"Thanks for shopping at store XYZ. Would you like to fill out our survey for a chance at winning [insert incentive here]?"
Eager to take the chance, the customer accepts his receipt with the printed URL or phone number and later, at home, logs in or dials to begin the survey. Forty-five minutes of mind-bending questions later, he wonders if he'll ever reach the end. When he does finish, he vows never to repeat the experience.
Unfortunately, such reactions to customer surveys are as common as the poorly crafted surveys themselves. Retailers need surveys that will provide helpful information without creating a chore for the customer. Here are four rules for building a superior survey:
1. Know what you want
What is the purpose of your survey? Planning is a must. Determine your specific needs and craft questions accordingly. Don't ask questions for which you already have answers. "The answer may be in your data bank, but you haven't put it together in a report," said Mary Grace Crissey, analytics technology marketing manager for biz intelligence and analytic software developer S.A.S.
And don't ask generic questions ("Was the employee polite?" "Was the service fast?"). Instead, take time to determine the questions your business needs answered and to learn how your product is different from competitors'. "You have to give the forethought as to how you want to distinguish your entire offering," said Gary Edwards, executive vice president of customer satisfaction consultancy Empathica.
2. Watch your language
If a company wants to be viewed as hip and fun, then the questions on its survey should be hip and fun. Don't use business jargon for a customer survey. Instead of asking "Did you receive fair market value for the product?" ask "Did you get a good deal?" Don't try to sound academic; doing so may lose the survey taker, but it also may damage your brand.
Every part of a retail experience should project the brand, even questions on a survey, says Empathica's Edwards. If a company wants to be viewed as hip and fun, then the questions on its survey should be hip and fun; they shouldn't read like a Ph.D. dissertation. "We are trying to capture the essence of the brand," Edwards said.
3. Keep it simple
A survey should be simple, so keep it short and keep it clear. Edwards says an ideal survey length is between 10 and 30 questions.
For clarity, keep survey questions brief. The more words in a question, the more likely the customer will be confused. So-called "double-barreled" questions, which ask two or more questions at once (e.g., "Did you receive your food promptly? Did it taste good?"), especially should be avoided.
4. The long and short of it
Many surveys rely on customers answering via a number scale. S.A.S's Crissey says surveys that rely strictly on number answers can go wrong in many ways — customers may misunderstand the scale and think 1 is excellent instead of terrible, for example. To avoid the possibility of confusion, supplement number scale answers with written responses. Longer responses make the survey more stimulating for the customer, Crissey says, as well as offer the added benefit of separating legitimate, helpful participants from crackpots.
Crissey added that questions that require long answers should be targeted and specific. Text mining software can help search through those responses.
Better beware
The Achilles heel of surveys is the professional survey taker — someone who takes more than six surveys per quarter, according to a definition by Forrester Research. Typically, such a person takes the surveys for the participation rewards offered, usually with an eye to completing them in quantity rather than providing quality responses.
From a survey of 4,913 people, Forrester determined that 14 percent take more than six online surveys a month, with 4 percent taking more than 20. Forrester recommends companies weed out these results-skewing pros by asking respondents how frequently they participate in surveys.
