Asking Why: How to Do Online Surveys

These days business owners are forced to act quickly to increase profits, create effective customer acquisition programs and reduce customer defection. While much attention has been placed on finding the right customer relationship management system, an equally important question has gone unanswered: Why? Why has a customer chosen to do business with you? Or equally important, why did a prospect not choose to do business with you?

Although CRM solutions help shed light on questions such as who, what, when, where and how much, they’re can’t offer insight into the “why” of the relationship. Online surveys can help answer that question.

For example, in the insurance industry, executives continually are faced with more competition, an overarching objective to boost their efficiency and cross-selling opportunities, and to improve their ability to evaluate risk. Web-based surveys can play a pivotal role in the value chain by capturing information from every link—including agents, brokers, adjusters, clients and partners—that can then be translated into business initiatives and processes.

Web polls can enable companies in a variety of industries to gain key feedback and hear the customer’s “voice.” Creating a survey that produces results capable of providing accurate information to make sound business decisions is more difficult than most people realize.

Typical questions an organization faces when embarking on such a poll tend to fall into five areas, each critical to the project’s success.

1. Determine the business process. The survey designer is faced with many important questions about the poll’s content and “look” which require consensus from all members of the team prior to execution.

They include:

•What is the survey’s purpose?

•What are we trying to measure?

•How many questions should be written?

•What type of rating scale should be used?

•How will we know that the survey worked? •Will the data be actionable?

2. Design the survey. Good survey design ensures that your organization will be able to get the results it needs. Some of these steps, although simple, often are overlooked. That can mean significant reductions in response rates. To be successful:

•Begin with a title and preamble. Explain the poll’s overall aim, either in the invitation or at the beginning of the survey itself.

•Balance the use of white space between both questions and sections. This greatly improves readability without unduly increasing the survey’s size.

•Provide general instructions to respondents at the beginning of each section and clearly define specific instructions associated with the different question types.

•Allow your respondents to bypass questions (or whole sections) that aren’t relevant to them.

•Allow “don’t know” and “not applicable” selections. When a large number of respondents choose such options, it’s time to examine whether the question is badly worded or in the wrong place on the questionnaire.

•Keep it short and simple. Enough said.

•Keep the reading level of the instructions and questions at the eighth-grade level or lower, without being condescending.

Another factor in designing the survey is to determine a rating scale. Scales are critical to your research’s success. Well-designed scales are easy to understand and accurately represent respondents’ true attitudes, preferences or opinions. However, two or three point scales usually aren’t distinct enough to rate the importance of various attributes. For instance, scales with four to eight points provide far more insight into the subtle distinctions and values of an attribute.

3. Select the sample. A sufficient sample size is an important requirement for a successful survey. If the size is too small, erroneous conclusions are possible. For example, a researcher may make a conclusion that no differences between groups exist when in fact they do, but were undetectable from an insufficient sample.

4. Conduct the survey. Response rate is the single most important indicator of how much confidence can be placed in a survey’s results. A low response can be devastating to a study’s reliability; therefore, testing is essential. What may seem obvious to the survey’s author may be completely unclear to a typical recipient. A difficult question will be misunderstood or skipped. What’s worse, a survey that’s too difficult to understand most certainly will be thrown away.

So how does one go about increasing those rates? One of the most powerful tools for accomplishing this is the use of follow-ups or reminders. Traditionally, between 10% and 60% of those who are sent questionnaires respond without reminders. However, these rates are too low to yield confident results. Following up, then, is imperative to a survey’s success.

5. Analyze and report the results. Response analysis should address two concerns: (1) the validity of the questions; and (2) the substantive business issues that were the survey’s purpose.

Validity can be assessed by examining the number of respondents who chose each response option. No single option should have more than 85% of the responses and none less than 5%. Business issues can be assessed by reviewing responses to individual questions and groups of questions on a single theme that are treated as separate measures. The inclusion of key demographics also can provide valuable opportunities for insightful subgroup analysis. D

Paul Squires is president of Applied Skills & Knowledge, a Morristown, NJ management consulting and outsourcing company. Arturo Coto is CEO of Inquisite, an online survey technology provider in Austin, TX.