27,329 Reviews Show Coaching Is the Weakest Skill Managers Have

If you have ever come out of a one to one feeling like nothing useful happened, there is data explaining why. Writing in Fast Company, the authors cite internal Stewart Leadership data covering twelve years of 360 degree assessments from 1,776 leaders and 27,329 evaluators, in which direct reports rated their leaders lowest on coaching. That is not a survey of opinions about coaching, it is the aggregated verdict of the people receiving it. At Metaintro, we cover findings like this because coaching is the single management behaviour most directly tied to whether the people below you develop or stall.

How bad is the problem, and how do we know?

Two independent bodies of evidence point the same way, which is what makes this more than a single survey result.

The first is a Gallup study of a nationally representative US sample of 2,729 managers and 12,710 individual contributors, each group assessing how they manage or are managed against a list of twenty managerial responsibilities. Both managers and individual contributors identified coaching related behaviours as the areas where managers need the most improvement. The agreement between the two groups is the important part. Managers are not being unfairly judged here, they are reaching the same conclusion about themselves.

The second is the 360 degree data already mentioned, where coaching ranked lowest across twelve years and nearly thirty thousand evaluators. A twelve year window matters because it rules out a passing effect from any particular period of disruption. This is a persistent structural weakness rather than a symptom of remote work, restructuring or any recent shock.

Put together, the picture is unusually consistent for workplace research. Everybody agrees coaching is the weakest thing managers do, including the managers.

Why do managers avoid it?

Because of what they think it is. Ask managers and employees what workplace coaching looks like and they will describe a formal scenario, such as a weekly calendar reminder that pops up mid task, a stiff conversation about performance during an annual review, or a manager stepping in to solve an employee's problem.

Each of those framings creates its own obstacle. A calendar reminder that interrupts real work makes coaching a competitor to the job rather than part of it. An annual review conversation loads a year of accumulated observation into one high stakes exchange. And a manager stepping in to solve the problem is not coaching at all, it is doing the work, which is the most common substitution and the one that feels most productive at the time.

The result is predictable. For managers, coaching feels like a time consuming, uncomfortable and high pressure responsibility, and that combination is a reliable recipe for postponement. Nobody schedules more of something that feels like all three, and the effect compounds where managers were promoted for delivery rather than for developing people, which is the pattern in why companies are cutting middle managers. That postponement has a price, visible in how skipping feedback costs you your best people covers what that avoidance produces downstream.

What actually changes the picture?

Broadening the definition. Coaching becomes a lot easier when you include all the coaching opportunities managers already have available to them, and coaching is not limited to annual reviews, weekly one to ones or formal conversations, since there are dozens of everyday workplace interactions ripe for growth and learning.

That is the idea behind in the moment coaching, which means weaving real time feedback into the fabric of the workday, turning small wins into confidence boosters and minor setbacks into lessons that help employees get better.

The shift is smaller than it sounds and more consequential than it looks. It relocates coaching from the calendar to the work itself. A thirty second observation immediately after a client call, delivered while the specifics are still fresh in both people's minds, does more than a scheduled conversation three weeks later attempting to reconstruct the same moment from memory. It also costs the manager almost nothing, which matters, because the honest reason coaching gets skipped is that it feels expensive.

There is a second benefit that the framing makes visible. In the moment coaching includes the wins, not just the corrections. Most managers, if they give feedback at all, give it when something went wrong, which teaches their teams that a conversation with the boss is bad news. Attaching feedback to small successes changes what the interaction signals. The cost of the alternative shows up in managers withholding praise, and what it takes to ask for it.

Why does this matter more now than it used to?

Because the skills employers are paying for have moved toward exactly the things coaching develops. PwC, analysing more than one billion job advertisements across 27 countries and territories, found AI increasing the emphasis on human skills such as judgement, creativity and leadership.

Judgement is the difficult one, because it is the least teachable through documentation. You cannot write a manual for it. It is built by making decisions, having them examined by someone more experienced, and adjusting. That is coaching, and there is no substitute mechanism.

The pressure is sharpest at entry level. PwC's analysis of 2.4 million entry level jobs in the United States found AI exposed entry level roles are seven times more likely to require traditionally senior level human intensive skills such as leadership, creativity or face to face interaction, with openings for those roles growing 35 percent since 2019 while other entry level roles shrank 10 percent. Junior employees are now expected to arrive with, or rapidly develop, capabilities that used to take years of routine work to build.

If the routine work that used to teach judgement is being automated, and the deliberate mechanism for teaching it is the thing managers are worst at, the gap has to close somewhere. That is the strategic case for treating coaching as a core competence rather than a soft extra, and it sits directly alongside the critical skills teams are losing to AI deskilling and why human judgment is becoming more valuable and the entry level squeeze created as AI raised the productivity bar.

What does the wider evidence say about manager development?

That the problem is structural rather than personal. Gallup describes a coaching habit as among the best predictors of employee engagement it has ever studied, and sets out what a routine coaching conversation should contain, namely goals, development, strengths and recognition of good work. Its own manager development guidance treats this as a trainable capability rather than a personality trait, which is the more useful framing for anyone currently bad at it.

The Bureau of Labor Statistics is a useful reality check on the incentives. Management roles are projected to keep growing, so the population of managers who were promoted for delivery and never taught to develop people is growing with them. The skill gap does not close on its own, and no organisation has ever accidentally coached its way out of it. Two professional bodies publish the standards worth measuring against if you want something more rigorous than an internal template, the Society for Human Resource Management on the employer side and the Chartered Institute of Personnel and Development on the professional standards side.

What does this mean for your career?

Five points, whether you manage people or report to someone who does.

If you manage, treat this as an opportunity rather than a criticism. Coaching being the lowest rated behaviour across 27,329 evaluators means competence in it is rare, and rare competence is what gets noticed. Most management advice tells you to improve at something everyone is already decent at. This is the opposite situation.

Start by auditing what you currently call coaching. If your honest answer is a recurring meeting and an annual review, you are using the definition the research identifies as the problem. Adding one observation to the end of a piece of real work, on the day it happens, is a bigger change than adding another meeting. Delivery is its own skill, covered in coaching your team with emotional intelligence, as are the practices that make team meetings work in an AI era covers the structural side.

Watch for the substitution. Stepping in to solve the problem is the most common failure, and it is invisible because it looks like helpfulness and produces a good outcome that day. The test is simple. Ask whether the person could do it alone next time. If not, you did the work rather than the coaching. The habit usually starts where accidental managers sink teams, promoted for output and never taught anything else.

If you are on the receiving end, do not wait to be coached well. The data says the average manager will not do this reliably, which makes asking a skill worth having. Extracting useful input from someone who is not offering it starts with getting career feedback from reluctant managers and with turning constructive criticism into career growth.

Finally, if you are aiming at a first management role, this is a specific and demonstrable differentiator. Being able to say how you develop people, with examples, addresses the exact weakness that both managers and their teams identify. What the role is becoming is the live question in whether managers are obsolete or more important than ever and in the manager readiness gap.

What does good coaching look like in practice?

Three characteristics separate coaching that works from a conversation that merely happened, and none of them requires a training course.

The first is proximity in time. The value of feedback decays quickly, because both people forget the specifics that make it useful. In the moment coaching exists precisely to close that gap, weaving real time feedback into the fabric of the workday. A correction offered while the work is still open can be acted on. The same correction offered at a quarterly review can only be absorbed as a judgement about the person.

The second is that it leaves the work with the employee. The failure mode named in the research is a manager stepping in to solve an employee's problem, and it is worth being honest that this is often the fastest route to a good result today. It is also how a manager ends up as the only person who can do the difficult parts of the job, which is a trap that gets tighter every time it is repeated. It is one of the hidden traps that sink strong teams, and the hardest to see from inside.

The third is that it treats success as material. Turning small wins into confidence boosters is listed alongside turning setbacks into lessons for a reason. A manager who only appears when something has gone wrong is training their team to avoid them, which removes the visibility coaching depends on, and quietly erodes the conditions described in building a speak up culture and psychological safety. It is the same mechanism running in reverse when managers withhold praise.

There is also a question of who does the talking. Coaching that consists of a manager explaining what they would have done is instruction, which has its place and is not the same thing. The version that builds judgement asks what the person was weighing, then tests the reasoning. That is slower in the moment and considerably faster over a year, and it is the practical mechanism behind developing critical thinking through deliberate practice.

Can AI do any of this for you?

Partly, and the boundary is worth drawing carefully because a lot of tools are being sold on the assumption that it can do more than it can.

What AI handles well is the recall problem. Coaching fails partly because managers cannot remember what happened three weeks ago in enough detail to be useful, and a system that keeps a record of work, decisions and outcomes genuinely helps with that. It is also reasonable at drafting, so a manager who freezes at how to phrase something difficult has somewhere to start.

What it does not do is the part that makes coaching work. The value is in a more experienced person examining a specific decision made under specific constraints by a specific colleague, then adjusting how that person thinks next time. That requires knowing the individual and carrying authority with them, neither of which is available to a tool. Workers are already turning to AI as a business coach for career advice, and there is a good reason for the counter argument that AI is not your only career coach in a tough market.

There is a risk worth naming for managers specifically. If coaching already feels time consuming and high pressure, a tool that appears to discharge the obligation is extremely attractive, and delegating it upward to software is a more comfortable version of the substitution the research already identifies. The team notices the difference immediately. Feedback that clearly did not require anyone to have been paying attention communicates the opposite of what coaching is meant to communicate.

The sensible split is to let tooling carry the administrative weight, the notes and the reminders, and to keep the judgement in the conversation. That is the same division the PwC data points to across the wider labour market, where roles that emphasise human judgement are growing at twice the rate and with 42 percent faster salary growth than those AI has made easier for non experts to perform. Coaching is squarely in the first category, which is also why picking a career coach applies the same test from the receiving side.

What should organisations take from this?

That a twelve year pattern is a design problem, not a personnel one. When the same weakness appears across 1,776 leaders, the explanation is not that those individuals were poorly chosen. It is that the system around them makes the behaviour hard.

Three things follow. First, if coaching is defined and measured as a scheduled event, that is what will be delivered and it will remain unsatisfying to everyone. Measuring the frequency of one to ones tells you nothing about whether coaching happened. Second, managers are rarely given time for this, since coaching competes directly with their own delivery targets. A manager assessed purely on output will rationally protect output. Third, most managers were never taught how, which is the gap executive coaching fills when it builds the leadership skills employers reward addresses.

The encouraging part is that the recommended fix costs almost nothing. In the moment coaching uses interactions that are already happening. It requires no new process, budget or meeting, only a different use of the thirty seconds after something goes well or badly. For an organisation trying to close a judgement gap created by automation, that is an unusually cheap intervention.

Related Articles

• The Attuned Manager and the Needs Your Team Actually Has
• How to Build a Team That Runs Itself
• Bottom Up Change at Work, a Playbook for ICs and Middle Managers
• The Career Development Gap Most Workers Blame on Themselves
• Your Best Employees Are Already Gone in Their Heads
• The AI Trust Gap and Why Managers and Workers Cannot Agree on Automation
• Burnout by Level and How to Beat It as an IC, Manager or Executive
• The Leadership Skill That Matters More Than Your MBA
• AI Skills for Managers Leading AI Augmented Teams
• What Makes Global Cross Cultural Teams Actually Work
• How ADHD Is Reshaping the Modern Workplace for Workers and Managers
• How to Build a Sabbatical Case Your Manager Will Actually Approve

People Also Asked

Q: Why are managers so bad at coaching specifically?

A: Because of how they define it. Managers and employees typically describe coaching as a formal scenario such as a weekly calendar reminder, a stiff annual review conversation, or a manager stepping in to solve a problem. Framed that way it feels time consuming, uncomfortable and high pressure, which makes it easy to postpone. The evidence for the weakness is consistent, with direct reports rating leaders lowest on coaching across twelve years of 360 degree assessments covering 1,776 leaders and 27,329 evaluators.

Q: What is in the moment coaching?

A: Feedback delivered inside the workday rather than in scheduled sessions. It means weaving real time feedback into the fabric of the workday, turning small wins into confidence boosters and minor setbacks into lessons that help employees get better. The underlying reframe is that coaching is not limited to annual reviews, weekly one to ones or formal conversations, because dozens of everyday interactions are opportunities for growth and learning.

Q: Does coaching matter more now than it used to?

A: Yes, because of what employers are paying for. PwC analysed more than one billion job advertisements across 27 countries and found AI increasing the emphasis on human skills such as judgement, creativity and leadership. Its analysis of 2.4 million US entry level jobs found AI exposed entry level roles seven times more likely to require traditionally senior level skills, with those openings growing 35 percent since 2019 while other entry level roles shrank 10 percent.

Ready to level up? Metaintro tracks the management and leadership skills employers are actually rewarding, so you can build the ones that move your career rather than the ones that sound good. Create your free Metaintro profile and see what the next role expects.