Evidence-based performance management uses measurable results, clearly defined behavioural competencies and relevant context to make performance decisions more consistent, transparent and fair.
It combines two essential dimensions of performance: what an employee achieves and how the employee works. Key Results (KRs) provide evidence of outcomes, while behavioural competencies describe the observable behaviours, ways of working and capabilities expected in a role.
The process begins with clear expectations, continues through periodic performance conversations and concludes with a structured assessment. Critical business KRs receive regular attention, while business-as-usual and developmental KRs are reviewed at a cadence appropriate to their nature.
The aim is not to eliminate managerial judgement. It is to make the basis for judgement explicit, reduce avoidable bias and create a culture where managers and employees work together to achieve results.
Key Highlights of Evidence-Based Performance Management Define measurable KRs, targets and behavioural expectations before the performance period begins. Distinguish between critical business KRs, business-as-usual KRs and developmental KRs. Review critical business KRs periodically, using weekly check-ins where appropriate. Use KR achievement data as the factual basis for assessing the results component of performance. Define behavioural competencies through observable indicators and proficiency levels. Combine KR achievement and behavioural competencies using a method established in advance. Retain the original KR achievement figure while considering significant circumstances beyond an employee’s control through a transparent normalisation process. Maintain regular performance conversations so that appraisal is not the first time performance is discussed. Shift from blame and year-end surprises towards collaboration, timely feedback, support and course correction. Introduction Performance decisions influence compensation, development, career opportunities and employee confidence. Yet these decisions can be vulnerable to incomplete information, inconsistent standards and personal impressions.
A manager may remember a major success from the last month but struggle to recall what happened earlier in the year. Another may give disproportionate weight to a recent mistake. A third may assess an employee more positively because their communication style or approach feels familiar.
These situations do not necessarily arise from poor intentions. They can result from the way people interpret information, particularly when expectations are unclear or evidence is incomplete.
This is where evidence-based performance management becomes valuable.
In my view, the starting point is not the appraisal form or the rating scale. It is the way an organisation defines performance, translates expectations into measurable outcomes and observable behaviours, and creates opportunities to discuss progress throughout the performance cycle.
When performance is discussed as work unfolds, appraisal becomes a consolidation of conversations that have already taken place; not a once-a-year surprise.
The objective is to improve the quality of performance decisions while creating an environment in which employees understand what is expected, managers can explain their assessments, and both can work together to improve results.
Evidence-Based Performance Management Defined Evidence-based performance management is an approach in which employee performance decisions are supported by agreed expectations, measurable outcomes, defined behavioural criteria, documented observations and relevant contextual information.
It connects three questions:
What was the employee expected to achieve? What did the employee achieve? How did the employee work towards those outcomes? The first two questions are addressed through Key Results and their associated measures. The third is addressed through behavioural competencies.
Understanding Key Results Key Results describe the measurable outcomes an employee is expected to achieve. They should have clear measures, targets and calculation methods so that achievement can be assessed consistently. The structure comes from Objectives and Key Results (OKRs) .
For example, a KR may relate to revenue generated, project milestones achieved, customer retention, reduction in defects, improvement in turnaround time or another outcome relevant to the role.
A KR should make it possible to establish the result against an agreed expectation. Where a KR is measurable, the achievement figure provides a factual basis for assessing the results component of performance.
Understanding Behavioural Competencies Behavioural competencies are the observable behaviours, ways of working and demonstrated capabilities expected from an employee in a particular role.
They describe how an employee works, complementing KRs, which describe what the employee achieves.
For example, two employees may achieve similar business results but demonstrate different levels of collaboration, problem-solving, customer orientation or ownership. These differences may be important to their overall performance and future contribution.
Behavioural competencies should not be expressed as vague personality traits or general impressions. Each competency needs to be defined through observable indicators and, where appropriate, proficiency levels.
Consider collaboration as an example. Rather than simply stating that an employee should be a “good team player,” the organisation could define observable indicators such as:
Shares relevant information with colleagues in a timely manner. Works constructively across functions to resolve dependencies. Seeks and considers relevant inputs before making decisions. Contributes to collective outcomes rather than focusing only on individual success. Proficiency levels can then describe the difference between basic, effective and advanced demonstration of the competency. The indicators should reflect the requirements of the role.
The competencies selected will vary by role. Collaboration may be particularly important in a cross-functional role, while problem-solving may be central to a technical role. Customer orientation may be critical in a client-facing position.
The purpose is to make the assessment of how work is performed as clear and evidence-based as possible.
Combining Results and Behaviour An overall performance rating may combine KR achievement and behavioural competencies. The organisation must define in advance how these two dimensions contribute to the final assessment.
The method may involve weightages, rating rules or another clearly specified approach. There is no single formula that every organisation must follow.
What matters is that employees and managers understand the criteria before the performance period begins and that the method is applied consistently.
Evidence-based performance management therefore begins with role-relevant expectations, measurable KRs, defined behavioural competencies and a transparent assessment method.
Where Bias Can Enter Traditional Performance Management Bias can influence what a manager notices, remembers, considers important and ultimately includes in a performance assessment.
It can also enter when expectations are vague, evidence is incomplete or different managers apply different standards.
A structured approach cannot eliminate bias completely, but it can reduce opportunities for personal impressions to dominate performance decisions.
Recency Bias Recency bias occurs when recent events disproportionately influence an assessment.
A strong achievement in the final month may overshadow earlier difficulties. Equally, a recent mistake may dominate the manager’s view of an otherwise consistent year.
Periodic check-ins and records of meaningful outcomes help bring the entire performance period into view. They reduce the dependence on memory at year-end and allow the assessment to reflect a broader body of evidence.
Halo and Horn Effects The halo effect occurs when a positive impression in one area influences judgements about other areas. The horn effect works in the opposite direction: one negative impression colours the wider assessment.
For example, an employee who delivers a major project successfully may receive an overly positive assessment of their collaboration or people management, even when there is insufficient evidence in those areas.
Similarly, one difficult interaction may influence the manager’s judgement of the employee’s overall contribution.
Separating KR achievement from behavioural competency assessments helps prevent one result or incident from defining the entire performance assessment.
Leniency and Severity Bias Some managers tend to rate employees generously, while others apply more demanding standards.
When rating criteria are open to interpretation, employees doing similar work may receive different ratings simply because they report to different managers.
Clearly defined criteria, behavioural indicators and proficiency levels provide a common reference point. Calibration discussions can help identify inconsistent interpretations and improve the application of standards.
Calibration should not become a process of forcing ratings into a predetermined distribution. Its purpose is to examine the evidence, question inconsistencies and ensure that the agreed criteria are being applied appropriately.
Similarity Bias Similarity bias occurs when managers respond more positively to employees whose style, background, interests or approach resembles their own.
An employee who communicates or solves problems differently may be judged less favourably, even when their contribution is effective.
Defining the behaviours expected in a role helps shift the discussion from personal preferences to observable evidence.
The question becomes: “What behaviour did the employee demonstrate, and how does it compare with the agreed expectations?” rather than “Does this employee work the way I prefer?”
Regular conversations, documented evidence and consistent criteria help reduce the influence of these biases. They do not remove the need for managers to remain aware of their own assumptions.
What Evidence-Based Performance Management Looks Like in Practice A practical system connects expectations, progress conversations, evidence and decisions. It does not require documenting every activity or creating an elaborate administrative process.
It requires clarity about what matters, the right conversations at the right time, and a consistent method of assessing performance.
An important starting point is to distinguish between different types of KRs because they do not all require the same level of attention or the same review cadence.
Three Categories of Key Results Critical / Business KRs Critical KRs are directly linked to important business priorities and outcomes. They may relate to revenue, profitability, strategic projects, customer outcomes, delivery commitments or other priorities that require close attention.
These KRs need periodic check-ins because delays, risks or dependencies may have a significant impact on business results.
The frequency should reflect the nature of the KR. Weekly check-ins may be appropriate for a time-sensitive project or a critical business commitment. Other KRs may require a different cadence.
The check-in should focus on:
Progress against the agreed measure or milestone. Actions taken since the previous discussion. Challenges, risks and dependencies. The employee’s response to those challenges. Support or collaboration required from the manager or other teams. Agreed actions and follow-up. The purpose is not to monitor every activity. It is to understand progress, identify red flags early and enable timely action.
Business-as-Usual (BAU) KRs BAU KRs represent the ongoing responsibilities required to run the business or deliver a service consistently.
Examples may include maintaining service levels, meeting routine delivery commitments, processing transactions accurately, maintaining quality standards or ensuring operational compliance. Many organisations track these as KRAs. See OKR vs KRA for how the two differ.
These KRs should have clear measures and targets. Their review cadence should be appropriate to the nature of the work and the frequency at which meaningful performance information becomes available.
A daily operational measure may be monitored frequently, while a monthly service-level outcome may be reviewed monthly. The aim is to maintain performance without creating unnecessary meetings.
Developmental KRs Developmental KRs focus on building future capabilities, skills or behaviours. They may support an employee’s ability to take on broader responsibilities, improve effectiveness in the current role or prepare for changing business requirements.
These KRs should define the intended development outcome, not merely the activity to be completed.
For example, attending a training programme is an activity. Demonstrating the ability to independently apply a newly learned skill in a work situation is a development outcome.
Developmental KRs should therefore be assessed against agreed evidence of learning, application and improvement. Completing a course or attending a workshop may contribute to the outcome but should not automatically be treated as proof that the capability has been developed, the same distinction the Kirkpatrick model of training evaluation draws between learning and behaviour.
Distinguishing these three categories helps organisations determine which outcomes need close business attention, which require consistent operational monitoring, and which support future capability.
Using Performance Data Alongside Manager Judgment Data establishes the result against an agreed measure, but it does not explain every factor behind that result.
A missed KR may reflect execution challenges, changing scope, dependencies, resource constraints, market conditions or decisions made elsewhere in the organisation.
The manager’s role is to understand the circumstances without losing sight of the agreed outcome.
For a critical KR, periodic check-ins help establish what is happening while there is still an opportunity to respond. They allow the manager and employee to discuss progress, actions, challenges and support required.
This is not about replacing data with opinion. It is about interpreting the data in context and deciding what action is appropriate.
The employee remains accountable for their contribution and response. At the same time, the manager and organisation have a responsibility to address barriers, clarify priorities and provide support where required.
Structured and Consistent Rating Criteria KRs describe results to be achieved. Behavioural competencies describe expected ways of working and demonstrated capabilities.
Both may contribute to the overall performance rating.
The organisation should specify in advance:
The KRs and measures relevant to the role. The targets and calculation methods. The behavioural competencies and observable indicators. The proficiency levels used to assess those competencies. The method for combining KR achievement and behavioural assessments. The process for considering significant contextual factors. The precise method may vary by organisation. Some may assign explicit weights to results and behaviours, while others may use defined rating rules.
The important point is that the approach is transparent, understood in advance and consistently applied.
Managers should not decide at the end of the year how much importance to give a KR or competency simply because the outcome is inconvenient or unexpected.
Documenting Performance Evidence Throughout the Review Cycle Periodic check-ins create an evidence trail of progress, decisions, obstacles, collaboration and actions.
A simple record, such as a goal tracking sheet , can capture:
The KR or competency discussed. Progress against the agreed measure or expectation. Significant outcomes or observable behaviours. Challenges, risks or dependencies. Actions taken and the employee’s response. Support provided or required. Agreed next steps. The aim is useful evidence, not paperwork.
For behavioural competencies, documentation should describe observable actions and their impact rather than rely on labels such as “not committed,” “difficult” or “not a team player.”
For example, instead of writing “poor collaboration,” a manager could record that a critical dependency was not communicated to the relevant team until a deadline was at risk. The discussion can then explore the circumstances, impact and expected behaviour.
This approach makes feedback more specific and gives the employee a clearer basis for improvement.
The Research Behind Evidence-Based Performance Management Performance assessment involves human judgment. A disciplined process improves the information available to managers and makes the basis for decisions clearer. Researchers have applied behavioural science to performance management to show how clear expectations and frequent feedback improve outcomes.
The principles discussed here—defined criteria, relevant evidence, timely feedback and consistent application—are intended to address common weaknesses in performance assessment. They should not be interpreted as a guarantee that ratings will become perfectly objective.
What Research Says About Performance Ratings A performance rating is an assessment based on information about work; it is not a direct measurement of a person’s entire contribution or potential.
Its quality depends on the clarity of expectations, the relevance of the evidence and the consistency with which criteria are applied.
When managers rely heavily on memory or broad impressions, assessments can be influenced by events that are recent, memorable or personally significant.
A structured process that captures performance evidence across the cycle can provide a more complete basis for the final assessment.
What Research Says About Feedback Feedback is more useful when it is specific, timely and connected to observable actions and outcomes, a finding supported by the CIPD’s evidence review on performance feedback .
Rather than saying “you need to improve ownership,” a manager can describe a situation, the action or behaviour observed, its impact and what is expected going forward.
Timely conversations also give employees an opportunity to respond while there is still time to act.
This is particularly important for critical business KRs. Waiting until the annual appraisal to discuss a significant delay or performance concern limits the opportunity for course correction.
What Research Says About Rating Bias No performance management process guarantees the complete elimination of bias. SHRM explains why most performance evaluations are biased and how to fix them.
However, clear criteria, defined behavioural indicators, regular evidence-based conversations and calibration can reduce avoidable inconsistency.
The effectiveness of these practices depends on how they are implemented. A well-designed form will not make a process evidence-based if managers continue to rely on personal impressions or discuss performance only at year-end.
The real change is in the discipline of defining expectations, examining evidence and having timely, meaningful conversations.
Data vs. Judgment: What Should Managers Use? My view is that performance ratings should be data-driven wherever work can be meaningfully measured.
If a KR is intended to measure an outcome, it should have a defined measure, target and calculation method. Without these, the organisation risks treating a subjective assessment as though it were a numerical fact.
Where the measure is sound, the percentage achieved provides a factual basis for assessing the KR result.
Numbers have no emotions. They do not favour an employee because of a personal relationship, nor do they remember only the most recent event. They establish what was achieved against the agreed measure.
However, numbers do not explain everything.
A result may be affected by factors outside the employee’s control. A major change in scope, a business decision, an external disruption or a dependency that was not resolved may materially affect the outcome.
This is where judgement and context matter.
Normalisation: Recognising Context Without Distorting the Evidence Normalisation provides a structured way to consider significant circumstances that materially affected KR achievement.
Suppose an employee was responsible for delivering an agreed business outcome, but a major organisational decision or an external event substantially changed the conditions under which the target could be achieved.
The original achievement figure should remain factual and visible. The organisation can then consider whether the circumstances justify an adjustment to the assessment through a defined and transparent process.
Normalisation does not mean changing a number simply because the result is disappointing or because a manager wishes to give an employee a better rating.
It means examining the circumstances, establishing their relevance, documenting the rationale and applying the agreed process consistently.
The original result, the contextual consideration and the basis for any adjustment should remain clear.
Assessing Behavioural Competencies Behavioural competencies require judgment because they assess how an employee works. But this judgement should be anchored in defined competencies, observable indicators and proficiency levels.
The manager should be able to explain the assessment through specific examples from the performance period.
A statement such as “strong ownership” is not sufficient on its own. The assessment should explain what the employee did that demonstrated ownership and how that behaviour compared with the expectations for the role.
Similarly, a concern about collaboration should be supported by observable examples rather than a general impression.
Combining the Two Dimensions The overall performance rating combines KR achievement and behavioural competency assessments according to the organisation’s pre-defined method.
The method may vary. What should not vary arbitrarily is how it is applied to employees.
The principle is straightforward:
Use measurable data for results. Use defined criteria and observable evidence for behaviours. Use informed judgement to understand relevant context. Apply the agreed method consistently. The objective is not to remove judgement, but to make it visible, explainable and disciplined.
How to Introduce Evidence-Based Practices Without Overcomplicating Reviews The answer is not more forms, more meetings or more documentation. It is a practical rhythm that supports performance throughout the cycle.
Agree on Expectations at the Beginning Before the performance period begins, managers and employees should establish the KRs, measures, targets and behavioural expectations relevant to the role.
The employee should understand what success looks like, how it will be measured and how the overall performance rating will be determined.
Identify Critical Business KRs Not every KR needs the same frequency of review.
Identify the KRs that are directly linked to important business priorities and require closer attention. Establish a suitable check-in cadence, which may be weekly for time-sensitive or high-impact commitments.
BAU and developmental KRs can follow review rhythms suited to their nature.
Make Check-ins Short and Useful A weekly check-in should be a working conversation, not a miniature appraisal meeting.
It should focus on progress, red flags, actions, dependencies, collaboration, appreciation and support.
The manager and employee should leave with a shared understanding of what needs to happen next.
Record Significant Evidence and Actions Capture meaningful outcomes, observable behaviours, challenges, decisions and agreed actions.
Avoid turning the process into a record of every email, meeting or activity. Documentation should help explain performance and support better decisions.
Address Concerns When They Arise If progress is falling behind or a behavioural concern emerges, discuss it in a timely manner.
The conversation should clarify the expectation, understand the situation, identify what needs to change and agree on the support or action required.
This reduces the likelihood of year-end surprises and gives employees a genuine opportunity to respond.
Consolidate the Evidence at Appraisal At the end of the cycle, review the evidence from the entire period.
Assess KR achievement against agreed measures, evaluate behavioural competencies against defined indicators, consider relevant context and apply the established rating method.
The appraisal should bring together the work and conversations of the year—not introduce a new set of expectations.
How to Build a More Evidence-Based Performance System An evidence-based system is built through a series of connected design choices. The organisation does not need to implement everything at once, but the essential elements should reinforce one another.
Define Performance Clearly Begin by establishing what performance means in the organisation and in different roles.
Define the outcomes employees are expected to achieve, the behavioural competencies relevant to their work and the evidence that can reasonably demonstrate performance.
Avoid using the same generic measures for roles that have fundamentally different responsibilities. HR OKRs offer useful examples of role-specific outcomes.
Design Measurable KRs Each KR should have a clear outcome, measure, target and calculation method.
Where a KR is directly linked to a business priority, identify it as a critical business KR and establish the appropriate check-in cadence.
Distinguish these from BAU KRs and developmental KRs so that the organisation can review each appropriately.
Define Behavioural Competencies Select competencies relevant to the role and describe them through observable indicators. Establish proficiency levels where needed, making clear what effective and more advanced demonstration looks like.
This creates a common reference point for managers and employees and reduces reliance on vague descriptions or personal preferences.
Establish the Rating Method in Advance Define how KR achievement and behavioural competencies contribute to the overall rating.
Clarify the role of any weightages, rating scales, decision rules, calibration and normalisation.
The method should be communicated before the performance period begins and applied consistently.
Build a Rhythm of Performance Conversations Create a practical cadence for check-ins and reviews.
Critical KRs may need weekly conversations; other KRs may be reviewed monthly, quarterly or at another appropriate interval.
The purpose of these conversations is to understand progress, recognise achievements, address challenges and agree on actions.
Enable Managers to Have Better Conversations Managers need to know how to discuss results, provide specific feedback, assess behaviours and distinguish evidence from assumptions.
They should also understand how to discuss circumstances beyond an employee’s control without removing accountability for the employee’s own actions and decisions.
The quality of these conversations is central to the effectiveness of the system.
Review Consistency and Improve the Process Calibration can help identify differences in how managers interpret criteria. Periodic reviews of the system can reveal whether measures are useful, expectations are clear and the process is creating unnecessary administration.
The organisation should use these insights to improve the design over time.
Technology can support goal tracking , check-ins, documentation and analytics. However, technology should follow the performance design; not dictate it.
Evidence-Based Performance Management Example Consider a project manager responsible for delivering a strategic project.
The organisation establishes three types of KRs:
Critical / Business KR: Achieve agreed project milestones within the approved timeline and budget. BAU KR: Maintain agreed standards for project reporting, risk management and operational governance. Developmental KR: Strengthen cross-functional stakeholder management, demonstrated through agreed improvements in coordination and resolution of dependencies. The project manager also has behavioural competencies, including collaboration, problem-solving, ownership and communication. Each competency is defined through observable indicators and relevant proficiency levels.
During the Performance Period The critical project KR is reviewed weekly because delays could affect business commitments.
The check-in examines milestone progress, actions taken, emerging risks, dependencies and support required. If a delay is likely, the manager and employee discuss the response rather than waiting for the annual appraisal.
The BAU KRs are reviewed at a cadence appropriate to project reporting and governance requirements.
The developmental KR is reviewed through evidence of application—for example, how effectively the project manager coordinates across functions and resolves dependencies. Simply attending a stakeholder management course would not, by itself, demonstrate that the capability had improved.
At the End of the Period The project manager’s KR achievement is assessed against the agreed measures.
Suppose a significant external change affected a project milestone. The original achievement figure remains visible. The organisation examines the circumstances and determines, through its defined normalisation process, whether the impact warrants an adjustment to the assessment.
The behavioural competencies are assessed separately against their defined indicators and proficiency levels.
The overall rating is then determined using the method communicated at the beginning of the period.
This example illustrates how results, behaviours and context can be considered without allowing one dimension to replace the others.
What Should Not Become Performance Data Not everything that can be counted is meaningful evidence of performance.
Emails sent, hours spent online, meetings attended or the number of tasks completed should not automatically be treated as indicators of contribution.
Such measures may be relevant in specific roles or circumstances, but only when they are clearly connected to role expectations and meaningful outcomes.
Poorly chosen metrics can encourage the wrong behaviour. For example, measuring the number of customer calls without considering their purpose or effectiveness may encourage volume rather than customer outcomes.
Similarly, measuring training attendance without assessing learning or application can create the appearance of development without demonstrating improved capability.
Performance data should be relevant, proportionate and connected to what the role is expected to achieve.
The organisation should also be mindful of privacy. Collecting more information does not automatically improve performance decisions. Data should be collected for a clear purpose, accessed appropriately and used in a way employees can understand.
An Evidence Hierarchy for Performance Decisions Different forms of evidence serve different purposes. A useful performance system recognises what each can and cannot establish.
Measured KR achievement and documented deliverables provide evidence of outcomes against agreed expectations. Where measures and calculation methods are sound, these are the factual basis for assessing results.
Check-in records provide evidence of progress, actions, challenges, dependencies, responses and support. They help explain how performance developed over the period.
Observed behaviours assessed against defined competencies provide evidence of how an employee works. The assessment should be linked to observable indicators and relevant proficiency levels.
Contextual information; such as market changes, resource constraints, scope changes, organisational decisions or dependencies; can help explain circumstances that affected results. Context should be relevant and supported, not used as a general excuse.
Unsupported impressions and personal preferences should carry little weight. Statements such as “not committed,” “not a team player” or “not leadership material” are not sufficient evidence without specific, role-relevant observations.
The hierarchy is not a rigid formula in which one type of evidence always overrides another. It is a way to distinguish facts, observations, context and opinion so that each is used appropriately.
Good performance decisions recognise what each piece of evidence can establish and where judgement is still required.
How NextAgile Brings Evidence-Based Design Into Performance Systems NextAgile approaches performance management as an operating system connecting goals, evidence, conversations and decisions, not simply as an annual HR exercise.
The work begins by examining how the organisation defines performance, sets measurable KRs, conducts check-ins, assesses behavioural competencies and arrives at ratings.
The objective is to establish clear expectations, practical performance rhythms and a transparent method for using evidence and judgement.
The approach can be structured around six stages:
Diagnose: Identify unclear or inconsistent expectations, evidence, conversations and performance decisions. Define: Establish measurable KRs, relevant behavioural competencies, observable indicators and assessment criteria. Evidence: Determine what information should be captured, how it will be reviewed and how it will contribute to performance decisions. Enable: Support managers in conducting useful check-ins, discussing performance concerns, providing feedback and agreeing on actions. Calibrate: Examine whether criteria are being interpreted and applied consistently, while retaining the distinction between results, behaviours and context. Improve: Use feedback and insights to strengthen the performance system over time. The design also considers the cultural shift required. Employees and managers need to move away from treating appraisal as a process of defending or justifying a rating. Instead, performance conversations should focus on clarity, progress, accountability, collaboration and improvement.
Technology can support goal tracking, check-ins, and analytics, but it should follow the performance design rather than dictate it.
The objective is better conversations and decisions without unnecessary administration.
For organizations reviewing their wider performance model, performance management consulting can help connect performance architecture, evidence, feedback, manager practices, technology, and decision-making into one coherent system.
The Delegation & Feedback Workshop can support the manager behaviors required for stronger performance conversations.
The same principle applies to accountability.
Employees should understand what they are expected to achieve, how their performance will be assessed, and what evidence will inform important decisions.
An ownership and accountability workshop can complement the broader performance design by strengthening the behaviors that make performance expectations meaningful.
The objective is not to create a more complicated HR process. It is to create a more disciplined way of making decisions about people.
Conclusion Evidence-based performance management is not about replacing people with numbers. It is about using measurable results wherever possible, defining behavioural expectations and making judgement transparent.
KRs establish what an employee is expected to achieve. Behavioural competencies describe the observable behaviours, ways of working and capabilities expected in the role. Together, they provide a more complete view of performance.
Critical business KRs require periodic attention so that progress, risks, dependencies and support needs can be addressed while there is still an opportunity to act. BAU KRs and developmental KRs require review rhythms and assessment approaches suited to their purpose.
At the end of the performance period, KR achievement provides the factual basis for the results component. Behavioural competencies are assessed against defined indicators and proficiency levels. Significant circumstances beyond an employee’s control may be considered through a transparent normalisation process, with the original achievement figure retained. The method for combining these elements should be defined in advance.
But the process is only one part of the change.
The larger shift is cultural: from blame to collaboration, from year-end surprises to real-time conversations, and from defending a rating to working together to achieve results.
Accountability remains essential. The difference is that accountability is practised through clear expectations, timely conversations, transparent decisions, appropriate support and follow-through. A clear accountability framework makes this explicit.
When performance is discussed regularly, employees have a better understanding of where they stand and what they need to do. Managers have a stronger evidence base for their decisions. Organisations can make performance assessments more consistent without pretending that human judgement can be eliminated.
The aim is not perfect objectivity. It is a fairer, clearer and more disciplined way to understand performance and to help people and the business achieve better outcomes together.
Frequently Asked Qusetions What types of evidence are most useful in employee performance decisions? Useful evidence includes measurable KR achievement, agreed deliverables, progress documented during check-ins, relevant outcomes and observable behaviours assessed against defined competencies.
The evidence should be connected to the employee’s role and the expectations agreed at the beginning of the performance period.
Can performance data eliminate bias completely? No. Data can reduce dependence on memory and personal impressions, but interpretation still involves judgement.
Clear criteria, regular evidence-based conversations, defined behavioural indicators and calibration can reduce avoidable bias and inconsistency. They cannot guarantee that every decision will be completely free of bias.
How should managers handle performance evidence that is incomplete? Managers should acknowledge the gap, seek relevant information and discuss the situation with the employee rather than fill gaps with assumptions.
Where evidence is unavailable, the assessment should reflect that limitation. Unsupported impressions should not be treated as established facts.
How should external factors affect a performance rating? The original KR achievement figure should remain visible.
If a significant circumstance beyond the employee’s control materially affected the result, the organisation may consider normalisation through a defined and transparent process. The circumstances and rationale for any adjustment should be documented.
Normalisation should not be used simply because the result is disappointing.
How can behavioural competencies be assessed fairly? Define the competencies and proficiency levels in advance, using observable, role-relevant indicators.
Managers should support their assessments with specific examples from the performance period. The organisation should also clarify how behavioural competency assessments contribute to the overall rating.
What is the difference between critical, BAU and developmental KRs? Critical or business KRs are directly linked to important business priorities and may require frequent check-ins, including weekly conversations where appropriate.
BAU KRs cover ongoing operational responsibilities and should be reviewed at a cadence suited to the work.
Developmental KRs focus on building future capabilities, skills or behaviours. They should be assessed against agreed development outcomes and evidence of application, not merely activity completion.
Why are periodic check-ins important? Periodic check-ins help managers and employees understand progress, identify risks, discuss challenges and agree on actions while there is still time to respond.
They reduce dependence on memory and make the annual appraisal a consolidation of ongoing conversations rather than the first discussion of performance.
Should every KR be reviewed weekly? No. The cadence should reflect the nature and importance of the KR.
Critical business KRs may require weekly check-ins when progress, risks or dependencies need close attention. BAU and developmental KRs may be reviewed at different intervals, depending on the work and the outcomes being assessed.
How should an organisation combine KR achievement and behavioural competencies? The organisation should define the method before the performance period begins. It may use weightages, rating rules or another transparent approach.
There is no single formula that suits every organisation. The important requirements are clarity, consistency and a clear explanation of how results and behaviours contribute to the overall rating.
Does evidence-based performance management work for creative and knowledge workers? Yes, provided the measures and evidence are appropriate to the work.
Creative and knowledge-based roles may require a combination of outcome-based KRs, quality of deliverables, problem-solving, collaboration and other role-relevant behavioural competencies.
The aim is not to measure every activity. It is to establish meaningful expectations and assess contribution using relevant evidence.
What is the cultural change required for evidence-based performance management? The shift is from blame to collaboration, from year-end surprises to timely conversations, and from defending a rating to working together to achieve results.
This does not mean lowering accountability. It means creating clarity about expectations, discussing challenges early, providing appropriate support and following through on commitments.
What privacy concerns arise when organisations collect performance data? Organisations should collect only relevant information, make the purpose of collection clear, restrict access appropriately and use the data consistently with their policies and applicable requirements.
More data does not automatically mean better decisions. The focus should be on meaningful evidence that supports performance conversations and assessments without unnecessary monitoring or administration.
OKR Coach & HR Consultant
I work with organizations to implement the strategy through OKRs & Performance Management for business success.
Skip to content 
