The cost of not training managers is not the training budget a company keeps. It turns up in resignations, in people who stay but are not engaged, in unnecessary work and stress, and in performance that never arrives. None of it gets booked against the promotion that put someone in charge of a team without preparing them for it.
Three of them can be found in recruiting spend, payroll and the engagement survey. The worksheet near the end turns two of those into a one-year range for your own teams, from records you already hold.
How most managers end up in the role
In the UK, a 2023 survey by the Chartered Management Institute (CMI) and YouGov found that 82% of workers entering management positions have not had any proper management and leadership training. CMI calls them accidental managers.
An accidental manager takes on work that an individual role usually does not include: handing work off instead of doing it, giving feedback someone does not want to hear, setting expectations, running one-on-ones. They learn those tasks in front of the team they are supposed to be leading. The mistakes first-time managers make tend to show up in exactly those tasks.
Who gets promoted is a separate question from how they are prepared, and a company can get either one wrong without the other. Gallup’s 2014 research on manager selection found that companies fail to choose the candidate with the right talent for the job 82% of the time. The matching number is a coincidence: Gallup is counting selection decisions, while CMI is counting new managers without proper training.
The same Gallup research estimates that about one in 10 people possess the talent to manage, a figure about people in general rather than the managers already in the role. The Peter Principle, the idea that people keep being promoted until they reach a job they cannot do well, is the older name for this problem.
What is the cost of not training managers?
The cost of not training managers lands in four places: the people who leave, the people who stay but are not engaged, the unnecessary work and stress that workers say poorly trained people managers create, and the performance their teams never produce. In the US, Gallup’s 2019 analysis of replacement costs found that the cost of replacing an individual employee can range from one-half to two times the employee’s annual salary. A 2020 Gallup analysis found that employees who are not engaged cost their company the equivalent of 18% of their annual salary.
Neither ratio is specific to managers, so the worksheet below applies both to the salaries of the people in a manager’s team. The first two channels can be sized that way from your own data. The other two cannot.
People who leave
Replacement cost is spread across recruiting, cover for the empty role and the time a new hire needs to get up to speed, which is why a ratio to salary is the practical way to size it. Your HR system already holds the record you need: exits by team over the last 12 months. Decide before you pull the data whether internal moves count as exits and whether to include leavers you did not replace, and apply the same rules to every team in the group.
In the UK, CMI’s 2023 survey of managers and workers found that half of those who say their bosses are ineffective plan to quit within the next year. A plan to quit is not an exit, so keep that figure out of the turnover line. For your own teams, the nearest measure is an intent-to-stay question in your engagement survey, if it asks one.
People who stay but are not engaged
An employee who is not engaged still draws a full salary and still counts in the headcount, so this cost never appears as a line of its own. The worksheet puts a number on it with Gallup’s 18% ratio, applied to the people your own survey classes as not engaged.
Pull the latest engagement survey results by team. Surveys report a share of respondents rather than a headcount, so the cautious count is the number of respondents classed as not engaged; if you extend that share to people who did not respond, say so next to the result. If your survey hides results below a minimum group size, pull the figure at the lowest level it reports and note that the group is wider than the teams you chose.
Two broader Gallup findings give context but stay out of the calculation. In the US, Gallup’s State of the American Manager research found that managers account for at least 70% of the variance in employee engagement scores across business units, a share of the differences between units and not of any cost. Globally, Gallup’s State of the Global Workplace 2026 report estimates that low engagement cost the world economy approximately $10 trillion in lost productivity, or 9% of GDP, a cost of low engagement overall and not a figure for managers.
Globally, manager engagement has dropped by nine points since 2022, according to the same Gallup 2026 report. Burnout is a separate question, covered in what burns managers out.
Unnecessary work and stress
In SHRM’s 2020 survey on people managers, 84% of American workers say poorly trained people managers create a lot of unnecessary work and stress. In the same US survey, 57% of American workers say managers in their workplace could benefit from training on how to be a better people manager. Half of the American workers in that survey feel their own performance would improve if their direct supervisor received additional training in people management.
Because these figures record what workers say and feel, and none of them puts a currency value on stress or unnecessary work, the worksheet leaves this channel out. In your own data, the nearest signals are the workload and manager items in your engagement survey and the reasons people give in exit interviews.
The performance cost nobody books
The three channels above leave a trace somewhere. This one leaves none. It is the work a team would have done and did not.
Gallup’s Q12 meta-analysis, 11th edition, published in 2024, pooled 736 studies covering 183,806 teams and more than 3.3 million employees across 90 countries. Comparing teams in the top quartile on engagement with teams in the bottom quartile, it reports median differences of 23% in profitability, 18% in productivity measured by sales, 14% in productivity measured by production records and evaluations, 32% in quality defects and 78% in absenteeism. Top-quartile teams score higher on the good outcomes and record fewer of the bad ones.
Those are comparisons between groups of business units sorted by engagement score, not between managers sorted by skill, and Gallup writes that the report “does not directly address issues of causality”. The manager enters by inference: the engagement scores that sort those teams are the scores Gallup’s own earlier research attributes largely to the manager. The chain is two associations long, so it supports the direction of the cost and not its size.
Manager-specific evidence points the same way. In the UK, CIPD’s 2023 analysis of about 6,000 workers sorted employees by how they rated their own line manager on nine behaviors, including feedback, support and fair treatment. Among employees with bottom-quartile managers, 38% were prepared to volunteer for duties outside their job description; among those with top-quartile managers, 74% were. That is discretionary effort, and it is the line executives never see, because nobody is failing to do their job. Both halves of the finding come from the same person on the same questionnaire, so it records opinion rather than output.
For output itself, a 2012 NBER working paper on the value of bosses used daily productivity records from one US services company, covering 23,878 workers and 1,940 bosses between June 2006 and May 2010. Replacing a boss in the bottom 10% of boss quality with one in the top 10% raised a nine-person team’s total output by more than adding a tenth worker would. It is one firm doing one repetitive task, and it compares the bosses a company already employs rather than trained managers against untrained ones.
Gallup treats this cost as separate from the others. The 2026 global report says its estimate counts lost productivity and does “not add together the economic value of separate worker outcomes that are related to the engagement of workers, including safety, turnover, theft and healthcare costs”, which is why it calls the figure conservative.
No published figure converts any of this into money for your teams, so the worksheet leaves it out, as it does unnecessary work and stress. What you can watch instead already sits in your own reporting: the output measures those teams are held to, rework or defect rates, days lost to absence, and whether anyone volunteers for work beyond their role.
Where the cost of bad managers differs
Bad managers cost a business in the same four places as untrained ones: exits, people who stay but are not engaged, wasted work and stress, and performance that never arrives. In the UK, CMI’s 2023 survey of managers and workers found that among workers who rated their manager as ineffective, just 15% felt valued and appreciated, while most of those who rated their manager as effective did.
A bad manager and an untrained one are different problems, though. An untrained manager is missing management skills that training can supply, such as giving feedback, delegating or running a one-on-one, the ground covered in a practical playbook for upskilling managers. A bad manager may be missing the same skills and may also be the wrong pick for the job, the selection problem in Gallup’s research above. For the day-to-day behaviors that drive people out, see the manager mistakes that push good people to quit.
For an HR team, the distinction decides which group to size. Managers promoted without training can be listed from HR records; managers rated ineffective by their teams come from upward feedback or engagement results. The two lists may not match, and the surveys cited here do not measure the overlap: CMI’s UK survey asks workers whether their manager is effective, and SHRM’s US survey asks about poorly trained people managers. The worksheet below runs on either list. If you size both, run it once for each group and keep the results apart, because a team on both lists would otherwise be counted twice.
Run the numbers for your own organization
Start by choosing the group of managers you want to size, for example managers promoted without training. Every input below is for the teams those managers run.
| Input | What to enter | Where it comes from |
|---|---|---|
| A | Average annual salary of the employees in those teams | Payroll |
| B | Exits from those teams in the last 12 months | HR records |
| C | People still in those teams whom your engagement survey classes as not engaged | Engagement survey |
| Line | Low end | High end | Ratio source |
|---|---|---|---|
| Turnover | B × A × 0.5 | B × A × 2 | Gallup’s 2019 replacement cost analysis, US |
| Not engaged | C × A × 0.18 | C × A × 0.18 | Gallup’s 2020 analysis of employees who are not engaged |
| Total for one year | Sum of the low ends | Sum of the high ends |
The result is a one-year range, in your own currency, for the whole group: the turnover and not-engaged cost in those teams, and nothing for the other two channels. It is not the share of that cost caused by missing training, and the sources behind the ratios do not divide it that way.
Report both ends of the range, with the choices behind it: the group of managers, the date of the survey, and how you counted exits. The not-engaged line is the same at both ends because Gallup gives one ratio for it.
Input A is an average, so every exit is costed at the average salary for those teams rather than the leaver’s own. If most exits were from junior roles, the turnover line runs high; if most were from senior roles, it runs low.
If you do not have one of the inputs, leave that line out; do not fill it with an outside average. The total is then the one line you have.
Count each person once. Input C covers only people still in those teams, so someone who left is counted in B and not again in C.
Where these numbers need care
The worksheet builds a range from your own numbers because the published research offers no reliable per-manager dollar figure.
Gallup’s finding on managers and engagement variance compares US business units. It is a correlation, and it does not show that training managers changes engagement.
Gallup calls its US replacement range a conservative estimate.
Your engagement survey may not define “not engaged” the way Gallup does, so input C is an approximation.
SHRM surveyed American workers, CMI surveyed managers and workers in the UK, and Gallup’s replacement ratio comes from its US research, so readers in India or elsewhere should treat these figures as directional. The sources also span 2014 to 2026.
Gallup’s findings on how often companies fail to choose the candidate with the right talent, and on how few people have the talent to manage, point at who gets picked for the job. Training does not change who was picked.
The quartile differences in Gallup’s meta-analysis sort teams by engagement score, not by their manager, and Gallup makes no causal claim for them. CIPD’s UK figures sort employees by their own rating of their manager and measure what those employees say about themselves, so both sides of each comparison come from one questionnaire. The boss study covers a single US company and one job whose output a computer measured.