In addition to simply not reaching the right respondents, here are four reasons your market research survey might be failing:
According to multiple sources, the number of questions on a survey can directly impact the number of people completing it. It’s been shown that the ideal survey length is 10 questions, which usually takes five minutes to complete. Bump up the number to 30, and the response time increases to 10 minutes. For surveys with more than seven or eight questions, survey completion rates can also drop 5% up to 20%. It’s not always the case that longer surveys are harder to complete; factors like topic interest and the perceived value of the relationship between the respondent and survey sponsor can affect how vested a person is with his or her answers. But why take the chance with possible respondent fatigue? Keep it short and sweet when possible.
Too often, survey questions don’t address what really needs addressing. They lack an understanding of research objectives and fail to target relevant issues. They can also be framed in manners that indicate a bias or that influence respondents in some way. Using overly complex sentences and loaded, leading or absolute language can impede respondent understanding and point them to answers they might not have given otherwise. There’s a difference in asking “How well did this product perform?” and “How do you feel this product performed?” It’s important to know it and design questions that prompt informative answers.
Many times, companies are in need of results quickly and cheaply. They might decide to partner with a company providing standard survey forms. However, using a prefabricated design, even one that can be tweaked, can never provide the integrity of data needed for meaningful analysis. Without robust introspection and careful consideration of unique consumers, markets and trends, a survey can’t be designed in a way that will appeal to respondents and won’t ensure their honesty and time. One size doesn't fit all. Skip logic, intelligent looping and drop-down menus that direct respondents to only those questions that are relevant to them can be ways to eliminate useless data and reduce the time it takes for the survey to be completed.
It’s somewhat of an oxymoron, but companies need specific ways to make standard measurements. But sometimes they just don’t get it right. For instance, if companies distribute a survey using different scaling formats, the data will never compute. There are no standard formats for scaled measurements: sometimes “1” means “excellent” and other times it means “poor;” sometimes surveys utilize five, seven or nine-point scales; sometimes it’s a simple yes or no. Respondents might be confused and inadvertently mark the wrong number. They might be forced to choose an option that doesn’t symbolize their reality. In this way, the data becomes less indicative of their experience and more representative of the survey’s own shortcomings. Companies must ensure that any metrics used are clearly defined, consistent across individual surveys, as well as company subsidiaries, and add value to the data collection.
Designing a survey takes skill and experience. Communications for Research (CFR) has both. Contact us to see how we can help you make the most out of your next market research survey project. You can also download our FREE eBook, "The Insider’s Guide to Successfully Using Market Research Online Surveys" for additional survey tips: