Motivating Today's Workforce: The Future of Incentive Program Design
- The Design Imperative
- Emergence of Conflicting Perspectives
- Resolving the Research Gap
- Influencing Employees to Better Performance By Design
- Return on Investment as Part of Design
- The Importance of Adverse Impact on Program Design
Acknowledgement and Context: This study represents the compilation of input from a Delphi Panel of incentive industry and measurement experts, secondary research, and related resources. The result is an effort to identify the evolving body of knowledge regarding the use of incentives and recognition programs to motivate today’s workforce. The purpose of this study is to stimulate dialogue and examine the need for advancing the traditional approach to the use of non-cash awards for motivating today‟s knowledge workers.
Disclaimer: The opinions, ideas and conclusions put forth in this paper do not necessarily represent the views of the Incentive Research Foundation nor those of its sponsors, staff or other constituents.
It is generally accepted that performance can be driven by the right mix of incentives and rewards. Until recently, few questioned the use of incentives in encouraging behavior change and higher performance. Research conducted by the Incentive Research Foundation (IRF) and others over the past twenty years has established the usage and performance improvement aspects of incentives in business. Yet today, where and even whether incentives and rewards programs have a place in the modern workforce is being questioned, not necessarily because skeptics believe they don’t motivate workers, but because they believe the program outcomes and impacts are becoming increasingly more difficult to measure, manage, and control. In other words, according to our research and the opinions of most experts consulted for this paper, the greater complexities of motivating creative, non-routine workers using incentives and rewards, has led to a higher frequency of poorly designed or misdirected programs. Poorly designed rewards, incentives and recognition programs can produce poor results, lack motivational appeal, or cause unintended consequences. When this happens, it is likely to cost the organization a significant amount of wasted time and money and perhaps lead to a cynical workforce, organizational damage, and, in some extreme cases, societal harm.
To better understand “state-of-the-art” in program design, we reviewed hundreds of research reports, books and articles. We also assembled more than fifty of the world’s foremost experts and leading practitioners in incentives, rewards and recognition into two “Delphi” panel sessions and conducted twenty in-depth expert interviews. Following what was learned in part one of this research: “Mastering Measurements: The Critical Performance Elements of Incentive Design”, we invited leaders in measurement and ROI to the discussions. Nearly all of our panelists and interviewees agreed that however one designs an incentive, reward or recognition program, they must apply proven, effective design principles by providing on-going communications, collecting key metrics throughout, measuring business impact where appropriate, and systematically monitoring, adjusting and improving the program.
More so than in the past, the artful design of incentive and recognition programs is critical to drive desired outcomes and avoid unintended consequences. Past thinking around motivating ourselves and others has been focused on the approach of offering external rewards such as money, trips, prizes or recognition. But many leading experts, educators, researchers, and practitioners in the incentive industry believe that these rewards alone no longer deliver.
One of the IRF’s landmark studies is “Incentives, Motivation, and Workplace Performance: Research and Best Practices.” This study identified the Performance Improvement by Incentives (PIBI) model1. This model is both a diagnostic and prescriptive tool for incentive program selection and implementation. Using this model, incentive program managers have a tool that will assist to:
Identify areas of importance and relevance to performance improvement
- Provide guidance on the step-by-step procedures of implementation
- Trouble-shoot and correct the system if it is not yielding desired results.
Despite the availability of tools such as the PIBI Model, too often incentive programs are designed, implemented and operated without employing effective design principles, support or monitoring. One of the persistent themes from the incentive experts on our Delphi panels was the need for additional rigor in the design and application of these programs.
Since the 1980’s there has been a continuing dialogue on the use and appropriateness of “intrinsic and extrinsic” rewards2. In fact, the development of these reward constructs has, in some instances, created a dichotomy for individuals. There have been arguments purporting that one or the other is more effective. What is interesting is the “locality” of the two perspectives: the stronger support of intrinsic motivation comes from academia, and the stronger support for extrinsic rewards comes from business. This study will examines both of these motivational constructs, looking first at intrinsic rewards.
The body of knowledge in support of the use of intrinsic rewards stems from the work of L.E. Deci, R.M. Ryan, and A. Kohn, among others3. The primary source of the research on intrinsic rewards comes from the field of education or academic laboratories. Most recently the intrinsic perspective and research has been highlighted in several well-publicized books. Dan Ariely’s book, “The Upside of Irrationality” (2010), and Dan Pink’s, “Drive” (2009) build on the previous works of Alfie Kohn (Punished by Rewards, 1993) and many of W. Edwards Deming writings (1900-1993). The authors make impassioned arguments against the use of extrinsic incentives and rewards – especially those that are offered before the fact in an effort to encourage specific behaviors.
They share a philosophy that workers – especially modern knowledge workers –cannot be counted on to predictably and reliably change their behaviors or perform better through the use of extrinsic rewards; they believe as well that the use of if/then extrinsic rewards are not only ineffective, but probably damaging to motivation in the long-term. This theory has been called “crowding out”4 because it posits that if you provide an extrinsic reward for a task that someone finds intrinsically interesting, their intrinsic motivation will effectively be “crowded out”. Their arguments state that creative workers will, in many cases, rapidly perform worse in the presence of extrinsic incentives and rewards (base pay and benefits excluded). Moreover, they remind us that in at least some cases, the wrong balance of incentives might lead to severe adverse impact on organizations and even society. Kohn’s 1993 bestseller, “Punished by Rewards” (revised in 1999), argues that tangible rewards, especially those offered contingent on doing a desired task, exhibiting a behavior or producing a result, are at best benign and usually destructive. In one of his milder statements, Kohn states: “Rewards are often successful at increasing the probability that we will do something. At the same time though, they also change the way we do it. They offer one particular reason for doing it, sometimes displacing the other possible motivations. And they change the attitude we take toward the activity. In each case, by any reasonable measure, the change is for the worse.” (Punished by Rewards, p.35).
Kohn doesn’t limit his criticism to extrinsic rewards, he also advises against the use of recognition and even praise. Kohn asserts that praise, whether from a parent, coach or supervisor, is given like any other reward: to exert control and manipulate behavior. Kohn concludes: “The bottom line is that any approach that offers a reward for better performance is destined to be ineffective.” (Punished by Rewards, p.119).
While few agree with Kohn’s position on praise and recognition, he is far from alone in his arguments against rewards and incentives. Edward Deci, Richard Ryan, Frederick Herzberg, Dan Pink and a host of others promote the abandonment of contingent rewards and incentives in almost every form. Among the most influential among them is W. Edwards Deming, the revered business thinker that transformed Japan after World War II. Deming said simply: “The emphasis should be on why we do a job,” inferring that all motivation is intrinsic. Deming preferred incentives aimed at teams, and, even then, only non-contingent recognition.
While the authors cited above highlight a view against the use of incentives, many behaviorists (those that believe in the usefulness of explicit rewards and incentives), provide strong voices that counter this argument, including researchers from academia. One such voice is Alyce Dickinson5 who in her response to Kohn’s “Punished by Rewards” wrote: “I fear Mr. Kohn has not told the truth and nothing but the truth. He argues that all rewards, including praise, undermine interest. When rewards are provided for successful performance, they enhance interest, of that there is no doubt. The results are consistent and robust (Dickinson, 1989).”
Stanford University researchers, Mark Lepper, Mark Keavney and Michael Drake further contend that “Certainly, there is nothing in the present line of reasoning or the present data to suggest that contracting to engage in an extrinsic reward will always, or even usually, result in a decrement in intrinsic interest…”6
Additionally, researchers Robert Eisenberger and Judy Cameron state: “The conventional view that reward has pervasive decremental effects on intrinsic task interest and creativity was challenged by our literature review (Eisenberger & Cameron, November 1996), which concluded that decremental effects of reward on intrinsic task interest occur under highly restricted, easily avoidable conditions and that positive effects of reward on generalized creativity are readily attainable by using procedures derived from behavior theory.7” With her colleagues, Judy Cameron has performed extensive research into reward/incentive experimentation over the past two decades. From her various meta-analyses over the years, she concludes: “What is clear at this time is that rewards do not inevitably have pervasive negative effects on intrinsic motivation.” (Cameron,Banko, Pierce, 2001.)
Cameron’s conclusions do not make the case for the use of contingent rewards in general, but she is able to provide evidence for organizations to consider before abandoning explicit rewards completely ─ even for non-routine and creative work. While Cameron’s results show a negative or neutral impact of extrinsic, contingent reward offers on influencing tasks themselves, she found a significant positive effect when rewards were offered for meeting or exceeding performance levels. In effect, Cameron asserts that a contingent, explicit reward can have a positive impact on interesting (heuristic, non-routine) work if the reward is tied to performance goals rather than task completion, solving problems, or hitting specific quantifiable targets.
So how do we rationalize these two perspectives on rewards? The answer may begin with a study by R.N. Kanungo and M. Mendonca8, in which they found that the critical attributes of organizational rewards are contingency, valence, and saliency. In other words, if the rewards are contingent on performance, have value to the recipient, and are “top of mind,” then the reward will have a significant impact on work motivation. In fact these researchers found that they were unable to categorize the organizational rewards as extrinsic and intrinsic since the subjects in the study were not viewing or able to define their rewards as such. This study seems to align with the arguments of Steven Reiss9 who posits that the diverse range of human motivations can’t be forced into the two categories of intrinsic and extrinsic motivations.
The result seems to be either to step away from the two motivational constructs or to accept that it is not one or the other, but the effective use of both. This second approach shines through Dan Pink’s conclusion that “… while a few advocates would have you believe in the basic evil of extrinsic incentives, that’s just not empirically true.” (Drive, p.49.). This leads to the insight from Dr. Reiss that “There is no reason that money can't be an effective motivator, or that grades can't motivate students in school,” he said. “It's all a matter of individual differences. Different people are motivated in different ways.10”
The actual workforce is where the 'rubber meets the road’ so to speak. By their actions, the majority of managers and executives assert their belief in the use of extrinsic incentives and rewards. Certainly, thoughtful pundits and practitioners believe not that extrinsic rewards should be eliminated, but that incentives are today becoming much more individually focused and situational. They also suggest that more attention to program design and implementation is required to create the ideal “total engagement” package for each employee and/or to encourage different outcomes and better performance. As Melissa Van Dyke, President of the Incentive Research Foundation puts it: “An incentive program or plan is much more than just the reward, and it is much more than a laboratory experiment. There are elements of communication and focused design to these plans that make them engagement tools.” (Delphi Panel, Aug 5, 2010)
Encouraging better “Design” itself is too broad and conceptual for most hands-on practitioners to find useful. The question is: How should one go about designing an effective rewards program for a knowledge-era workforce? The PIBI Model, previously introduced (See Appendix B) is one method. And Dan Pink has constructed a flowchart tool that aids in program design (See Appendix C). At a basic level, Figure 1 below offers additional rough guidance that may be useful in taking a range of work into consideration when designing program models to encourage better performance. Note that not all occupations fit nicely within the guidelines offered in Figure 1. Sales positions, for example may be quite creative yet in many cases, salespeople are greatly motivated by extrinsic, if/then type rewards.
Figure 1: Use of Extrinsic Rewards and Intrinsic Reinforcers
Assuming an organization has finite resources with which to drive employee performance, the scale above might guide an organization that hopes to motivate or engage different tasks and workers. Doing so, the designer would balance and craft extrinsic rewards and intrinsic reinforcers into the ideal combination for the greatest impact. The best blend for tasks associated with “Job A” might skew that person’s total rewards package slightly toward extrinsic incentives (cash bonuses, reward travel, consumer items, etc.). For “Job B” the right balance might rely more on intrinsic motivators (projects of interest, more autonomy, flexible work, recognition, etc.), and contingent extrinsic rewards (aside from regular pay & benefits) might be offered for general performance and outcomes rather than for behavior or discrete tasks. Figure 1 does not necessarily imply that if the use of intrinsic reinforcers increases, extrinsic motivators must decrease or vice-versa. In general, the more creative the work, the less effective might be long-term reliance on cash and tangible type incentives (and vice-versa). That said, every employee must receive extrinsic rewards for performing well over the long-term (competitive pay at minimum) and, despite Kohn’s assertions, employees should expect the intrinsic reinforcement provided by a competent leader who knows how and when to provide praise and feedback.
Note also that extrinsic type incentives might well be appropriate for Job B, for example where many sales roles are concerned. However, care must be taken to align extrinsic rewards with the incumbent’s intrinsic motivations. For example, if a travel incentive is used, rather than a week at a resort, the designer might substitute the opportunity to travel to the organization’s overseas operation for work on an interesting project (or some combination of both).
As work becomes more creative, motivation becomes more intrinsic and the balance of incentives, rewards and reinforcers grows more complex. After all, most creative workers cannot easily be induced to follow a prescribed process that lends itself to simple measurement and reward. For organizations to realize their full value, creative workers should be given the freedom to innovate. Within reason, the emphasis should be more on encouraging measurable outcomes and less on the behaviors that might get them there. All of this implies far greater emphasis on the design of programs or packages that hope to measure and encourage performance.
Designers should also bear in mind that the right mix of incentives, motivators and reinforcers is a moving target and may be unique to every individual. Doubtless, for a new model to be practical, it must recognize that our industrial age focus on changing behaviors should give way –within reason and through careful design – to a creative age focus on outcomes. Therefore, there is a need to create a greater linkage between the work context, the individual/team, and the available motivational factors – creating harmony between intrinsic and extrinsic motivators.
In their 2002 book Driven: How Human Nature Shapes Our Choices, authors Paul Lawrence and Nitin Nohria argue that all managers and business leaders need to get things done through people. “It is hard to be a great leader without understanding what drives and motivates the people you need to lead.” The authors go on to state that leaders have several structural devices they can use to promote a balance between the four drivers they have identified as essential to employee engagement, workplace harmony and overall productivity - acquire, learn, bond, and defend11. For example, the authors advise leaders to balance financial and symbolic rewards for both individual achievement and teamwork.
Designers should consider the business impact of incentive, reward and recognition programs. How is performance lifted or enhanced, and how does it impact the organization, what is the Return on Investment (ROI)?
Our Delphi panelists confirmed the need for better measurement of programs. Sherri Schafer, of Southern California Edison said, “We are screaming for this. What … is very common in this area [are] the rewards, trips, etc., and we put them out and see how many use them and whether we are satisfied with [them]. But we need to measure the performance we are seeing. We simply measure employee incentive satisfaction, how much it costs and how many used it, but we don’t get the levels of business impact and ROI.”
Dr. Srinath Gopalakrishna, Professor of Marketing at the University of Missouri – Columbia, agreed. “Corporations spend millions on these programs, so there has to be a viable method to document the ROI for them. Resources are tight, and accountability is the key word. We must show value. If we do, then partnering between corporate entities in industry will improve.” He goes further to say, “Corporations are sitting on huge volumes of data and don’t really know what to do with it. It’s imperative for the industry to say – „don’t take our word for it – just look at what the data says.’ If the industry sets itself up for systematic data collection, and if these programs are indeed providing value, the science and the research behind the analysis will deliver significant long-term payoffs.”
Dr. Jack Phillips, noted author and Founder of the ROI Institute, advocates keeping the analysis as simple as possible and that systematic processes and checklists should take the user all the way through the highest levels of ROI. Phillips recommends that designers consider a proven ROI Methodology (Figure 3 below) in designing incentive programs, this approach, combined with best practices from the work of the IRF since 1992 are captured in Part One of this paper, “Mastering Measurements: The Critical Performance Elements of Incentive Design”
Figure 3: The ROI Model from the ROI Institute
As discussed in Part One of this study12, unintended consequences or adverse impact is a key consideration in program design. Unfortunately, designers have to live with the fact that they can never predict every potential consequence of their programs. As Paul Hebert, Managing Director and Lead Consultant at I2I-Align, told us: “One of the problems is that we don’t always know what they are because they are unintended. We do have to predict what participants will do and protect against those activities. Trying to identify potential problems is always good and something professionals should do. But, from a selling point of view, we don’t want to talk about all the negatives. On the other hand, we don’t want to leave the client not knowing all the factors. That’s just professional behavior.” (Delphi Panel, July 28, 2010)
Following on the arguments made in Part One and throughout this paper for careful incentive program design and evaluation, consider that many, if not most incentive plans, still target sales today. One reason is that sales targets are concrete and measurable; another is that the success of a sales incentive plan is sure to benefit the entire organization through increased revenue. Yet, the idea that more sales always equates to success is a dangerous notion.
The IRF’s 2007 paper, Assessing the Impact of Sales Incentive Programs, illustrates the hazards of ignoring adverse impact in determining the success of an incentive program.
In that paper, the researchers outlined what program designers should consider and measure to include the impact of increased sales on other parts of the business:
- Finance: The short–term sales focus of an incentive program may generate higher sales volume, but the accounts acquired may pay their bills later, increasing the drain on accounts receivable, hurting cash flow and thus profitability. Unless inventory and accounts receivable are given additional investment, margins can be significantly impacted.
- Operations: Additional sales change the demand pattern, because new customers entering the sales system may order products on erratic schedules, creating imbalances in finished goods inventory and increasing shipping costs. Careful planning is a central need to ensure that excess inventories and shipping issues will not adversely impact cash flow and thus profitability.
- Prourement: More products may be produced to handle the sales increase, which places greater demands on procurement. Additional investments may be necessary in procurement in order to avoid the higher prices the company might have to pay given the spot market material purchases. Perhaps (for example) longer term contractual arrangements are needed with suppliers.
- Production: Higher production volume that would be needed to meet a short-term sales increase might entail additional labor hours to produce the extra goods. Finished product defects might also increase because less skilled, temporary labor may be required.
- Human Resources: Planning for additional workers (even though temporary labor) involves considerable expense. Beyond the cost of hiring, the cost of training the new workers must also be considered.
Applied to a real case, the researchers demonstrated that after the impact of the above was taken into consideration, the sales incentive program in question could be expected to net a negative 92 percent ROI.
Clearly, then, if the program was to be successful, it had to address some of the potential adverse impact on the rest of the organization. As such, the designers expanded the goals of the program from sales only (increase sales by $1 million) to include a balanced emphasis on gross margin (from 30.4 percent to 32 percent). Next, several steps were taken to predict and measure ROI accurately:
- Stage One: Establish Reasonable Benchmarks for Key Indicators.
- Stage Two: Examine Actual Sales Levels.
- Stage Three: Assess How The Incentive Program Would Impact Cash Flow.
- Stage Four: Compare Actual Results to Benchmarked Projections.
The researchers were able to predict a swing of 176 percent ROI through better incentive program design. In other words, by considering the adverse impact increased sales might have on other parts of the business, the team was able to craft additional goals, which led to a better balanced program and a forecasted positive ROI of 84 percent13.
Organizations are more sensitive to the unintended consequences of incentive programs than in the past. “After the international financial meltdown of 2008, caused in part by poor incentive design,” Dr. Thomas Haussmann, Variable Pay Leader for The Hay Group, says, “We are now asked to help make these programs risk proof.” He advocates looking at the measures and objectives to determine if they are short term, if they fit into the long-term objectives, and if the overachievement of the short-term goal is going to be detrimental to the long-term success and viability to the company. “The standard procedure should be: Does the over-achievement or achievement of objectives entail risks that are not intended? if you find them, address them in your program design.” [Return to Top]
There is little doubt that incentive program design and implementation, including measurement and ROI, is critically important in today’s workplace environment. And while the incentive plan designer must consider the overall context, including the type of worker or team they are attempting to motivate, it is far from agreed that in designing an effective rewards program ─ even for knowledge workers ─ that one or the other of intrinsic or extrinsic, contingent rewards must be used. We are seeing the evolution of an effective blend of both, or a more inclusive approach of any appropriate reinforcer that is contingent, valued, and top of mind.
What is clear from our research, including the opinions of the great majority of our experts, is that incentive, reward and recognition programs must be more tailored today than in the past. Careful design must make allowance for the many different ways in which workers are motivated.
Yet there are also consistencies and best practices to guide designers. Knowledge workers, for example are more likely to be driven by recognition (now/that type rewards) than contingent incentives (if/then rewards). Where incentives are used with creative workers they should be designed to drive desired outcomes rather than to encourage specific behaviors. Extrinsic, contingent rewards should be tied to performance goals rather than task completion, solving problems, or hitting specific quantifiable targets.
In general, designers should balance the best use of extrinsic incentives versus intrinsic rewards and recognition depending on the type of worker, the assignment, the work environment, the desired outcome and the duration of the program amongst other factors. In every case, designers should build in the appropriate levels of measurement, in many cases including ROI analysis.
In summary, incentive and reward program designers should:
- Take advantage of the PIBI Model (Appendix B) by using it as a checklist of key questions that need answers. They should also consider the Pink flowchart (Appendix C) and the guidelines we propose in Figure 1 and Appendix D.
- Articulate and consider what is the program trying to achieve or address? More sales? Lower absenteeism? Quality? Safety?
- Think about the program context, including which workers the program is targeted to - Factory floor? Administrative? Sales? Researchers? Executives? (And whether they are primarily creative or task oriented).
- Think about the unintended consequences and potential adverse impact of the program. Will the outcomes it aims to drive impact other parts of the organization? Is the program susceptible to “gaming”? Will it encourage undesirable behaviors?
- Consider whether the program is in balance with other key drivers of performance and employee engagement such as those identified by Lawrence and Nohria - Acquire, Learn, Bond, and Defend .
- Include communications, well-planned implementation and on-going monitoring and refinement.
- Build in metrics and ROI measures from the beginning so that you can credibly evaluate the program and make adjustments throughout.
- Define for your organization the terms “incentive” and “recognition,” and then clarify when and where they are most effective based on research and best practices.
As this paper illustrates, the optimal design of Rewards, Incentives and Recognition programs is complex problem. Our research brought dozens of thought leaders together and considered hundreds of papers and articles on the topic but more ideas, opinions and examples are needed. We hope you’ll join our discussion at: http://theirf.blogspot.com/