"HOW TO KEEP ME"--RETAINING TECHNICAL PROFESSIONALS.

These 15 predictors of retention can help, but first you need to collect data from your own scientists and engineers.

Staff turnover is a concern almost everywhere as the demand for talent begins to outstrip the supply. However, turnover in R&D is particularly damaging because of the disruption to projects and because technical and scientific staff are hard to replace. The key to retaining scientific and technical staff is to understand the real cost of turnover and the causes

By technology and scientific professionals, we mean knowledge workers whose work is governed primarily by their own expertise rather than by a routine or system. We exclude executives, who typically have very different responsibilities and a very different reward system than professionals. We also exclude administrative staff, customer service and other job families that may exist in an R&D environment but have less professional discretion or less portable intellectual capital. Our definition of scientific and technical professionals includes both individual contributors and lower and middle management. Included are software designers, research scientists, most types of engineers, and product and project managers.

Our perspective on managing turnover of technical and scientific staff comes from consulting with companies in pharmaceutical, computing, networking, medical products, electronics, and other industries. We have also gained insights by conducting a study of North American workers to understand how employees view the rewards of their work. This study, The Rewards of Work[SM]: What Employees Value, was co-sponsored by Nextera, Sibson Consulting Group and WorldatWork (formerly, the American Compensation Association). It focuses on the attitudes of full-time employees in the private-sector work force, investigating scientific and technical employees' attitudes in detail. Additionally, the Industrial Research Institute and Sibson conducted an online survey of IRI member companies to assess their turnover rates in R&D and the actions they are taking to control turnover.

The Changing Environment

The basic nature of the technical and scientific professional has not suddenly changed, but the environment around them has. Three key changes have made these professionals more important as a corporate asset:

1. Technological and scientific advances are coming to market more quickly, and the company that misses the window loses the pricing and profit premium available to market leaders. In many companies, science and technology has eclipsed marketing, finance and even sales as the critical employee segment. These professionals can create the franchise for company growth and are increasingly sought after by established corporations or pre-IPO start-ups.

2. Supply has not kept up with the demand for these professionals, and it will likely get worse as the baby boomers begin to retire.

3. The web has made career mobility and pay information so easy to access that even the most passive job hunters can hardly avoid learning about more lucrative positions.

These factors have changed the expectations and realities of technology and science professionals, and employers must respond in order to avoid an exodus of these key talents.

Underestimating Turnover Costs

Many companies have difficulty measuring the cost of turnover because the information is hard to get and the true costs hit many different budgets. This may be one of the reasons why it is hard to find out who is responsible for managing turnover in some organizations. The R&D manager knows that it is difficult to meet milestones with positions vacant, but has no way of estimating the impact of turnover on revenue and earnings. This sometimes causes R&D managers to use turnover to help stay within salary budgets even while the corporation is hungry for the revenue from new products. The human resources department may consider the cost of turnover to be the cost of finding a replacement, but probably has no way of knowing exactly how much the turnover is costing in delays in time to market, in contractor fees, in relocation of existing staff, and so on.

Standard estimates of turnover costs range from one times annual salary to seven-times salary, depending on which source is being quoted. Few companies are willing to act on these broad and general external cost estimates. Companies that have made a positive return on investment in retention have done so by understanding their real turnover costs and how the costs for R&D turnover vary from other employee segments. Cost studies we have conducted within individual companies have revealed that the cost of turnover for scientific and technical professionals differs from other staff. This is due to many factors, including the rate of turnover, the cost to replace, the time to replace, the cost of the vacancy, including the intellectual capital attributed to the departing staff, the contribution to revenue and earnings attributed to R&D staff, and other factors, including the rate of company growth. The difference in the cost of losing a scientific or technical professional can be three to six times the cost of losing an administrative professional, such as from human resources, finance or facilities.

The true costs of turnover are critical for each company to know because they are the key to knowing how much to invest in retention. We find that most companies are under-investing in retaining some employee segments like R&D and Sales, and may be over-investing in retaining other staff. Their overall rate of turnover may be relatively even throughout the company, but in certain parts of the business it may cost a huge amount in terms of expense and lost revenue.

Turnover Rates

Often, the turnover rate for R&D is lower than for other segments of the organization and from other industries such as retail and services. A survey of IRI member companies in September 2000 revealed that the 54 responding R&D organizations experienced turnover well below the 15 percent average for all industries in the United States. Table 1 shows that the computer and electronics industry segment had the highest median turnover rate of 9.2 percent and that R&D organizations that are part of general manufacturing had the lowest median turnover rate of 3.6 percent.

Table 1.--Turnover Rate Among IRI Member Companies
by Industry (Aug. 2000, 54 Respondents).

                           Turnover Percentage

                    25th                          75th
                 Percentile                    Percentile
                     of          Median of         of
  Industry       Respondents    Respondents    Respondents

Chemical             4.5            5.0            7.0
Manufacturing        1.6            3.6            5.0
Computer/
 Electronic          8.1            9.2           13.0
Other                3.1            6.5            8.0

When interpreting these turnover rates, we encourage employers to be concerned if they are above the median rate, because they are incurring turnover costs that are both significant to the business and controllable. Turnover rates at or below the 25th percentile, on the other hand, sometimes mean that the organization is not putting enough pressure on the low contributors on its staff. Although R&D in general has lower turnover than other business segments, R&D managers often find the loss of a single individual much more costly than do managers in other industries because of the disruption to projects and the difficulty of replacing that person.

Many companies see a spike in turnover immediately after a major organizational change, such as a merger, reorganization or a downturn in business prospects; these almost seem to push employees out of the company. Other times, although nothing significant has changed inside the company, turnover increases because employees experience greater pull from outside the company.

Understanding Causes of Turnover

How can managers understand the causes of turnover? How can companies minimize the forces that push scientific and technical employees out of the organization while maximizing the forces that motivate staff to remain?

Push and pull factors can be understood through surveys, interviews and focus groups with current employees. Several science and technology companies are now using frequent web-based surveys of candidates, new hires, incumbents, and exiting employees to keep on top of their ability to attract and retain staff. In this way, the company's managers can spot any trends before they have a chance to create widespread turnover. Other companies are finding that their means of gauging the pull factors are too slow or too limited to spot trends. For example, relying on annual or biannual salary surveys to set salaries is being replaced with more frequent means of checking competitiveness of compensation.

The key to understanding the causes of turnover is the "employee value proposition," or EVP. A company's EVP is the total set of rewards that the company offers in exchange for continued employment and dedicated effort. The EVP includes monetary rewards, such as salary and incentives, but also many other types.

The Rewards of Work study examined the EVP for the U.S. workforce as a whole and scientific and technical employees specifically. The study sample consisted of 1,218 full-time U.S. workers: 1,008 regular (non-high-tech) employees and 210 high-tech workers. They were selected for a telephone survey using a random-digit dialing methodology. All of the non-high tech sample and about half of the high-tech sample were identified through the random sample, with candidates for the rest of the high-tech sample identified from marketing sources (such as magazine and journal subscription lists). Because statistical tests showed the responses of both parts of the high-tech subsample to be highly similar, we are comfortable that the full high-tech subsample reflects the views of the general population of high-technology professionals in the United States.

It should be noted that job title, rather than company or industry, defined whether a respondent belonged in the high-tech subsample. Thus, a research scientist working for General Motors would be counted as a high-tech worker, while a secretary for Cisco would not be included. Statements in this article about the attitudes of the U.S. workforce as a whole reflect a weighting of the high-tech subsample that is appropriate to its percentage of the U.S. workforce. Based on the comparison of the sample's demographics compared to those of the U.S. workforce as a whole, the sample is representative of North American workers, with the exception that it is slightly more middle-aged and better educated than the workforce as a whole.

Five Types of Reward

Figure 1 depicts Sibson's EVP model. It identifies five types of rewards that the firm can offer to employees. These include:

* Direct financial rewards, including all monetary rewards.

* Indirect financial rewards, that is, benefits and perquisites.

* Career rewards, that is, the long-term opportunities for development and advancement.

* Work content, that is, the satisfaction that comes from the work that employees do.

* Affiliation, that is, the feeling of belonging to an admirable organization that shares the employee's values.

[ILLUSTRATION OMITTED]

The study found that, for both the U.S. workforce as a whole and for scientific and technical employees in particular, each of the five types of rewards are important in retaining talent. Over 60 percent of the scientific and technical sample reported that all five types of rewards were highly important in their decision about whether to remain with their current employer. This means that there is no magic bullet for keeping retention levels high. Rather, management must look at the whole picture, examining the wide range of factors that make up the EVP.

Both scientific and technical and other employees reported that work content is the most important type of reward for retention purposes. Figure 2 shows the results for both the U.S. workforce as a whole and for scientific and professional employees. Fully three-quarters of scientific and technical employees reported that work content was "very important" or "extremely important" in determining whether they remained with their current employer. Professionals want to do interesting work that challenges them and uses their skills and talents. Repetitive, narrow work with little individual discretion repels professionals.

[ILLUSTRATION OMITTED]

The other four types of rewards were roughly equal in importance, with 62 to 65 percent reporting that each type of reward was highly important in their decision to remain with their current employer. Interestingly, a higher percentage of scientific and technical than other employees reported that all five types of rewards were highly important. This may reflect the tight labor market for scientific and technical talent. Scientific and technical employees may have grown used to demanding more, and getting it, on all fronts.

We went further and analyzed data from the Rewards of Work study to understand the specific drivers of retention within each type of reward for the high-tech subsample. We examined the relationship between intention to quit and the specific variables comprising each reward type, using multiple regression analysis. Because of sample size and the statistical properties of the data, it was inappropriate to include all potential predictors in one analysis. Therefore, we conducted a separate regression analysis for each of the five types of rewards of work, examining a set of drivers within each category.

Figure 3 and Table 2 summarize some important results of these analyses. The table displays predictors within the five reward categories that are significantly related to turnover intentions in order of their relative strength (that is, their beta weights). The variance explained is for the set of predictors within each type. Because all variables could not be assessed in the same regression, the total variance explained exceeds 100 percent. The results indicate that all five types of rewards are important in explaining turnover, and 15 specific variables are significantly related to turnover intentions. Because variables from different categories were not included in the same regression, differences in the percent of variance explained are only a general guide to the relative strength of each set of predictors.

[ILLUSTRATION OMITTED]

Table 2.--Predictors of Employee Retention

                                         Variance in
                                           Turnover
                                          Intentions
                                         Explained by
                    Significant           Predictors
  Reward         Predictors within          within
   Type                 Type               Type (%)

Work           Feedback from                  20
 content        coworkers/
                supervisors
               Job responsibility
               Skill variety needed
Indirect       Time off                       22
 financial     Level of benefits
               Healthcare benefits
Direct         Pay raise                      23
 financial     Pay system
               Pay level satisfaction
Career         Career opportunities           37
               Training and
                development
                opportunities
               Supervisor style
               Job security
Affiliation    Organizational                 46
                commitment
               Organizational
               support

15 Turnover Predictors

The most important conclusion from the analyses is that there are significant predictors of turnover in each type of reward, and that all five types of rewards help explain turnover. Moreover, the specific predictors of turnover identified are actionable. That is, firms interested in reducing turnover can find ways of addressing all of these predictors. Overall, the results are similar to those for the American workforce as a whole, but there are also some interesting differences. We consider the most important findings from our analysis next.

Direct Financial Rewards

Nothing more clearly demonstrates that money is not the only factor in retaining scientific and technical employees than the finding that actual pay level is a less important predictor of retention than feelings about pay raises and the process used to administer pay. The way pay changes are made matters to employees, and it matters a great deal. People want to understand how the pay system works, and they want communication about pay that effectively tells them how they can earn pay increases. However, pay level is a significant predictor as well, and companies pay noncompetitive wages at their peril in a hot labor market.

Other analyses with the study data indicate that scientific and technical employees have a culture that is very receptive to stock options, are more likely than others to be motivated by stock opportunities, and are much more likely to base retention decisions on option opportunities.

Indirect Financial Rewards

Benefits are an increasingly expensive component of total compensation cost, reaching 40 percent or more of salary in many organizations. Benefits have become important in the battle to retain employees. In an era of two-income families, and when scientific and technical employees often work at a pace that invites burnout, time off is more important to employees than any other indirect benefit in predicting retention. Many science and technology companies offer sabbatical programs and more short-term ways to provide such a benefit.

For the U.S. workforce as a whole, but not for scientific and technical workers, benefit process (the way benefits are administered) is more important than the actual level of benefits. Scientific and technical workers seem to care less about how benefits are administered than about the value of benefits. Healthcare benefits are also a significant predictor for scientific and technical employees.

Career Rewards

Careers yielded more significant predictors of retention (four) than any other type of reward. Career opportunity was the most important predictor, followed by satisfaction with training opportunities and the employee's relationship with his or her supervisor. Job security was significant, but less important than for the workforce as a whole, probably reflecting the hot job market in science and technology. Interestingly, job title was a significant predictor of retention for the workforce as a whole, but not for the scientific and technical sample.

Work Content

Scientific and technical workers care greatly about the work they do. The most important predictor in this category is feedback from co-workers and supervisors. This reflects the work in science and technology, which is often ambiguous and difficult to measure, especially in R&D. Such conditions can leave employees hungry for good performance management and other systems that provide useful performance feedback from their supervisor and peers.

The amount of job responsibility and skill variety needed for the work are also significant predictors of retention. Autonomy on the job is an important predictor for the U.S. workforce as a whole, but not for scientific and technical employees. This may reflect the relatively high level of autonomy in professional work, which reduces the variance and thus the explanatory power of this variable.

Affiliation

Two strong predictors of retention in this category reflect two sides of the same coin: organizational commitment, or the employee's feeling of attachment to the organization, and organizational support, which is the firm's degree of support for the employee. These are also the two most important predictors of affiliation for the U.S. workforce as a whole.

Targeting the Remedies

Depending on what the company's real costs and causes of turnover are, the remedies may be very different. For example, after a merger, staff usually are nervous about job security and whether they will receive the assignments they want; the way to minimize turnover may be to rapidly clarify the organization and job assignments and get everyone settled into place. In a low-growth company that does not offer the career advancement opportunities of faster-growing companies, the solution may be a pay system that pays for scientific or technical growth. In addition to solving specific turnover problems, we have found that there are several remedies that are particularly suited to scientific and technical environments:

Better Jobs

In most cases, staff don't quit a company, they quit a job. Improving job design is one of the least used, yet most effective ways to reduce turnover in the long run. "Enriching jobs" was one of the five most effective solutions to turnover according to the IRI/Sibson survey. A few companies have made special efforts to reengineer their R&D jobs to eliminate, automate or outsource routine tasks, and to make sure that staff have real decision-making rights and work in a collegial atmosphere. Another trend is to make sure that there are ways for staff to change assignments at least as easily as they can change employers, and to reduce the ability of a manager to hold staff in an assignment that they wish to leave.

One of the top five most effective retention practices reported by IRI members is allowing staff to participate in spin-offs or new ventures. This approach allows staff who are attracted to the excitement of a start-up to experience it without leaving the company.

Managers Matter

We have seen many technical organizations that ignore the notoriously bad manager, the one everyone knows but no one wants to remove either because he or she is a technical or scientific star, or simply because they are so difficult to deal with. Often it is not until a new senior manager comes along that these obvious problems get dealt with, either by moving the individual to a non-management position or by easing them out of the company altogether. In almost all the cases we have seen, the loss of the bad manager's expertise is far exceeded by the better retention and contribution of other staff. Additionally, scientific and technical employees like to work for managers who understand what they do and can help when needed. In most cases, a purely administrative manager is ill-equipped to prevent turnover of scientific and technical staff.

Flexible Work Life

Note that this does not say "work/life balance"--working less is not necessarily the solution. Restrictive part-time policies have caused many talented professionals to leave who would have stayed if they could have had a part-time role. The trend toward helping employees work in ways and times that they find more acceptable is undeniable and is a natural outgrowth of a tight labor market. Flex hours, telecommuting and other worker-friendly policies are becoming more the norm than the exception.

The IRI/Sibson survey revealed that the least effective practice for retaining staff is the sabbatical. While it may be a very effective practice for "recharging the batteries," R&D organizations often find that staff do not return from their sabbatical. A policy that seems to be effective is additional vacation time, such as giving new hires credit for years of experience rather than just time with the company.

The easiest part of flexible work life solutions is adopting more flexible policies. The hardest part is changing the culture to allow the policies to be used while maintaining or increasing the quality and output of the work.

Pay What It Takes

Many companies are finding that they need to be much more flexible in the compensation types and levels they offer new hires. Likewise, companies are finding that they may need to be more generous and innovative in what they offer their incumbents. But the reality of cost control means that companies cannot reward all their scientific or technical staff at the same level. Knowing the true cost of turnover can help justify higher levels of compensation, but an across-the-board increase does little to lock in the highest performers. Pay vehicles being used successfully include broader pay ranges, special cash "stay bonuses," special stock option grants to the highest performers, and expansion or renewed use of dual-career tracks. Three of the top five most effective practices as reported by IRI companies involve direct compensation: "equity or royalty sharing," "hot skills premiums," and "more aggressive pay increases."

Focus on Stars

In today's competitive environment, many employers are finding that they must be clearer on the contribution level of each employee and must differentiate their highest contributors. This need comes from several sources. First, it is impossible to sustain zero turnover, so companies need to make sure they know who they are most and least willing to lose. Second, high performers may leave if they perceive that they are treated the same as low performers. However, stars must be managed artfully, not only to keep the stars but to avoid alienating others. For example, an overly evaluative environment can cause staff to become anxious and to leave for another company that appears to value them more or evaluate them less.

Keeping Turnover in Check

Managers of scientific and technical employees have a lot to consider as they attempt to limit the turnover of valuable employees. First, they need to understand the competitive opportunity that arises from limiting costly employee turnover. They need to understand the costs of turnover, trends in the retention of employees in their organization, and the risks that turnover may become much worse. Second, they need to appreciate the employee value proposition offered by their firm--that is, the full range of rewards that employees receive in exchange for their membership and effort. We have identified 15 predictors of retention for scientific and technical employees. Managers may need to collect data from employees, through interviews, focus groups, or attitude surveys, to understand the importance of a wide range of variables for limiting turnover. Finally, they need to create an action plan to address the factors that push employees out of the organization and pull them into the firm.

Given the complexity of the retention problem, the solutions usually will not be simple. However, most managers in the science and technology arena will not be intimidated by the difficulty of the problem. After all, solving complex problems is what science and technology management is all about!

Jim Kochanski is a principal with Nextera, Sibson Consulting Group in its Cary, North Carolina office. He leads Sibson's talent management practice, which helps organizations ensure they have the right number, quality and type of people needed to create and execute business strategy. He works extensively with companies that are experiencing talent shortages or changes in talent requirements. He received a master's degree in human resources development from the American University. jkochanski@nextera.com

Gerry Ledford leads Nextera, Sibson Consulting Group's employee performance and rewards practice, which helps organizations align human capital systems and practices with strategic corporate goals. A recognized authority on improving organizational effectiveness and employee well-being, he is the author of seven books and 70 articles, and for 16 years was on the faculty of the University of Southern California's Marshall School of Business. He received his M.A. and Ph.D. in psychology from the University of Michigan. gledford@nextera.com