Some employees make the performance review easy. Their results are visible, their behaviour is unambiguous, and the written comments almost write themselves. Then there are the others. The quiet achiever whose impact is real but hard to quantify. The technically brilliant contributor who leaves friction wherever they go. The person whose performance swings from excellent to absent, and whose manager has never quite understood why. These are the reviews that get written at midnight, full of vague phrases that mean nothing and commit to even less.
This post is built for those conversations. It is a phrase library and a framework, grounded in what performance management research actually says about assessing people whose contribution resists easy measurement. Every example here is designed to be specific enough to cite, honest enough to matter, and direct enough to hold up if the employee pushes back.
What do good performance review examples actually look like for hard-to-assess employees?
Good performance review examples for hard-to-assess employees share three features: they describe observable behaviour rather than inferred personality, they cite specific evidence rather than general impressions, and they separate what happened from why it happened. The result is a comment the employee can read without feeling judged on factors outside their control.
The standard advice is to be "specific and constructive". That is correct but not sufficient. Specificity without context produces a list of complaints. Constructiveness without specificity produces praise the employee cannot replicate. The examples that actually work do both at the same time, within a single comment.
Consider the difference between these two approaches:
"Mariana tends to work in isolation and does not always communicate proactively with the team."
Versus:
"During the product launch in March and the system migration in September, Mariana completed her deliverables ahead of schedule but did not flag blockers or share progress updates until the final handoff stage. Two team members independently noted they had to request status updates rather than receive them. The work itself was strong. The visibility of the work needs to improve."
The second version is longer. It is also the only one that gives Mariana something specific to change. The goal of a written review comment is not to describe an employee - it is to give them a legible account of what they did, what it cost or contributed, and what a better version would look like.
This applies at every level of the organisation. The framework does not change. Only the evidence changes.
How do you write a performance review for a quiet employee whose impact is real but invisible?
Write about output, not personality. A quiet employee whose work is consistently strong is not a performance problem. The review should reflect that plainly, then identify the one or two areas where greater visibility would multiply their impact. Do not pathologise introversion or describe quietness as a gap unless it is genuinely limiting outcomes.
Gallup's research on employee engagement consistently shows that employees who feel their contributions are not recognised are significantly more likely to disengage. Quiet employees are especially vulnerable to this dynamic, because their outputs often accrue to the team without attribution. A performance review that names the specific contribution fixes that, at least for the cycle.
Here are three worked examples for the quiet high performer:
Example 1 - Consistent delivery, low visibility: "Over the past twelve months, the data quality on every report this team produced was anchored by the work coming from this role. Errors that would have surfaced downstream were caught at source. The business impact of that is not small. What would make next year stronger is bringing this same level of analysis into the planning conversations earlier, before the work is already defined."
Example 2 - Strong output, limited peer influence: "The technical output this year has been genuinely excellent, and the documentation standards set in Q2 have improved how the whole team hands off work. The next level of contribution involves actively sharing that methodology, not waiting to be asked. One structured knowledge-sharing session per quarter would make this person's impact felt across the team rather than just within their own workstream."
Example 3 - High delivery, low recognition footprint: "This role quietly held a large amount of operational continuity together during a period when several processes were changing at once. That stability was noticed by the people who depended on it, even if it did not appear in any formal recognition cycle. That gap - between what was contributed and what was acknowledged - is worth fixing going into next year."
Notice that none of these examples frame quietness as a defect. They name the contribution, they note the gap in visibility, and they make a concrete forward-looking suggestion. That is the whole structure.
How do you write a performance review when results are inconsistent across the year?
Name the pattern honestly, tie it to conditions where possible, and separate the person from the circumstances. Inconsistency is a signal, not a verdict. The review's job is to surface the signal clearly so the employee and their manager can interpret it together and build the action plans needed to address it.
The worst version of this review is the one that averages out the year and calls the employee "meets expectations". That approach obscures the variation, which is the most important information in the data. It also fails the employee, because they leave the review without understanding which conditions brought out their best and which ones did not.
Here is how to write it honestly:
Example - Peaks and troughs with identifiable conditions: "The first half of the year was among the strongest on record for this role. Every milestone was met, stakeholder feedback was positive, and two process improvements came directly from this person's initiative. The second half was notably different. Three deliverables were late, one was returned for significant rework, and two stakeholder escalations landed on the team. It is worth understanding what changed. The external context of the Q3 restructure is relevant, but so is the question of what support was in place and whether it was adequate. That conversation needs to happen before next year's targets are set."
Example - High output, low reliability on collaboration: "The individual outputs from this role this year have been consistently above standard. The reliability as a collaborator has been less consistent. In cross-functional projects, three separate team leads raised concerns about responsiveness and adherence to shared timelines. The technical contribution is not in question. The question is how to bring the same discipline to joint commitments that this person already applies to solo work."
Bold the key nuance here: an inconsistent performance record is not the same as a weak performance record, and the review should make that distinction plain. One produces a development conversation. The other produces a performance improvement plan. Conflating the two is unfair to the employee and confusing to everyone who reads the review later.
How do you write a performance review for someone technically strong but interpersonally difficult?
Address both dimensions explicitly and in separate paragraphs. Do not let strong technical output absorb the interpersonal concern, and do not let the interpersonal concern overshadow genuine competence. HR teams frequently make one of these mistakes because writing both clearly, in the same document, feels contradictory. It is not. It is accurate.
Research published in Harvard Business Review on feedback and performance makes the point that people improve when they understand specifically what behaviour is producing which outcome. Vague observations like "can be difficult to work with" produce defensiveness, not change. The comment needs to describe the specific behaviour, the specific context, and the specific impact.
Here are two worked examples:
Example 1 - High technical output, recurring friction on shared work: "The modelling work produced this year was among the most sophisticated the team has seen, and the accuracy rate on the Q4 forecasts was exceptional. That is a genuine strength and it has been relied on. At the same time, during the two major cross-team projects this year, three colleagues raised concerns about how disagreements were handled during the design phase. The pattern described was of positions being defended past the point where other perspectives could land. The technical credibility this person has earned is an asset. Using it to bring people along, rather than to end conversations, would significantly expand the influence available to this role."
Example 2 - Expertise that creates dependency rather than capacity: "This person is the most knowledgeable member of the team on the legacy system, and that knowledge has been essential this year. The concern is that the knowledge tends to stay in one place. Requests for documentation have not produced usable outputs, and two junior team members have noted that asking questions can feel unwelcoming. This is not about personality. It is about whether the team is stronger or more fragile as a result of how that expertise is held. The expectation for next year is a structured approach to knowledge transfer, with specific milestones in the action plan."
Both examples carry recognition and a development challenge in the same breath. That balance is what makes them credible.
How do you write a performance review when you lack direct visibility into the employee's work?
State your data sources explicitly and be honest about the limits of your view. Do not guess. Do not synthesise vague impressions into confident-sounding statements. The employee deserves an accurate account, and an honest admission of limited visibility is more useful than a fabricated one.
This situation is common on hybrid teams, in matrixed organisations, and anywhere that employee sentiments are gathered through tools rather than direct observation. It is also one of the strongest arguments for running regular pulse surveys and building continuous feedback loops between review cycles. If a manager has twelve months of feedback data to draw on, the year-end review is synthesis. If they have six weeks of memory, it is guesswork.
Here is what an honest review looks like under these conditions:
Example - Manager with limited direct visibility: "My direct visibility into this role across the year has been limited by the split reporting structure. This review draws on three sources: the outputs delivered directly to me, structured peer feedback gathered in October, and the two mid-year check-ins we held. Where I can speak with confidence, I have. Where the picture is incomplete, I have said so rather than filled the gap. That means this review is narrower than it should be, and I want to name that upfront. One of the commitments I am making for next year is a more regular check-in cadence so that the year-end conversation has a fuller dataset behind it."
Bold the key insight: real time analytics and regular communication tools between manager and employee are not a convenience feature in a performance management system. They are the evidence base that makes performance reviews credible. Without them, managers are reviewing their own memory, not the employee's actual year.
What performance review examples work for employees who are strong on potential but weak on current delivery?
Separate potential from performance clearly, and be honest that the gap between them is itself a performance issue if it persists. Recognition of potential without accountability for delivery is not kindness. It is a message that standards are negotiable, which undermines company culture and demoralises the people delivering consistently right now.
McKinsey research on talent and performance consistently identifies the gap between assessed potential and realised performance as one of the most consequential and least addressed risks in talent management. High-potential employees who are not delivering are expensive. The review is the right place to name the gap.
Here are two worked examples:
Example 1 - High potential, delivery gaps: "There is genuine talent here, and it has been visible in the moments this year when the work came together: the client workshop in June and the internal strategy session in November both showed what this person is capable of. The challenge is that those moments were isolated. The day-to-day delivery across the year was inconsistent in ways that the team absorbed at a cost. The next cycle needs to close that gap. The expectation is not potential. It is consistent, reliable output that matches what we have already seen is possible."
Example 2 - Strong in visible moments, weak on sustained commitment: "This employee performs exceptionally in high-visibility situations. Project kickoffs, client presentations, and all-hands contributions are consistently impressive. The concern is the sustained work between those moments, where timelines have slipped and quality has varied. The business does not run on highlights. It runs on the work in between. Next year's review will assess that work directly, with agreed milestones reviewed quarterly and built into the action plan from day one of the cycle."
Both of these examples hold two truths at once. They do not let potential become an alibi. They also do not dismiss real strength. That is what fairness looks like in a performance review.
How do continuous feedback and pulse surveys change what managers can say in a performance review?
They give managers a documented, longitudinal record of employee performance instead of a compressed memory of recent events. When HR teams build continuous feedback and pulse surveys into the performance management cycle, year-end reviews become more accurate, less biased, and more defensible. The employee also arrives at the review having already processed most of the feedback, which changes the dynamic of the conversation.
The recency effect in performance evaluation is well documented. Without a continuous feedback record, managers systematically overweight the last six to eight weeks of the review period. Everything before that gets compressed or forgotten. Gartner's research on performance management identifies this as one of the primary drivers of perceived unfairness in annual reviews, and it is one of the reasons continuous feedback and real time analytics have become central to modern HR systems.
Practically, what this means is:
A manager who has conducted monthly check-ins, documented key moments of performance and contribution, and reviewed pulse survey results across the year can write a review comment that spans the whole twelve months with specific evidence. A manager working from memory writes about the last quarter and calls it a year.
For the hard-to-read employee - the quiet one, the inconsistent one, the technically strong but interpersonally challenging one - this longitudinal record is especially valuable. It converts a subjective impression into an evidenced account, which is the difference between a review that holds up and one that falls apart the moment the employee asks where the evidence is.
The implication for HR teams is clear. Investing in communication tools and feedback infrastructure between review cycles is not an administrative cost. It is the foundation that makes the review itself credible.
Frequently Asked Questions
What should a performance review say about an employee who is quiet but consistently delivers?
Acknowledge the output directly, then name the behaviours that produced it. Write something like: "Consistently delivers accurate, on-time work with minimal oversight. The team relies on this reliability, even when it goes unannounced. The next step is to bring this same standard of thinking into team discussions, where the perspective would add real value."
How do you write a performance review for someone whose results are inconsistent?
Be specific about the pattern rather than the person. Name the periods of strong output, name the periods where delivery fell short, and connect both to observable conditions rather than attitude. Avoid words like "unreliable" and instead describe what happened: "Q1 delivery met every milestone. Q3 saw three deadlines missed, two of which coincided with the restructure period."
What are good performance review examples for someone who is technically strong but difficult to work with?
Separate the two dimensions clearly. Address technical output in one paragraph with concrete evidence, then address collaboration in another. Write: "The analytical work this year was consistently among the strongest on the team. At the same time, three project leads raised concerns about communication during handoff stages. That gap needs to close in the next cycle."
How do you write a performance review when you do not have enough data on an employee?
State what you can verify and be honest about the gaps. Do not invent impressions. Write: "My direct visibility into this role has been limited this cycle, so this review draws on the work that came to my attention, peer input, and the output I can verify. A more regular check-in cadence would strengthen next year's review for both of us."
How long should a written performance review comment be?
Between 80 and 150 words per competency or theme is enough for most roles. Shorter than 80 words risks feeling dismissive. Longer than 200 words per section tends to dilute the core message. The goal is one clear observation, one piece of evidence, and one forward-looking statement per section.
Should performance review examples include ratings alongside written comments?
Written comments should always accompany ratings, never replace them. A rating without commentary leaves the employee guessing. A comment without a rating can feel evasive. The combination - a clear rating and a specific comment explaining it - is what gives the employee something to act on and what protects the manager if the rating is later questioned.
What is the difference between a development-focused and an accountability-focused performance review?
A development-focused review centres on growth, learning, and future potential. An accountability-focused review centres on whether the employee met agreed expectations and what the consequences are. Most reviews need both. The mistake is letting one crowd out the other: praising growth while ignoring a performance gap, or cataloguing failures without a path forward.
How do pulse surveys and continuous feedback improve year-end performance reviews?
They give managers a documented record of employee sentiments, challenges, and wins across the full year rather than the last six weeks. When HR teams run regular pulse surveys and encourage continuous feedback between reviews, the year-end conversation becomes a synthesis of real data rather than a memory exercise. That changes both accuracy and fairness.
Build the Evidence Before the Review Begins
The examples in this post work because they are specific. They are specific because they are built on a continuous record of observation, feedback, and documented evidence. Without that record, even the best-written performance review comment is a reconstruction, and reconstructions are where bias lives.
Kodecrew gives HR teams and managers the infrastructure to build that record throughout the year: pulse surveys that capture employee sentiments in real time, continuous feedback tools that document meaningful moments of performance as they happen, real time analytics that surface patterns before they become surprises, and action plans that connect review outcomes to tangible next steps. The year-end review becomes the easiest conversation of the year because the evidence was already there.
