Performance Management: What is Forced Ranking?

Reader Sriram from Chennai (India) asks,

A multinational recently acquired our 35-employee software testing company. Our personnel department sent an email on how this purchase affects us. The email mentioned a new forced ranking system for performance evaluation. Can you describe this system?

Every organization needs a formal approach to track individual contributions and performance against organizational goals and to identify individual strengths and opportunities for improvement. Typically, this system involves placing employees along a performance curve or classifying employees into categories of percentiles for performance.

Bell curve for forced ranking / performance management

Jack Welch, General Electric’s former CEO, is often associated with a 20-70-10 distribution: the top 20 percent is rewarded for best performance, the middle 70 percent is rated ‘average’ and the bottom 10 percent is coached for improvement. The ‘rank-and-yank’ system, also associated with Jack Welch, automatically terminates employees in the bottom category, allowing organizations to purge the worst performers.

Although an individual’s supervisor conducts the formal performance review discussion, management higher-ups assign the individual’s ranking following debates on performances of comparable individuals from across the organization. Often, these higher-ups are not knowledgeable enough of an individual’s performance. An individual’s ranking then depends on the supervisor’s willingness to fight on behalf of the individual. The ranking is ‘forced’ because an individual may be ranked in a lower category regardless of whether the direct supervisor (and hence the most knowledgeable reviewer) would have rated the individual that way on his/her own.

In intent, the forced ranking system is an excellent method for rewarding top performers and setting specific deadlines for improvement for poor performers. Despite its appeal, the system has several drawbacks. For instance, the system promotes individual performance over teamwork and often leads to dissatisfaction among ‘average’ and poor performers. In my opinion, most of this dissatisfaction stems from poor administration of the system at the ground level. I will cover this in another blog article.

Comments

  1. says

    Nagesh,
    Your insight is right on the mark. The system is flawed in the favor of upper management and allows them to make decisions based on statistics, not the opinions and expertise of frontline supervisors.

    It lacks any common sense, promotes political environments, and continually leads to poor personnel decisions.

    How does a person have any confidence in their leadership?

    Dave

  2. krishna rao says

    Dear Nagesh,

    In the present performance review systems, we are being evaluated by higher ups and in some cases by 3600 feed back also. These reviews are mainly dependent on perceptions. But, “Perception is the paradox of mind”. Is there a way out to overcome the perception based evaluations? Yes, let us look into the following approach.

    Through scientifically proved self assessment psychometric tools coupled with behavior psychologist’s interventions, the individuals can be evaluated and the data base can be captured.

    All the activities of the organizations can be logged on to credit points, based on the efforts being required for each activity [like work studies on white collar activities also]. SAP system can track the activities and credit points associated for the individual for carrying out those assigned activities.

    Since team work is essential for success of any task, the team members will be playing their assigned roles and the associated activities to gain maximum credit points with synergy. In many cases, the individual is a team member in many team tasks. So, the gross credit points gained by the individual are a direct measure of performance, without having any perceptional ambiguity in appraisal.

    It is the role and duty of the administrators to design a system to give appropriate opportunities to the individuals to gain maximum credit points.

    Basically, work is an out let for self expression. So, our development processes should be based on assessing individual’s interests, motivations, personality traits, strengths and weaknesses, capabilities, knowledge and skill levels. The efforts of the administrators should be to match the individual inventory to the work. Thus, capabilities are to be tapped with proper work assignments and indiscipline should be punished/ behavior to be modified by social disapproval approach.

    Thus, performance appraisal is to be used for development of individuals by matching individual inventory to that of work and credit points based performance data base, by objective oriented humane administrators for long term gains.

    Yours,
    KRISHNA RAO

  3. Jawad Mansoor says

    Forced distribution is useful for jobs that are structured with strict quantitative criterias. For example, sale of FMCG goods can be measured in terms of $ day by day, week by week, month by month of individual SKU and can be plotted on bell curve for performance evaluation. But in case work is dependent on multiple criterias and people then it is discouraging thus, promotes competition rather than cooperation.

  4. says

    For everyone who has spent even a minuscule amount of time working in a corporate environment, the words “Bell Curve” “Normalization” would have become one of the most un-understood words that they came across in their lifetime.

    I am not saying “un-understood”, just for the heck of it.

    Having spent close to 10 years in the corporate world and being a part of the team that does the normalization and well as being a part of the team that gets the result of the normalization, I completely believe that I do not understand the concept. If you think you understand what does a normal curve mean, please try to read the article in Wikipedia (http://en.wikipedia.org/wiki/Normal_distribution)

    Normal Curve or the Bell Curve, is a statistical term that was theorized by mathematicians to determine how a large set of numbers, when plotted on a graph, would look like.

    So organizations use this tool to determine, how do they review your performance and what do you get in the end after your review is done.

    I know I am not a mathematician, but one thing that I have learnt as a child is that there is a threshold on numbers, below which any statistical analysis does not hold any significance.

    Consider for an example, you visit a website to get the review of a new camera that you plan to buy. You realize that there are only 2 people who have provided the reviews. So what is your instant reaction? You look for another place where more people would have provided their reviews, or at the least you try to visit multiple review sites to validate the reviews. What you in-turn are doing, is to reach that threshold number beyond which you can trust the reviews.

    Now consider a corporate team, do you have a threshold before you say that now this set can fit into a normal curve? What is that threshold? It is obviously hard to determine and I have no information of is someone has determined it. However, one thing that I am confident of, is that, a group of 20 people CANNOT fall under the NORMAL CURVE!

    The dismal situation is that this normalization of people’s performance goes on rampantly without people even trying to understand that whether this normalization is justified or not! I have seen people doing normalization for a team comprising or 4 members! This was INSANE, I banged my head against it, but as expected, the effort was futile.

Leave a Reply

Your email address will not be published. Required fields are marked *