Pay-for-performance, merit pay, and annual reviews have not worked out well. W. Edwards Deming started as a statistician but became one of the greatest business thinkers in human history. His ability to penetrate common business issues and get to the fundamental truths and fallacies behind them was amazing. His research clearly illuminated what most of us had already felt, if we took the time to think about it: pay-for-performance and periodic performance reviews, while often yielding us pay increases and other rewards, almost always left us feeling mistreated and angry, and sometimes in competition with our colleagues – not a good feeling. Why is this?
Why do pay-for-performance, merit increases, and periodic performance reviews fail? Bad feelings are inevitable when people are tested or evaluated and then ranked against each other. The sense of being ranked against others is persistent and divisive. Even where the company claims that people aren’t compared to each other, or when peer reviews and other modern approaches are implemented, the process of being evaluated and rewarded, retrained (punished), or even let go leave a bad taste in people’s mouths. Sometimes when management has tried to be more constructive with the evaluation scheme and disconnect it from pay raises, the imposition of budget specifically intended for pay increases has undermined the attempt and left employees even more unhappy and angry than usual. Just the thought of being evaluated is sufficient to cause stress, and the idea that one would be penalized for work performance when one is doing the best one can under the circumstances is itself unfair and distressing.
The individual is not solely responsible for his or her results – the system and culture are involved. Dr. Deming proved this to us quite powerfully with his Red Bead Experiment, in which workers tightly constrained by the system are rewarded or penalized based on results predominantly influenced by random factors in the system. This exposes one of the critical flaws behind individual evaluations: they don’t take the circumstances (including artifacts of the systems and culture in play) into account in any systematic way. Thus, annual reviews and pay-for-performance fail to deliver the promise of improved performance and completely ignore the systemic nature of work in any human organization, while frustrating and demoralizing employees and putting them in competition with each other instead of instilling a desire for collaboration and company success.
Why performance reviews? Because they are easy. Performance reviews have been pervasive because the concept is simple and seems intuitive. The problems come from the fact that reviews and evaluations don’t require management to face the problems their people are experiencing, either on or off the job, and from the fact that periodic reviews focus only on the outcomes, not the systems and processes (part of “the circumstances”) that govern how productive people can be. That’s like driving by looking in the rear view mirror – you can make corrections but you won’t know if they are the right ones, and by the time you do know it’s too late because you could only see them when they were in the past. So what did Dr. Deming recommend as a replacement for annual reviews?
As usual, homework done “up front” is worth far more than people realize. By homework I mean actions management can take before and directly after hiring an employee that will ensure better performance by the employee and the organization over the long term. Taken directly from Out of the Crisis, Dr. Deming’s seminal work, here are his eight recommendations for replacing pay-for-performance and systems that rank employees by their individual performance, interlaced with my comments:
1. Institute education in leadership, obligations, principles, and methods. *1
Understanding the fundamental factors defining a productive, team-oriented workplace is of highest importance. Few people are taught the fundamentals of systems thinking and organizational behavior today. Once a person learns the fundamental principles of how systems work and the impact of organization culture on performance (important leadership training), however, they see their work and the organization around them in a new light. Understanding that the expectations and guidance given the employee by the organization around them fundamentally governs and predicts the employee’s performance is an epiphany for many, a real wake-up call, and essential knowledge for effective management. Leadership involves guiding systems and enabling employees to be more productive and creative. Management is more about monitoring and managing outcomes, and taking corrective action after the quality of the results is determined. Leadership style is a big factor in determining the culture, which indirectly but powerfully influences performance by individuals and the organization as a whole.
2. More careful selection of people in the first place. *2
It’s exceedingly difficult to make a silk purse from a sow’s ear, so concentrate more on hiring “silk”. For an employee to do their best they need not only to have the appropriate skills and an understanding of the job they are to do. They also need to have a personality that will enable them to interact effectively with the rest of the people and conditions in the system. Most hiring is done either based on a narrow set of rules and requirements delivered to a department responsible only for finding people with the right skills and without an understanding of how the business actually works, or by an even less rigorous “gut feel” on the part of one or more managers, or through some combination of both. A manager armed with a fundamental understanding of the systems, culture, and other details of the business as well as a decent understanding of human nature can do much better. Part of that understanding is gained by working closely with subordinates, peers, and superiors to understand the challenges the organization faces and how those challenges are felt and dealt with at different levels of the company.
Increased emphasis on hiring high quality workers who fit in with the culture produces lasting though hard to measure improvement. Increased focus on the hiring process and finding candidates more in tune with the culture, pace, and nature of the business, etc. can go a long way towards producing lasting improvements in organizational performance, yet it may be exceedingly difficult to attach those results to particular individuals. (Identifying star performers in an organization may be misleading and unfair to those with whom they work, and may be a sign that teamwork and team spirit are lacking.) Without such knowledge many managers view hiring as a painful necessity and a bother, as something that takes them away from their job of running the business. If this kind of clueless hiring and evaluation continues, eventually the company can be saddled with a large proportion of ill-suited and poorly chosen workers who will hinder performance and profitability in subtle and immeasurable ways for years.
3. Better training and education after selection. *3
Being hired and then “thrown into the pool to learn to swim” is a poor substitute for proper training. How many times have you been hired into a firm and given a desk, only to flounder for days or weeks trying to provide value to the organization while feeling frustrated and lost? In my experience over approximately five different careers, it takes several weeks after starting with most companies before one can contribute substantially, and three to six months to “know the ropes” and be able to function in relative comfort and with good output. This is wasted productivity that could be improved, and the ramp-up time reduced, with the aid of training for all new employees the first weeks after they sign on.
Training immediately after hiring can reduce unproductive time and increase worker effectiveness. A good grounding in the fundamentals of the business, the history and culture of the organization, principles of leadership and systems theory (including control limit theory), communication and record-keeping systems, and the procedures, standards, and guidelines employed by the firm can be acquired in a few days. Usually those early days would have been of only marginal productivity anyway, and such knowledge will allow the worker to reach a good level of quality and throughput in her or his job much sooner than otherwise. In addition, the worker’s morale will be enhanced by the knowledge that the firm is willing to invest in them by providing such education, and they will meet people, build relationships, and absorb the less-quantifiable aspects of the culture in the process.
4. A leader, instead of being a judge, will be a colleague, counseling and leading his people on a day-to-day basis, learning from them and with them. Everybody must be on a team to work for improvement of quality in the four steps of the Shewhart cycle shown in fig. 5 on page 88. *4
Real leaders are collaborators and coordinators rather than “generals” giving orders. The most effective managers I’ve worked for have been collaborators, genuinely interested in the people they work with and who work for them, and focused on enabling those who work for and around them to do their best work by eliminating barriers, identifying need for and providing training, and keeping everyone focused on monitoring and improving the systems by which the work is done so that the output was of highest possible and ever-increasing quality. The Shewhart Cycle is, of course, the famous “Plan – Do – Check – Act” loop which is expressed in so many ways in various popular business improvement schemes, but which is fundamentally always the same.
5. A leader will discover who if any of his people is (a) outside the system on the good side, (b) outside the system on the poor side, or (c) belonging to the system. The calculations required (illustrated on p. 114, ch. 11, and elsewhere) are fairly simple if numbers are used for measures of performance. Ranking of people (outstanding down to unsatisfactory) that belong to the system violates scientific logic and is ruinous as a policy, as may be clear from the text and from chapter 11. *5
A leader understands how to evaluate performance and improve operations without demoralizing workers or misdirecting them to focus on their personal results at risk of undermining or sub-optimizing the results of the team as a whole. Focusing on the sub-processes and the quality of inputs at each stage of each process, and giving employees the ability to not only see the results but to actually take the measurements and suggest improvements themselves, is the way to maintain teamwork and achieve improvements in the quality and profitability of the entire organization. Workers with even rudimentary training in the use of control limit calculations, and with the ability to measure key parameters involved in their work processes, will improve their work.
Ranking people sends a destructive message. Once workers think they are being ranked, or that they are being rewarded or penalized based on any measured aspect of the system, they will focus not on company results but mostly on improving that specific measurement, possibly degrading results of surrounding, interdependent processes and the organization as a whole. The determination of whether results are within the workings of the system or a result of some external cause is made by using the Shewhart control limit calculations, which establish control limits showing the inherent variability of the system when it is running normally and which make it easy to detect external “special causes”. Note that these special causes must be eliminated before the system can be said to be stable, and it cannot be systematically improved unless and until it is stable. This understanding is fundamental to be able to make meaningful and lasting improvements in company results.
6. The people of a group that form a system will all be subject to the company’s formula for raises in pay. This formula may involve (e.g.) seniority. It will not depend on rank within the group, as the people within the system will not be ranked No. 1, No. 2, No. last. (In bad times, there may be no raise for anybody). *6
Pay raises are best determined by a common formula, applied to everyone for fairness’ sake, that is NOT performance-based. While most of us are used to the old, clueless and destructive way of performance measurement, people will understand, with education, why a non-performance based system is used: it produces happier workers and better company results. An organization that administers pay increases and other rewards in this way will thus avoid the bad effects from ranking.
7. Hold a long interview with every employee, three or four hours, at least once a year, not for criticism, but for help and better understanding on the part of everybody. *7
Periodic interviews are more than, but a positive addition to, practices such as “management by walking around.” Much has been written about “management by walking around” or MBWA, a concept pioneered at the infant Hewlett Packard Corporation in the heydays of Silicon Valley. This concept gets managers out of the office and out among the workers and processes they need to understand and guide. Adding regular discussions to learn of problems affecting the employee’s work, and of their suggestions for improvement, taps into their expertise and allows them to provide significantly increased value to the company. 7*
Socialization into the organization’s culture provides significant but hard to measure influences on company success. When employees of differing authorities see each other frequently and interact in a social as well as purely-work related context they not only form productive relationships but also better understand their places and functions within the overall system. “Putting a face on” things is a powerful way to get people to relate to them, and people will, for instance, better know what impact they might have on another worker or group if they cut corners in their implementation of any of the company’s processes. This will increase stability in the overall process and results and enable real improvement, as well as improve the work-lives. A manager who understands the strengths and weaknesses of their employees can better utilize those strengths and more appropriately assign people as new work is available and new responsibilities are needed. Employees who have personal contact with their supervisors and maintain an open and trusting relationship with them will feel much better about their jobs than otherwise, and this will show up in their commitment to the work and the company, as well as in specific areas such as reduced absenteeism.
8. Figures on performance should be used not to rank the people in a group that fall within the system, but to assist the leader to accomplish improvement of the system. These figures may also point out to him some of his own weaknesses (Michael Dolan, Columbia University, March 1986). 8*
This final point captures the fundamental reason for adopting this new practice instead of regular evaluations and pay-for-performance. People are the basis of every business, and the true source of lasting success and profitability. If people are engaged in improving the process, and allowed to experience and be proud of their contribution to the good of the team, they will perform far better than if they are not. The fundamental concepts involved are striking in their simplicity. Anyone with a basic understanding of statistical methods can master the control limit theorem and related calculations, and then, with management support, use them to provide significant benefit to the organization. These methods enable continuous improvement of product and process, which leads to greater pride and job satisfaction for all employees as well as better results for customers and suppliers alike. A company that adopts these methods and strives for ever-better understanding of its business fundamentals by all employees will advance faster than its competition, and will most likely carve out a stronger market position and establish a pattern of lasting success over time.
Unfortunately, much of Dr. Deming’s teachings was allowed to fall by the wayside, though it is still there to be gained and used. While much of the western business world ignored Dr. Deming’s work, those few that took it to heart had significant periods of success above and beyond their competitors. Unfortunately, management, worker, and ownership turnover allowed much of this fundamental knowledge to be lost to some of those organizations. Parts of the knowledge Dr. Deming gave us are still in use, renamed and “productized” under names like “Six Sigma”, but usually without the fundamental core that produces the most positive effects. Fortunately, the knowledge is still there to be gained and used, and those who learn and internalize it will be better workers and organizations as a result. I can’t recommend enough that anyone who works in or is interested in business obtain Dr. Deming’s books, tapes, and videos and gain a good understanding of them. The knowledge there is fundamental, head-smacking stuff that will provide a different and more logical approach to business improvement and success.
For more depth and understanding, please read Professor Deming’s books, and even the wikipedia entry on W. Edwards Deming gives some of his most important ideas. As always, I welcome your comments. — Tim
*1 – p. 117, Out of the Crisis, W. Edwards Deming, 1982, Massachusetts Institute of Technology, Center for Advanced Engineering Study, Cambridge, MA 02139
*2 – ibid
*3 – ibid
*4 – ibid
*5 – ibid
*6 – ibid
*7 – ibid
*8 – ibid