Trying to figure out what your customers really want? There’s no need to guess. And it doesn’t take long. Put your client database to work by distributing a well-designed online survey.

Take this recent case-in-point from a TriComB2B client, Crane Energy Flow Solutions (CEFS). A critical 2008 initiative for this organization is to revamp its Web site – a tall task for a company with thousands of global customers, a large, complex sales channel and thousands of products. To start defining its Web strategy, TriComB2B helped this client reach out to more than 7000 stakeholders.

"We had only a couple days to get results for an important meeting," said Doug Clendenin, Marketing Manager, Crane Energy Flow Solutions. "TriComB2B met our deadline, but more importantly, they helped us obtain concrete results we could literally touch and feel."

In only a few days, TriComB2B delivered concrete results from nearly 500 respondents, including:

• Usability likes and dislikes
• Most desired Web site content
• Must-have functionality
• Best and worst Web sites within peer group

Improve Response Rate Through Question Development

When drafting survey questions, the most important fact to remember is that clarity is king. Do not ask loaded questions and be sure to define any terms that may not be understood, including acronyms. Keep wording concise to limit the length of the survey, but again not at the expense of clarity. If your respondents do not understand what you are asking, you will get both inaccurate data and a poor response rate.

Try to personalize the survey by incorporating the word “you” into questions where appropriate. This draws respondents in and keeps them motivated to complete the survey. In addition, utilizing an “easy” question as the opener compels the respondent to continue as they are now involved. It also establishes a sense of self-efficacy.

Answers for each question should be both mutually exclusive and exhaustive. Mutually exclusive means that there is only one answer for each question and the exhaustive means that the list of answers provided is complete. Furthermore, depending on the type of analysis you wish to conduct post survey completion, determine the level of measurement appropriate for each set of answers. Each level — nominal, ordinal, interval or ratio — has particular statistical tools available that will aid in follow-up analysis.

How to Effectively Structure an Online Survey

To improve response rate follow these easy steps when developing the structure of your online survey.

Step 1: Construct a single objective. Make sure content of your questions adheres to the established goal. Testing too many variables at once leads to inconclusive data.

Step 2: Create an introduction. Provide a short explanation of the purpose of the survey and include a time approximation for completion.

Step 3: Limit the number of questions. Ideally, a survey should contain no more than 10 questions and take less than 5 minutes. Recipients do not have the time or patience to spend responding to your questionnaire.

Step 4: Sequence questions appropriately. Avoid bias by not leading with questions that alter responses to subsequent questions. Also, try alternating “easy” questions with those that may be more difficult for the respondent to answer. This keeps them motivated.

Step 5: Obey the 80:20 Rule. Eighty percent of questions should be quantitative in nature and the other 20 percent open-ended. Placing these open-ended questions at the end of the survey results in a higher response rate.

Avoid These Common Pitfalls When Conducting an Online Survey

Avoid…

• A poor first impression. You must grab attention immediately with the subject line. This should be no more than 60 characters in length. Text and graphics should then follow to compel recipients to respond.
• Allowing each department of your organization to contribute questions. This often results in not only longer surveys, but also ones with multiple objectives. Ultimately this leads to lower response rates.
• Complicated formatting. Using drop-down boxes containing multiple answers or having respondents rank a long list can become confusing and/or tedious for users. Keep it simple.
• Spamming your audience. Online surveys should be sent from a credible source that utilizes personalization logic, such as custom salutation. Remember most recipients are looking for a reason to delete your e-mail, not read it.
• Poor timing. Sending surveys at the start or end of the week result in lower response rates. Before and after holidays is also a poor time to survey.
• Not pretesting. Use a small test sample to ensure any technical glitches have been resolved. This also reveals whether or not respondents understand the wording of the questions.

Chris Eifert
chris.eifert@tricomb2b.com

View articles about online survey best practices

• Improve Response Rate Through Question Development
• How to Effectively Structure an Online Survey
• Avoid These Common Pitfalls When Conducting an Online Survey

Marketing 101

• Planning Made
  Simple(r)
• Rigid Flexibility
  Part One
• Rigid Flexibility
  Part Two
• Ouch! Do You Really Think That?
• Reduce Your Print Expenditures
• Take a Look at Public Relations
• Marketing Admin
  Part One
• Marketing Admin
  Part Two

TriComB2B is focused on helping its clients achieve meaningful results – even when there doesn’t seem to be enough time. Contact TriComB2B to learn more about our e-marketing capabilities.