Editor's note: Gayle Kaplan is senior research analyst with National Computer Systems, Minneapolis.

National Computer Systems (NCS) is in the business of helping customers collect useful information about customer and employee satisfaction. Among other products, NCS provides customer satisfaction and employee attitude research services.

NCS knows that many companies are doing customer/employee satisfaction surveys as part of their quality program, an important first step to discover customer and employee opinions and needs. But NCS has found that some companies run into difficulties translating their survey information into action plans which operations or marketing staff can use. NCS decided to sponsor a roundtable discussion at the 1993 American Quality Congress on companies' experience transforming survey information into action plans.
NCS invited a group of conference attendees to participate in the discussion. Members were invited based on whether

  • their company conducted customer or employee satisfaction surveys; and
  • they were involved in conducting the surveys.

Eight people from companies in different industries and across the country participated in the discussion, as well as two NCS research staff members.

All the participants were part of their firm's quality department. Their titles included quality control coordinator, quality specialist, corporate manager of quality and director of quality. All the companies have conducted employee and/or customer surveys; most have done both. Group participants were in charge of or involved with conducting the surveys at their companies. These facts lead us to believe that quality departments are usually involved with customer and employee surveys and often direct them.

Employee surveys

Participants raised several issues. Some companies are concerned that conducting employee surveys raises employee expectations - "We might need to do something." The companies are unsure they can meet employee expectations, especially in a short time frame; they don't want employees to become more dissatisfied as a result of the survey.

Other group members responded that the basis of a successful employee (or customer) survey is determining how to use the survey data before conducting the survey, so that the company can set employee expectations, not be driven by them. For example, the company may decide that each plant manager will work on the top four issues at his or her respective plant. Some also suggested that the company should decide upfront to conduct structured employee feedback sessions. These sessions let employees know the results of the survey, overall and for their group and ask for the reasons behind the responses. Respondents agreed that feedback sessions were very useful, maybe even the most useful part of an employee survey. Group participants also agreed that employee surveying has to be a process, not an event, in order to maintain management credibility with employees.

Confidential and anonymous

Another issue was employee distrust of management and concern about the anonymity and confidentiality of their input. These concerns may decrease response rate or bias results. In one case, employees knew they could be identified; the results were very positive even though the company had just gone through a very difficult restructuring.

All the roundtable members acknowledged that employee confidentiality and anonymity is an important issue. Beyond that, however, there was a range of responses. Some said that survey results should be broken out only along very broad areas, such as type of job (e.g., management, professional and line worker), number of years at company and division, and the smallest breakout group should be 50. With this broad analysis, the company gets useful information but the results are too general for an individual employee to be identified. Other companies wanted reports on each management group so they could evaluate managers. For employee feedback groups, some employees did not want their managers or human resources staff in the room.

Buy-in necessary

Participants agreed that top management buy-in is a necessity for a successful employee survey program, not only to conduct the survey but to implement actions and resurvey to determine the results of those actions. One suggestion to help attain buy-in was to involve top management in the design of the survey instrument. Explicit executive buy-in also helps to increase response rate by showing that the survey is important to top management. For example, a pre-notification or cover letter to survey respondents should be signed by the division vice president or the company president.

Another issue was how general or specific the questionnaire should be. Problems arose when asking employees' opinion of areas that participants knew they could not change -- or areas that management didn't want to ask about. Several participants suggested that the first survey should be more general and cover issues that could be changed in the next couple years. Once actions have been taken as a result of the initial survey, future questionnaires could deal with the tougher questions. On-going surveys should always ask:

  • if the company communicated the results of the previous survey;
  • if employees have seen changes as a result of the previous survey and;
  • if the survey is a useful method for tracking employee opinion.

Communication from management is often a major area of concern. Sometimes it's possible to improve overall employee opinion just by improving communication to employees.

Another issue was whether to survey all employees, i.e., a census, or survey a sample of employees. A census gives all employees the opportunity to participate and allows for detailed data analysis at the department or group level. A sample is less expensive and time-consuming to administer.

Summary

Overall, the discussion suggested that an effective employee survey program should include the following steps:

  • Obtain top management buy-in; if possible include top management in the design or review of the survey instrument.
  • Make sure survey results are anonymous and confidential, and make sure employees know it.
  • Decide how the company will analyze survey data and implement recommendations before the first survey is sent out.
  • Consider designing the first questionnaire to determine general areas of concern or more easily changed areas. Ask about thornier issues in future surveys.
  • Set up employee feedback sessions to communicate survey results and find out the reasons behind the responses.
  • Make changes based on the most important concerns.
  • Communicate plans and actions taken to employees.
  • Resurvey employees every 18 to 24 moths to determine if survey results have been communicated and if employees have noticed any changes.

Customer surveys

In general, participants felt that companies are usually more interested in doing customer satisfaction surveys than employee surveys because customers bring in money and are seen as more likely to leave than are employees. In short, companies are more interested in keeping customer than in keeping employees.

Some customers are not interested in doing continuous surveys unless they know what the companies are doing with the data. They want the company to "do something significant with it" -- they want to know that someone is interested in the survey results. Some business-to-business companies have found their customers require survey feedback; they share their survey results with customers. Other companies find it hard to determine how to communicate survey information or general action plans to their customers.

Companies differed widely in how and how much they have used results from customer surveys. The extremes ranged from throwing the data away to restructuring the company based on one key question. One company ignored any negative input. Some companies have started planning how to use the data but have not yet implemented action plans.

Of the possible actions a company can take, one action is to call every customer who gives a negative response to a key question or questions to find out the reasons behind the negative rating. The company then gathers additional detailed information and can also fix whatever problem led to the poor rating. Another action is to send a letter of apology and/or compensate the customer for the bad experience.

One participant recounted his experience as a Marriott customer. He received a personal letter from the president and a free night's stay, which changed his attitude from totally negative to semi-positive.
Participants agreed that it is "not an easy process" to find out what customers think and the reasons behind their opinions. Several participants suggested conducting focus groups or personal interviews to uncover the causes for the survey results. Company staff need to "get into the trenches" to find out what customers really want and determine how to solve their problems. One suggestion was to train internal people on problem-solving methods to help them meet customer needs.

Some companies evaluate employees or compensate them based on the results of the customer satisfaction surveys. Incentives can include recognition or monetary rewards. However, problems may result if too much emphasis is placed on one or two key questions. Staff may ask customers to rate them highly on those questions or may focus all their service efforts on getting a high rating on just those questions.

Other companies do not use incentives but do require their managers to develop and implement action plans based on survey results within a certain time frame. In one case, businesses were expected to take customer survey input and incorporate it into their strategic plan.

A few participants had conducted suppliers surveys with very positive results. They obtained high response rates (more than 90 percent) and useful information. In one case, the information was "earth-shattering." The speaker said that most employees at his company thought the company paid its suppliers very slowly and that this slow payment caused problems with suppliers. However, in their supplier survey, no one indicated that slow payment was a problem at all.

Almost no participants compared or correlated the results of customer surveys to employee surveys. One company informally compared findings through its audit procedure, while another used the same underlying constructs but did not tie the two surveys together.