On this page
TL;DR: Weaker results are not enough to conclude that an employee is disengaged. First you need to check whether the goal was clear, the resources were available and the task was within their skills. A diagnostic conversation is about understanding the causes and agreeing on support. A performance improvement plan makes sense when the problem is significant, persistent and requires clearly described improvement that is reviewed regularly. Formal employment consequences are a separate matter.
- Compare results with a previously agreed standard and the conditions that were in place.
- Do not confuse a competence problem with a problem of tools, priorities or workload.
- Agree the expected outcome, the support provided and the dates of joint reviews.
- Keep developmental support separate from formal HR proceedings.
How does Empatyzer help you hold a conversation about performance?
Em helps you prepare feedback based on facts and specific tasks, and then structure a conversation about the causes of the problem, support and clear expectations. It is worth starting with comparable examples and finishing with a plan that both sides understand in the same way.
Features that will help you:
- Talk to Em about a specific person: Practising how to give difficult feedback without judging the person.
- Talk to Em about your team: Taking into account constraints on collaboration and how tasks are allocated.
- Personalised micro-lessons: Practising corrective conversations and following up on agreements.
Before you judge the person, check that you are measuring the same work
Two people handle customer service. One closes more cases; the other mainly handles difficult complaints and needs more time. The number of closed tickets alone can therefore lead to the wrong assessment. Before you start a corrective conversation, make sure the standard was known, measurable and appropriate to the conditions. Did the task have a set deadline? Did the employee receive the tools, decisions and information they needed? Was their scope of responsibilities changing from week to week? Research by Gilboa and colleagues on work stressors found that role ambiguity and situational constraints are associated with poorer performance. This does not mean that poor results never depend on the employee. It means that an assessment made without checking working conditions is incomplete. It is worth starting with comparable examples and asking the other person to point out any barriers their manager may not be aware of.
The first conversation looks for the cause, not confirmation of blame
A manager might start like this: “Over the last four weeks, three out of five reports came in late. We agreed on Wednesday by midday. I want to understand what is making it hard to meet that deadline and what needs to change.” Then they should listen, ask follow-up questions and check examples. If the employee does not understand a tool, it is worth offering to walk them through the process or arranging short training. If the task is being held up by decisions in other departments, the manager needs to help unblock them. If the problem is repeated oversights despite the right conditions being in place, a change in working methods is expected. Research on performance appraisal does not justify judging someone’s “attitude” on the basis of a single behaviour. In a review of a hundred years of research, DeNisi and Murphy highlighted the many purposes of appraisal and the limitations of the appraisal system itself. That is why the conversation should not turn into reading out the boxes on a form. Its purpose is to describe the problem clearly, understand its causes and agree the next steps.
Feedback can help – but it can also make performance worse
You often hear that more frequent feedback always improves results. A meta-analysis by Kluger and DeNisi covering 607 effect-size estimates painted a more complex picture: on average, feedback helped, but in more than a third of the interventions analysed, performance deteriorated. The authors pointed to the importance of directing attention to the task rather than to a threat to the person’s self-image. This study does not provide a ready-made script for every conversation; it covered many kinds of situations and interventions. In practice, it is better to say “Three items were missing from the report, so the finance department couldn’t close the month” than “You’re unreliable.” Then ask how the employee checks that the report is complete and what control mechanism could be introduced. Feedback should lead to action that can be checked, not to someone having to justify their own character. If the manager cannot point to an example, they should establish one first.
When to create an improvement plan rather than just another conversation
A plan makes sense when the problem concerns key duties, is entrenched, has been clearly pointed out to the employee and has not improved after adequate support – or when it requires a swift, structured intervention. The document should set out the current situation, the expected outcome, the scope of support, how progress will be measured and the review dates. For example: “For six consecutive weeks, at least nine out of ten tickets are to be closed on time, based on the current, agreed allocation of cases.” The figures here are an example of what might be agreed, not a universal norm. The plan must also state what the employer will do: provide access to tools, arrange a consultation with an expert or remove process blockers. The plan should not include goals the employee has no influence over. If the standard depends on a legally regulated procedure or on safety, the organisation’s relevant requirements must be taken into account. The plan on its own does not determine whether later HR decisions comply with the law.
Developmental support and formal assessment are two separate processes
A manager may spend many weeks helping an employee improve the quality of their work and yet, at some point, still have to make an HR decision. It is worth being honest from the outset about the difference between a development conversation and the formal proceedings provided for by the organisation and by employment law. You must not promise that the plan will have no consequences if the company does in fact intend to assess how it has been carried out. Nor must you use the threat of dismissal as a motivational technique. Formal documents should comply with the applicable legislation, including the rules on grounds for terminating employment contracts where these apply. In Poland, the point of reference is the Labour Code (Kodeks pracy), but the answer depends on the specific situation and the type of contract. The manager holding the conversation should prepare the facts, deadlines and scope of support, and consult the appropriate person in the organisation on formal matters. A good process allows the employee to understand what depends on them.
The most important test: do both sides know what to do tomorrow?
After a difficult conversation, the employee should be able to repeat the expectation in their own words, and the manager should be able to say what support they are providing and how they will recognise improvement. If all that comes out are slogans such as “more proactivity” or “a better attitude”, the conversation remains unclear. It is better to write down: “Report by Wednesday, all six items complete; if data is missing, flag it by 3 pm on Tuesday.” Then agree a short review after a week and a broader one after a month. When results improve, it is worth acknowledging this rather than constantly reminding the person of past mistakes. When nothing changes, you need to look at the causes again and consider the next step that is formally appropriate. A process like this does not guarantee success for everyone, but it provides a fairer basis for decisions than the label “doesn’t deliver”. The conversation becomes work on the task, not a public trial of the employee.
Weaker results call first for a check of the standard and the conditions, then a conversation about the causes, and only if necessary a performance improvement plan. A well-designed plan describes a measurable outcome and the support the company will provide. Formal assessment remains a separate decision, which must not be replaced by applying pressure during feedback.
How Empatyzer helps you prepare a conversation about weaker performance
A manager has to tell an employee that the timeliness of their reports has been slipping for two months. Instead of starting with a judgement about engagement, the manager describes to Em the actual examples, previous agreements and information about working conditions. Em helps phrase a message that focuses on the specific task and prepare questions about possible barriers, missing skills or a misunderstanding of the standard. If both people use Empatyzer in the appropriate context, relationship tips can make it easier to choose the right tone and way of delivering feedback, while respecting the privacy of the other person’s answers. The manager can rehearse how to respond to anger, withdrawal or the words “Nobody ever told me that before.” They then ask Em to structure a plan for the conversation: the expected action, the support and the date of the next meeting. With Em, the manager can prepare the conversation better, while formal HR decisions are made in line with the company’s procedures. Micro-lessons make it possible to practise regular corrective conversations before difficulties turn into a months-long conflict. The practical result is greater clarity about exactly what needs to improve and what each side will do about it.
Sources
- Kluger AN; DeNisi A (1996). The Effects of Feedback Interventions on Performance: A Historical Review, a Meta-Analysis, and a Preliminary Feedback Intervention Theory. Psychological Bulletin. https://doi.org/10.1037/0033-2909.119.2.254 10.1037/0033-2909.119.2.254
- DeNisi AS; Murphy KR (2017). Performance Appraisal and Performance Management: 100 Years of Progress?. Journal of Applied Psychology. https://doi.org/10.1037/apl0000085 10.1037/apl0000085
- Gilboa S; Shirom A; Fried Y; Cooper CL (2008). A Meta-Analysis of Work Demand Stressors and Job Performance: Examining Main and Moderating Effects. Personnel Psychology. https://doi.org/10.1111/j.1744-6570.2008.00113.x 10.1111/j.1744-6570.2008.00113.x
- Sejm Rzeczypospolitej Polskiej (2026). Kodeks pracy, tekst jednolity Dz.U. 2026 poz. 1245. Dziennik Ustaw. https://eli.gov.pl/eli/DU/2026/1245/ogl Kodeks pracy, tekst jednolity Dz.U. 2026 poz. 1245
Find an article
Enter at least 2 characters. Minor typos are OK.