What's the best way to find out what people think of your business and where they think you need to improve? Just ask them, Sherlock.

Regularly surveying your customers can provide a direct insight into how happy your products and services make your customers, what deficiencies hurt your bottom line, and on what kind of new product development you should focus your efforts. Professional surveyors talk a lot about the concept of "fit:" Is your business meshing with your desired audience? If not, you might as well be throwing out money. 

"Quality is in the end much more important than price in terms of determining overall satisfaction," says David VanAmburg, director of the American Customer Satisfaction Index. "It's very easy to attract customers by offering discounts. At the end of the day, if you don't offer a quality product that taps into what the customer feels is really a good fit for him or her, it doesn't matter where your price is."

Some corporations use customer surveys to impress their stockholders, others, such as public utilities, show the information to regulatory commissions, while still more use it to track trends over years. Customers who rank themselves as "completely satisfied" are worth three to six times more than those who say they are just "satisfied" or "dissatisfied," says Jeffrey Henning, founder and vice president of strategy of Vovici, an online survey management company that has worked with Marriot, Cisco, and many other large companies.

Most customers, VanAmburg says, are eager to share their opinions, and show a great deal of savvy for rating products. They're just waiting for you to ask.

Writing a Customer Survey: Identify Your Goals

Before you even start thinking about what questions you want to ask customers, survey professionals say you should ask yourself: What am I trying to learn, and what am I going to do with that information?

"If customers are happy, you really want to know that. If they're not happy, you can't hide," says Howard Deutsch, CEO of Quantisoft, a survey and consulting company. "You need to know why and you need to take action, or you're going to go out of business."

Deutsch says growing companies should strive to conduct a customer survey once or twice a year.

Don't ask customers a question without a plan for how it will be used to provide insight for you company's stakeholders, says Gina Pingitore, the chief research officer for J.D. Power and Associates, a global customer satisfaction research firm best known for its automotive quality rankings.

"What's the impact of being a yes or no on satisfaction scores?" she says. "You should have a very clear understanding of how you're going to analyze the data."

Deutsch says companies usually turn to two types of surveys. Self-service questionnaires through web services such as Survey Monkey, through which a company can write its own questions and then be presented with the raw data. The second type is to go through more individualized professional survey services that have their own survey methods and present you with analyzed data, charts, graphs and detailed comment reports.

Experts say common goals of surveys include:

  • Measuring customer loyalty
  • Helping human resources departments train staff or execute new staff initiatives
  • New product development
  • Direction for new financing
  • Gauging customer service effectiveness

Writing a Customer Survey: Crafting Quality Questions

Designing a questionnaire is a more complicated science than most people think, Pingitore says.

"People get their Ph.D.s in it," she says. Every aspect of the survey can affect the outcome, including the phrasing used to pose the question, the order the questions appear on the survey and the options for how to answer, such as whether respondents are asked a yes-or-no question, or to rate their response along a scale.

Professionals say to keep these tips in mind:

  • Don't write questions that are ambiguous. Make them as specific and targeted as possible.
  • Don't write "double-barrel" questions, such as asking, "How easy and timely" an experience was. "They're two different constructs," Pingitore says.
  • Use scalable questions that ask customers to rank their responses on a numerical or qualitative spectrum. The ASCI uses a "multiple indicator" approach that creates scores based on responses to three different questions that relate to customer satisfaction. By asking customers, for instance, 1) how satisfied they were with an experience, 2) to what degree did their experience exceed or fall short of their expectations; and 3) how that experience compares with their ideal, the results create a weighted three-dimensional picture, VanAmburg says. The ASCI uses a 10-point scale for its qualitative questions as well, which allows for more grey area than a more narrow scale, he says.
  • Ask a lot of questions. "The more questions that you ask, the more facets of it you get at, the more you minimize the margin of error, the more you minimize the noise in any survey," VanAmburg says.
  • If repeating a survey, make sure to keep the questions identical from year to year so the results can be compared.
  • Include at least a few open-ended questions. They allow for broader feedback you may have left out of your survey. "If a question is worth asking, it's worth putting a comments field in after each section," Deutsch says. "They'll pour their guts out in many cases."  Open-ended questions are tricky because the answers often aren't specific, Pingitore says. But as coding software even for large-scale surveys has improved, these kinds of questions help round out your survey and add more flavor to the outcome.

Writing a Customer Survey: Choosing the Best Format

Figuring out how to distribute your survey depends on your type of business. Phone surveys used to be the standard in the industry back when all customers had land lines, but in the era of the cell phone and do-not-call list, it's less reliable. When Vovici recently tried to do a national phone survey, halfway through the survey process the company realized it hadn't reached a single person under the age of 24. If phone surveys are your only option, keep the questions short and make it clear right away that you're not trying to sell anything, Pingitore says.

Online surveys are now the preferred method because they are the most cost-effective, efficient means of producing data quickly, experts say. They also eliminate the human error from a surveyor keying in data over the phone. 

If you only have customers' mailing addresses, it's more cost-effective to mail a postcard directing people to an online survey rather than send the whole questionnaire, Pingitore says.

Some stores also have success handing out a survey at the register or printing a link to a survey on the receipt, Deutsch says.

Writing a Customer Survey: Work Toward Getting a Great Response Rate

It's a good idea to plan ahead for a survey and start building a database of customer contact information.

"If your list isn't good it doesn't matter how good your survey is," Pingitore says. It might even be smart to send out a test e-mail to your contact list to see how many addresses bounce back before investing in a survey.

Surveying by e-mail also means you'll have to format it so it doesn't get marked as spam, or disregarded as e-mail marketing. Survey companies can help you tailor the keywords in your subject line and body of the message so that the purpose is "simple, clear, and motivating," she says. 

If you're doing a snail mail survey, trade up a bit: Anything that distinguishes the survey from direct marketing will increase your success rate. First-class postage is more expensive but also more effective than third-class, Pingitore says. Use a laser printer to make the address look hand-printed. Make the form attractive with white space, large fonts and a clear description of how the information will be used. Allowing respondents to remain anonymous also helps, but if personal information is collected, you should clearly describe how it will be used.

Picking a random sample is important to make your survey credible, Henning says. "If you have true randomness, then you only need to talk to 400 to represent population in the millions," he says. Smaller companies have a harder task: if you only have 100 customers, you'll need to talk to 80 of them to get a significant confidence level," he says.

Many stores are able to get contact information through club card or frequent buyer program: 40 percent of people will usually give their address when asked at the register, Henning says.

"I actually think that justifies having a loyalty program alone," he says.

Writing a Customer Survey: Interpreting the Results

One sure way to annoy your customers or clients is to ask for their opinion but then do nothing with the feedback.

"If you don't take action, you're better off actually not doing the survey," Deutsch says. "You're really disappointing them, and chances are they may go elsewhere."

Pingitore cited a recent study in Germany that looked at the impact of consumers who took part in satisfaction surveys. Customers who believe companies take action based on the feedback feel better about the company and are more likely to respond to surveys.

Once you receive the results, circulate the data widely throughout the company to make sure everyone knows what's on customers' minds.

"Too often management will see them and no one else will see them," Henning says. Publicizing the results can also help too: it paints your business as responsive to customer concerns, and willing to make pragmatic changes.