A system is an operating entity that links and integrates elements, parts, structures, processes and sub-systems into a functioning whole.
For the CEO of the local enterprise mentioned in the Preface, a management system was necessary for his organization to differentiate itself, compete better and grow faster.
However, not all management systems ensure functionality. Some can be more dysfunctional than effective and can induce unproductive behavior in the workplace. This can be the result of system weaknesses, poor implementation or a combination of the two. Therefore, choosing the appropriate management system to get work done through and with people to achieve results is an important management undertaking.
The following discourse with a training manager on her organization's training needs led to a review of a common management system, the organization's performance-appraisal system.
I was working as an external consultant with a large organization, Resu Co., whose Training Manager, Tracy Chan, was briefing me on the organization's training needs and programs for the ensuing year. Her assessment included the need for line managers and supervisors to be better trained on how to conduct performance-appraisal interviews and make appropriate ratings.
"We have conducted several training sessions which were well rated," Tracy explained,"but many of those who have been trained still experience difficulty each year." To illustrate the difficulties, she produced the following list highlighted by line managers:
Difficulties with making appraisal ratings using the current system
No matter how well defined, the descriptors for the appraisal ratings are subject to interpretation (or misinterpretation) by appraiser and appraisee alike.
Some appraisers are more lenient, and some have high standards and strict expectations. This affects staff morale, as people sometimes compare their appraisal ratings with one another, even though these are meant to be private and confidential.
Some appraisal factors are traits rather than performance. They should be used for assessing potential for development, rather than for rewarding contributions to the previous year's results. Considering them together distorts the actual performance results of the past year.
Appraisal of "Job Knowledge" depends on the appraiser's wealth of knowledge or specialization. If an appraiser is appraising someone whose field of specialization is not similar to the appraiser's, naturally by comparison, the specialist job holder will get a higher rating. On the other hand, if an appraiser is more knowledgeable than the appraisee, the appraisee may not get a high rating.
"Job Knowledge" and "Attitude" contribute to a person's performance and are means to an end, not ends in themselves. They are best assessed separately for development purposes, rather than for rewarding for the previous year's contributions.
Factors like "Resourcefulness" and "Teamwork" are difficult to rate even though definitions are given.
Ratings tend to take into account performance based on recall. If an appraiser is transferred to manage a new department during the year, the appraisee's performance prior to the change may not be taken into account, or it could be forgotten or lost.
There are also the proverbial "recency effect" (when the appraiser remembers only recent events) and the "halo effect" (when one outstanding event overshadows other work done) to consider.
Difficulties in conducting year-end appraisal discussions using the current system
In most appraisal discussions, both the appraiser and the appraisee end up trying to recall events and incidents to justify or disagree with a rating on a factor.
Assertive subordinates, whose perceptions about their own performance may be different from the appraiser's, often put the appraiser on the defensive. By the same token, a domineering appraiser can also put the appraisee on the defensive.
There is nothing much to discuss if the appraisee is reticent or the appraiser is inarticulate, particularly with people whose first language is not English.
With tense and confrontational situations, training on how to conduct interviews has limited effectiveness.
An appraiser's overall rating is sometimes adjusted by a counter-signing superior, whose appraisal is sometimes more impressionistic or based on hearsay, rather than on first-hand knowledge. This renders the appraiser's original ratings ineffectual and meaningless.
I reassured Tracy that the line managers were not alone in facing these difficulties if their appraisal system used the behavior-anchored rating scales (BARS). While training in appraisal interviewing skills may alleviate some of the difficulties, it may not eliminate them if it is a systemic weakness. I then asked to look at the company's appraisal form, as shown in Table A. This utilized a behavior-anchored rating scale (BARS) system.
I could see that all the descriptors were intentions, rather than specific, measurable results. Measurability enables the appraiser to give an unequivocal "yes" or "no" as to whether the appraisee has or has not achieved the intention as written.
Table A. Employee Appraisal Form Using a BARS System
Appraisal Rating | Wt | Description | Appraisal Rating | Wt Score | ||||
|---|---|---|---|---|---|---|---|---|
5 | 4 | 3 | 2 | 1 | ||||
Quantity of Work | 4 | Consistently high output Varying high output Does work normally expected Works less than normally expected Low output | ||||||
Quality of Work | 5 | Very accurate and thorough Exceeds expectations for accuracy and thoroughness Meets expectations for accuracy and thoroughness Below expectations for accuracy and thoroughness Inaccurate and incomplete | ||||||
Resourcefulness | 4 | Exceptionally enterprising—goes beyond routine work Seeks and accepts responsibility Occasionally goes beyond job responsibilities Seldom assumes greater responsibility Does routine job with little original thought and shuns responsibility | ||||||
Job Knowledge | 3 | In-depth knowledge of all aspects of job Good working knowledge of job details and tries to acquire more Adequate knowledge for normal requirements Has limited knowledge. Needs to improve Knowledge inadequate and makes little effort to learn more | ||||||
Teamwork | 5 | Maintains excellent relations with peers and goes out of the way to assist all the time Good team worker. Assists without being asked Works and cooperates reasonably well with most Generally works well with people but not always cooperative Uncooperative and helps reluctantly when asked | ||||||
Weighting (Wt) 4 to 5 Very Important 2 to 3 Quite Important 0 to 1 Less Important | Performance Categories (Weighted Score) A – Outstanding Performer B – Excellent Performer C – Good Performer D – Satisfactory Performer E – Poor Performer | Above 100 80 to 100 60 to 79 40 to 59 Below 40 | ||||||
I am not saying that appraisers cannot have a good intuitive feel, perhaps based on long years of experience and observation, to make a valid judgment of someone's behavior. However, validity and reliability are two key considerations in measurement. Thus, when we rely on line managers with varying experiences and an intuitive feel of things to evaluate performance, this usually results in the difficulties listed above.
But the company's difficulties in this regard were by no means limited to Tracy's list and she recounted for me the experience of the Recruitment Manager who, she said, was upset that he had missed half a month's bonus by just one weighted average point in the performance-appraisal rating.
It transpired that the Recruitment Manager was rated a Good Performer with a weighted score of 79, just missing the Excellent Performer rating by one point. He was rated four in all factors except for Teamwork, for which he was rated three.
He believed that he was a good team worker. He attended all departmental meetings, contributed suggestions, and even allowed the unused part of his expense budget to be used by others.
He was told that he should have interacted and lunched more frequently with his peers, as this was considered part of team spirit. He explained that his interviewing schedules sometimes did not permit him to have leisurely lunches; in fact, he sometimes had to skip lunch altogether. The fact that he often ate alone and hurriedly was seen as him not being with the group.
However, since his superior had been most helpful and encouraging in all other respects, the Recruitment Manager did not argue with him, even though he believed that he deserved an excellent rating. That solitary point about not lunching and interacting with others probably cost him his bonus.
Logically, if there is a range, there must be a cut-off point. But missing out on a bonus by a single point can be very difficult to accept emotionally. There also seems to be arbitrariness about what score to award based on the descriptors. Why is this a three and not a four? The Recruitment Manager probably gave good reasons why he should be awarded a four, just as the supervisor could justify awarding a three, based on his interpretation of teamwork.
So the perceived subjectivity of evaluation or appraisal every year often culminates in appraisees becoming cynical and appraisers finding it an annual chore. When this happens, the purpose of a performance appraisal loses its significance.
There is another phenomenon which many HR managers and directors are aware of. This happens when the HR department collates all the appraisal-rating returns and looks at the distribution of ratings, across the organization, by division, or by department. There are some typical distributions of ratings.
The skewed distribution shown in Figure A is a common phenomenon. With conventional BARS-based appraisal systems, where ratings are discussed with appraisees in "open appraisals," there is a tendency for raters to skew their ratings to the positive. When asked about this, most appraisers will say that their people are good, and that's why most of them have good ratings. If bonuses and increments were to be based on the generous ratings as given, then wage costs would escalate without commensurate increases in productivity.
The real reason behind the skew is that appraisers will be put on the defensive when ratings are based on behavior and personality traits. They will have to explain to their appraisees, often not convincingly, their unfavorable ratings.
For example, if you rate someone "satisfactory" or "poor" on resourcefulness or teamwork, you will have to justify the rating with examples. The appraisee may come up with examples why he should be rated better. So it is one person's interpretation versus another's. Favorable ratings will be more acceptable to the appraisee, and most (but not all) appraisers will tend to shy away from unfavorable ratings to avoid arguments. They may even be doing so with the idea that this will motivate their subordinates.
Given the skewed ratings, most organizations have to moderate the returns, requiring a forced ranking into a normal or "bell-shaped" distribution. This means using a statistical device to correct inconsistencies of judgment. Employees, however, see this as setting a quota for ratings, thus causing them to perceive the appraisal ratings by individual appraisers as irrelevant.
When asked for an explanation by a disappointed appraisee, appraisers often use the forced-ranking results to relieve themselves of the responsibility for an unfavorable assessment. Appraisees are then left to wonder why they bothered going through the process in the first place if those who are most familiar with their work can be overruled in this way by a moderation committee.
Another BARS-based rating distribution (see Figure B) has appraisals in the form of confidential reports or closed appraisals. Here, countersigning superiors usually endorse or approve the ratings to provide a form of check and balance to the ratings. In such situations, there is a tendency to bunch ratings around average. This is because very unfavorable ratings would have to be discussed with the appraisee, and very favorable ratings would have to be justified to the countersigning superior, especially if the latter is likely to think otherwise.
Such bunched ratings will result in good contributors not feeling recognized and rewarded; consequently, talent may be lost.
Weak performers, on the other hand, will not see a need to improve, especially if most people are getting about the same rating and reward for doing the minimum. They will not go beyond their normal routine, because there is no great incentive to improve their performance.
Companies that use this conventional method will find that the issues and contention still persist, no matter how many times they revise the forms and redefine the factors or behavioral descriptions. More training in performance-appraisal interviewing skills simply treats the symptom rather than the cause of the tensions over behavior/personality-based appraisal systems.
To change the perception of individual bias that accompanies the BARS system, some organizations supplement this with the 360-degree appraisal system. This involves more appraisers who have work contact with the performer, but they use more or less the same behavioral and traits factors.
Although appealing in concept, in practice the 360-degree appraisal system does not seem suited for a pay-for-performance reward system. Apart from the logistics involved, in some organizations the 360-degree appraisal system has turned out to be an annual popularity or opinion poll. There is also a tendency for ratings to skew towards the favorable, because those who have to appraise or provide feedback are usually conscious of the impact of their appraisal on the appraisee's bonus or increment.
Also, supervising or line managers may tend to minimize their responsibility in evaluating and giving feedback on their subordinates' performance, since others also have a part to play in deciding performance levels of the appraisee, regardless of the extent and intensity of inter-functional work interactions. This notion of shared responsibility may result in line managers abdicating their responsibility and authority. However, multi-rater feedback is useful for assessment and development of potential and talent.
A results-based appraisal system, though not perfect, is less contentious and more valid and reliable than the systems outlined above. A results-based appraisal system is derived from the Management by Objectives (MBO) system advocated by Peter Drucker in The Practice of Management in the 1950s. However, Drucker advocated setting objectives, or specifying desired results, to achieve better results from an organizational and operational perspective. It was not a performance-appraisal system per se.
A results-based appraisal system allows an appraiser to evaluate performance and award appraisal ratings based on a balanced set of specific targets. Such systems use objective measures that are more specific than descriptors, thus allowing less room for divergent interpretations. In this way, people can focus on the purpose of performance appraisal, which is to achieve greater productivity. Under this system, performance appraisals are not mere annual rituals to justify pre-determined bonus awards or increments.
Table B provides a quick comparison of the common appraisal systems that may be used in the context of wage restructuring that has become necessary in light of rising operating costs that demand commensurate improvements in productivity.
After working with several organizations on this issue, I feel we need to look at the system globally as a results-management system, to give a greater emphasis to results and productivity.
Table B. Performance-Appraisal Systems Compared
Appraising results is just one aspect of the system for enhancing productivity, and this is the concept that line managers can understand and connect with. The other aspect is the managing of performance.
A great deal of emphasis is on achieving results through and with people, and that implies substantial management of performance, besides the measuring of performance results.
In Tracy's case, although the apparent need was to train line managers on how to conduct year-end appraisal interviews, the real need was to train line managers to implement a results-management system. The thinking underlying this is that if people are managed well, performance measurement and appraisal will be made easier. Employees will be more inclined to conduct self-appraisals to evaluate their own performance and, in the right circumstances, they will generally be harsher with themselves than you might expect.
Companies have realized that wage costs represent the bulk of their operational expenses. The lower the fixed costs associated with base salary increments, the lower their fixed overheads. High overhead costs hinder competitiveness.
Consequently, there are moves to mitigate base salary increments and focus on awarding higher variable payments for good organizational and individual performance. This is to ensure that rewards are based on tangible performance results which can vary from year to year. In this way, an employee's reward is more closely linked to his real performance and the performance of the organization.
With the conventional appraisal system, productivity cannot be measured more definitively to match increasing wage costs. One well-regarded HR director once said to me, "If there's going to be a paradigm shift in the reward system, there must be a complementary paradigm shift in the current performance appraisal system."
This shift will require establishing individual performance measures, periodically reviewing results to correct and improve performance, evaluating final results, and emphasizing pay-for-performance to identify and retain talent. It is crucial then to provide accurate inputs to the compensation system for it to work effectively.
Ideally, the distribution of ratings should look like a normal distribution or bell-shaped curve to reflect the validity and reliability of the ratings (see Figure C).
In short, if you have a management system that incorporates individual performance measurement, performance management and performance evaluation to make individual contributors feel that they are managing a "business" within a business, then many of the usual difficulties and distortions will be reduced, if not eliminated.
It will not be an appraisal system per se, but a holistic results-management system that essentially looks at performance measurement, performance management, performance reward, talent identification and retention.
Calling it a results-management system (RMS) eradicates the notion that managing for results is doing HR's work, complying with policies and procedures and filling forms. Getting results through and with people will be seen and accepted as part of a line manager's key functions.
A strong management system is a prerequisite for effective people performance.
A results-management system can reduce many of the difficulties and confusions associated with conventional, behavior-based appraisal systems.
It is more cost-effective to focus training on the overall real needs of a system than on single-item apparent needs.
An effective measurement system can help an organization to achieve wage cost-effectiveness by differentiating between good performers and weak performers so as to compensate them accordingly and retain talent.