I'm Sorry I broke Your Company

4. Standardized Human Asset Management Is a SHAM How Performance Management Demoralizes the Performers

While the '80s and '90s saw the rise of strategy and process consulting, the new millennium brought with it a focus on human asset management. The last piece of the command-and-control structure of the strategy, processes, and metrics framework is the people required to enact it. Although «human asset management» and «human capital management» were the original names given to the practice of managing a company's people, they have largely been replaced with «talent management» due to the industrial connotations of human assets. Some people don't react well to being labeled similarly to a line item on a balance sheet. However, the legacy of that term is the ubiquitous corporate declaration of «People are our greatest asset!» which you’ll find on every company website and in every annual report.

Human asset management systems include a slate of human resources processes and methods, including performance management, incentive compensation, competency development, career planning, leadership development, career and leadership coaching, succession planning, and learning management. In this chapter, I will cover only performance management and incentive compensation, and I will discuss other aspects of HAM in the next three. (I couldn't resist the SHAM title.)

Even though the fields of strategy development and process reengineering consulting are crowded, I am often overwhelmed by the multitudes of consulting, coaching, and software companies offering solutions for all your talent management needs. I hate the copy on these companies'websites and in talent management promotional materials because it has such a dire tone urging use of these systems. Much of this tone can be traced back to the book The War for Talent published in 2001 by three McKinsey consultants who coined the term «the talent war.» The book was written during the dot-com bubble with the premise that because aging baby boomers would leave the workplace faster than young people could enter it, companies would have a talent crisis in the upcoming decades wherein they would not be able to fill vital leadership roles. Of course, the authors didn't predict the economic crisis of 2008 wherein young people wouldn’t be able to enter the workforce at all. (I'm trying really, really hard not to say «You can't predict the future!» Oh, darn.)

However, despite the high unemployment rate, the tone of crisis still exists, and without this set of methods and practices in place, a company will be unable to identify and groom future leaders, will lose all It's best people to competitors, and will eventually shrivel up and die. It reminds me of the tone of many parenting magazines that play on mothers' fears — if you don't follow the recommended practice, your baby will be in danger. You don't want to put your company in danger, do you? I think the reason the written material has such emotional urgency is because rationally, many of these methods make no sense.

Of the whole alphabet soup of management methodologies, from competitive strategy to competency development, if I had to choose just one that has done the most damage to companies and their employees’ lives, the modern performance management system is the winner. In It's current state this system is an automated tool that cascades goals and measures into an employee appraisal and that results in an overall employee rating tied to incentive compensation. With the input of various employee parameters and the click of a few links, an employee’s history can be boiled down to a few numbers and her future career with the company ordained. I can’t really trace the origins of this system to a book or a method. It seems to have evolved from the management-by-objectives movement that tied compensation to goal achievement. Then someone added competencies and development planning, and the whole thing got automated.

My first exposure to management by objectives was in the early ’90s when I was consulting on a business transformation project that involved both strategy and process work. Although we were already accustomed to creating strategic objectives with targets and breaking those down into individual performance metrics, this was the first project in which we would tie these specifically into compensation, which we called «pay-for-performance.» (I apologize for all the jargon, but this particular field is rife with jargon.) This was also the first time we would base all our work on the balanced scorecard. Fortunately for the company, the software to do this automatically had yet to be developed, so this particular system was paper based and applied only to the top levels of management.

When the HR consulting team explained how they were creating a compensation system based on target achievement, I was captivated. The compensation piece was what was missing from my other projects, and this was the linchpin that ensured the whole organization did as it was supposed to do. This way, clients had to implement our recommendations; otherwise, they wouldn’t get their bonuses. It was perfect! Plus, it was great for consulting firms. Strategy consultants could bring in their other arm, operations consultants, to develop the metrics, the targets, and the collection and reporting processes while HR consulting teams could design the compensation systems, goal development, and performance evaluation processes. Then the IT consultants could automate the whole thing. Over the next decade, I observed the growth of performance management systems from big players like PeopleSoft and SAP to specialized companies like Success Factors and Halogen and the consulting firms that could help implement them. Everyone I know who works in a large company is subject to some version of a performance management system.

In 2000, I left consulting and joined a division of a Fortune 100 company that was acquired by another in 2007. I got to experience firsthand all the havoc we consultants had been wreaking for years with this system. Performance management at both places was an automated process that consumed the last six weeks of every year. This is work that does not contribute to developing new products or services or helping customers. Yet every year, I spent a good part of November and all of December writing up lengthy reviews, reviewing and approving others'reviews, attending several daylong meetings to haggle over employee ratings, ranking them, identifying high performers, and then having to inform direct reports of their ultimate rating — the number spit out by the system based on their weighted goal achievement, their mastery of leadership competencies, and a comparison with their peers. What started out as a way to implement strategy turned into a system that rates, labels, and tracks employees the same as if they were inventory. No wonder it was called «human asset management.» It is!

An elegant methodology that theoretically would eliminate nonstrategic work and ensure an aligned and motivated workforce ended up in practice to have the opposite effect. This model is a sham in so many ways that It's difficult to know where to start enumerating It's flaws. Even if we pretend that the assumptions upon which it is based are true — people are motivated only by money, employees improve their performance after being judged and graded by a superior, the whole system is fair and objective, and no one games it for his benefit — the sheer amount of effort, time, and costs involved in implementing, administering, and maintaining such a system is enough of an argument against it.

Lets start with what's involved in implementing this type of compensation system, assuming that It's done with every effort to be fair across the company. Typically, the first step is to determine the salary increases, bonuses, stock options, and stock grants by job and performance level. (Or better yet, the first step is to hire a consulting firm to do this for you.) For instance, under a new reward system, a manager performing at a satisfactory level would get a certain percentage of bonus and salary increase and, perhaps, a package of stock or stock options. That sounds easy enough, but it assumes that the company has standard job levels across all functions. In my experience, a manager in one department is very different from a manager in another and even more so if the company has more than one line of business. If the company has grown as the result of mergers or acquisitions — and seriously, what company hasn't — then job levels are probably really inconsistent.

The company therefore needs to implement standard job levels company-wide or at least group the current levels into a set of standards. This is the only way to ensure fairness. Anyone who has experienced standardizing titles and job levels knows that this is a highly emotionally charged and political process. Slotting every employee into a level is a huge undertaking, and It's aftermath is a vast swath of disgruntled employees. A senior manager doesn't enjoy being demoted to a manager (it rarely goes the other way), nor does an associate director like finding out her job level is the same as a senior managers. All those years of clawing your way up that extra rung of the corporate ladder vanish with a stroke of the delete button.

The next step is to standardize the performance appraisal forms and ratings. Again, if the company has silos or has grown through acquisitions, a variety of forms and scales are likely in use. Some departments will want a 1 to 10 scale while others will want 1 to 5. It takes another company-wide effort to agree on one system and method, but with some cajoling and compromise, this can be done in a couple of months.

The third step is to develop a simple process governing when goals need to be written, how goals will trickle down from the corporate goals, when appraisals need to be completed, and when rewards will be distributed. And the last step is to communicate the new process, calendar, and form to all employees and offer training if needed.

However, after a paper-based system is implemented, a few wrinkles appear:

• Not all goals are created equal. Some people have a knack for writing easily achievable goals. Other goals cant be readily quantified or measured for achievement. Plus, managers' expectations of their direct reports vary widely. A standard method of writing goals is needed.

• Performance is very subjective. What one person judges as «meets expectations,» another person judges as «exceeds expectations.» This also calls for some standardization.

• Developing goals at the beginning of the year to be rated on at the end does not allow enough flexibility to respond to changing business conditions. People need the ability to change goals during the year, but changes would have to be subject to approval or else people would constantly update their goals to what they have already done, defeating the whole purpose of setting goals.

• Waiting until the end of the year to review goals does not give anyone the opportunity to course-correct if goals are not being met. Managers and employees will need to review goals more than once a year.

• A tremendous amount of paperwork is involved that needs to be managed!

At this point, most companies realize that they cant manage this process with paper. Today, tons of software packages are available, so It's likely that this process will be automated. This does require a whole new project, however, to both implement an IT system and to address the deficiencies above, which means several new initiatives are needed:

• Development or purchase of an information system.

(Cue the information technology consultants.)

• Creation of the infrastructure needed for the system, including the hardware and the staff for development, maintenance, and user training and support.

• Company-wide training and communications on how to write SMART goals.

• Manager training on how to rate performance objectives fairly.

• Business processes to ensure fair assessment across departments, like cross-functional calibration meetings.

• HR command-and-control functions to ensure fairness of goals and ratings and the accuracy of information.

• Periodic reviews to update goals and course-correct and documentation regarding these periodic reviews.

What is the result? An organization that spends a significant amount of time writing goals, reviewing goals, revising goals, ensuring SMART compliance, approving revised goals, summarizing performance to date, reviewing performance with managers, attending calibration sessions to compare ratings and force-rank employees into a normal distribution curve (which results in another revision of goals and ratings), and approving periodic reviews in an information system. And lets not forget about updating and maintaining the performance review system and, of course, learning how to use the information system and how to comply with all the steps involved. Basically, most of the last quarter of the year is devoted to this process. If the reviews are done semiannually, add two weeks in the summer, too. In addition to losing weeks of work to this rating process, the company has added overhead in the form of HR and IT personnel needed to manage the system, staff a help desk, and collate reports as well as the ongoing costs of hardware and software. Worst of all, managers have less time to spend with their direct reports because of all the requisite paperwork and mandatory meetings.

Remember the point of the pay-for-performance system? It was to focus everyone on the corporate goals to ensure the entire organization was carrying out the strategy. Instead, we have a huge administrative burden that doesn’t actually contribute to executing the strategy, unless your strategy is managing goals and implementing incentive compensation. This wouldn’t be so bad if it actually helped improve employee motivation and morale and, hence, performance, but in my experience, it does the opposite. The reality is that such systems are designed to dole out money, not to provide feedback to employees or help managers coach their direct reports. Instead, they take time away from relationships and impose a set of standards into which one is required to fit people.

Lets review the assumptions upon which these systems are based:

1. These systems are fair and objective.

2. Being evaluated and graded by a superior improves performance.

3. People are motivated by money.

4. People wont game the system (covered in the previous chapter).

These systems don't even come close to being fair and objective. At most, they just pretend to by using numbers, checklists, and formulas. The use of SMART goals is supposed to be a best practice to ensure objectivity and fairness. However, in the last chapter, I showed how numerical targets contribute to game playing, working at cross-purposes, and limiting judgment. The other reality is that not every job responsibility lends itself to the SMART format. In fact, many jobs themselves are ill defined and depend on responsiveness to customer demands, competitor actions, or other changes in the environment. Many companies cite «being responsive to customer and changing marketplace needs» as an important value, but how do you predict and measure these responses to add them to your goals?

Here's an example of a problem I once had. One of my direct reports was responsible for responding to employees'calls for help with computer problems. To rate her performance, I was planning to ask some of the people she helped for feedback. However, the HR team insisted that the goal be written in the SMART standard. They suggested I add «helping x number of employees in a month,» but I only wanted her to help those who called and not have her solicit calls or be penalized if fewer people called. Then they suggested that we implement a customer satisfaction index by surveying everyone who called. Everyone who called for help would now be burdened with filling out a survey, thus annoying people and inhibiting calls. (Don't you hate those customer service surveys?) Not to mention, tallying the survey results would be a new job responsibility for her and would have to be added to her goals. This particular goal caused so much consternation with HR that I had to make it a minor part of my employees official rating even though it was a big part of her job, which, when you think about it, really wasn't fair to her.

The other problem with SMART goals is that they cannot address the differences in managers' expectations of their employees. Some people just naturally expect more from their direct reports and others, less. For example, one year, my colleagues and I were discussing why one woman was receiving the highest rating allowed. She was a low-level financial analyst who had recently been promoted from an administrative assistant position after obtaining her bachelor's degree. Formerly my AA, she and I were still friendly. She often complained to me that her boss treated her as if she were incompetent. Even though she had been an AA, she had worked in finance for almost twenty years and had picked up a lot of knowledge. Yet her manager treated her as if she had no relevant experience. Due to her colleagues' absences from a confluence of maternity leave, jury duty, and illnesses, she was the only one in her group during a quarter closing, which she completed unassisted. For her, it was a lot of work but no big deal. She had been doing as much for years, yet her manager was blown away. Her written goals consisted mainly of basic data entry and administrative tasks. She had far exceeded the expectations, which had been way too low. In her manager's eyes, she should have been ecstatic with her rating. However, she felt insulted by the process. An employee was demotivated by receiving the highest possible rating on a performance appraisal!

That year, I had someone at the same level as the financial analyst accomplishing much more complex work who ended up receiving a lower rating to accommodate this situation in the forced ranking of employees. Again, this wasn’t fair. Even though everyone was working with SMART goals, lots of unfairness and subjectivity were going on.

These issues are not exceptional, but represent what typically happens during these ranking sessions. The intent of these meetings is to objectively compare goals and ratings, impartially align them, and systematically slot people into the normal distribution curve. But how can we be detached and objective if we are personally invested in the success of our employees? How can we be good managers if we are not personally invested in the success of our employees? Instead of unemotionally cranking out ratings, we sit there endlessly wrangling, arguing over tenths and hundredths of a percentile to give one employee an edge over another. When we calculate the compensation impact of what we are arguing about, and It's about $10 a month, we wonder why we are debating. The debate is not about the money. It's about that number. My employees are inherently better than yours simply because they are mine, and It's my duty to fight for the best number possible. Far from being fair and impartial, these are the most emotional and biased meetings I have ever attended. Unfortunately, the most eloquent managers with the most passionate arguments win. I pity the employees of managers with poor debating skills.

Probably the most biased part of this process is the evaluation meeting, also called the «performance appraisal.» Although we like to think that the actual review is objective, because we use SMART goals and lists of competencies, research in the field shows that the evaluation is subject to multiple biases. Among some of the documented biases are these:

• Favoritism—We give higher ratings to the people we like.

• Shared values and social styles—We like people who are like us so we give them better ratings.

• Age, race, and sex discrimination—Again, we give higher ratings to people who are like us or according to perceived stereotypes.

• Halo/horn effects—Good or bad performance in one area carries over to our perception of performance in unrelated areas, for example, someone who looks disheveled is perceived to be a poor public speaker and rated accordingly.

• Our own treatment—We use the rating we receive as the baseline for the ratings we give others.

I can speak to the shared values bias from my career. After a reorganization, I took over a position formerly held by a perfectionist. Perfectionism is at odds with my own philosophy of accomplishing a great many things to about 80 percent satisfaction. My view is that the world changes too fast to try to be perfect. One of my new subordinates, also a perfectionist, who had become accustomed to being praised for her attention to detail, suddenly found herself being coached for improvement! Her less-detailed colleagues, on the other hand, were now better performers than when under their previous boss. No ones performance had actually changed. It was being judged from a different set of criteria, one that wasn't on the appraisal form but was in my head. The performance appraisal process is not fair nor objective in any way, shape, or form. How can it be? It's an evaluation or judgment, the very definition of «subjective.»

Before companies had automated performance management and reward systems, the purpose of performance management was to improve employee performance and provide a baseline to compare employees. We still had performance appraisals, but these didn’t have scores and weren’t tied directly to rewards. Your manager was given a lot of leeway in determining your salary increases and your overall appraisal. If you had a good manager, you reviewed what went well and what could be improved. In a really good manager-employee relationship, the meeting was a true dialogue where you both reflected on what went well and what didn’t, as well as how you could work together better. If you had a poor manager, he used the session to point out all your flaws. (I always figured that this was what was done to him, so he was going to do it to you.)

There is something unnatural about being evaluated by someone else. We don’t meet with others to judge their competence in any of our other relationships. I’ve been married for over twenty years with no annual performance appraisal. I tried it once with my teenage sons, but that didn’t work out so well. The problem is that appraisal meetings smack of paternalism. They are based on the assumption that the manager has superior judgment and knows better than the employee what is good for her. The employee’s self-appraisal doesn’t count in the final assessment.

Now that we have a number on the bottom line of the appraisal, the nature of the conversation can’t be anything but a paternalistic evaluation of another being’s worth. Because of the SMART goals, listed competencies, and weighted averages, the appraisal appears to be objective, but it can’t ever actually be anything but a highly subjective process. It’s a game of pretend, a real sham, the worst part of which is that the number prohibits any kind of meaningful dialogue between the manager and employee. The numerical rating is the first thing an employee wants to hear and barely hears anything before that. A real dialogue on how to work better together is replaced by anticipation of the grade.

Regrettably, most employees will be unhappy with what they hear. Because the system forces people into a bell curve to identify high and low performers, the majority of employees will be ranked as average. This conflicts with how we rate ourselves. We all think we are above average. This is a well-documented cognitive bias that has many names: «the better-than-average effect,» «illusory superiority,» «superiority bias,» and even «the Lake Wobegon effect» where all the children are above average. According to Tom Coens and Mary Jenkins in their book Abolishing Performance Appraisals, «Virtually all employees see themselves as excellent workers. Most people are disappointed with their ratings and rankings unless they are at the highest level. In fact, studies show that some 98 % of workers rate themselves in the top half of performers while 80 % rate themselves in the top quarter.»

No one likes being told that he is average, especially good performers. To compound this issue, even above-average ratings aren't high enough for the best performers. When I had an appraisal meeting with one of my top performers and informed her of her 4.15 out of 5 rating, an astronomically good rating at this company, which forced most people to a 3, she wanted to know why she wasn't a 5 out of 5. She wanted to know what she did wrong. Instead of having a conversation around what went well, what could have gone better, and what we should do differently, I was stuck explaining the number and trying to convince her that a 4.15 was a really good number and that no one actually got a 5 out of 5. She stewed for days afterward. Again, here was someone demotivated by a really good rating!

I have seen many happy, motivated employees become demoralized by this process. Our underlying assumption number two is wrong: performance is not improved through the appraisal process — rather, just the opposite occurs. The appraisal process deflates people. Performance improves by the coaching and feedback that occur in the daily interactions of people. The communication that occurs between the manager and employee (and with peers) is the vehicle for improving performance. Our current performance management systems are all about filling out forms, calculating scores, doing rankings, and allotting dollars— at the detriment of relationships. Many managers are lulled into thinking that if you follow the process and provide an assessment of strengths and weaknesses with an action plan, then you’ve got good management covered. Add a few comments about what your direct reports could do better, and, bingo, your job is done until next year. You don’t need the relationships because you have the process!

If you think about it, the ultimate purpose of the entire system is to allot money. The end result is a grade with a dollar amount attached to it. The monetary distribution is the actual final product. It’s not a performance management system at all; it’s a compensation allotment system. This brings us back to assumption number three — people are motivated by money, that is, tying compensation to goal achievement makes the employee more likely to achieve that goal.

This assumption has been written about a great deal lately. In addition to numerous articles, this topic has accounted for two best-selling books: Alfie Kohn’s Punished by Rewards and Daniel Pink’s more recent Drive. Punished by Rewards shows how rewards and incentives may have very short-term benefits but in the long-term inhibit learning and intrinsic motivation. Daniel Pink cites studies showing that employees are happiest with a slightly-above-industry-average fixed salary and that the external rewards of bonuses, stock, and so on, have little effect on knowledge workers. Both authors agree that extrinsic reward systems tend to make people less creative and less enthusiastic about their work by destroying their inherent desire to learn, add value to society, and do a job well. By attaching a monetary value to a task, we unwittingly communicate that the task has no value in and of itself. The other effect it has is to limit our thinking. Because we are now so focused on the immediate task, we miss other information that could help us, and we shut down our lateral thinking. In short, the evidence against using incentive compensation as an employee motivator is overwhelming. The only case where it works well is when we want someone to complete a simple, manual task faster, like rewarding factory workers for completing more pieces. However, even that scenario poses a whole set of risks. I’ve already talked about the harmful consequences of rewarding quantity produced over meeting order commitments. The other consequence is poor quality.

I don't want to rehash all the popular literature on incentives and motivation. You can refer to the bibliography for a few of the studies. I do want to mention the research of a friend of a friend, Marc Hodak, an executive compensation consultant, who in 2006 studied S&P 500 companies and their executive compensation packages and was able to ascertain some best and worst practices. Here are some of his findings:

• Companies that rewarded performance based on balanced scorecard measures underperformed the S&P by 3.5 percent. By focusing on so many measures, executives felt that they were not able to focus on anything.

• Rewarding for a particular measure, like revenue growth, improves that measure but doesn’t necessarily translate into profits or shareholder value. Often, individuals will game the system, and meeting the targets for that particular measure will come at the expense of others, like revenue increases at the expense of profit margins.

• People are not motivated by stock options or grants because they perceive that the value of stock is mostly due to economic factors that are out of their control and not due to internal initiatives.

Ultimately, this means that all the effort to align employees around measures and targets and provide money and stock as a motivator has absolutely no proven benefit to a company; rather, it has been proven to have an adverse impact. All that time, money, and effort devoted to standardizing job levels, developing performance criteria and common competencies, agreeing on rating scales and bonus and salary targets, developing forms and processes and automating them, and then filling out forms, conducting calibration meetings, fitting workers into normalized distribution curves, arguing over ratings, attaching dollar distributions to the ratings, and meeting with each employee to discuss the persons strengths and weaknesses and the overall rating and compensation increase has deleterious effects on both worker motivation and a company's performance. So why is this considered a management best practice? Because a few management gurus came up with some interesting models that caught on with management consultants, who thought it seemed like a good idea to meld them together and set about convincing their clients of the supposed benefits without any real knowledge of the consequences. Everything looks so nice and tidy on paper!

Of all the management practices that are unsound and ultimately harmful, this one needs to be eliminated as quickly as possible. In It's place, companies should consider a number of alternatives because there really isn't one best way to manage employee performance. When it comes to compensation, my opinion is to go with Daniel Pink's recommendation of an above-average salary and a very simple profit-sharing plan. Offer the same percent reward to everyone when the company meets a profitability threshold. This creates a «we're all in it together» mentality. Stock options and grants are meaningless as part of an expected compensation package but can be used as rewards after the fact for a job well done. This means you can throw all the weighted goal calculations and calibration ranking sessions out the window and just offer a fixed percentage of profits to everyone when the company succeeds. An added benefit is that all the gaming of the system gets thrown away. This method is inherently fair because everyone gets the same amount. SMART goals and normal distribution curves become irrelevant. Everyone can be above average.

When you get rid of the ratings and evaluations, managers and their employees are free to find the best way to improve their performance together. Without the formal system, managers and employees can decide how often, if ever, they want to meet. I like one-on-one sit-down meetings every other month. Sometimes this reflection is better accomplished with a team to discuss team performance or as part of a project postmortem review. Some manager/employee ratios don’t allow the supervisor to really know the employees, let alone provide coaching. In those cases, peer feedback may be a better way. Now that we’ve freed up almost two months, we have time to conduct all sorts of meaningful review and feedback processes like meeting debriefs, lessons learned after major milestones, and team development sessions. Because these can be done throughout the course of business, and not bottled up and saved for the end of the year, they are more accurate, more timely, and more likely to be enacted.

Employees can write their own goals and make them as challenging as they want without getting penalized. In fact, Peter Drucker, who first came up with the concept of managing by objectives, believed that it was critical for people to develop their own goals. People are much more motivated to work on objectives they created themselves than objectives that were handed down to them. I know I prefer working on the projects I want to work on rather than working on the projects you want me to work on. Plus, goals can change to meet changing market conditions without impacting the system. Of course, the managers and the employees would be responsible for ensuring the goals are aligned with the corporate goals, but if you cant trust your managers and employees to work toward corporate goals, then you have a bigger problem than goal alignment.

The only way to address individual performance is individually. This means the one-size-fits-all approach with standardized forms, meeting agendas, checklists, and formulas works only for the standardized employee. I have yet to meet one. Together, management and staff can figure out how to work together better to achieve the company's goals. Isn't that a much better use of everyone's time than rating, ranking, sorting, and labeling employees like some kind of inventoried human asset?