The following article appeared in
Strategic Communication Measurement, June/July 1998
Melcrum Publishing Ltd., London

.

.

The Linear Connection Between Employee and Customer Satisfaction

By Angela D. Sinickas, ABC

 I have heard a lot lately about a linear relationship between employee satisfaction and customer satisfaction.  Can you explain how communicators can play a part in this process?

Sears conducted some interesting statistical analyses of their employee satisfaction and customer satisfaction surveys.  They found that an improvement in a group of employee measures results in an increase in customer satisfaction, leading to fewer lost customers, which results in a positive bottom-line impact (see SCM April/May 1998).

Communicators can play a significant role in this type of research because their jobs cross over the silos that typically separate the Marketing and Human Resources departments:

• Make your senior management aware that such an analysis can be done;
• Work with the Marketing Department to identify in focus groups which of the critical customer satisfaction factors are related to interactions with your organization's employees;
• Work with Human Resources Department to include on the employee survey the employee behaviors/attitudes that relate to the critical positive and negative customer interactions identified in customer focus groups;
• Include questions about communication in both employee and customer questionnaires.  If they're not included, the analysis won't be able to correlate any communication impacts;
• Make sure that the demographic breakdowns for the employee and customer research match, or it will be harder to conduct a correlation between the two.  For example, if customer satisfaction at a bank is measured by branch, employee satisfaction should be as well.

How can employee satisfaction measures be tied in to continuous improvement efforts?

• Involve employees in focus groups leading to survey design.  For example, ask employees, for each continuous improvement standard, "What are the underlying attitudes and knowledge needs?" Put these on the survey.
• Track employee perceptions of the quality of your products or services and their timely delivery (see illustration).  Compare these against your customers' perceptions and your operational measures of actual quality and timeliness. Communicators can help educate employees about any gaps, which can be the first step in changing their behaviors.  As long as employees think everything is fine, they will continue to do what they've always done.  You may also find that different employee subgroups see the issues differently. At one of my clients, the sales and management employees' were much closer in their assessment to customers' evaluations.  Production employees and managers, who have the most direct control over quality and timeliness, were the least likely to think there was a problem.
• Involve employees in developing and implementing improvement initiatives for negative issues identified in the survey.  Many of these initiatives should be developed at the work group level.  Task forces of employees from a cross-section of your organization can also provide input on enterprise-wide issues.
• Identify and measure on a survey various employee communication and satisfaction issues that relate to your company's ISO 9000 standards.

The attitudes of employees in service companies can be especially critical in determining the quality of their interactions with customers.  One of my clients, an ambulance company, identified through focus groups about 14 managerial behaviors (about five of them related to communication) that led to employees' satisfaction.  Then employees were asked to evaluate individuals in their management chain, by name, on those 14 characteristics every six months.  Manager bonuses and promotions were tied to getting good scores on those questions.  Not surprisingly, employee satisfaction increased significantly even within the first six months, and so did customer satisfaction.

How many people do you need to send questionnaires to if you want to pull a random sample? 

You'll need to talk to a statistician for the precise answer in each situation.  Here's why.  In most cases where the entire population of your group is over a couple of thousand, you need to receive back only about 400-600 surveys to have pretty reliable results for the group as a whole.  However, if you want to be able to compare subgroups against each other or against the whole group's average, then you may need about that many from each subgroup.  If you have some subgroups that are relatively small and others that are large, you may need to survey a bigger percentage of the small groups than the large groups.  This is because in a small group, one person's 'vote' makes a bigger change in the percentage for that group.  Finally, you need to look at the at the response rate you are likely to receive.  If you find that you need 1,000 completed surveys, and you are usually able to get a 50% response rate on your surveys, you should invite 2,000 people to participate in your current survey. 

What is a good response rate?

A good response rate is one where you receive enough responses so that you can be about 95% confident that a different random group of respondents would provide results on any question that are ±5% of the original results.  This means that you need to plan for a worst-case response rate when you decide how many survey respondents to invite.

Do you have any tips to improve response rates for surveys?

Most employee surveys of reasonable length delivered at work can get about a 30% to 40% response rate.  This can go up to about 60% to 80% if you have people identified in each department or location to encourage participation.  This will go down if employees are expected to complete a work survey on their own time at home.  Response rates tend to be higher if a survey is delivered electronically, as long as the questions are not perceived to be confidential since individuals feel too easily identifiable on an electronic survey.

Survey response rates for external audiences tend to be much lower, usually in the single digits, unless some type of financial reward is offered.

Should I offer an incentive to people who complete surveys? 

With external surveys, that's a good idea.  With internal audiences, there are two hazards: people might feel you can identify their individual surveys if you can identify people eligible for an incentive, and your organization will have difficulty getting good response rates for future surveys unless you continue to offer new incentives.  One technique that avoids both of these hazards is to have a reward for the departments or locations with the highest response rates overall.

(return to text)

Angela Sinickas, ABC, is president of Sinickas Communications, Inc., a communication consultancy specializing in helping corporations achieve business results through targeted diagnostics and practical solutions. You can visit her new website, CommToolbox.com, to see the automated planning, measurement, and benchmarking tools she has developed based on her manual, How to Measure Your Communication Programs.