Deliver Actionable Survey Data
Through Effective Survey Design

Anyone can ask a question and get a response. Knowing the survey questions and response options to use is critical to generating actionable data. When creating your survey, follow these proven guidelines:

1. Avoid the “no-brainers” that are universally agreed upon. Do you really need to ask your customers if they prefer to be treated with respect? Who would respond with anything other than an enthusiastic “yes?”
2. Identify the survey’s objective, and structure questions and response options to achieve the objective.
3. • If gauging customer loyalty, utilize the proven Net Promoter Score® (NPS®) approach to compare results to documented industry benchmarks.
   • Utilize Likert statements to measure service performance. Positive statements (e.g., “The representative answered my question.”) followed by “Strongly Disagree” to “Strongly Agree” options reduce inherent positive bias through the use of negative to positive scales.
   • Use multiple-choice questions when the options can be clearly articulated and easily perceived by the customer. (e.g., my call took less time, more time, or about the amount of time I expected). This is most useful when the selected option can be directly correlated to an observed product or service attribute.
   • Avoid 10-point rating scales. Customers respond accurately to fewer response options, and analysis frequently focuses on “top two boxes.” In these scenarios, 10-point scales are essentially converted to five-point scales. NPS® (0-10 scale) is the lone exception to this guideline.
4. Beyond questioning techniques and scales, survey effectiveness as measured by accuracy of results can be enhanced by:
   • Accurate customer perceptions relative to reliability or ease of use based on experience with the product; avoid the bias of customers still euphoric about a new purchase (honeymoon reviews).
   • Avoid survey fatigue. When a colleague from perhaps another department asks if you have room in your survey for another question, the answer should be a resounding “No.” Excessively long, highly granular surveys frustrate the customer and result in fewer completions and lower reliability of results.
   • Test the survey with a small sample of customers (in person or via phone, optimally) to ensure that survey question language is not confusing. Revise language accordingly for the audience.
   • Ask essential survey questions early in the process to ensure that critical data is collected even on incomplete surveys (e.g., loyalty and overall satisfaction questions).
   • Drill down to understand potential root causes when customers provide negative feedback. Avoid the arrogance of predetermined lists, and allow customers to provide free-form comments. “You don’t know what you don’t know” may be a cliché, but it is true. Take the time to catalogue and analyze free form comments to understand and learn from the actual, valuable voice of the customer.