Six in Ten or Nearly Nine: Japan's Post-Hire Gap Surveys

An interview is where the company describes the job, and the first month is where the new hire checks the description. Employers tend to assume that if someone is disappointed after joining, the candidate did not do their homework. In a 2026 survey by en Japan's job site, 61% of respondents said their most recent employer turned out worse than the picture they had before joining. When recruiters were asked why, three in ten said the company had shown only its attractive side during selection.

Who is counted

There is no official statistic on the "post-hire gap" (入社後ギャップ). What exists is a set of surveys that recruiting companies run on their own members, each worded differently. I use four sources here, so the coverage of each comes first.

The first is the 61%. It comes from the 136th user survey of En Tenshoku (エン転職), en Japan's job-change site, run from 3 to 31 August 2026 with 857 valid responses. It asks people who have changed jobs as regular employees about their most recent employer: 61% in total reported a bad gap against their pre-joining image, and 15% a good one. The page does not give the number of people who answered that question, and the detailed breakdown exists only as chart images, so I use only the figures stated in the text.

The second is also from en Japan: 929 users aged 39 or under of AMBI, its scouting service for young high earners, surveyed in May and June 2024 and published on 15 January 2025. The question is whether they have ever felt a gap after joining a company, any company, good or bad. 87% said yes.

The third asks a different group. 171 recruitment consultants who use the Middle no Tenshoku (ミドルの転職) platform were surveyed from 25 June to 1 July 2025, published 23 July. 43% said that many, or somewhat many, mid-career candidates feel a gap after joining.

The fourth is a government sample survey. The individual part of the Ministry of Health, Labour and Welfare's 2025 Survey on Job Changers covers people who had joined an establishment with five or more regular employees within the previous year, as of 1 October 2025, with 7,310 valid responses. It does not ask about gaps. It asks about satisfaction.

The first three draw on members and users of recruiting services. People browsing a job site are more likely than average to be unhappy where they are. Read those numbers with that discount.

Is it six in ten or nearly nine in ten?

It depends on which employer you ask about and whether good surprises count. Most recent employer, bad gaps only: 61%. Add the 15% with good gaps and it is 76%. Any employer, ever, good or bad: 87%.

Change the question and the number changes character. The ministry asked recent job changers how satisfied they are with their working life now. 54.7% were satisfied or somewhat satisfied. 10.7% were dissatisfied or somewhat dissatisfied. On job content alone it was 68.7% satisfied and 10.9% dissatisfied.

More than six in ten found a gap. One in ten is dissatisfied. You cannot subtract one from the other, because the populations and the questions differ. Still, one thing is visible. "This is different from what I was told" and "I am unhappy here" are separate questions, and most people who find a gap stay.

Asked "was there a gap", six to nine in ten say yes; asked "are you dissatisfied", one in ten says yes

Four surveys with different populations and questions; they cannot be subtracted from one another. The first four bars are online surveys of members, users or recruitment consultants of en Japan services (%); the last is the MHLW 2025 Survey on Job Changers, individual survey (sample survey, %)

• Ever felt a post-hire gapAMBI users aged 39 or under, 2024
  87%
• Bad gap at latest employerEn Tenshoku members, 2026
  61%
• Good gap at latest employersame survey
  15%
• Say many mid-career hires feel a gap171 recruitment consultants, 2025
  43%
• Dissatisfied with working lifeMHLW, within a year of changing jobs, 2025
  10.7%

"There was a gap" and "I am dissatisfied" are two questions. Check which one the survey asked.

Where do the gaps show up?

Most often in the work itself and in the people. In the En Tenshoku survey the most common gap was job content, followed by company culture and then relationships with colleagues.

The AMBI survey is more detailed. "Worse than expected" was most often job content, at 39%, and workplace atmosphere, at 38%. "Better than expected" was most often the same two: atmosphere 40%, job content 31%. "As expected" was most often work location, at 37%.

Put those three lines side by side and here is my reading. What matches is what was written down. Work location is in the written notice of working conditions, so there is little to be surprised by. What fails to match is what was said aloud in the interview. What you will do all day, what the team feels like, what the boss is like. None of that fits in a contract. The candidate has only the interviewer's word.

The consultants point the same way. The gaps they see most among mid-career hires are relationships that did not work, at 38%, ways of working and process that did not fit, at 32%, and job content that differed from what was assumed, at 29%.

Did the candidate just fail to research the company?

Partly, though most candidates do not see it that way, and three in ten consultants put the cause on the company.

AMBI asked whether the gap could have been prevented by better research before joining. 42% thought so. 52% thought not. Asked what they wished they had checked more, the top answer was a conversation with current employees, at 39%, ahead of job sites and review sites at 34% each and a workplace visit at 33%.

The consultants were asked for causes, multiple answers allowed. Insufficient research before joining: 42%. Expectations set too high: 32%. No information other than the attractive points offered during selection: 30%.

That last one is testimony from the person in the middle. An interview is half sales. The interviewer is assessing a person and selling a seat at once. Telling only the good half is not quite lying. The other half turns up by itself after the start date.

What the company says in an interview is a claim

My own habit when I research a company, set out in my ground rules, is this: whatever a company says about itself is recorded as a claim, and it becomes a fact only when a real event or an independent source backs it up. The glossary entry on claim versus fact says the same thing. I normally apply that habit to other companies. These survey numbers suggest applying it to your own interviewers.

"You'll have a lot of autonomy." "The team gets along well." "Overtime is light." In the interview room these are claims. Within a month of joining, each one has become a fact or has collapsed. The new hire does not need to investigate. They sit there every day and see it.

What the employer can do is run the check before the interview. Attach one event to each selling point. Autonomy: what did the person in this seat decide alone last quarter? Team: how many people are on it now, and how many left in the past year? Overtime: how many hours was it last month? If no event attaches, leave the sentence out. And since a conversation with current employees was the top item people wished they had checked, the company can simply offer one.

AMBI also asked what happens otherwise. 34% had changed jobs because of a gap, and another 33% had started looking without moving. The gaps that triggered a move were workplace atmosphere and job content at 34% each and pay at 33%, close to a three-way tie. I worked through what it costs to hire and employ one person in what one hire in Japan really costs. When a gap is big enough to move someone, you pay that again.

There is a small result on the other side. In the En Tenshoku survey about seven in ten said they feel they are contributing to their company, and those who reported a good gap tended to feel it more. Promise a little less than you can deliver, and the difference shows up later on your side of the ledger.

What these numbers cannot show

The ministry's satisfaction figures have a built-in tilt. They come from people still employed on the survey date. Anyone who left after three months is out of the denominator. The job-site surveys tilt the other way, since the people answering are often already looking. The real rate sits somewhere between, and no survey I found can say where. I also found no survey in which employers counted how many of their own interview statements held up after the hire. Every survey I saw asks the person who joined, which is a little odd, since the employer is the one who knows exactly what was said. The candidate's side of the check is in your first 90 days in Japan.

So for now the count has to be done in-house. Take a few recent hires, list what each was told in the interview, and mark which lines an event has since confirmed. It is a small exercise, and it is the only version of this statistic that describes your company.

---