 |
 |
|
|
| Tapping Into People: How One Company Involves its Employees in HR and Business Strategy by Brien Palmer and Mike Ziemianski This article originally appeared in the April 2000 issue of the Quality Progress magazine When Respironics, Inc., a manufacturer of medical equipment, decided to assess general employee well-being by undertaking a survey of its 500 associates, nobody could have guessed the breadth of the topics that would be covered, or the uses that it would eventually serve. In 1995, Respironics made a commitment to be viewed as the area's best place to work. A group of people from the human resources and the quality departments thought that an employee survey would provide valuable input toward this goal. Also, we felt that instituting a survey process, with new surveys completed each year, would provide a systematic approach for employee relations. Planning and designing the survey Selecting and preparing the survey team To manage the survey process, a team was assembled consisting of a human resources manager, a telemarketing specialist, a staffing specialist, a statistician and a quality consultant. To make sure that the team had a good understanding of surveys, two team members attended professional training on customer survey techniques, while others benchmarked best practice employee survey processes at several companies, including Federal Express, Pepsi, Bob Evans and others. Determining what to ask and how to ask it Our background research impressed on us the importance of using a good design: the time spent planning would allow a rich and meaningful analysis of the survey results. Also, since we intended to treat Respironics associates as customers, we would use survey techniques that applied to customer surveys. One of our first issues was determining what questions to ask. After some interesting sessions during which we discussed the topics to cover, we decided to ask our customers. We selected individual associates at random and held several focus groups. At least two members of the survey team participated in each focus group, one acting as the moderator and one as the scribe. Some of the sessions seemed to start out slowly, but soon we were getting as much input as we could handle. It was a challenge for the moderator to maintain a neutral tenor and not influence the discussion, and for the scribe to record the input as fast as it came! We used the results of the focus groups to identify the topics that we should address on the survey. For consistency and ease of analysis, we chose to use the same syntax for all questions. Specifically, we chose to express all of the survey "questions" as statements, phrased in a positive manner. For example, we used "My workload is reasonable" as one question. We kept this convention throughout&emdash;that way, "strongly agree" always indicated a favorable result and "strongly disagree" always meant bad news. We also decided to ask very direct questions. (A person responsible for one particular area of employee benefits asked us, "You're not really going to ask that, are you"?) For recording the results, we chose the Likert five-point scale based on its ease of use and its general acceptance in the literature and the benchmarked surveys. In addition, each question offered an "N/A" (not applicable), even though we felt that we had taken care to select topics that applied to everyone. During the training sessions that some of us had attended, we had become aware of the power of the "importance" dimension in surveys. While surveys typically ask for information related to "satisfaction", a few go further and ask the participants to rate the importance of the topic. For example, employees could register low satisfaction with the parking situation, but still consider this relatively unimportant. We thought it made sense to assess both the satisfaction and importance dimensions of each topic, so that we could most effectively apply our resources to the issues raised by the survey responses. We considered several ways of doing this, including asking, "How important is this?" after every question. Finally, we came up with a simple convention that proved effective. See Figure 1 for an example. In order to break up the questionnaire, we grouped the questions into ten different topic areas, plus one summary area: Quality of Worklife (Refer to Figure 2 for some sample questions.) In addition to the questions scored with the five-point scale, we decided to use some open-ended questions. This would give people an opportunity to address any topics that we had not included, and might add another perspective to the results. (See Figure 3 for these questions.) After we had drafted a series of questions based on focus group input, we took it to senior management for its input. They liked the questions, but wanted to add others that focused on important strategic business issues such as use of information technologies, effectiveness of company leadership, deployment of quality policies, etc. We held several meetings with managers and added the resulting topics to a new section called The Business. Answers to these questions would provide robust input to the Respironics' strategic planning process. Demographics and anonymity In addition to designing an effective structure for the survey questions, we needed to capture demographic information. This would let us edit the responses to see whether there were any patterns of dissatisfied customers in different demographic areas. Accordingly, we gathered information on the following demographic categories: Location Capturing this level of demographic data required sensitivity to anonymity. Earlier, we had decided that we would leave the results anonymous so as to encourage candid responses. However, we realized that detailed demographic information would allow some people to be singled out. For example, how many people would fit the description of a female customer service manager, 31 years old, serving at one particular location, with two years tenure? We decided to do three things to preserve anonymity. First, we attempted to keep the demographic categories broad enough to preclude groupings of less than about five people (although we recognized that this was not always possible). Second, we limited access to the survey results&emdash;we would not permit anybody to use the data to isolate survey results from individuals or small groups. Third, we made public our commitment to confidentiality. (Judging from later feedback, this seemed to work.) Conducting the survey In order to encourage a high response, we used publicity, confidentiality and careful survey design. However, we got the best help from management. In the planning stages, we struggled to keep the number of questions as low as possible, since the response rate generally correlates inversely with the length of time needed to complete the survey. When we pointed this out to the company president, he told us not to worry about the time, that everybody would take it on company time. We consequently scheduled series of one-hour meetings dedicated to completing the survey. Analyzing the detailed survey data The response rate was 351 of 450 associates, or 78%. We considered this acceptable for a first-time survey, but certainly capable of improvement. During our planning phase, we had noted that several companies received lower-than-hoped-for responses in the first year (some below 30%). Several sources told us that to expect several years of surveys to achieve responses as high as 90% or more. The design of the survey made it possible to translate data into meaningful information, although this proved to be a far more time-consuming process than originally intended. Each question on the survey was analyzed against each possible demographic and organizational component. The intent was to become familiar enough with the survey results from a descriptive perspective so that sense could be made of the statistical analysis. In the first year of the survey, this helped establish potential benchmarks and provided a first time view of how associates perceived various issues. The actual results of the survey data were translated back into spreadsheets for ease of distribution to line managers. One file was created for the overall results of the entire company showing the feedback from all associates, and then breaking the data further into exempt and non-exempt employees. A separate file for each major organizational group was also created along this same format. The all-company results were copied onto these spreadsheets. By subtracting the organizational group's average from the all-company average for each question, it was easy for line managers in charge of various organizational groups to quickly see where they had significantly different responses, both positive and negative, from the rest of the organization. This helped to identify unique, area-specific issues that would have been hidden if only company-wide data was considered. An example of this type of hidden information was apparent in safety issues. Although the organization as a whole had good scores concerning work related aches and pains from a physical perspective, the view of this issue for maintenance and production workers was substantially different from the rest of the company. General results The first year, we manually entered results on a spreadsheet, which we imported to a commercially available software package. Our statistician set up the computer program to report survey results in quadrants, plotting both the importance rankings and the satisfaction rankings (see Figure 4). The upper left quadrant contains results expressing dissatisfaction and perceived importance: a definite area of interest. The lower left quadrant contains results expressing dissatisfaction but not importance. The lower right quadrant contains items perceived as satisfactory but not very important. Finally, the upper right quadrant contains the good news--issues perceived as important and with which people were well satisfied. This graphical structure proved very helpful in presenting results. Anybody could see at a glance the general result of the survey, and could quickly focus on outliers. The pattern shown was generally positive: most questions registered in the upper right quadrant, indicating relatively high satisfaction coupled with relatively high perceived importance. All questions were rated over the middle value for importance, indicating that the team had done well in selecting questions to ask. Also, most questions showed high satisfaction levels. The mean score for importance was 3.75 (of 5), and the mean score for satisfaction was 3.51. To present the results, the team ranked the questions in order from best news to worst news, and, for added convenience, listed the top 20 and bottom 20. Also, we created histograms based on the responses. The quantified results provided a reference point from which to historically track trends over time and establish a baseline measure of overall employee satisfaction. These results also served to rank issues of importance to the organization so that relevant information could be fed into the annual planning process. That way Respironics could have broad input from the organization for the strategic planning process. An overall employee satisfaction index was derived from the composite scores of the ten summary questions in the last section of the survey. The intent of this was to hold a constant set of questions that can be tracked over time, providing a trend line for measuring changes in overall employee satisfaction. Strategic planning The survey team presented the results of the survey at a strategic planning meeting attended by all Respironics' managers. Since the questions covered such a wide range of topics, including issues important for business strategies, it seemed natural to use the results as input to the strategic planning process. It is common practice for a few managers to meet to assess a company's Strengths, Weaknesses, Opportunities and Threats (SWOT), producing an analysis from which to formulate a business strategy. In this case, however, we had in effect a whole-company SWOT analysis, that is, an analysis by the management staff using input derived from the whole organization. This gave Respironics a broader perspective on issues of importance to the business, and perhaps avoided some insularity. By analyzing employee input, various themes were developed. Teams of managers discussed the issues presented to them and later provided feedback to the overall group. This feedback was then incorporated into an annual strategic plan for the upcoming fiscal year. This plan was then communicated to all associates as part of the formal goal-setting process for the entire company. In a systematic fashion, issues were presented to each organizational area. Managers then were asked to develop one or two initiatives to address the concerns raised in the areas they managed. In addition to these programs, Respironics developed some cross-functional projects to address some company-wide areas of interest uncovered by the survey. For example, the company undertook projects for: Improving its 401(k) plan, Using a systematic process, the strategic plans and high-level projects were rolled down to specific initiatives on the part of departments and individual employees. Thus the efforts of departments and individuals over the next year were aligned with the strategic plans. Communicating results Members of the survey team helped publicize the results of the survey and the nature of the improvement projects on a company-wide basis. This helped disseminate the survey results and also raised expectations for improvement actions. In addition to the general publicity, Respironics' human resources manager met with each department manager to debrief them on issues related to their department. This gave the managers the opportunity to initiate improvement actions that might not necessarily apply to the broader employee population. Several managers did institute improvement projects based on these reviews. To their credit, managers did not attempt to trace low satisfaction results to specific individuals, or otherwise compromise the anonymity of the survey. Measuring progress over time The survey team viewed the survey project as a process, not an event. As planned, we repeated it the following year (1996). The experience of having done it once made the second survey easier to conduct. Participation in focus groups drawn from the general employee population, providing feedback to key management members and communicating the results to all associates had almost become an expected part of conducting business within Respironics. In order to track progress over time, the team kept the wording of some questions constant, including an employee satisfaction index. This provides the ability to measure responses over time, while adding or dropping other questions to meet evolving business needs. Results of the second survey Departmental groups that diligently worked on their initiatives showed substantial improvement in their scores. Groups that failed to develop and implement plans effectively showed a decrease in their scores. This demonstrates the importance of follow-up surveys with effective actions, particularly since the expectations will have been raised on the part of the survey participants. The cross-functional project on compensation also showed a shift in the scores for measures pertaining to pay. Although compensation was still a concern to people in the organization, the implementation of a gain sharing plan, better communications of company pay philosophy, and changes to the 401(k) plan were all positive outcomes from the first survey that shifted the relative importance of pay down the list when compared to other issues. However, some employee concerns did in fact carry over to the second year. For example, employees still reported some dissatisfaction with communications in the company, and with the pace of integration of information technology upgrades. These scores no doubt reflect the length of time required to improve these areas. The overall satisfaction index score improved by 2.3% in the second year. Continuously improving the survey process Since the survey team intended the employee survey as a vehicle for continuous improvement, it made sense to by improve the survey process itself. To this end, we made several changes. The second year, we used scanning technology to enter data, rather than doing it by keystroke. We added some questions to better understand some issues and to respond to subsequent focus groups. We consolidated some other questions. We asked an educational institution to review our methodology and got a favorable review. Finally, we established and tracked two benchmark metrics that relate to the survey process itself. The first was the level of employee participation, and the second was the level of employee confidence that improvement actions would actually make a difference. Both metrics showed slight increases in the second year. What we learned about the company The surveys gave a rich insight into how the employees perceive Respironics and the business, and how satisfied they were with the quality of work life. In general, employees reported that they felt they knew the business. They were generally satisfied with their benefits and their work life. They perceived management as generally approachable, and held the company in high regard. The survey results also provided many opportunities for improvements. The company will continue to work on its flexibility to change, its communications, its information systems and its facilities. What we learned about surveys Use good design principles. Good survey design makes a survey look simple, supports a high return rate, and makes analysis more robust. Good design is mandatory. However, it takes time and effort. If you plan to use employee surveys as an ongoing process, expect to invest considerable effort in planning. Ask bold questions. You will be surprised at what people will tell you if you ask them directly. Survey data provides robust input to strategic planning. Survey data provide a broad perspective on a wide variety of business topics, particularly those related to internal strengths and weaknesses. Employee surveys provide excellent opportunities for collaboration between Quality professionals and Human Resources professionals. The survey process provides an excellent vehicle to address both quality topics and people topics, and to provide input into strategic issues affecting these areas Don't start something you can't finish. Recognize that a survey consists of not just the survey itself, but more importantly, the projects undertaken as a result. A survey is a beginning, not an end. SIDEBAR 1 Quality Principles In retrospect, the survey process showed several quality principles in action: Continuous Improvement: The survey process provides a vehicle for continuous improvement, one of the company's core values. Customer Satisfaction: Surveys provide a vehicle for measuring and monitoring customer satisfaction, another core value. Employees as Customers: In this case, we treated the employees as customers. Employee Involvement: Quality practitioners have for some time championed the idea of a high-involvement work force as a way to improve quality and business performance. The survey process lets employees provide input on a range of issues, from benefits to strategic planning. Use of Effective Survey Techniques: With the help of the "importance" dimension, we were able to assign logical priorities to improvement opportunities. Also, we used the detailed demographic data to focus on opportune areas in the organization. Strategic Planning: By including all employees' perceptions on issues of strategic importance, the input to the strategic planning process became richer and more robust. SIDEBAR 2 The People Side of Quality Instituting the survey process has let us see several areas where the interests of the quality professionals and human resources professionals converge: Common Goals - Both quality and human resources professionals are seeing a greater focus on business performance, and on how quality and HR practices can positively affect business performance. While quality practitioners have traditionally focused on business systems and processes, and human resources have traditionally focused on people systems, the two areas clearly have a common interest in achieving good business performance and a satisfied work force. Teamwork - Both fields have traditionally recognized the importance of teamwork. Aligning with Core Values Both quality systems and HR systems must align with an organization's core values before it can reach peak performance. The survey process gives one way to measure this alignment. Human Resources in Strategic Planning - Increasingly, business leaders are recognizing the importance of a "people strategy" in achieving quality and business performance objectives. The Baldrige award criteria specifically recognize the importance of human resource systems in achieving quality and business results. Return to Articles |
|
Home | Services | Products | Articles | People | Contact | Your Growth | How We Work | Testimonials | Clients | |
 |