Grey Matter

Confidential to Betsy Levine, branch general manager at Arrow Electronics circa 1997: Quit your job.

I’m sure by now she already has. Betsy’s management performance review is one of five exhibits evaluated in Harvard Business School Case Study 9-800-290, Compensation and Performance Evaluation at Arrow Electronics.

Arrow’s industry has dramatic turnover, up to 25 percent annually. Its employees, and those of its competitors, feel no attachment to any particular company. When these companies need to build market share quickly due to pressure from suppliers, they do it by stealing customers from another distributor via job-hopping salespeople. Even the company’s 43 branch general managers only have an average life span of three to four years.

The main reason salespeople jump ship: pay. There appears to be nothing particularly compelling about Arrow’s culture – in fact, its culture is “no one stays.”

Betsy’s drama began when Steve Kaufman, Arrow’s CEO, recognized that “problems” existed with the company’s Employee Performance Review system, or EPR. No one was happy with the three-year-old system: not employees, not managers, and certainly not Kaufman.

Kaufman was in a state of shock that “no one” in his entire organization received a 1 or 2 (the lowest scores) in one of the seven performance areas. As he put it, “that can’t be right.” Unfortunately, that’s the wrong mindset. He wonders why it is so difficult to get his managers to evaluate their employees “accurately.”

Well, because they can’t. An employee evaluation such as Arrow’s, which ranked employees on measures such as judgment and initiative, is highly subjective. What’s accurate to one manager is inaccurate to another.

Kaufman may suffer from his own educational background and experience: He’s a quant*. He has an engineering degree from MIT, an MBA from Harvard, and 11 years with McKinsey as a strategy consultant. He’s completely overlooking the impact of emotion and subjective judgment on the managers’ assessments of their employees.

Kaufman even decreed that “Every employee must receive a 2 on at least one of their seven areas of evaluation. No exceptions.” This silly, arbitrary assessment of course rubbed his managers the wrong way, let alone the employees. Part of Kaufman’s desire was to determine who was really worthy of promotion. But determining leadership potential from a subjective rating alone is faulty. What if the employee’s manager feels threatened? Even more important: What about the employee’s intrinsic desire to lead? Wouldn’t it be better to ask people if they want to lead and then assess their potential and provide training?

The CEO wants to “upgrade” his team — rather a loaded word, don’t you think? Any new hires need to be “better than the average of the existing team,” so that’s why he needs to measure performance “accurately” via the EPR. Well, wouldn’t the most expedient way of increasing the average be to increase the performance of every individual member of the team, rather than hire one or two better-than-average superstars?

So in addition to the written case, we’re given five written Management Performance Reviews to evaluate as exhibits. Three of them are extremely sparse and vague – one says to “develop some continued educational goals.” About what? Another refers to “developing direct reports” under position-specific knowledge and skills. How is that position-specific knowledge for a manager? Three of them were extremely poorly written, including a reference to a “pier group.”

And then we get to Betsy’s review. Poor Betsy. She exemplifies the problem Kaufman wants to solve: She’s a newly promoted branch general manager in her role for less than a year – exactly the person who’s likely to leave in another year or two, and exactly who Arrow wants to retain in order to grow its leadership team in years to come.

So what kind of review does she get? A coal raking. A mean-spirited, highly critical rant that at the same time acknowledges that there is “little to no turnover” in Betsy’s area – a key metric Arrow needs to improve. And that Betsy goes out of her way to help others in the company, which she’s dinged for. And that her customers, both internal and external, are satifisied. She even gets downgraded for the opinions of those who are not her direct reports! Her evaluator goes off topic on many of the questions, filling in what he wants to cover when he (I’m assuming it’s a “he”) is supposed to be addressing something else entirely.

And that underlies the fundamental problem with Kaufman’s EPR process: the pre-fab structure used to evaluate employees doesn’t focus on what’s really important to the company, just what’s most important for filling out the form. And the variation in the narratives – some are inscrutable, misspelled one-sentence responses, while Betsy gets a three-page diatribe – points to a flawed process poorly executed.

Kaufman made a fundamental error: The “problems” are not with the EPR. The problems are with the system as a whole.

* Quant: A person who works in finance using numerical or quantitative techniques.

Jeffrey Pfeffer is right about the essential issue with performance-based incentive programs: “Most companies have production processes and objectives that are way too complicated to be adequately captured in any incentive scheme.” As Pfeffer points out, the answer to complicated production processes is to make performance measurement more complicated. However, the more complex it gets, the less likelihood it has of guiding behavior.

I used to receive an incentive payout based in part on the overall patient satisfaction score for a 528-bed acute-care hospital. For a hospital with more than 20,000 annual admissions, there are literally millions of individual interactions and transactions between patients and employees that would eventually comprise the score, including how the survey itself was conducted and the response rate. Patient satisfaction scores are based on a huge range of variables, everything from how the patient was greeted at the front desk to whether the food was hot on Tuesday. Weeks after the fact, the patient will get a bill, which will also affect how he or she perceives the experience. And of course, even the physician can’t guarantee the patient will get the cure he or she seeks.

While it’s great in concept to make us “severally liable” as a team, from a practical standpoint no one individual has a significant effect on this score. Employees are not guided in their moment-by-moment choices by what’s going to be on their annual review. They’re guided by what they’ve been trained to do, what their culture expects them to do, what their peers do, guidance from their supervisor and their own judgment and experience.

Am I saying this measure, patient satisfaction, isn’t important? Of course not – it’s crucial. But what really guides our behavior as employees? Broader internal forces, such as organizational culture and behavioral modeling by superiors and peers, are more likely to have impact on the day-to-day actions of employees than a once-a-year measurement process.

So why is it there? It’s the path of least resistance. It’s easier for leaders to spend one week or so tweaking the performance measure process for the year than it is work daily to change organizational culture, to really look at how to improve the work environment and communication. Managing requires developing a clear vision and goals, communicating that vision to employees, and repeating those goals over and over again. The ideal is for employees to have such clarity of mission and purpose that it guides them almost intuitively to make decisions and take actions which lead directly to achieving the business goals. They know the patient wants to be greeted warmly at the front desk, receive hot food and an easy-to-understand invoice, and they act accordingly.

“Sins of Commission: Be Careful What You Pay For, You May Get It” is excerpted from Stanford Management Professor Jeffrey Pfeffer’s book What Were They Thinking? Unconventional Wisdom about Management.

The anti-Nordstrom

Case Study HR6 from the Graduate School of Business at Stanford University asks us to review SAS Institute’s need to recruit a talented work force in order to build and maintain intellectual capital.

Could it, and should it, maintain its unique approach to pay and practices, and could it reasonably expect to thrive?

To answer that question, we need to determine if SAS succeeded because of its management practices. Or is its success simply due to being in the right place at the right time?

While the former certainly plays a role, I think there’s a bigger reason for the company’s success, and the reason WHY its management practices exist in the first place: It’s privately held.

Its need to please no one but its relatively few owners meant it could create a “benevolent dictator” such as Jim Goodnight who then creates benevolent employee policies and practices. With an internal locus of control, the company can eschew long-term planning and create a truly customer-driven development process. Its licensing structure, which forces the company to work to KEEP customers yet trades off margins in the short term for greater market penetration, would be less successful in a company focused on the drive for quarterly earnings. Although SAS doesn’t have a long-term planning function — wisely, since its industry changes so rapidly — the company certainly takes a long-term view in how it treats employees and their contributions to its success.

SAS has no grand policies about how to treat employees, just four basis principles, one of which emphasizes intrinsic motivation. The company aims to “deemphasize financial incentives as source of motivation.”

It’s really the anti-Nordstrom: even its salespeople are not commissioned. According to SAS leadership, sales commissions do not encourage an orientation toward taking care of the customer or building long-term relationships. Its HR leader commented “People are constantly finding holes in incentive plans.” Yes, no kidding. Just look at Nordstrom!

Does this kind of compensation system make it difficult to attract and retain talent? No – it eliminates the “sharks” and encourages people who really want to serve the company and its customers. A commission-oriented, extrinsic motivation structure means employees will move as soon as that motivation looks better somewhere else.

Can SAS succeed without the stock options common in Silicon Valley? Resoundingly yes. Everyone wants off the treadmill, to some degree. SAS counts on the “value of a return compounded annually” – getting its employees to stay for the long term. It’s a better guarantee for the employee than stock options, which are inherently risky. Yeah, you might get rich – but then again, you might not. Best of all, when SAS employees develop new ideas they will probably stay rather than forming their own startup, keeping intellectual capital within the company.

There’s no question that SAS will continue to thrive with these practices. In fact, they are thriving. In 2010, 12 years after the case study was published, SAS was named the No.-1 company on the Fortune 100 Best Companies to Work For 2010 list – after appearing on the list every year since its inception.

The anti-SAS

Harvard Business School Case Study 9-191-002 questions the compensation policies and practices of Nordstrom, which are inextricably linked with its high-touch customer-service culture. After rapid expansion in 1980s, this publicly traded retailer was lambasted by labor unions and the news media for its practices, which allegedly led to employees underreporting actual work hours in the interest of keeping both their commissions high and the best floor schedules, as well as currying favor with management.

Nordstrom, which started out as a family business in Seattle, grew rapidly in the 1980s. Its sales force grew by six times in less than 10 years as the company expanded geographically beyond its home base of Washington and Oregon.

Its reputation for “superior customer service” is considered a strong competitive advantage and source of its financial success, including sales per square foot double the industry average. However, that’s a strategy that can be copied readily. “Superior customer service” is only part of the equation in retail: No matter how good your service, if you don’t stock the clothes and shoes customers want to buy, they’ll shop elsewhere.

From the case study, there is little evidence of a strong management emphasis on creating a “great culture,” as at Southwest Airlines.  Although publicly held, the Nordstrom family owns half of the company. The family seems a little entitled, as witness this comment from Jim Nordstrom: “People don’t put in enough hours during the busy time.” What?

Nordstrom’s caliber of salesclerks seemed to withstand the pressures of rapid growth, but apparently it didn’t have a strong hiring screening program or training, as at Southwest. Most importantly, it didn’t seem to really care about treating employees well and backing it up with fair rewards.

Nordstrom’s policy was to “to pay employees for time,” but its practice was something different. The service behavior for Nordies is almost codified, such as the expectation that they make home deliveries. Compensation is driven by sales per hour, so if an employee does any work that’s not ringing up the register, he or she is effectively penalized. Why can’t they just put those activities against non-selling time? Or adjust the commission structure? Its culture made it clear what the “right” thing to do is, yet without breaking any laws.

A management memo defined tasks such as writing thank yous or attending a meeting “selling.” But writing a customer a thank you is no guarantee that customer will make next purchase from YOU! That should not be classified as a selling activity, but rather as a post-sale activity. The company should pay for that, not the employee, since the company overall is most likely to benefit from the next purchase. [Bruce Nordstrom compared this non-selling time to an advertising salesperson, or an insurance salesperson. But there’s a big difference in the business models: The next sale will probably go DIRECTLY to that person in those businesses.]

There’s no question that the United Food and Commercial Workers union, which represented about 5 percent of employees, was behind a lot of the complaints. It reported Nordstrom’s practices to the National Labor Relations Board and the Washington Department of Labor and Industries. It clearly wanted to grow its membership by representing a greater percentage of employees, and what better way to rally them by telling them their employer is a cheat?

Although not every employee is unhappy – in fact, a fair number seemed to have stood by the company, and they later ousted the union – at the root there is something inherently wrong with the company’s compensation structure. The old saw “Where there’s smoke, there’s fire” lends credence to this case study.

However, the union’s claims were not entirely true, and this is why Nordstrom in essence prevailed: Employees DO get compensated for selling work. It’s just that the underlying system discourages reporting the hours. The irony: If there were no union, would any of this have come to light?

When Denny Flanagan performs outrageously for United, he’s probably not being compensated. However, he’s an outlier within the company – no one is expecting him to do that, or marking him down if he doesn’t. At Nordstrom, however, everyone is expected to perform outrageously, but unlike Southwest, the environment seems punitive, rather than fun.

Nordstrom tells employees, “This is your business, treat it like your business.” No, it’s not their business – they don’t get the profits! The family and shareholders get the profits.

In the end, the Department of Labor and Industries order that Nordstrom compensate employees was rather silly, because it did not address the systemic and cultural issues at the company. What will change? Yeah, now employees have a written time sheet, but will they use it?