One of the most common challenges faced by our clients is writing good survey questions. After all, if you aren’t asking the right questions, you aren’t going to get the data you need.
At aytm, we pride ourselves on having transformed the way consumer insights are gathered. We created a platform where you can design surveys that seamlessly get you the information you're looking for, saving time, and hassle for both your team and your panelists. But, our platform can only take you so far. It starts with designing an effective survey to engage your target audience.
Let’s take a look at the top five things you should consider to make sure you’re sharing solid survey questions that will yield invaluable consumer insights.
1. Include screening questions
Screener, or pre-qualification questions, take your survey to the next level when it comes to reaching your target market by helping you connect with a qualified audience at the beginning of your survey.
Perhaps you ask, "Which of the following have you purchased in the past six months?" along with a list of options to choose from. If you only want people who have recently purchased a smart TV to take your survey, this pre-qual question would only let those who respond accordingly through. Screening questions not only respect the survey taker’s time but also ensure your final results only include your target demographic.
However, try to avoid “yes/no” screening questions. People tend to be biased towards choosing positive responses when presented with yes/no questions, even when their genuine opinion leans more toward a neutral response. To combat bias, offer a list of possible answer choices with no indication that one answer is preferred, like above.
2. Keep questions unbiased & unloaded
You want to avoid leading or biased survey questions that suggest which answer the respondent should select. For example, "What do you like most about this logo?" This is a leading question because the respondent may not like anything about the logo, so a list of attributes may not accurately reflect their honest opinion. This question could be improved by allowing the respondent to express that they don't like the logo.
Biased questions have no objectivity and lead respondents to answer your question in a certain way based solely on the way you asked it, making your data unreliable. Keep your survey questions clear and concise, and avoid leading them to your preferred answer.
Likewise, loaded questions force survey respondents to choose an answer that doesn't reflect their opinion, thereby making your results misleading. For example, "How often do you feel stressed when watching the news?" is a loaded question because the respondent might not watch the news at all, or the news may not make them feel stressed. But by including answer options along the lines of "I don't watch the news" and "watching the news doesn't make me feel stressed," the question becomes more objective.
3. Mix things up
Try to offer a variety of question types in your survey. But also understand that the kinds of questions you choose will affect the answers you receive and the types of analysis you can do.
Quantitative data is great for understanding a large quantity of people at a given time, but never rule out the value of qualitative or open-ended questions. These questions give customers a chance to respond with their own words and descriptions. There may be a new perspective on your product that you haven't considered yet. Open-ended questions can also reduce the risk of survey fatigue, prompting respondents to really think about their answers.
When designing a survey with ranking, matrix, and multiple-choice questions, be sure to shuffle your answer choices. We're naturally inclined to select the first option we're presented with in a series of answer options, so shuffling the order of the answer choices can go a long way toward reducing bias in your panelists' responses. But for timeline questions, when the answers relate to one another, you'll want to keep them in a logical order to avoid confusion.
4. Take environmental factors into account
Since the outbreak of the COVID-19, aytm has been tracking how consumer behaviors, sentiment, priorities, and attitudes have been changing in direct correlation to the pandemic. Our weekly reports have given our clients valuable insight into the lives of their current and potential clients.
A major finding in week 13 was that a majority of Americans — whether by choice or need — continue to cut discretionary spending and alter their consumption habits in other ways.
Prior to the pandemic, you may have had a target audience that was highly interested in international vacation experiences. However, that same audience may now be unable to engage in those opportunities due to circumstances outside of their control.
If you want to write good survey questions, this type of environmental consideration is key data to consider ahead of time.
5. Write questions to solve problems
You want your product or service to solve a problem, meet a need, or fulfill a desire for your customer. So, you need to investigate that area of their lives.
Is your product a caffeinated beverage that can fuel them through a busy day or working a night shift? Ask questions that probe sleep and work habits. Do you offer business solutions to help small business owners navigate uncertain times? Ask what challenges business owners are facing right now. You get the idea. Make sure you know what customer problem you’re hoping to solve before you get started writing your survey questions.
Good survey questions are essential to the success of your consumer insights study, but they are just one factor among many others when it comes to high-quality data collection, research, and implementation.