Managing Underperformance in Japan: Documentation That Matters

The complaint is consistent across companies: the employee has been struggling for more than a year, the manager has tried to work with them, and everyone on the team knows the situation is not working. The performance problem is real and well-understood internally. What is not ready — often discovered only at the point someone asks “can we let this person go?” — is the paper trail that would allow the company to demonstrate all of that to someone outside the room.

In Japan, that outside audience matters acutely. A performance-based dismissal under Japanese law is not just a personnel decision that stays internal; it is a decision the company may have to defend, retroactively and in detail, before a labor tribunal or court. What the company did to support the employee, what it communicated, what it documented, and in what form — these are not administrative formalities that can be reconstructed after the fact. They are the substance of whether the dismissal was lawful, and courts look at them carefully.

This article focuses on the documentation side of that equation: what courts look for, what the record needs to contain, how to design the process so that what is produced will actually hold up, and — for companies that arrive at this question with thinner documentation than they would like — what can still be done before the situation reaches a formal endpoint.

Why “He Just Isn’t Performing” Is Not Enough to Dismiss Someone in Japan

Foreign HR teams sometimes arrive at the performance dismissal question with an implicit model: the employee is not meeting the standard, the company has decided the situation is unworkable, and the legal question is simply how to execute the departure cleanly. Under that model, the decision itself is the hard part; the law is expected to accommodate a reasonable business judgment once it has been made.

Japanese labor law does not work that way for regular employees. Article 16 of the Labor Contract Act provides that a dismissal is void if it lacks objectively reasonable grounds or is not considered appropriate under general social norms. In practical terms, the employment relationship is treated as having continued, which can expose the company to claims for unpaid wages for the period following the dismissal. For performance-based dismissals specifically, courts apply this standard in a way that gives foreign companies considerably less deference to their own assessment of an employee’s capabilities than they might expect. The employee’s manager concluded the performance was insufficient; a Japanese court will assess that conclusion against what the company actually did to support, guide, and give the employee a real opportunity to improve — and “we decided he wasn’t performing” is not, on its own, an answer to that inquiry. The general structure of Japan’s dismissal framework and the full range of risks it creates for foreign employers is set out in Employee Dismissal and Labor Law Risks for Foreign Employers in Japan.

What specifically distinguishes performance-based dismissals from other dismissal categories is that courts look hard at whether the insufficiency of performance could have been remediated by the company, and whether the company genuinely tried. An employee terminated for gross misconduct presents a different legal profile than one terminated for failing to meet targets or lacking certain skills — the latter triggers inquiry into whether the company’s own conduct before the dismissal was adequate. That inquiry is what the documentation record either supports or fails to support.

What Japanese Courts Actually Look for Before Accepting a Performance-Based Dismissal

When a performance-based dismissal is challenged, courts generally assess it against a cluster of questions that foreign companies do not always anticipate as distinctly Japanese concerns, because in many home markets some of these questions are simply not part of the legal analysis at all.

The first is whether the performance gap was real and specific. Vague assessments — “attitude problems,” “not a culture fit,” “performance below expectations” — without concrete, documented examples of what was expected, what was delivered, and how the gap was measured do not carry the weight that a precise, fact-based performance record does. Courts are unimpressed by descriptions of poor performance that cannot be traced to specific incidents, outputs, or measurable criteria applied consistently.

The second, and often the decisive factor, is whether the company provided the employee with adequate guidance, clear instruction, and a genuine opportunity to improve before the dismissal decision was made. This is the element that most directly requires a documented pre-dismissal process — not because the process is a formality, but because the existence and quality of that process is itself what courts are evaluating. An employee who was managed out within weeks of a performance concern being raised, without a record of structured feedback, instruction, or a defined improvement period, presents a much more difficult position for the company than one where a full documented process preceded the dismissal. Performance management within the legal framework Japan establishes for this is addressed specifically in Performance Management in Japan: Legal Limits and Practical Steps.

The third consideration is proportionality: was dismissal genuinely the last available resort given the circumstances, or was a less drastic outcome — reassignment to a different role, additional training, or a further improvement period — available and not genuinely attempted? A company that moved directly from identifying a performance problem to dismissal, without exploring intermediate steps, is in a weaker position than one that can show a progression from support, to warning, to structured improvement process, to departure.

There is a category where the analysis shifts in a specific way: employment agreed for a defined, limited role type (職種限定合意) or where the original terms specified a particular job function. When a company hired someone explicitly for a specialized position and that person lacks the core competencies the role requires, courts tend to apply a different standard on the proportionality axis — the expectation that the company should address underperformance by offering a role transfer is weakened where the hire was made precisely because of specialization in that role. Whether a role-type limitation exists is assessed on the facts of the specific employment relationship, including the job description, what was communicated at the time of hire, and the actual practice of the parties — a domestic job title alone does not establish it. Similarly, where the employee provided materially false information about their qualifications at the time of hiring, the analysis of the company’s obligations before dismissal also shifts. These situations do not eliminate the documentation requirement, but they change how heavily the alternative-options analysis would otherwise weigh against the company, and they should be flagged at the outset when assessing the strength of the overall position.

The Documentation Gap That Causes Companies to Lose Labor Cases

The problem most companies face when a performance-based dismissal is challenged is not that the underlying performance concerns were not real — they almost always were. The problem is that the record the company produced to substantiate those concerns and demonstrate its pre-dismissal process looks, from the outside, like it was assembled for the purpose of defending the dismissal rather than generated in real time as part of managing the employee. That distinction matters enormously to a labor tribunal.

Several specific documentation patterns create particular vulnerability. The first is a record that begins only when someone decided the employee was going to be let go — nothing in the file for the prior eighteen months of struggle, then a sudden burst of performance documentation that coincides precisely with the decision to terminate. The timing itself signals that the record was constructed to support an already-made decision rather than to genuinely manage the employee’s performance.

The second is a record composed entirely of HR system entries and annual review ratings without any contemporaneous narrative — numbers in boxes, but no written account of what was discussed, what instruction was given, what the employee said in response, or what specific steps were agreed to. Rating an employee below standard in an annual review system does not, by itself, constitute the guidance and instruction courts are looking for. What courts need to see is evidence of real-time engagement: the manager raised a specific concern, the employee understood what was expected, and both sides knew what improvement looked like.

The third pattern — perhaps the most consequential — is the absence of any acknowledgment by the employee that the severity of the situation was communicated to them. A manager who raised performance concerns in the gentlest possible terms, in the style of a supportive coaching conversation without any explicit statement that continued underperformance would have consequences, has not created a record that demonstrates the employee was warned. The concern, the instruction, and the consequence of non-improvement all need to be communicated with enough clarity that there is no reasonable ambiguity about whether the employee understood the situation.

How to Design a PIP That Holds Up — and the Version That Creates More Risk Than It Solves

A Performance Improvement Plan can be one of the most valuable pieces of documentation in a performance dismissal case, or it can make the company’s position substantially worse. The difference lies entirely in whether the PIP was designed as a genuine improvement process or as a compliance wrapper for a decision already made.

A well-designed PIP in the Japanese context does several things. It defines specific, measurable performance objectives tied to the role’s actual requirements — not aspirational targets set impossibly high, but concrete tasks and quality standards the employee is expected to meet within a defined period. It provides for regular check-in meetings during the plan, documented in real time with notes covering what was discussed, what the employee said, what feedback was given, and what the agreed next steps are. It is communicated to the employee as a genuine effort to support their improvement, with clear language about what successful completion looks like and what the consequence of failing to meet the plan’s requirements will be. And it runs for a period that gives the employee a realistic opportunity to demonstrate improvement given the nature of the concerns — a timeline set too short relative to the complexity of the performance issue makes the process look predetermined.

A PIP that creates risk rather than reducing it is designed to fail. The version foreign companies produce most often in this category is one where the targets are set at a level the manager already believes the employee cannot reach, the check-in meetings are cursory and not documented, and the entire plan exists to create a paper trail that ends in a predetermined outcome. Courts and labor tribunals in Japan have seen this pattern. A PIP that reads, in retrospect, as a procedural cover for a decision made before it started is not likely to be treated as genuine evidence that the company provided a meaningful opportunity to improve. Worse, if the circumstances suggest the PIP was designed to produce a predetermined failure — with targets set at levels the company itself did not genuinely believe the employee could meet, or with check-ins that were cursory and not genuinely aimed at improvement — the PIP may be characterized as a form of pressure to resign (退職勧奨) or as evidence that the company did not provide a genuine opportunity to improve before dismissal. Either characterization may weaken the company’s position and, depending on the circumstances, may affect the assessment of whether a resulting dismissal is lawful.

The safest framing — both legally and practically — is to design the PIP as if the company’s genuine intention is for the employee to succeed. Clear targets, real support, documented check-ins, and an honest assessment of whether improvement occurred. A company that follows that process and the employee still fails to meet the plan’s requirements is in a materially stronger position than one that ran a perfunctory PIP and is now trying to characterize it as a meaningful opportunity.

Warning Letters and Performance Reviews: What to Say, What to Keep, and in What Format

Formal documentation — warning letters, performance review records, meeting notes — serves a specific evidentiary function in a Japanese labor case. Understanding what that function is shapes what the document needs to contain.

A warning letter in the Japanese context is not primarily about discipline in the punitive sense. Its evidential purpose is to demonstrate, with a contemporaneous record that both parties can point to, that the employee was clearly informed that their performance or conduct was insufficient, what specifically needed to change, and what the consequence of failing to change would be. A letter that reads diplomatically — “we would like you to take your performance to the next level” — does not serve this function. A letter that states specifically what was inadequate, what the expected standard is, and that continued failure to meet it will result in further action up to and including dismissal, does. The difference between these two versions of what looks like the same document is the difference between having a warning on the record and not having one.

Performance reviews require the same precision, applied over time rather than in a single document. A review that rates an employee below standard, states why in specific terms, records what improvement is required, and documents that the content of the review was discussed with the employee in a meeting where the employee had an opportunity to respond creates a contemporaneous record that is difficult to dismiss as after-the-fact construction. A review that applies low ratings without explanation, is filed without any meeting record, and is never discussed with the employee until it is referenced in a termination discussion does not.

Format matters in a specific way: documentation generated in the company’s own HR systems and file records, signed or acknowledged by the employee where possible, and date-stamped contemporaneously is more credible than the same content reconstructed in a memo written after the fact. An email sent to the employee summarizing the feedback from a performance discussion — sent the same day, before the conversation has receded — is worth more as evidence of what was communicated than a set of meeting notes written from memory three months later. Asking the employee to acknowledge receipt or respond in writing is also worth the discomfort it may cause the manager: an employee who signs off on or replies to a performance improvement notice has created evidence that they understood the message.

When the Record Is Thin: What Companies Can Still Do Before It’s Too Late

This is the section that addresses the situation most companies are actually in when they first ask this question: something has been building for a long time, the informal conversations happened but nothing was written down, and now someone at HQ is asking when the company can move to a formal endpoint.

The honest answer is that a thin record cannot be converted into a strong one by adding documentation that is obviously retrospective — backdating, reconstructing, or papering over a gap with documents clearly prepared in the shadow of a dismissal decision is worse than having a thin record, because it can itself become a problem. What the company can do, starting from where it actually is, is build forward from the current point with documentation that is real, contemporaneous, and appropriately serious.

Concretely: the company can still introduce, at this stage, a genuine PIP designed and communicated as described in section 4 — not as a pretextual process to reach a conclusion already decided, but as a legitimate final opportunity that the company would be satisfied with the employee completing. It can begin generating the kind of documented check-in meetings and written summaries that section 5 describes. It can deliver, in writing, a clear statement of what the performance concerns are, what is expected, and what will follow if the situation does not improve. None of this retroactively creates the eighteen months of documentation that does not exist, but it can begin to demonstrate, from this point forward, the kind of process a court would expect to see before a dismissal is treated as lawful.

The alternative that is often the more realistic path, particularly where the documentation gap is large, is a voluntary exit rather than dismissal — resignation encouragement (退職勧奨) conducted within legal limits, aimed at reaching a mutually agreed separation. That route does not require the same record needed to justify a disputed dismissal, although the voluntary separation process itself should still be documented carefully to demonstrate that the employee’s decision was genuinely voluntary and that no pressure or coercion was applied. The design of a voluntary exit process that stays on the right side of the pressure/coercion line is addressed in Taishoku Kansho: How to Encourage Voluntary Resignation in Japan, and for companies with a thin documentation record, it is worth understanding both paths before deciding which to pursue.

Conclusion

The underlying performance problem is real, the business judgment that the situation is unworkable is often correct, and the frustration with a legal framework that requires all of that to be proven rather than simply accepted is understandable. But what Japanese labor law asks for is not proof that someone was eventually let go — it asks for proof that before they were let go, the company gave them a real chance to succeed.

Documentation is what makes that proof. Not the existence of a PIP checkbox or an annual review in the HR system, but a contemporaneous record that shows the manager raised specific concerns, the employee understood what improvement looked like, both sides knew what the consequence of continued underperformance would be, and the company genuinely tried before concluding it could not work. That record is far easier to build over the course of the performance management process than to reconstruct after a dismissal decision has already been made — and the companies that build it in real time are the ones that end up in a materially different position when a dispute arises.

Managing an Underperforming Employee in Japan?
Our team regularly advises foreign companies on designing documentation-backed performance management processes in Japan — including PIP design, warning letter format, and when to shift from performance management to a voluntary exit approach.

If you are managing an underperforming employee in Japan and want to ensure your documentation process is designed to hold up if the situation escalates, our team can help you structure the approach. Contact the TSL Partners – International Business Desk