management role, and make regular development conversations part of how managers lead, not something saved for performance reviews.

That work is easier in a coaching culture, where coaching is an expectation across the organization rather than something individual managers choose to do. When it’s the norm, managers can ask probing questions or give honest feedback without it landing as criticism, and leaders can ask for help without worrying it signals weakness. Leaders at every level ask questions before offering answers. People seek out feedback instead of waiting for it. Teams take time to reflect on what worked and what didn’t. And development gets as much attention as performance.

Leaders with manager and AI coaching are 8X more likely to rate their development as high quality than leaders with no coaching

A coaching culture also makes AI coaching more valuable. When leaders are used to reflecting, asking for feedback, and practicing, an AI coach becomes one more place to do that work, not a substitute for conversations that aren’t happening.

2. Layer in AI coaching to give leaders more room to practice

AI coaching can give leaders more ways to build their skills between conversations:

• In the moment: guidance right before a difficult conversation, or a discussion guide when something unexpected comes up.
• For practice: a low-risk place to rehearse a skill they're building with their manager or in training, like delivering tough feedback.
• For reinforcement: prompts that keep what leaders learned in development programs active in their day-to-day work.

That benefits managers, too. When leaders have already worked through everyday questions on their own, one-on-ones can focus on bigger conversations about growth, feedback, and career direction.

Because an AI coach responds to the situation a leader is facing right now, it can help with a barrier leaders often name: development that isn’t personalized to their needs ranks among the top five reasons they don’t make time for it.

Not every AI coach is built for this. When evaluating AI leadership coaching, ensure it is:

• Grounded in proven leadership science. Guidance should come from validated leadership models and research on what changes behavior, not from whatever a general-purpose tool turns up.
• Aligned to your organization. The coaching should use the same leadership competencies, language, and expectations leaders hear in your programs and from their managers, so everyone works from one standard of good leadership.
• Built for practice, not just answers. Leaders should be able to rehearse real conversations, get feedback, and try again, not just receive advice.
• Connected to the rest of development. The best AI coaching links to the courses, assessments, and development plans leaders already have, so it picks up where formal learning leaves off.
• Designed with clear guardrails. Leaders need to trust that what they share stays private, so they can be candid. The tool should also recognize when a situation calls for a manager, HR, or another person.
• Insightful for HR. An aggregated view of leaders’ capabilities can help HR spot common skill gaps and shape development priorities.

HR teams that pair AI coaching with formal development see it in their results: those that use both instructor-led training and AI coaching extensively are 1.3X more likely to report improved business performance than those using instructor-led training alone.

3. Keep AI coaching connected to the manager

AI coaching works best as an extension of the manager's coaching, not a separate track. When a leader and their manager agree on what to work on, AI coaching can carry that work into the days between conversations, while the manager stays involved in the leader's progress.

For example, a leader working with their manager on giving tougher feedback might rehearse with an AI coach before a real conversation, then debrief with their manager afterward about what went well and what to try next time.

 A development plan gives that work a shared direction. Among leaders who receive AI coaching, those with an up-to-date development plan apply 76% of what they learn on the job, compared with 47% for those without one. Organizations seeing strong results from AI in development show the same pattern: they’re 1.9X more likely to have leaders who regularly review their development plans with their managers.

The question for most organizations isn't whether to use AI coaching. It's how to make it count. AI coaching built on proven leadership methods can help more leaders get coaching more often. But it delivers the most value when a manager on the other end knows the leader, cares about their growth, and helps them turn practice into progress.

4. Target external coaching where it adds the most value

Focus external coaching where an outside perspective makes the biggest difference: senior leaders facing complex business decisions, and high-potential leaders being prepared for those roles. It’s especially valuable during transitions, such as a leader’s first months in a senior role, when expectations change quickly, and they’re still building the relationships and context they’ll rely on. It also supports ongoing development for senior leaders already in their roles, and it can help high-potentials get ready faster by focusing on the specific capabilities their next role will demand.

Organizations that rate their strategy for identifying and developing leaders as highly effective are 5.6X more likely to use external coaching extensively than those who rate their strategy lower. That suggests external coaching delivers the most as part of a well-planned leadership development approach, not as a standalone fix. Even at the top, it works best alongside the coaching leaders get from their own managers.