I love navigation apps. Especially the way Siri narrates directions with authority. Sure, she has a specified route out of my neighborhood that makes very little sense, but out on the open road, she warns me with exuberance about my turn in 300 feet.

The great thing about navigation apps is they provide us with a map from our starting point and directions to the destination of our choosing. From there, it is up to us to remain on course.

Market research survey design is similar. Without clear directions for how to approach unfamiliar territory, those new to primary research may not feel comfortable taking the driver’s seat.

Our Research Quality GPS ensures you, our delivery team, and the survey participants all have a smooth journey, collect the highest quality data possible, and avoid any bumps in the road. In our Navigating High-Quality Data series, we will touch on three key components to conducting high-quality research, starting with this blog on design.

Survey design is the engine that drives research quality.

In research, the survey is essentially the engine for quantitative research. A well-designed survey engages respondents, functions well on all types of devices, and fulfills research objectives, thereby ensuring all arrive at their final destination – high-quality data.

To drive a well-designed survey, you need to:

• Determine the research objectives and target participant.
• Develop a survey that is clear, concise, and intuitive.
• Design with all device types in mind.

Determine the research objectives and target participant.

Online surveys are a great vehicle for collecting information and are utilized to solve a myriad of business problems. The more precisely business challenges are outlined and then translated into research objectives, the more accurate and actionable the insights produced from the research.

In most cases, multiple research objectives will evolve from broadly conceptualized questions if a strong focus is given to the specific information needed to answer the business decision in question. Well-defined research objectives are critical to achieving the desired outcome from the research as they dictate the types of questions participants are asked and also the target market for the research.

For example, ‘will our customer base look different in 3-5 years?’ can be converted into specific and measurable research objectives like these:

• What is the short v. long term commitment to the brand?
• What is the impact of new technology on customer usage?
• Are heavy users more loyal than light users?

Once you have clearly defined the objectives of the research, you must then determine the target audience. For example, at times it may be important survey the general population to get a broad read on the consumer. Other times a product is regionally based and we will only gain intelligent insight from those that live in the northeast. In business (B2B) surveys, in particular, it can be crucial to the research to define the ideal person to interview since knowledge can be specific to the title or industry of the participant.

Develop a survey that is clear, concise, and intuitive.

Regardless of the audience, minimizing survey length and non-essential content reduces participant fatigue, which can negatively impact data quality. Essentially, save participant brainpower for the truly important elements: thoughtfully answering your survey questions – not deciphering the survey itself.

Does version 5 of your survey have twice as many questions as version 1? Are they all essential, or were some added as ‘nice to haves’ that stayed along for the ride? Survey design is an iterative process and it is important to refer to the research objectives often to validate that survey questions will net you the necessary data to solve the business problem. Cut questions that do not bring you closer to this goal.

• Use simple sentence structure (avoid acronyms and slang)
• Ensure respondent instructions are clear (i.e., “Please select all that apply.”)
• Use transition statements to provide breaks that allow a respondent to refocus on the next section of questions (i.e., “Now, thinking about the brand of toothpaste you use most often…”)
• Use the same or similar scales throughout the survey (i.e., use 1-5 scales, where 1 is consistently the low end of the scale)
  • Avoid respondent fatigue with limited use of text or long sections

Finally, review the questionnaire as if you were taking the survey yourself. Does it flow well or have you veered off the path? Do all that you can to optimize the user experience as it will help to improve data quality. Thoughtful questionnaire design helps to eliminate the possibility of either under or over cleaning the data. No quality measure or data cleaning is as effective as screening questions that qualify the right participants and questionnaire design that provides robust intelligence.

Design with all device types in mind.

In today’s mobile-heavy culture, a growing number of people are using a smartphone or tablet as their primary means of going online. Designing a survey that ensures all participants have an engaging experience is critical for achieving a truly representative view – and for reaching survey participants who may opt to only access surveys on a mobile device.

Remember that a survey participant on his or her mobile device is not only working with smaller screen space, but is also much more likely to be on the go or multitasking. Through experience, we’ve established some basic best practices on Why Mobile Matters for Conducting Representative Research to consider as you design your survey for mobile users.

Just as navigational apps require the occasional update to include newly paved roads, updates are required in our approach to quantitative research to ensure we are embracing new interfaces and technology. Although it’s tempting to stick to well-worn roads, we can pave new roads to more efficiently arrive at our destination – high-quality data.

Stay tuned for part 2 in our blog series for a deeper look at the importance of survey screening.

This blog is the first installment of a 4-part blog series by Research Now. To check out parts 2, 3 and 4, click the hyperlinks.