“Good Question!”
October 9, 2025
If you want valid answers, you need to ask good questions. At Epinion, that’s exactly what we do for a living.
It’s easy to ask questions and get representative answers. All it takes is a survey in one hand, a bag of money in the other, and a quick stop at an online panel — and voilà, you’ve got yourself a “what do the Danes think” poll.
But asking good questions and getting accurate answers is harder. That requires solid survey craftsmanship. In a world that’s increasingly data-driven, and where ordinary citizens are becoming more data-critical, the ability to ask the right questions is key to producing valid results and credible measurements that don’t collapse under poor survey design.
At Epinion, we make a living out of asking good questions, and there are several reasons why. Four, to be precise.
- First, we’re methodology nerds. We take professional pride in crafting surveys we can stand behind — surveys where questions are unbiased and objective.
- Second, we care deeply about delivering advice based on accurate insights — on facts. Poorly designed questions lead to misleading results and a weak foundation for sound decision-making.
- Third, results based on bad questions distort public debate on important social and political issues. Discussions risk either being built on a misleading foundation or getting lost in methodological quarrels about how the questions should have been asked.
- Fourth, we care about our respondents. They’re a scarce resource. And with more and more feedback and satisfaction surveys popping up after every purchase, product use, or app experience, competition for their attention is fierce. Respondents don’t want to answer bad questions, and they tire quickly from poor user flow and irrelevant content. That’s why we always keep the respondent in mind when designing surveys.
So how do you do it?
Deciding whether a question or survey is “good” can sometimes feel subjective. But it’s not. At Epinion, we draw on extensive research literature — including language sensitivity and order effects — and combine it with 25 years of hands-on experience designing and developing surveys. Survey craftsmanship is still relevant, because we can’t create meaning out of noisy data — no matter how smart our consultants are, or how advanced the machine learning or AI we use. If the data is noisy from the start, there’s no clarity later.
In survey terms, noisy data means respondents don’t answer what they actually think. There are many reasons for this; the list of biases is long, and context often determines what works. That’s why we’ve formulated a set of guiding principles for how to help respondents give meaningful answers.
Good survey design starts with good question formulation. Questions must be clear, precise, and free of bias. Response scales must match the question naturally and allow respondents to express their opinions without unnecessary constraints.
About to launch a survey and need some advice?
In our daily advisory work, we often see how easy it is to point out what doesn’t work in a survey, but harder to suggest what would work better. That’s why Epinion’s survey design principles don’t just outline the theory — they offer practical solutions to the most common design problems we encounter.
And the good news? We’re happy to share.
Would you like to get access to the same good survey checklist our consultants use every day? Just write to kros@epinionglobal.com with a short note on what you need it for, and we’ll send it to you.
If you need expert advice on your survey, Epinion’s consultants are always ready to help. With us, you’ll get well-founded guidance that’s based on the latest research in survey design — and always objective, unbiased, and methodologically sound.
