The limitations of subjective manager evaluations and multisource feedback
If you read most performance reviews, you'd be convinced that the vast majority of companies are doing great. In one business Gallup has observed, 95% of all employees exceeded their managers' expectations. In another organization, almost all employees -- with very few exceptions -- received nice year-end bonuses.
The problem is, neither of these companies is performing very well. Their market share has declined, service quality has deteriorated, and they have exceeded their budgets. Memo to management: Look again. The operation may have succeeded, but the patient died.
Unfortunately, scenarios like these are hardly exceptional. Most performance management systems are significantly flawed and don't deliver the results executives seek. The first article in this two-part series will review the flaws inherent in two of the most common performance evaluation approaches. The second article will cover two evaluation methods that deliver more positive results.
Analyzing performance management systems
When analyzing the various practices -- both effective and ineffective -- start by looking at three distinct components of any performance management system. The first component is evaluation -- how can a company capture, quantify, measure, or evaluate an employee's performance? Second, how is that performance rewarded? Is there a direct link between performance and rewards? And what range of reward is being offered (financial, recognition, growth opportunities, etc.)? Finally, what is the company's employee development approach? How are employees being developed and supported to deliver superior performance?
Each of these components has a powerful impact on that system's effectiveness. It's also important to review how the performance management system functions within the company's culture and how it supports corporate goals and strategies. What works for one business won't necessarily work for another. Nonetheless, it's possible to identify best practices common to top-performing workgroups and other practices that rarely ever work. And these observations often hold true across industries, countries, and cultures.
Effective and ineffective methods of evaluating performance
There are many ways to quantify or evaluate employee performance. This two-part series reviews the four most common methods:
- manager evaluations, or the traditional
performance review conducted by an employee's boss
- multisource feedback, such as 360s, in which
multiple constituencies -- including managers, peers, direct
reports, and/or customers -- are asked to evaluate an
- staffing reviews, in which a management team
ranks, evaluates, and discusses a group of employees in their
- objective performance outcome metrics, for
which an employee is held accountable
These four approaches are not mutually exclusive, and most companies use combinations of them, as is reflected in the graphic below.
In most organizations, the performance appraisal process begins when a manager rates an employee on a series of attributes. For example, in one business, managers rated their front-line employees on whether they "demonstrated positive energy by handling customer issues on the spot." In another company, supervisors were evaluated on their ability to "articulate a compelling vision for the future" and "build relationships across organizational boundaries."
Gallup consultants are often asked to statistically validate performance appraisal ratings against objective performance criteria. In many cases, the ratings don't correlate in any meaningful way to measurable outcomes. Often, there is no correlation, or worse, a negative one. In some cases, employees with the highest performance levels received the lowest ratings, and as a result, the least rewards. The end result is that the system weeds out top performers while rewarding mediocre ones.
Why does this happen? Because manager ratings are inherently subjective -- and this subjectivity only increases when the appraisals are linked to financial incentives such as merit pay raises. In one company, employees referred to the performance appraisal as "that form you need to fill out to give a person a raise."
Systems like this have serious unintended consequences for a company. When managers control their employees' ratings -- and thus, their year-end raises -- employees will focus on currying favor with their managers rather than trying to engage their customers or increase their productivity. In essence, the system rewards toadies, not high performers.
Why then, do so many organizations rely on manager appraisals? A major reason is that they're easy to use -- any employee can be rated on the same scale using the same form. And it would be a terrific solution -- if only it would work.
In an effort to eliminate the subjectivity of manager evaluations, many organizations have introduced multisource feedback, such as 360-degree surveys, in which managers, peers, or subordinates rate an employee on a fairly extensive list of attributes.
There is nothing fundamentally wrong with 360s if they are executed effectively and professionally. Indeed, they can generate useful feedback on employees' strengths and weaknesses that can encourage their personal and professional development.
But in most organizations, poor execution limits the usefulness of 360s. Here are a few common problems:
- Occasionally, employees get to choose the people who
evaluate them. This self-selection inherently biases the
approach, as many employees tend to pick colleagues who like them
or will give them good ratings.
- Companies usually don't train managers to interpret 360s or coach employees on the results of 360s. Rather, the reports are simply dumped on managers' desks, and the managers are left to figure out the reports on their own. This is no easy task: As 360 surveys can be quite lengthy, it takes skill to sort through and analyze the plethora of data and give employees clear and actionable feedback.
The first two problems occur when companies attempt to make the 360 process scaleable. Many companies start off using multisource feedback successfully with a small group of leaders. But in their effort to apply this time-intensive process to a wider group of managers and employees, they may put the 360 surveys online or automate the reporting. This may leave managers and employees with a morass of data but no clear idea of how to interpret it -- or what to do with it.
- Most 360 reports don't measure performance; they measure
style. Unless they are validated against the actual business
outcomes required for success in a role, the reports may be a
better reflection of a desired work style rather than desired
results. Executives might assume that people who score highly or
improve their ratings on the survey are becoming solid performers
when this is actually far from the truth. Unless the report results
are validated against business results, it's much safer and
effective to use this tool for development only, and not
for accountability or performance management purposes.
- The natural tendency when using 360s in the development
process is to focus on fixing an employee's weaknesses.
Inevitably, the person receiving the report -- and often, the
person helping him or her interpret it -- immediately fixates on
the employee's problem areas while overlooking his or her
strengths. Yet Gallup's research indicates that this is a backward
approach, because a person's greatest opportunities for development
are in his or her areas of strengths, not weaknesses. This means
that a person's biggest opportunity for development has been
completely overlooked. (See "The Four Disciplines of Sustainable
Growth" in See Also.)
A fundamental questionDespite their myriad problems, the two performance appraisal methods outlined above are an integral part of how organizations evaluate employees. It's time for executives to stop, step back, and consider if their performance evaluation system is meeting the organization's business needs. They should ask themselves: Does the system improve employees' productivity, safety records, customer focus, or the quality of their work? If not, the next step is to borrow some ideas from performance evaluation methods that are far more effective: staffing reviews and objective performance measurement. The next article will cover those methods.