The short answer
When it comes to keeping people from quitting, the thing to look at is not
- more drinking parties
- a company sports day
- getting the young employees to be friends
- saying "come talk to me anytime"
- telling everyone to "stay motivated"
None of those is forbidden. They're just not the core of the problem.
The core is this:
Catch the warning signs in workload, autonomy, support, role clarity and change before someone burns out, and then fix the structure.
Put another way, preventing turnover is
not about talking someone out of leaving. It's about shrinking the things that make people want to leave.
Drinking parties and sports days can help a little with relationships if they're done well. But what if the workload is far too heavy, nobody knows who's responsible for what, speaking up changes nothing, the one capable person always ends up with the gray-area tasks, and the people who say the right thing get disliked for it? Add more events on top of that, and you may simply be piling unpaid after-hours tasks onto someone who's already on fire.
This isn't a cold-hearted argument. It's the opposite.
If you really care about your people, you have to design the job so they don't burn out.
How the WHO sees burnout
In ICD-11, the WHO's international classification of diseases, burnout is not listed as a disease. It's described as an occupational phenomenon. What matters is that the WHO doesn't treat burnout as laziness, weakness or a lack of grit.
The WHO describes burnout as
a syndrome conceptualized as resulting from chronic workplace stress that has not been successfully managed
and says it has three main features:
- Feelings of energy depletion or exhaustion
- Increased mental distance from one's job, or feelings of negativity or cynicism about it
- Reduced professional efficacy
This part is important.
Burnout isn't a story about someone's personality going bad. It isn't about someone suddenly becoming unserious, either.
It's the state you end up in when workplace stress goes unmanaged and your energy, engagement and sense of efficacy get worn down.
So if someone is showing signs of burnout and all you say is
- "Try to be more positive"
- "Be more aware"
- "You're a leader, act like one"
- "Pick up the phone, please"
- "Let's all work together"
- "Let's talk it out over drinks"
that's pretty shallow.
What you should look at isn't the attitude. It's what had been building up for a long time before that attitude showed up.
Translating the three burnout elements into workplace language
1. Energy depletion
This isn't just "I'm tired."
- Mornings feel heavy
- Rest doesn't bring you back
- Even a small question drains you
- Just looking at the chat app is exhausting
- A ringing phone is enough to make you dread it
- Work stays in your head on days off
- You get overly irritated by minor remarks
That's what it looks like.
Telling this person to "be more aware" is like telling a car with an empty tank to "press the gas harder."
2. Mental distance and cynicism
This is often misunderstood on the job.
As burnout progresses, people start pulling their hearts away from work.
- "I'm done with this."
- "Do whatever you want."
- "There's no point in saying anything."
- "Oh, here we go, another 'mindset' lecture."
- "Yeah, yeah, good luck with that."
- "Taking it seriously only costs me."
- "Why am I the only one getting drained here?"
You hear more and more of these.
From the outside, the person seems
- to have a bad attitude
- to be cold
- to be uncooperative
- to have become harsh
- to not answer the phone
But sometimes what's really going on is that if they engage any further they'll break, so their mind is keeping its distance.
Mental distance isn't just a bad mood. It's a defense reaction.
3. Reduced professional efficacy
This is the loss of the feeling that
"I'm producing something meaningful at work."
- There's no point in doing it
- All I get is the blame anyway
- I improve things and nobody rewards it
- I do it right and still get yelled at
- I don't know what the right answer is
- Maybe I'm the problem
- Bare minimum is fine now
That's what it looks like.
Once it gets this far, it's dangerous.
The energy is already drained. The heart has already started to pull away. And now even trust in their own ability to do the job is going.
When someone quits in this state, the people around them say "they left out of nowhere." But for the person themselves, it wasn't sudden.
They'd been burning for a long time.
Why "more drinking parties" and "a company sports day" miss the point
Whenever the talk turns to preventing turnover, the same ideas come up:
- Let's have more drinking parties
- Let's hold a company sports day
- Let's bring back the company trip
- Let's get the young employees to mix more
- Let's run cross-department events
Of course, not all of these are bad.
If participation is optional, people feel psychologically safe, it doesn't add to their workload, those who skip aren't penalized, and it's honestly fun, it can support relationships.
But that's strictly a support buff, like in a video game.
It isn't the core of burnout or turnover.
Say someone is close to burning out because
- the workload is far too much
- their scope of responsibility is vague
- there's nobody to make decisions
- talking to someone doesn't lighten their work
- the same person always picks up the gray-area tasks
- the more they say what's right, the more they're disliked
- there's no time for improvement
- nobody can cover for them when they're off
- they get hit with blame after the fact
If you add more drinking parties to that situation, what happens?
You may well be
adding after-hours events on top of someone who's already being ground down by work.
A company sports day is the same.
"Let's build a sense of unity" — I get it. But from the point of view of the person about to burn out, it sounds like
"Before the unity, please look at my workload and my scope of responsibility."
What a person on fire needs first is a fire extinguisher. Not a fire-safety festival.
Seen through the HSE's six work-stress areas, the limits of event-based fixes are clear
The UK's Health and Safety Executive (HSE), the government body for workplace safety, focuses on the following six areas when managing work-related stress:
Demands Workload, deadlines, interruptions, work pressure.
Control How much say the person has in how they do their work.
Support Help from managers, colleagues and the organization; training; someone to turn to.
Relationships Conflict, being hammered by others, isolation, harassment, cooperation.
Role Whether their responsibilities, expectations and decision-making authority are clear.
Change How organizational changes, policy changes and sudden shifts in the work are handled.
Across these six areas, drinking parties and sports days touch, at best, part of "Relationships."
And if they feel forced, they can even backfire:
- Demands: extra non-work load
- Control: less freedom to choose whether to join
- Relationships: people who skip get left out
So company events aren't evil. But they only address a small slice of the six work-stress areas.
That's why making events your main anti-turnover strategy is risky.
What you really need to look at is:
- Is the workload too heavy?
- Does the person have any say?
- Is there enough support?
- Are relationships wearing them down?
- Is the role clear?
- Are changes being dumped on people carelessly?
Why quality assurance leaders burn out so easily
Take the quality assurance (QA) leader at a manufacturer, for example. They burn out easily.
The reason is simple.
QA structurally puts you in the squeeze between everyone.
- Customer demands are heavy
- The shop floor finds you annoying
- When a defect slips through, the responsibility is heavy
- The more you say what's right, the more you're resented
- But if you don't say it, you get a quality accident
- You're asked to drive improvements
- But there's not enough time or staff for improvement
- Phone calls and requests keep flying in
- Everyone around thinks "QA will deal with it"
In that situation, the QA leader turns cold. Stops picking up the phone. Their reactions get sharper. "I'm done with this" shows up more and more.
Seeing that, the people around them say:
- Bad attitude
- Not cooperative
- You're a leader, act like one
- Get more people on board
- Stay motivated, please
But that's not what we should be looking at.
What kind of load is it that this person can no longer keep a good attitude under it?
That's the question.
This isn't about fixing a person with a bad attitude. It's about fixing the structure that pushed them until their attitude got bad.
Burnout detection means catching exactly these people early.
The more conscientious you are, the more you burn
People who burn out weren't unserious to begin with.
If anything, the opposite.
The people who burn out most easily are usually the ones who:
- notice gray-area tasks
- can't leave someone who's struggling
- have a strong sense of responsibility
- absorb problems before they happen
- deal with things on their own before they say anything
- fill the holes nobody else notices
- try to put things back in order
- try to handle everything themselves
In the short term, these people are convenient for an organization.
But what happens when the organization starts relying on that convenience?
The capable person becomes the shock absorber.
And only after that person turns cold, pulls away or quits do people say,
"They've changed lately, haven't they?"
No.
They didn't change. They ran out of fuel.
Detecting people about to burn out is not "surveillance"
A word of caution here.
When you say "detecting people about to burn out," it might sound like
- monitoring individuals
- reading their chat logs without permission
- flagging "dangerous people" from attendance data
- HR slapping mental-health labels on people on its own
That's not what this is.
Burnout detection isn't individual surveillance. It's anomaly detection for job design.
What you should look at is not
- whether the person is weak
- whether they're motivated
- whether they have a bad personality
but rather
- Is work piling up on this one person?
- Do they carry responsibility without authority?
- Are they in a state where raising problems changes nothing?
- Do they permanently absorb the gray-area tasks?
- Has their role been left vague?
- Do they hold work that stops if they take a day off?
In Japan, the Ministry of Health, Labour and Welfare's Stress Check system, a screening questionnaire to prevent mental health problems, is likewise meant to stop problems before they start. Under the Industrial Safety and Health Act amendment promulgated in 2025, the obligation extends to workplaces with fewer than 50 employees, taking effect on April 1, Reiwa 10 (2028).
For society as a whole, too, the idea that
"we're a small company, so we don't need mental health measures"
no longer holds.
That said, stress checks and burnout detection have to be built on privacy protection. They should be used not to watch individuals, but to improve the workplace.
Checklist for spotting people about to burn out
To detect people who are close to burning out, it's easiest to combine the WHO's three elements with the HSE's six areas.
A. Signs in the individual
| What to look at | Early sign | Shallow interpretation | What you should really look at |
|---|---|---|---|
| Energy depletion | Slow replies, lifeless face, drained by small questions | Lacks motivation | Workload, interruptions, overtime, lack of rest |
| Mental distance | More "I'm done with this" and "do what you want" | Bad attitude | Has engaging become a losing move? |
| Cynicism | More sarcasm toward "mindset" talks and improvement activities | Not a team player | Did they speak up in the past and see nothing change? |
| Reduced efficacy | More "there's no point" | Negative | Is there a structure that actually rewards results? |
| Avoiding consultation | Dodges calls, requests for advice and meetings | Running away | Has asking for help become extra workload? |
B. Signs in the work itself
| What to look at | Early sign | Structure to examine |
|---|---|---|
| Demands | The same person is always working overtime | Uneven workload, deadlines and interruptions |
| Control | Has responsibility but can't decide anything | Mismatch between authority and responsibility |
| Support | Nobody to ask when in trouble | Missing consultation routes, training materials, backups |
| Relationships | People who point out real problems are disliked | A culture that makes the person raising problems the villain |
| Role | Lots of tasks nobody knows who owns | Gray-area tasks left unattended |
| Change | Sudden changes arrive with no explanation | Weak change management and priority management |
C. Signs in the team
- Nobody says what they really think in meetings
- Requests for advice always go to the same person
- More and more "let's just ask so-and-so"
- No handover documents
- No scripts or procedure manuals
- Some work stops when someone's off
- Lots of problems pointed out, but no one is named responsible
- More talk of "mindset," "initiative" and "being conscientious"
- Event ideas come up, but workload never does
If you've reached this point, it's dangerous.
What to actually do to prevent turnover
Preventing turnover is not a program for holding on to people's feelings.
What you should really be doing is a design job, like this:
1. Make gray-area work visible
List out everything that "isn't anybody's job, but somebody ends up doing."
- Who is picking it up
- How much time it takes
- Who is actually supposed to own it
- What happens if it isn't done
- Whether any of it can simply be dropped
Sort all of that out.
If you leave gray areas gray, the conscientious people soak them up.
2. Draw the line on responsibility
Don't mix up "support" and "responsibility."
For example,
- helping to create materials
- helping to write scripts
- advice on how to proceed
- building a framework for training
all of that counts as support.
But if
- running the project day to day
- being accountable for results
- making the final call
- deciding on corrections
- managing other members' mindset
come flying in uninvited, that's responsibility being dumped on someone.
3. Decide the route for asking for help
"Ask someone if you're stuck" is weak.
Decide
- what to ask
- whom to ask
- by when
- through which channel
- at what level of detail
- how far to think it through yourself first
before asking for help.
4. Make a STOP list
For preventing turnover, you need to decide what to stop before deciding what to add.
- Unnecessary meetings
- Reports with little meaning
- Events with no clear purpose
- Improvement activities with no named owner
- Corrections after the fact
- Training that depends on one person's verbal instructions
- Dumping everything on the capable person
Stop these.
5. Make people replaceable
Make sure things keep running even when someone's off.
Create:
- scripts
- procedure manuals
- checklists
- decision criteria
- templates
- past logs
- handover documents
This isn't cold toward people. It's a system so that people don't break.
Instead of "Are you okay?", ask this
If you just ask someone who's close to burning out, "Are you okay?", most of the time you'll hear, "I'm fine."
That's because the people who are really struggling have no energy left to explain.
If you're going to ask, ask something more specific.
- What's the heaviest task for you right now?
- Is there any work we could stop this month?
- Whose decision are you stuck waiting for?
- Are you picking up any work that isn't really yours?
- Which work would stop if you took a day off?
- Has there been anything you raised that never changed?
- Can this be solved by your own effort, or is it a problem of authority, headcount or rules?
This is burnout detection.
Don't only ask about feelings. Ask where the work is getting stuck.
Company events and outside events are not the same
It's worth separating these two.
Company drinking parties and sports days, unless they're well designed, turn into an extension of the current job.
- There's pressure to attend
- Saying no feels awkward
- You're tied up outside work hours
- Your boss is watching
- You can't speak your mind
- You can't get away from your work relationships
Once that happens, it's not recovery. It's extra load.
Outside events are different.
An outside event lets you step out of your current workplace's yardstick, even if only once.
- People understand what you're saying
- You get exposed to outside standards
- You find out how your skills look on the market
- You get some distance from the closed evaluation of your current job
- You can get back the feeling that you're still useful
- It can connect to a job change, a side project, an app or an article
Of the three burnout elements, this especially protects professional efficacy.
Say your energy is drained at your current job and you're already feeling mentally distant. But if outside events, apps, articles, using AI or job hunting let you confirm that
your abilities aren't dead,
that counts for a lot.
It's dangerous to leave your whole self-evaluation in your current workplace's hands. When a closed workplace doesn't value you, it's easy to think "maybe I'm the problem."
But once you step outside, you realize:
Your current job's "operating system" just doesn't fit you, and your abilities are worth something outside.
So outside events aren't just recreation. They're a refuge for your sense of efficacy.
The "mindset team" versus the "design team"
When a problem comes up, the mindset team says:
- Be more aware
- Be more proactive
- Be more serious
- Look around more
- Ask for advice more
- Cooperate more
- Let's have more drinking parties and get closer
- Let's build unity with a company sports day
The design team looks at it this way:
- Is the workload reasonable?
- Is the scope of responsibility clear?
- Who is the decision-maker?
- Is there a route for asking for help?
- Does the work keep running when someone's off?
- Who is absorbing the gray-area tasks?
- Is anyone carrying responsibility without authority?
- Are there scripts, procedures and checklists?
- Have we decided what work to stop?
Which one is kinder to people?
At first glance, the mindset team looks warmer. But in the long run, it's the design team that protects people.
Because willpower alone can't prevent burnout.
Mindset is the final fine-tuning. It's not the foundation.
The foundation is workload, autonomy, support, role clarity, relationships and change management.
Getting new graduates to join and stay is the same problem
It's the same with hiring new graduates.
You're tempted to say,
"Young people don't apply." "They join and then quit." "Kids these days." "We should promote the company's appeal more."
But according to figures from the Ministry of Health, Labour and Welfare, for those who graduated in March of Reiwa 4 (2022), the rate of leaving within three years of starting work was 37.9% for new high school graduates and 33.8% for new university graduates.
In other words, a new graduate joining isn't "mission accomplished." You have to design for retention on the assumption that a considerable share will leave within three years.
So
- recruiting PR
- company events
- mingling with senior colleagues
- drinking parties
aren't enough on their own.
What you need is a design where:
- they know what's expected of them
- it's easy to ask for help
- feedback is specific
- training doesn't depend on one person
- they can learn even when they fail
- the scope of responsibility is clear
- people don't each say something different
- those about to burn out are caught early
Young employees don't quit because there's no sports day. In most cases, they quit because they start getting stuck in the work itself.
Mendoi-chan's conclusion: events are a support buff, burnout detection is the defensive spell
Drinking parties and sports days aren't evil.
But they're a support buff.
What you really need is
the defensive spell of burnout detection.
Putting out a fire after it starts is too late. Holding an exit interview after someone says they're quitting is too late. Cracking down on someone after their attitude goes bad is too late.
Notice when the smoke first appears. Don't ask who's about to burn out. Ask which job design is burning people.
What you need for that:
- making workload visible
- making gray-area work visible
- clarifying scope of responsibility
- clarifying routes for asking for help
- scripts, procedure manuals, checklists
- backups and handovers
- a STOP list
- matching authority with responsibility
- securing a sense of efficacy outside work
Preventing turnover is not the art of holding on to employees.
It's the art of removing, in advance, the structures that make employees want to leave.
So before you start thinking about drinking parties or a sports day, ask first:
Who in this workplace is about to burn out? What job design is burning them? Can we put out that source of fire with structure, not mindset?
Turnover prevention that ignores this ends up as nothing more than a fire-safety festival.
Quick one-line summary
- Preventing turnover is not about persuading people who are leaving. It's about shrinking the structures that make people want to leave.
- Drinking parties and sports days are support measures, not the core of burnout prevention.
- Burnout isn't a lack of grit. It's the result of chronic workplace stress that wasn't managed.
- Before looking at the person with a bad attitude, check whether they're under a load that makes a good attitude impossible.
- The more conscientious someone is, the more gray-area work they pick up, and the more easily they burn out.
- Detecting people about to burn out is not individual surveillance. It's anomaly detection for job design.
- Outside events can be a refuge that protects the sense of efficacy worn down at your current job.
- Before raising mindset, build a structure that doesn't burn people out.
