What’s the biggest secret about performance appraisal? It’s this: performance appraisal actually works! Too many companies remain in denial about the benefits that a well-executed performance management system generates. They may articulate the importance of transforming their stale, “best-effort” culture into a tough-minded, results-driven one, but fail to fathom that performance appraisal is the best tool available for muscle-building an organization. Some companies do appreciate just what a well-designed, forcefully managed performance management system can do to assure the execution of organizational strategy. Early in 1999 I agreed to serve as subject-matter expert for a national benchmarking study of best practices in performance management that the American Productivity and Quality Center (APQC) and Linkage, Inc. were undertaking. My first task was to identify those companies that were in fact doing stellar work in performance appraisal, and then convince them to share their processes and techniques with the 17 sponsor organizations that were each ponying up $16,000 to learn the secrets. Identifying the companies that are performance management masters wasn’t that difficult. But convincing them to reveal what they were doing? A different story — several flatly refused. Many of America’s most admired companies just said no when offered the chance to be recognized as best-practice models in performance management. Each one gave the same reason for declining to share their forms and procedures. They saw their performance management processes as a genuine source of competitive advantage and were unwilling to let any outsider peek. One Human Resources VP put it bluntly when he turned me down: “We would no more show our performance appraisal form to a bunch of outsiders than the Coca-Cola company would let you come in and look over the secret formula for Coke.” What’s different in these places? What are America’s best organizations doing in performance management that the rest can learn from?
There’s probably no management process that has been the subject of more Dilbert lampoons than performance appraisal. But at America’s best-managed corporations, performance appraisal is no Dilbert joke — it’s serious business. Probably no one better exemplifies the tough-minded manager than GE’s Jack Welch. And there’s no stronger advocate of demanding performance appraisals. "You have to go along with a can of fertilizer in one hand and water in the other and constantly throw both on the flowers," he said. "If they grow you have a beautiful garden. If they don't, you cut them out. That's what management is all about." General Electric is known for its insistence that managers force rank all of their people according to their talent and potential. Welch is never far from the big books that hold the vitality rating on 3000 GE professionals. GE breaks performance into five categories, numbered 1 through 5. "Ones are the top 10%,” says Welch. “These are the top people. Twos are the next strongest 15%. Threes are the middle 50. The ones in the middle have a real future. The Fours are the caution 15%. They can move to the left. Fives are the least effective 10%. We've got to get rid of them. We don't want to see these people again.” Cruel? Not according to Welch. What’s really cruel, he counters, is falsely telling a manager that he’s doing fine for twenty years and then firing him at 53 when he’s got two kids in college. GE’s demand that managers force rank their people is not uncommon at many top firms. Andy Pearson, Harvard Business School professor and former PepsiCo president, used to engage every manager in the soft drink giant in frank, tough discussions of each subordinate — and insist that they then carry the same demanding discussion process down. Sort your population into four groups, poor to superior, and then ask for a specific plan for the people in each group, Pearson recommends. “Always focus first on the bottom group. Rooting out the poorest performers will foster a climate of continual improvement. If everyone in the bottom quartile is replaced, the third quartile becomes the new bottom group and the focus of subsequent improvement efforts.” Constantly refine your gene pool. Upgrade the quality of the herd. Find the best; cull the rest. Many of the organizations that perennially make Fortune’s most-admired list take that blunt approach. Through its employee-appraisal system, Microsoft annually weeds out about 5% or so if its work force. Intel Corp. also has a tough succeed-or-out program. In 1998 Ford put in motion a plan to offer a buyout program targeted explicitly at poor-performing salaried employees or those "average/solid employees with limited potential." About 10% of Ford's 55,500 white-collar workers were marked. But no performance management process is more fiercely resisted. Their recently hired vice president of human resources abruptly left EDS last year when the CEO ramrodded a quartiling plan over much employee resistance. Spotlighting talent also-rans can destroy the congenial environment that many organizations enjoy. Managers complain that force-ranking systems, like EDS’s quartiling approach or Texas Instruments’ 10-10 system, where supervisors have to finger their best and their worst, is inherently unfair, particularly if they’ve already cut the unit’s fat and are now expected to scalpel bone and muscle. Glenn Lovelace, president and CEO of Austin’s rapidly-growing TManage, Inc., argues that the appropriate model for top organizations is Harvard’s medical school: “Harvard Medical School is incredibly difficult to get in . . . incredibly selective. But once you’re in, you’re in. Nobody gets flunked out because you’ve got the best — it’s a collegial environment.” Leadership Darwinism it may be, but Welch, Pearson, and many other CEOs argue that a merciless push to upgrade human capital is vital. No one enjoys delivering bad news, but good managers understand that it’s critical to long-term success. Does it demoralize? Actually, just the opposite. Top performers relish working in an environment free of non-contributors, and what genuinely demoralizes is a climate that tolerates mediocrity. Supervisors, each the captain of an individual ship that en masse makes up the organizational flotilla, must ask of each member of their crew: “Is this person a sail or an anchor?” Sails must be rewarded and retained; anchors must be confronted and, in the absence of immediate and dramatic change, eliminated.
In spite of encountering several decliners, our APQC / Linkage benchmarking study did identify many “best-practice partners” — organizations that have developed genuine performance management breakthroughs and are forcing rigorous assessments of talent and potential. But even among these standouts, two of the five that agreed to participate and were site-visited by representatives from the funding sponsors still requested anonymity in all reporting of results. Why the secrecy? Again, their conviction that their performance management practices provide them a competitive leg up. In the best-practice organizations we identified, along with many others that take the process seriously, performance appraisal is the acknowledged central tool for spearheading a change in their corporate culture. “We’re skilled at managing toward mediocrity,” the VP-HR of one anonymity-requesting consumer goods company admitted. “Now we have to manage toward excellence.” Traditional approaches to performance management and people development — like promotion from within based primarily on job tenure — are no longer good enough. A company that uses experience as its primary criterion for advancement is encouraging organizational hardening of the arteries. “Culture-change” is a term commonly bandied about, but at these organizations it’s serious business. In the past, as long as Charlie was diligent and dependable and Jane was loyal and industrious, that was all that a company felt it could ask. Now loyalty and diligence are viewed as threshold factors — taken for granted, the price of admission — and continued tenure with the organization is contingent on delivering the goods. Both behaviors and results are assessed in the performance management systems of top-tier organizations. Performance appraisal’s role in changing corporate culture from a best effort to a results-driven culture was one of the chief findings of the APQC best-practices study. Another finding used tough language to make a similar point: “Best-practice organizations are using their performance management systems to directly target poor performers for termination.”
Over the past several years, one of the significant advances in the technology of performance appraisal has been the identification of specific “core competencies” by organizations. Limited in number and critical to organizational success, competencies define for all members of the organizations the critical behaviors, skills, attributes and proficiencies that every organization member is expected to possess and display. The performance management system plays two roles here. First, it frequently serves as the mechanism for identifying the finite number of competencies against which employees will be assessed. Creating a new performance management system may force the organization to determine just which attributes or factors are truly core to the organization’s success. Second, the performance management system can be the primary driver to assure that these competencies are fully understood and institutionalized. One of the key differences between the best-practice companies in the APQC study and a group of 24 others that served, as a control group was the significantly greater emphasis placed on the identification and assessment of competencies by the best. The process for determining which competencies to assess turns out to be exactly the opposite as the one used to set goals and determine objectives. In setting goals and objectives, the manager and the subordinate are the key players with an enormous amount of work to do. They must determine the specific accountabilities of the subordinate's job and identify the various roles, which the individual is expected to play. For each of those roles or accountability areas they must determine the goals and objectives to be achieved. In some cases the objective will be merely a maintenance one: keep things running as smoothly in the future as they have run in the past. In other cases the objective will involve significant stretch and growth. Having determined goals and objectives, they will set checkpoints along the way and determine what satisfactory performance will look like. Then, over the course of the year, they will revise objectives as missions are accomplished and strategies change. The opposite is true of competencies. Competencies are determined on a corporate basis and apply to all; individual raters and ratees may — at most — determine which ones to particularly emphasize. Goals change; competencies do not. Competencies tend to be permanent; objectives are ephemeral. As a result, it is vital for the organization to choose wisely when it publishes the list of competencies against which individuals will be assessed. In putting forth its list of competencies, the organization is telling its members that these few are the most important attributes that we seek in members of our team. Of course, there will be other attributes expected of corporate citizens — no one will argue that any list of competencies, no matter how long, is exhaustive. But whatever items do not appear on the list must necessarily be less important than those factors that do make the cut.
Competency-identification is a powerful organizational exercise: it forces top management to determine what really is important for the company’s success. Dozens of important attributes and proficiencies can be offered up for consideration. One study concluded that there are 67 discrete factors, which correlate positively with job success; another found over 100. Every effective performance evaluation system focuses on both competencies and results. But which competencies should the organization pick to assess? One practical and highly enlightening way to help top management identify the most important competencies for their organization is to list all of the possibilities, together with a definition or explanation, on separate index cards. Each senior officer is then responsible for arranging the cards into three equal piles: Must, should and nice. TManage, Inc. is a rapid-growth technology company that sets up and manages the telework operations of huge corporations. Today a company of 55 people, by the end of the year a payroll of 300 is anticipated. Since the employees of TManage will themselves be telecommuters and only infrequent visitors to the corporate office, creating an effective performance management system for their unique operation was one of the founders’ top priorities. For two hours one morning earlier this year, the eight senior officers of TManage worked at their conference table. Each had been given an identical set of cards with 21 different competencies: Accountability, Customer Focus, Results Orientation, Developing Talent, Ethics/Integrity, and sixteen others. Their task: sort the deck into three piles of seven cards each. Was “People Management” a Must, a Should or a Nice? Was “Risk Taking” most, middle or least important for corporate success? The results of the first round were predictable: most cards landed in the Must pile with only a few in the Should. Nice was empty. Each manager struggled to refine the cards so only 7 appeared in each pile. The demanding discipline of identifying only a few crucial competencies forced serious discussion and decision-making. And the resulting determination of the competencies to be assessed also helped top management focus on what was truly pivotal to the fledgling organization’s success — and what was important but secondary. Two other sophisticated techniques in the development of the TManage performance management system added to its success: First, the use of mastery-level descriptions of performance for each competency selected; second, the creation of a behavioral-frequency rating scale.
A glaring error on too many performance appraisal forms is that desired performance is only defined, not described. It’s easy enough for a form developer to stick in a Webster’s definition of a word like Accountability or Leadership or Teamwork. But what does it look like in practice? How do you know it when you see it? If you watched a teamwork or leadership master at work, what would you see? Far more useful than definitions are mastery-descriptions: narrative portraits of behavior that one who has mastered the area will likely engage in. While they are much more challenging to create, mastery descriptions give the appraiser a benchmark against which to compare the actual activities of the individual he is assessing. Even better, it provides the appraisee with a clear picture of exactly what the organization expects. For example: Customer Focus: Knows his/her customers and can describe their expectations. Meets all of the expectations of both internal and external customers. Gains customers’ trust and respect. Manages customer expectations. Actively seeks customers’ feedback on quality of service he/she provides. Puts the customer first. Ethics/Integrity: Respects and maintains confidentiality. Avoids rumor, gossip and subjective opinions in decision making. Admits mistakes in spite of the potential for negative consequences. Presents unpleasant or disagreeable facts in an appropriate manner. Keeps promises; meets goals and deadlines. Avoids situations and associations that could be considered inappropriate. Honest in all dealings. A good role model. Trusted by all. Positive Attitude/Enthusiasm: Displays a positive attitude and optimism about the work to be done, the people he or she works with, customers, management, and company policies. Has a constructive sense of humor. Doesn’t spread gossip or rumors. Acts as a positive influence on others. People like being around this person. Is courteous, cooperative, and helpful. Once the core competencies have been identified, an equally difficult challenge awaits: How should these competencies be evaluated? What kind of scale should we use? Picking the scale values to be used to evaluate people generates more grief than almost any other aspect of performance appraisal. If a numerical scale is chosen (1-2-3-4-5) people whine about being reduced to a number. If a comparison-against-standard system is used (Failed to meet expectations / Fully met expectations / Greatly exceeded expectations), then low-ranking performers will demand of hapless appraisers that they specify precisely what the expectations are. And if an absolute-judgment label scheme is used (Marginal / Fair / Competent / Superior / Distinguished) then everyone will complain about the middle rating being a connotation of mediocrity. (Mike Vent, COO of TManage, Inc., still fumes over a “Wholly Adequate” rating he received twenty years ago.) There is a better way. Use a behavioral frequency scale. Instead of forcing the rater to judge the individual being assessed, behavioral frequency scales ask raters to indicate how frequently the appraisee behaved like a true master. To see how it works, read any of the three descriptions above. Next, think of someone whose performance you are responsible for assessing, and then ask yourself, “How often does the person do all the things listed in the description? Does she do them rarely, occasionally, frequently or regularly? That’s the essence of a behavioral frequency scale. It avoids absolute judgments, and instead asks the rater to do the far more palatable job of describing how often true mastery is exhibited. Another advantage of the behavioral frequency approach is that it directly guides performance. Instead of having to figure out examples of each of the items on a corporate competency list, all the manager needs to do now is review the list with the subordinate and say, “Just do the things on this list and you’ll be a fully acceptable performer.” Finally, behavioral frequency scales make for easier, less defensive reactions when bad news has to be delivered. Instead of forcing the manager to call Sam a 1 or a Marginal performer, the manager can say, “Sam, in this area of Customer Focus, only Rarely do I see you doing the things listed here. What do you need to do so that the next time around I can report that I see you doing this all the time?” Of course, there are some assessment items that just don’t lend themselves to evaluation through a behavioral frequency scale. Particularly when it comes to assessing how well the individual met her job description responsibilities, or achieved the goals that were set in the initial planning meeting, absolute judgments are required. But best-practice organizations work hard to come up with rating verbiage that reduces resistance. Alcon Laboratories calls its middle rating GSP — their shorthand for “Good Solid Performer.” Who could object to being called a good solid performer? Another highly palatable term for the center of the scale is “Fully Successful.”
One of the primary reasons that writings about performance appraisals are such eye-glazers is that the advice proffered is so often either stale or wrong. Triteness abounds in performance appraisal literature. “Listen to what the individual has to say,” is perpetual advice appraisers are given, but nobody tells them exactly what it is they should listen for. “Typist” is offered up as a job example — but the content of a typist’s job rarely bears any relation to the real jobs a manager has to assess. The banal acronym SMART is ceaselessly served up, each time by an author who assumes that the counsel to create objectives that are Specific, Measurable, Attainable, Realistic and Time-bound is fresh and enlightening. “No surprises in the appraisal interview!” managers are continually admonished — but often the trigger of a performance appraisal deadline crystallizes a manager’s vague concerns about a subordinate’s shortcomings and forces a conversation that otherwise would not have occurred. Perhaps the worst advice appraisers are urged to swallow is: Be objective! Keep your judgments and personal feelings out of the assessment process. The flaw in the objectivity admonition is the bogus notion that objective means quantifiable, and that if it can’t be counted then it isn’t objective and therefore shouldn’t be used. That’s nonsense. The talent of a pianist is not measured by the number of notes she plays. The quality of a priest is not a function of the number of confessions heard. The issue isn’t quantifiability, its verifiability. Numbers just happen to be easy to verify. While quantifiable, numerical, countable measures would be nice to have for every objective, the search for meaningful quantitative measures is often fruitless. When I was designing the performance appraisal system for the National Security Agency two years ago, we confronted this issue of “objectivity” directly. NSA hires more linguists and translators than any other organization in America, public or private. How do you evaluate the performance of a translator? Do you measure the number of words translated? Better not. What NSA really wants from its translators is the ability to capture nuance, and no quantifiable measure of that capability exists. But it certainly can be described and evaluated: a skilled native speaker can easily sort translations into separate piles of those that read like machine generated transliterations and those that capture the soul of an author’s words. NSA also employs thousands of programmers. Should the quality of their performance be assessed by the number of lines of code they write? Caution: danger ahead! What the agency needs from its programmers is the ability to write elegant and parsimonious code. There’s no quantifiable measure, but a practiced systems analyst will know it when she sees it. A glance in the dictionary reveals what “objective” really means: “1. Uninfluenced by emotions or personal prejudices: an objective critic. (See synonyms at fair.) 2. Based on observable phenomena; presented factually: an objective appraisal.” And what’s the true meaning of subjectivity? “Taking place within a person's mind, unaffected by the external world. Moodily introspective. Existing only in the mind; illusory.” Writing a person’s performance appraisal requires the manager to be fair, objective and unprejudiced. But the fair and objective requirement does not mean that the manager is restricted only to quantitative, numerical resources in completing the assessment. The manager’s feelings, opinions and judgments are precisely what is demanded by the appraisal process. Managers are paid to make judgments even when — or particularly when — all of the facts are not available. In every other area of managerial activity, the ability to act appropriately on the basis of limited and occasionally conflicting data is celebrated and rewarded. Only in the case of performance appraisal are we uncomfortable about the fact that non-quantitative, experience-based information is used. Actually, people don’t want “objective” information. What they want is their boss’s opinion, plain and unvarnished. In the appraisal setting each appraisee is asking: “How am I doing, boss? Tell me the truth. What does my future look like here? My family’s happiness depends on your perception of my contribution and potential — skip all the ‘objective’ stuff and tell me what you really think.” Managers need training less in how to create and deliver an objective appraisal and more in how to summon the courage to tell it like it is; to talk straight from the heart. Giving someone a performance appraisal is like being in the Olympics of management. This is not day-to- day stuff. Very few people ever get to do it; fewer still do it well. Training is critical.
Organizations with world-class performance management systems do things that the also-rans don’t. They insist that all managers maintain consistent, demanding standards for everyone — and they keep raising those standards. They work relentlessly to identify their highest-potential managers and professionals and develop them quickly. They move marginal performers aside so they don’t block the path of talent; they eliminate non-contributors swiftly. They treat their Human Resources departments as partners, staff them with the highest caliber talent available, and insist that they be active agents for change.
Dick Grote is a management consultant in Dallas, Texas, who specializes in helping organizations design effective performance management systems and build leadership excellence. He is the author of the management classic, Discipline Without Punishment, The Complete Guide to Performance Appraisal, and The Performance Appraisal Question and Answer Book. His most recent book, Forced Ranking: Making Performance Management Work, was published by the Harvard Business School Press. Grote Consulting offers clients expertise in employee performance appraisal, employee performance improvement and talent management. Dick is also the developer of the GroteApproachSM web-based performance management system. He can be contacted at dickgrote@groteconsulting.com.