fbpx

Performance Evaluation: What is it? How does it work?

Renan Araújo

produto-de-sucessao

Performance appraisal is a structured assessment of a team member’s individual performance over a given period of time, the goal of which is to improve the company’s performance by improving employee performance.

The performance appraisal is the culmination of a well-structured performance management cycle – and performance management is a systematic process of improving an organization’s performance through improving the performance of its teams and people.

But structuring an efficient performance evaluation that produces the desired results (i.e.

that makes the company improve) is a bumpy road full of forks and pitfalls.

In this article, you will see some of the basics and the main choices of the process. Check it out!

The relationship between performance management and performance analysis

Performance management is a process that takes place in cycles: it begins with the establishment of performance expectations (which are goals, indicators, competencies, and behaviors) and ends with the assessment of that performance and the subsequent start of a new cycle.

Performance appraisal is the closing of the performance management cycle, where the intention is to measure and evaluate the employee’s performance in relation to the expectations established at the beginning of the cycle.

The context of performance evaluation

The organizational/industrial psychology literature devotes much of its effort to the study of performance appraisal. Much of the literature is focused on how to structure performance appraisal questionnaires that:

  • increase satisfaction with the process; improve the perceived fairness of the process;
  • increase the accuracy of the process;
  • Moreover, traditional literature deals almost exclusively with performance in its individual aspect.
  • However, there are several occupations and performance areas within organizations where measuring individual employee performance is much

more difficult and even sometimes undesirable nowadays, as in the cases of companies that organize themselves into agile teams, such as Spotify.

“Performance” in performance appraisal

The vast majority of experts divide a professional’s performance into two major dimensions:

results; ∑ behavior.

An easy way to understand the difference between results and behavior is to think that results are “what” the team member does (i.e. the results he achieves) and behavior is “how” the employee produces those results.

The two dimensions of performance are important because they form the backbone of the performance management cycle – which we will discuss later – and are how the vast majority of performance appraisals are organized.

The performance management cycle consists of two phases: setting expectations and assessing performance against the set expectations.

The main objectives of performance evaluation are:

  • measuring employee performance for decision-making purposes;
  •   provide inputs for the team member to develop, that is, to improve his performance in the next cycle.

The structure of a performance appraisal

A performance appraisal usually consists of a series of questionnaires. There will be questions to be answered in them by one or more appraisers about an appraisee, addressing aspects related to the two dimensions of performance we have discussed: behaviors (“how”) and results (“what”).

The history of behavior in performance appraisal

Behaviors, which form one of the two dimensions of performance, are assessed in a portion of the performance appraisal that is commonly called “competency-based appraisal.”

The use of competencies and behaviors in performance management gained momentum in the 1970s and 1980s, when companies were coming off a few years of huge focus on the use of goals and objectives as the sole measure of performance.

The use of goals (as we will see later) was born with the Management by Objectives fad. But at some point, organizational psychologists (the theorists) and human resources departments (the practitioners) missed some behavioral component that measured how people had achieved the results set forth in the MBOs – in particular, whether the results had been achieved ethically, collaboratively, and in line with the company’s culture.

Thus, the use of competencies and behavior was born, although they had already been widely used in job and occupation analysis, for performance management purposes.

The behaviors today

In modern performance appraisals, there is usually one component of the appraisal that deals with “competencies” and another that deals with “results”.

Here, we will use the terms “competencies” and “behavior” interchangeably.

Types of behavior are usually assessed one by one and can be grouped based on common themes.

In some companies, a composition is made between behavior derived from the values, the critical competencies, and the functional competencies of the job. In other one, which opt for simplicity, only the critical competencies are assessed.

Scale in performance appraisal

Another defining point of performance appraisal, especially in its behavioral component, is the rating scale to be used.

The scale has three major variables to be defined:

  • the scale to be used;
  • the label for each of the options on the scale;
  • the use of behavioral rules anchored in behavior.

Some companies discuss the merits of an even-numbered scale (for example, of 4 options), because on an even-numbered scale there is no “middle ground,” as being a way to avoid the central tendency of the ratings – which leads most raters to choose the central, or middle, options in their ratings.

So the first decision is to define whether there will be an even or odd number of options in the scale. Other companies argue the merits of scales with more options vis-a-vis scales with fewer options. The argument for using larger scales, such as the 5-option scale, is to make the ratings more accurate as each rater can be more precise in differentiating his ratings.

Labelling of scales

A second important aspect is the labeling of the chosen options. In some cases, companies choose to leave the options with numeric labels (for example 1, 2, 3, 4 and 5).

Other companies choose to replace the numbers with a scale of concepts (for example, “much lower than expected,” “lower than expected,” “within expected,” “higher than expected,” and “much higher than expected”). Some companies also choose to keep numbers and concepts side by side.

This brings us to a third set of options, related to the type of scale chosen. There are two

major types of scales: relative scales and absolute scales.

Relative scales require the appraiser to evaluate the appraisee relative to something. It can be, for example, relative to what is expected of the job (which we find quite efficient), as well as it can be relative to peers – that is, the rater evaluates the rater’s relative performance position relative to the group (which we find much less efficient as a rating scale label).

Absolute scales, on the other hand, require the evaluator to evaluate his or her evaluated in an absolute way, without any relation to what is expected or to other evaluated people. This is the case with the “bad”, “average” and “good” scales).

Anchored scales

More sophisticated companies can make use of Behaviorally Anchored Rating Scales (BARS), which are nothing more than descriptions of the behaviors observable in each of the ratings/concepts of the rating scale.

It would be like, in the case above, using the description “Think like an owner in most situations, defining strategies from beginning to end and considering all possible risks and threats” tied to the score 5 of the “Think-Like-an-Owner” behavior.

Behavioral rules can increase the accuracy of the assessments being made, but are very difficult to do without a huge time commitment and the assistance of specialized psychologists.

The results of the Performance Evaluation

Results are the most controversial dimension of performance to be evaluated. Simply put, if the company is at a more basic stage of maturity in its appraisal process, the appraiser can evaluate the responsibilities and activities outlined in the job description on a given rating scale.

Thus, a financial analyst could be rated on “Produce the cash position reports without errors and in a timely manner” on a 5-option scale that evaluates his performance against

what is expected of his position. In this case, the “results” portion of the performance appraisal ends up looking a lot like the “behaviors” or “competencies” portion.

In cases where team members’ responsibilities are measured by indicators (for example, in the case above, the employee might be evaluated on the quantity and quality of reports produced), these indicators ultimately serve as a proxy for the employee’s performance and are used in this portion of the evaluation.

Many companies, however, end up simplifying the performance evaluation process into just one section composed of competencies and behavior, without a clear distinction between what is “result” and what is “behavior.” For these companies, management maturity is often lacking so that there are enough indicators (and especially the measurement infrastructure needed to produce these indicators) to feed the performance management process.

Tips for starting a performance appraisal process

If your company has never gone through a performance evaluation process, it is reasonable to think that we can assume a few things about it:

  • employees have low maturity: followers have low professional maturity, and leaders have, most likely, low maturity as people managers;
  • the company has low maturity: there is little culture of measuring results, which is observed through well-defined KPIs per area, management by sight, cadenced management processes (such as weekly planning meetings and monthly result meetings);
  • the company does not have (or has only recently had) a well-defined job and salary program or corporate competencies.

It also makes sense to assume that what you need now is performance development. In that sense, you need to bring about clear, direct conversations about performance – that generate clear, actionable feedback and feedforward – aimed at the following benefits:

  • “the company” (the magical ensemble of its people) gradually understands what is expected of each one;
  • it creates a development culture;
  • it improves its performance by improving people’s performance.

In other words, inputs for decisions such as bonuses, promotions, and career planning are absolutely secondary in this incipient phase of performance management.

So it makes very little sense to start doing a performance evaluation with something complex. And by complex, you will see, we mean just about anything other than a few text questions.

The model suggested by Qulture.Rocks

What we suggest at this stage is to start your performance appraisal with a set of 3 questions that each manager should answer for his or her subordinates. We call this process the performance check-in.

Manager’s evaluation

In what can the leader improve (results and behaviors)?

In this question, the manager must make an analysis of what is not going so well, how this is negatively impacting his or her subordinate, the team, and the company, suggesting clear and objective changes (that can be transformed into action plans) that the subordinate must adopt to reverse the situation. So remember, the answer to this question has three parts:

1. What is not going so well?

How does this negatively impact the individual being led, the team, and the company? How can it be improved in practice?

2. What is the leader doing well (results and behavior)?

Here, things work analogously to question 1. What changes is that the manager must identify the points that are going well:

  • How does this positively impact the individual being led, the team, and the company?
  • How to keep doing well in this activity/area/delivery, or how to leverage the results of this “strength” through mentoring with other employees?

What should be the priorities for the next X months (deliverables, results and employee development)?

In this question, leader and protégé define high priorities for the next work cycle. If the company works with OKRs (Objectives and Key-Results) or goals, much of this discussion revolves around what the leader’s goals should be. If there is no formal practice, the manager should point out priority areas of focus or projects and deliverables that the led should accomplish in order for the company to get closer to its goals or mission.

On the behavioral side, one or two focus areas should be identified where the leader should focus their development efforts (probably one or two themes identified in question 1, “can improve”). Something like the MVP of an Individual Development Plan.

Self-assessment (Optional)

An additional possibility to the three questions above is some kind of self-assessment, which brings more reflection and self-knowledge to the leader in the process. This is the only valid motivation for the company to include some kind of self-assessment in the process. In our view, the self questions would be very similar to the manager evaluation questions:

1. What were the main results achieved and deliveries made in the last X months?

Here is a retrospective of the last X months (where X is the time period between performance appraisals), in which the employee should discuss the results he has achieved and the efforts he has made to achieve them.

Results are practical impacts on KPIs (sales, rates, ratios, etc.) and project deliverables (implementations, studies, etc.). A discussion ofMission-2-Metrics should be encouraged: how the leader understands that he has helped the company get closer to its goals and mission with his individual efforts and results.

2. In what can the led/you improve (results and behaviors)?

The same answer structure as for questions 1 and 2 for the manager should be applied.

3. How well are you/leaders doing (results and behaviors)?

Ditto.

Now that we have discussed what we think your performance evaluation should look like, let’s get to the FAQs related to our suggestion.

There is no “note”.

There are no notes. Go ahead and deal with that.

First of all, think about it: do you really need grades or concepts? Any attempt to make performance evaluation more quantitative is based on the need to differentiate employees

from each other. Differentiation is for decision making: who should be promoted, who should get raises, who should get bonuses, etc.

If your company has no bonus program, or if it is small (fewer than 500 employees), or if it has few openings or few internal candidates for the openings that are opening up every cycle, you don’t need a structured differentiation process. Managers, executives and HR know who is prepared or who has been performing well enough for the few people decisions to be made.

Second, your impression that grades make performance evaluation less subjective is just that: an impression, and a very subjective one at that. Want an example? “Rate from 1 to 5, with 1 being a little and 5 being a lot, how subjective you find an evaluation without grades…”

Subjectivity

Performance appraisals are always subjective. And making them less subjective (with People Analytics, lots of training, and calibration sessions or collegiate evaluation) takes a very high investment. Too high for this stage of your company’s performance management journey.

If the company doesn’t have significant maturity regarding what is expected from each one, what are the main result levers, and what is the evaluation “ruler”, the quantitative evaluation process becomes nothing less than a garbage-in, garbage-out process, that is, the final numerical results are worth very little.

Thirdly: what matters for performance management (measuring and improving) is the quality of the feedback and feedforward being exchanged: how, in practice, the employee can and should improve. Knowing it’s a 3, or 4, is not actionable. On the contrary, it is most likely a major motivation killer.

Common questions about performance appraisal

Check out the main questions related to performance evaluation!

How often should I do the performance appraisal?

According to Jeff Immelt, CEO of GE since 2001, “in this day and age, doing anything once a year doesn’t make sense. It’s just bizarre. Unless you work in one of the most stagnant government offices, a year is too long for any process.

Even more so with a young workforce, composed increasingly of millennials, a year is time for your employee to die of anxiety.

So our suggestion is that to start doing performance appraisal you do your cycle (which we suggest to be called performance checkin) at least 4 times a year. Depending on the stage of your company (the younger, the more frequent) or average employee demographics (the younger, the more frequent), a shorter cycle might be even better.

Who should evaluate whom?

To begin with, managers evaluate subordinates. If you are inspired, let your employees evaluate themselves. This increases their degree of self-knowledge.

Peers should give feedback to peers, but wait a while to insert this 360-degree feedback component into your performance management process: it can, and should, come later, only after the basic process is well established and mature.

What about the values of the organization’s culture?

In a very simple place: ask your leaders to take into account the values and behaviors of your culture when asking the “doing-well” and “could-do-better” questions. Ask them to identify examples of situations in which they represented – or should have represented – the values, and ideas for how they can do so more often and more intensely.

Our intent here was to help you learn how to get started with performance appraisal. We hope we have helped you!

Calibration of results

A very important step in a performance evaluation process is the calibration of its results.

Goals of results calibration

Calibration of results is a series of discussions among the appraisers that are intended to make the appraisals more accurate. In many companies, calibration is a meeting in which managers discuss the evaluations they have done on their subordinates with each other, so that any discrepancies in the evaluation criteria used are brought to light.

When comparing “live” the evaluations made by different managers about their subordinates, possible injustices can become very evident, such as some managers tending to be “nicer” with their team while others tend to be more demanding with theirs.

Results of the evaluations

In some companies, the results of the evaluations are plotted in some kind of “curve” that shows what percentage of the evaluated received each band of grades and concepts. For example, at this time it can be identified excessive allocation of appraisees in the “right part” of the curve, of high performance and that may not necessarily reflect the reality, especially when considering an appraisal scale relative to what is expected of the position.

It is very unlikely that a company will have, for example, more than 50% of its employees performing “above” or “far above” what is expected for their respective positions.

In this spirit, some companies choose to “force the curve,” that is, to force the distribution of scores and concepts in a performance evaluation to have a certain statistical format.

Case Study: GE

This concept was born in GE, under the leadership of Jack Welch, where the 70/20/10 curve is used. In it, only 20% of the organization can be high-performing employees in a given year. At the other end, 10% of the employees need to be classified in the worst performing portion and the remaining 70% are in the “middle”.

In these cases, it is common that a ranking of all the appraisees of the process is made according to their concepts or evaluations. After the ranking, if the forced curve says that only 10% of the appraisees can have a maximum concept, the first 10% of the ranking (for example, the first 20 in a company with 200 employees) receive their maximum concept, and so on up to the last ones.

The forced curve

The forced curve is a great source of heated discussions and possible unfairness, precisely because it forces the employees of a company to obey an obligatory performance distribution. This leads to differentiation where there are not necessarily differences in performance.

Thus, the use of the forced curve is declining among organizations and human resources areas. However, its use can still be useful in cases of turnaround and cultural change, where it may be important to terminate a large portion of a company’s worst-performing employees.

The product of this calibration exercise is the eventual revision of the evaluated grades and concepts that may be “out of calibration”, so that the evaluation results are as fair and accurate as possible.

Case Study: Ambev

Some companies, such as Ambev, conduct the results evaluation in a manner completely independent from the competencies and behaviors portion. There, the results assessments are carried out at the end of the year and have as their main product the payment of variable remuneration in the form of bonuses to employees.

The competency assessments, on the other hand, are carried out in May, and serve as input for the company’s succession planning discussions and promotions (vertical and lateral).

Still speaking of Ambev, the portion of “results” held at the end of the year is absolutely objective and mathematical: if a company employee has beaten 90% of his market share target, for example, he has a performance equivalent to 90% in that criterion/target. It is worth noting that performance triggers and accelerators can be defined in goals, which define, for example, that below a certain market share threshold the employee has 0% performance, and above a certain ceiling, has 120% performance.

Advantages of the Ambev model

The advantage of this model is that in it the rules are extremely clear and there is little room for favoritism and cognitive biases in the definition of variable remuneration.

The great advantage of using goals in evaluating a team member’s results lies, on the one hand, in the intuitive connection between goals and the company’s strategic objectives, derived from the breakdown of these goals, and on the other hand, in the adoption of more objective evaluation criteria, therefore less subject to cognitive biases, personal preferences, and injustices that can hinder the accuracy of the evaluators.

Disadvantages of the Ambev model

It can be unfair to the employee

On the other hand, this direct and mathematical relationship can be extremely objective but often unfair to the employee and his or her colleagues.

Understand: we can think of a company employee that beats his goal and has a high performance in the evaluation, but in order to reach this goal he had to neglect some important aspect of the company’s business.

The long-term sustainability of the brand, for example: imagine an employee who sells his products at no profit to reach his market share goal, thus undermining the company’s cash position and bottom line. Or, again, an employee who had a goal that is no longer relevant (due to a strategic change), which leads to the employee’s harm.

Equal goals, different difficulties in practice

Another risk of the direct connection between goals and performance is that the goals may look the same but have extremely different difficulties in practice.

Consider, for example, two salespeople who have equal sales quotas of R$ 100 thousand in a given cycle. On the other hand, consider that in the middle of the cycle the client portfolio of one of them enters a severe recession caused by macroeconomic aspects, which end up, differently, favoring the client portfolio of the second salesperson.

Now imagine that both deliver their (BRL) R$100,000 quotas at the end of the cycle. Who performed better? For these and other reasons, it is often advantageous to add a manager’s judgment aspect to the evaluation, so that potential injustices and behavioral/purpose deviations are compensated for and evaluated.

In order to solve cases like these, in many companies the manager’s judgment is used to define if the goal was really achieved. We lose a little numerical objectivity, but we gain precision and fairness in the evaluation when managers are well trained, have the right motivations, and count on the support of a close HR area.

Communication of results: the so-called “feedback”

The last stage of the performance appraisal process, and therefore of the performance management cycle, is the feedback process. This is a meeting in which manager and employee (and eventually someone from the human resources area, such as a business partner) participate and where they discuss:

  • The results of the assessment, that is, the observed behavior and the achieved results;
  • Inputs for the development of the team member for the next cycle;
  • Decisions on promotion, compensation, etc.

One of the main trends that has been developing in human resources areas is the separation of this feedback meeting into two distinct meetings: one where the results of the process and development inputs are discussed, and another where the decisions made in the cycle are communicated.

On Google

At Google, for example, these two meetings are separated by at least 1 month. According to Laszlo Bock, former Vice President of People Operations at Google, employees close their ears to development topics when they are in anticipation of decisions affecting their future and their pocket (or for that matter, when they are digesting decisions that may have been communicated). Therefore, two different meetings are held.

You may have noticed that we leave the term “feedback” in quotes when it is used to denote this meeting. This is because we think that these meetings – and Google apparently agrees with us on this – are much less about feedbacks and much more about process feedback.

Cycle duration

Finally, an important consideration about the design of a performance management cycle is its duration. Traditionally, most companies have structured their performance management cycles according to their companies’ fiscal – and therefore annual – calendar.

More recently there is a huge trend toward shortening performance management cycles to semi-annual and even quarterly periods. However, complex processes of setting expectations and evaluating performance can be extremely costly to the organization and therefore impractical to conduct more than once a year.

So some companies do simpler cycles, usually just focused on employee development and performance, more frequently, but maintain a super annual cycle where people decisions are made and from which the company’s talent management process is carried out.

We have come to the end of our article. We hope you have reached the goal we had for this text, to give a solid foundation for management and human resources professionals to

understand what performance management is, what performance appraisal is, and how these two topics are related.

Do you want to understand how Qulture can contribute to your performance evaluation?

Schedule na appointment with one of our professionals!