Consistency in performance feedback – Challenges and Techniques

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Performance management is the process of creating a work environment or setting in which people are enabled to perform to the best of their abilities. Performance management is a whole work system that begins when a job is defined as needed. It ends when an employee leaves your organization. Many times the consultants are using the term "performance management" as a substitution for the traditional appraisal system. It is a broader work system context. A performance management system includes the following actions:

Develop clear job descriptions.

Select appropriate people with an appropriate selection process.

Negotiate requirements and accomplishment-based performance standards, outcomes and measures.

Provide effective orientation, education and training.

Provide on-going coaching and feedback.

Conduct quarterly performance development discussions.

Design effective compensation and recognition systems that reward people for their contributions.

Provide promotional / career development opportunities for staff.

Objective of the study

Main objective of the project is to understand the common fallacies in performance evaluation and documentation and find techniques to weed out the roadblocks.

The output of the project is to identify techniques that would be beneficial to the organization in removing the inconsistencies in performance management. The project is to apply this technique on some subjects and establish the effectiveness in applying the technique to remove inconsistencies.

In Scope

Performance management including employee appraisal, participating in moderation / laddering process, performance feedback documentation etc will be in-scope for this project. All the common fallacies in performance management will be studied and recognized within the sample group identified.

Techniques including training, policy / procedure changes, skill assessment of supervisors will be considered and analyzed to identify the solution to remove performance bias. The effectiveness of the recommendations will also be measured in the identified sample size, if possible.

Out of scope

Appraisal feedback communication, identification of improvement areas or strengths is not part of this project.

Organizational Context:

XYZ Services P Ltd is a multinational company operating across 60+ countries providing IT, ITES, Consulting and Outsourcing services for several prestigious clients. With more than 190,000 people serving clients in more than 120 countries, the company generated net revenues of US$21.58 billion for the fiscal year ended Aug. 31, 2009.

XYZ has diverse workforce with employees from various cultures, continents and contributions. The employee population is more than 100,000 strong. Being an IT organization, people are its biggest asset; hence performance management is a very important and critical factor for XYZ's success. Performance appraisal is done in XYZ for multiple purposes. The goal of the performance management system is not merely to rate the employee for compensatory purposes but to focus on his development and to groom him for a leadership role in the future. It also enables recognizing the employees' abilities and charting out an appropriate career path for him.

The various stakeholders in the performance evaluation process are:

HR Analyst




XYZ has a structured, yet tailored performance management process absorbing all the best practices from the HR world. Performance management process consists of 4 distinct steps:

Step 1 - Set Objectives:

XYZ employs results methods for performance appraisal and enforces Management by Objectives technique. MBO is a philosophy of management first proposed by Peter Drucker in 1954 that has employees establish objectives through consultation with their superiors and then uses these objectives as a basis for evaluation. MBO is a system involving a cycle that beings with setting the organization's common goals and objectives and ultimately returns to that step. The goal setting process flows down from the organization through departments (department heads) to individual employees.

A significant feature of the cycle is the establishment of specific goals by the employees, but those goals are based on a broad statement of employee responsibilities prepared by the supervisor. Employee established goals are discussed with the supervisor and jointly reviewed and modified until both parties are satisfied with them. The goal statements are accompanied by a detail account of the action the employee proposes to take in order to reach the goals.

Step 2 - Review Results:

The results or outcome achieved against each objective is reviewed during the following intervals:

Project end feedback - Individual's objectives are set based on both project success factors and individual's goals. Due to the IT nature of the organization, the resource can frequently move between projects or departments based on the nature of demand - supply at that point. The employee documents the results achieved in the project along with the supervisor, when he moves out of the project.

Midyear feedback - To avoid recency effect or bias and to have a checkpoint for the employee to understand where he stands against his objectives, the midyear feedback discussion is conducted. This acts as an effective checkpoint where the employee can evaluate on where he stands against the objectives set at the beginning of the year and what needs to be done to achieve or excel the set objectives. Supervisors use this arena as the discussion ground to reflect on employee's contribution till mid of the year and sets further expectations for the rest of the year. The employee and supervisor are free to re-look at the objectives and reset it, if required. This may be due to the changed requirement or business strategy of the unit, high or low objectives set at the beginning of the year, or to promote specific behavior from the employee.

Year end feedback - This is a critical phase to document the feedback of the employees for the complete year. Supervisor has a deep and thorough review of the results achieved by the employee during the complete year.

During periodic reviews, as objective data are made available, the progress that the employee is making toward the goals is then assessed. At the conclusion of a period of time (either end of project or six months whichever is lesser), the employee makes a self appraisal of what he or she has accomplished, substantiation the self appraisal with factual data wherever possible. The interview is an examination of the employees' self-appraisal by the supervisor and the employee.

Other than MBO, XYZ also partially uses some of the other appraisal techniques:

Other than documenting results achieved, XYZ also tries to understand the traits exhibited to achieve these results. Though the methodology employed is not strict Graphic Rating scale method, but a similar approach is followed. The favorable trait to be exhibited by the employee is listed to the supervisor and he needs to rate whether the listed trait was established below, equal to or above expectations of the supervisor. This way, XYZ emphasizes not only on results, but also on the traits exhibited or through which the results were achieved.

Essay method is also employed to document the strengths and areas of development for the employee. Unlike rating scales, which provide a structure form of appraisal, the essay method requires the appraiser to compose a statement that best describes the employee being appraised. The appraiser is usually instructed to describe the employee's strengths and weaknesses and to make recommendations for his or her development. Essays provide additional descriptive information on performance that is not obtained with a structured rating scale.

Step 3 - Relative rating post moderation (laddering process):

Relative Rating process is done on an annual basis to decide on the relative performance of the employees and to promote meritocracy. During a series of rating meetings (at the project/team and deployed-to entity levels), individual contributions are discussed and compared with peer group to determine a final, and relative contribution rating for the year.

XYZ promotes relative rating methodology to support philosophy of meritocracy - enabling identification of top and bottom performers to continuously improve the aggregate performance of the organization. This also provides a mechanism to differentiate rewards based on performance. This process is called the laddering process to arrive at relative rating of the employees.

Step 4 - Create Development Plan and Objectives for the next year:

As part of annual performance feedback process, a new development plan is created for the employee based on the contributions and action items agreed by the employee and the supervisor.

Common bias from supervisor during the performance appraisal:

Halo Effect - The halo effect is the tendency to rate someone high or low in all categories because he or she is high or low in one or two areas. Results in appraisals that do not help develop employees, because they are two general or inaccurate as to specifics. Evaluating someone lower is sometimes also called the "devil effect".

Standards of Evaluation - If you are using categories such as fair, good, excellent, etc be aware that the meanings of these words will differ from person to person. In any event, the use of these categories is not recommended because they do not provide sufficient information to help employees develop.

Central Tendency - The habit of assessing almost everyone as average. A person applying this bias will tend not to rate anyone very high or very low.

Recency Bias - Tendency to assess people based on most recent behavior and ignoring behavior that is "older".

Leniency Bias - Tendency to rate higher than is warranted, usually accompanied by some rationalization as to why this is appropriate.

Opportunity Bias - Ignoring the notion that opportunity (factors beyond the control of the employee) may either restrict or facilitate performance, and assigning credit or blame to the employee when the true cause of the performance was opportunity.

False Attribution Errors - We have a tendency to attribute success or failure to individual effort and ability (at least in North America). So when someone does well, we give them credit, and when someone does less well, we suggest it's somehow their fault. While there is some truth in this, the reality is that performance is a function of both the individual and the system he or she works in. Often we misattribute success and failure and assume they are both under the complete control of the employee. If we do, we will never improve performance.

Rating definition in XYZ:

The employees are rated on contribution areas (results oriented) and performance factors (behavior / trait oriented). The objectives of the employee are set against contribution areas and the performance factors. Hence, the appraisal process identifies how the employee fared in each of the contribution areas.

The final rating of the employee, which is relative rating within the delivery or business unit, is fit into 5 bands. Each employee is rated to belong to a performance band in conjunction with their performance for the year.

Approach followed for the project study:

The objective of this project is to identify the common bias between supervisors and thereby identify the dissimilarities and inconsistencies in the performance appraisal process.

Identification of sample to be studied

Both supervisors and employees where sampled for this project. To identify the appropriate sample, the supervisors and employees who had rated or been rated exceptionally high or exceptionally low have been considered.

HR personnel were also consulted to identify their perspective on the appraisal process and its consequences.

Interviewing sample subjects

All the sample subjects were interviewed to identify their experience in the appraisal process along with what went right or wrong in the process. The subjects were briefed that this was a pilot project and the results will be anonymous. This made sure that employees were able to share what was running through their minds.

Analysis of interview results to identify top 3 common bias of supervisors / concerns from employees / HR perception of issues in the process

Identification of technique to root out common bias of supervisors

Application of the technique on supervisors and understanding the feedback of the technique along with effects

Arrive at conclusion and recommendations for XYZ.

Sample subjects identification

Sampling process outlines a six-step procedure that can be followed when drawing a sample of a population. In the first step, the target population is defined as that segment of population about which inference has to be made.

In the second step, the sample selection process identifies the sampling frame. Sampling frame is the listing of the elements from which the actual sample is drawn. One of the main tasks in sampling is to develop an appropriate sampling frame when the list of population elements is not readily available.

In the third step, in selecting a sample procedure depends on what the researcher can develop for a sampling frame.

In the fourth step, the sample selection process requires the sample size to be determined.

In the fifth step, the researcher is required to do the sampling, i.e. choose the elements that will be included in the study.

Finally, the researcher needs to collect data from the designated respondents.

Step 1 - The target population was decided to be the employees and supervisors in one particular engagement within XYZ who had taken part in the recently concluded performance cycle.

Step 2 - Sampling frame was finalized to be

employees or supervisors who were or who rated employees at the top level or bottom most level within the delivery or business unit.

Supervisors who have high deviation in their absolute rating of employees

Few reviewers to understand the perspective from the leadership and identify the bias factor from supervisor.

HR personnel who are the co-coordinators and who aid in successful completion of the performance cycle.

Step 3 - Sampling Procedure was decided to be non probability samples. Non-probabilistic samples are used for two reasons:

They involve personal judgment somewhere in the selection process.

Given that the sample is non probabilistic, sampling error cannot be assessed, which in turn means we cannot place bounds on the precision of estimates.

Judgment samples were used for this study. Judgment samples are purposive samples, the sample elements are handpicked because it is expected that they can serve the research purpose.

Step 4 - Since the technique used to collect data is through interviews, a small but representative sample size was chosen. In the 5 band performance ratings, representation of employee and supervisor from each band was chosen. Only one HR was taking care of the engagement, and hence his inputs were considered. Few reviewers were also interviewed to take the leadership perspective of the process.

Step 5 - The employees and supervisors chosen were of different criteria. Since the engagement involved work both from offshore and onshore (near to the client), employees and supervisors from both categories were chosen. To rule out any deviation in the procedure, both experienced and fresh supervisors (who are doing the performance appraisal duties first time) were chosen.

Due to the sensitive nature of the topic, identification of sample subjects was a crucial step. Some subjects who are known and present in the system for a sizeable duration where coming forth to provide inputs. However, some subjects had their own reservations to take part in the interview process. They were counseled about the nature of the project and the anonymity of this exercise. Care was taken so that this exercise does not affect the trust of the subject in the interview process.

Step 6 - The actual interviewing process for the sample size was done.

Interviewing the sample subjects:

Performance appraisal being a very sensitive topic, the most difficult phase of the project was the interviewing step. For each rating band, the corresponding supervisor and employee were interviewed. The employee and supervisor did not know that the other person was being interviewed. Interview process was done in a secluded environment that would provide anonymity and trust for the interviewee.

Interview questions include:

Employee Questions:

Are you satisfied with the rating provided to you?

Did the supervisor communicate clearly to you the justification behind the rating?

Are you convinced about the justification provided and the rating?

Do you have any feedback or suggestions for the performance management process in XYZ?

Do you know your areas of improvement that would empower you to grow in your career path?

Do you know what is expected out of you in the next performance year?

Supervisor Questions:

Was the employee satisfied with the rating provided by you?

Were you able to clearly articulate the justification behind the rating?

Were you able to own the final rating and represent the same to the employee?

Do you have any feedback or suggestions for the performance management process in XYZ?

Does the employee have clearly laid out areas of improvement that would empower his growth in the next performance year?

Do you think you were forced to take a decision due to any of the appraisal bias?

Did you have enough information to understand the performance of the employee?

How clearly were the objectives documented? Were they quantifiable?

If yes, how clearly was the employee able to articulate his performance against the objectives?

Do you see any constraints in arriving at the right performance decision for the employee? If yes, what are they?

Reviewer Questions:

Do you see any appraisal bias from the supervisor end?

During the moderation / laddering session, was the supervisor able to represent his employee with full details and data?

Was the performance feedback fully documented and was it showing the depth of supervisor's understanding of the employee's contributions?

Do you see any constraints with the supervisor's decision of the rating given to the employee?

Interview data analysis and identification of problem areas:

Following are the statements from employee, supervisor and reviewer during the interview process. Each person had their own perspective of the performance management cycle complete and how they went through with the process.

Employee perspective:

Objectives or goals were not clearly set at the beginning of the performance cycle.

Objectives or goals were not quantified, and hence it was difficult to prove that the objectives were met.

Supervisor did not spend enough time during the year to communicate concerns and areas of improvement.

Supervisor did not provide enough time to improve after communicating the performance areas that are lacking.

Supervisor does not have clear idea of my day to day contributions.

Supervisor is generalizing my performance based on one or few instances.

Supervisor did not consider my contributions in the earlier portion of the year.

Supervisor did not provide enough opportunity in this performance year to prove my potential.

I was doing my job right; however the outcome was impacted due to external factors which are beyond my control. Supervisor refuses to recognize that the root cause of the outcome failure is external factors and not my performance.

Good performance ratings are awarded to people in a specific, high visibility project.

Objectives are result oriented, but the feedback was on how the results were achieved.

Supervisor perspective:

Performance appraisals are result oriented. A job not done cannot be declared to be done with any justification provided.

Employee is not ready to take the feedback constructively.

Employee is biased with his ideas and not able to understand the performance factors objectively.

I don't care how other supervisors rate their employees. I am very strict about the way I perform appraisal for my team.

It is easy to please the employee during appraisal feedback and hide behind the laddering process later.

Employee is leaving my project, Iet good benefits go to the resources continuing in my team.

It is just not enough on what job was accomplished rather it is important to consider how the job was done.

I need to take care of my team in the laddering irrespective of how significant the other team members' contributions are.

Reviewer perspective:

Employee and supervisor need to connect more frequently on performance feedback than the current frequency.

Supervisor must understand each employee's contribution and must have the data ready for each employee.

Supervisor should have clear instances of employee contributions for each performance factor.

Employees must be more mature to take the feedback constructively and apply it in the next performance year.

Supervisor must be able to articulate the employee's contributions in such a convincing way that the employee must be able to understand and accept his resultant rating. For this, supervisors should be prepared well with data and instances.

Supervisors must do a concise and effective performance documentation that would help them in representing their team during the laddering session.

Circumstantial data not being considered in case of failure in outcome.

On further analysis of the summary statements from all three participant groups, following are the high level problem areas identified. Each circle represented in the Venn diagram represents the bias and other factors in the performance appraisal process as perceived from each participant group.

Recency bias

Halo effect

High Visibility Bias

The How Surprise


Results vs. Reasons

High Bar Mania

Blame it on laddering attitude

Staying angels - leaving demons

Readiness of Supervisor

Frequency and timely feedback

Opportunity bias

Defensive attitude of employees

Clarity of Objectives

Clarity of Objectives - Clear and quantifiable objects are not set.

Recency bias - Supervisor remembers recently happened events and introduces bias into the performance feedback.

Opportunity bias - Ignoring the notion that opportunity (factors beyond the control of the employee) may either restrict or facilitate performance, and assigning credit or blame to the employee when the true cause of the performance was opportunity.

Halo effect - Tendency to rate someone high or low in all categories because he or she is high or low in one or two areas

High Visibility Bias - Resources with high visibility or in high visibility projects get better performance feedback.

The How Surprise - Supervisors setting up only "what to be achieved" objectives and missing out "how to be achieved" objectives. The employee gets surprised when the supervisor talks about "how-to" during performance feedback.

Readiness of Supervisor - Supervisor is not fully exposed to employee's contributions. This is especially true in projects where employees work closely with client and with minimal supervision from source organization.

Frequency and timely feedback - Supervisor is not providing frequent and timely feedback to the employee.

Defensive attitude of employees - Employees do not take the feedback constructively for their improvement. Employees do not look at feedback objectively for their improvement but go into a defensive mode.

Protectionism - Supervisors take up a protective mode during laddering sessions to make sure that their team gets the maximum benefit even if a person from another team is more deserving.

Results vs. Reasons - Supervisors are concerned more on results and not reasons on why the objective is not met. An objective not met cannot be justified with any number or reasons.

High Bar Mania - Some supervisors set so high bars and evaluate the resources in a strict sense that employees come under pressure, especially when they compare with their peers.

Blame it on laddering attitude - Some supervisors find it convenient to hide behind the laddering process, and not owning the rating decision taken. The performance feedback session is done very amicably, however the supervisor provides a different picture during final feedback communication hiding behind the moderation / laddering process.

Staying angels - leaving demons - Some supervisors tend to lean towards and provide higher ratings for people staying back in the team, even if the person leaving or left the team is more deserving.


Further data scrubbing was done to identify the mode value for the top few concerns raised by different stakeholders. Apart from this, any new type of bias which is previously not known is also listed here. Recommendations for each type is covered along with the root cause:

High Visibility Bias

Root Cause:

Resources that have high visibility due to their personal reasons or charm or being part of high visibility projects are at a better advantage than their peers. Everyone in the organization sees the positive story of the resource and hence always feels that the resource should be performing above expectations.


Reviewers play a very crucial role in this process. Reviewers must be able to identify and question the rating justification for such high visibility resources. HR should also play a critical role here by first identifying such high visibility resources and understanding their performance feedback through the documentation provided. In some cases, leadership team takes decision to reward a particular set of project team due to the criticality of their contribution during that year.

The How Surprise

Root Cause:

MBO philosophy is to create smart and quantifiable objectives during the initial goal setting process. These objectives are altered periodically depending on project movement or role change. During this process, both supervisor and employee concentrate on defining result oriented objectives, and most often forget about the path and means to achieve it. During the feedback session, the employee gets "the How Surprise" when the supervisor pulls up the employee for the process or means by which the objective was achieved is not in consistent with organizational practices or expectations.


Objective setting process should be modified to enforce setting objectives on the "how-to" end and not just on "what" needs to be achieved.

Readiness of Supervisor

Root Cause:

In some cases, due to the nature of the engagement with client, supervisors are not fully aware of the critical contributions their employees are making to the client organization. The client, on the other hand, is not willing to share any performance feedback for the resource as that is one of the reasons he has a vendor in his team (to avoid people management). Hence, the situation goes into a loop with no proper representation for the resource in the organization.


Leadership team should request and encourage client members to give at least year end written feedback that would help his team member. Leadership should make sure that the supervisors are engaging with the clients to receive verbal feedback every month. This needs to be percolated to the employee for his improvement.


Root Cause:

This is primarily due to the selfish attitude of the supervisors to see their teams successful and in turn they getting out of bad books of their team members. Though the inner intention is good from the team perspective, from the organization standpoint, the best person needs to be rewarded. Hence, it becomes important for the supervisor to understand the importance of the performance management cycle in the organizational context encouraging meritocracy.

There are cases where supervisors give higher rating to their team members just to fill in for the quota available, though the persona might not be at that band.


Organizational value of meritocracy has to be reinforced within the supervisory band of resources. Moderators in the laddering session such as reviewers and HR should take a proactive approach in identifying the protectionism attitude exhibited by any supervisor.

High Bar Mania

Root Cause:

Few supervisors set high bar for their team members and award those who are able to cross the high bar. These supervisors are very stringy in awarding performance rating to their team members. During the laddering session, when the employee is moderated with another employee coming from a lenient supervisor, it becomes hugely difficult to compare and understand.


XYZ has started looking at the performance numbers of a supervisor to check if the supervisor is setting a high bar or very lenient. The individual ratings given to each contribution area is consolidated to bring out a single number that would be the performance score of an individual. When performance scores are compared across two supervisors, it becomes very easy to identify the leniency bias vs. high bar mania supervisors.

Blame it on laddering attitude

Root Cause:

Supervisors feel that it is easy to pass the buck on the laddering process as it is easy for them to hide behind the laddering process. The employee is not available in the laddering session and hence does not get to know on what actually transpired. This gives the benefit of doubt to the supervisor who can easily hide behind the decisions taken inside the room.


HR has started educating supervisors to own the rating decisions taken inside or outside the moderation table. Ownership being a very critical trait of an employee, any attempt by the supervisor to relinquish the rating decision will be treated as a black mark in the supervisor's performance appraisal. Setting up objectives for the supervisor to own up rating decisions can also resolve this problem to some extent.

Staying angels - leaving demons

Root Cause:

Supervisors are more aligned towards benefitting the resources that are staying back in the team, even though the outgoing employee might be more deserving.


Moderators like HR and reviewers should take care that outgoing employees are not at a disadvantage due to their untimely exit. Contributions for both the employees are compared forgetting that the employee is quitting the project.