What Makes a Great Survey? 10 Ways to ACE Survey Creation

A survey is just a conversation; you ask the questions, the respondent answers. The trick is knowing what questions to ask and who to survey. We’ve pulled our Head of Client Experience away from helping our clients for just long enough to quiz her on the 10 factors that make great survey.

If you’re new to the world of research, creating surveys can seem more daunting than anything else you might encounter on a regular work day (unless you’re an astronaut, matador or skydiving instructor, perhaps).

But you don’t need to worry, a survey is just a formalised conversation between two people; you ask the questions, the respondent answers. You have conversations all day every day, and probably ask dozens of questions in that time, in truth, you’re already a pro. The trick is knowing what questions to ask and who to survey, that’s where we can help.

We’ve pulled our Head of Client Experience, Renata, away from helping our clients achieve their research goals for just long enough to quiz her on the top 10 factors that go into making a survey great.

Renata’s team, ACE (the Attest Centre of Excellence), assist clients in drafting surveys for all types of business needs every day, and have tens of years of experience in the market research industry between them. So there’s no one better placed to help us answer this question:

What makes a great survey?

1. A goal

Any great survey serves its purpose. That’s what makes it great – it does exactly what you need it to, it delivers the answers to your questions. And so, the first thing you need is a plan of the question(s) it needs to answer.

“Ask yourself, what are you trying to find out? While consumer research should be thought of as a conversation between your brand and your consumers, it’s a fixed conversation. It’s not like a face to face interview where the conversation can take twists and turns. With surveys you need actionable data, so you create questions to produce that for you. You can, of course, follow up with respondents, but this should be a positive decision and not a necessity caused by failing to ask all the questions you really needed to the first time around.”

Understanding the goal for your research allows your survey to go forth and do all the hard work, collecting the answers you need.

2. Or at least an idea

Lots of research is conducted retrospectively: to find out if an action worked, why it worked, or how to fix something that’s gone wrong.

But a great survey builds on the hindsight you’ve gathered and then looks forward, testing and iterating ideas before they launch or informing strategy for the future.

“It can be easier to set a goal for backwards-looking research: what worked, why, what lessons should we learn? Creating a survey aiming to gather foresight might be vaguer, but it is a huge bonus to brands, it helps you step ahead of the competition. All you need is a hunch to test.”

3. Two contributors

While the temptation might be to invite lots of your colleagues to contribute, the optimum number of writers is in fact two: you (the person tasked with answering the business question) and one editor to offer objective eyes.

“When you’re writing a book you have one writer and one editor. The more people you introduce to the process, the more confused the tone becomes.”

The editor doesn’t even need to be a colleague. In fact, the further removed from the business goal they are, the more objective their opinion will be. The ACE team are expert editors, with eagle eyes for implicit biases, leading questions and mistakes. If you’re an Attest subscription client, make sure you’re making use of the ACE team’s expertise.

4. A human element

We’ve said it before and we’ll say it again, a survey is a conversation. If you wouldn’t phrase the question to a colleague in the way you’ve drafted it in your survey, then reword it.

Typing questions into an online platform can mean you forget that it’s real people at the other end of the line, answering your questions. They’ll be more likely to understand the question, and answer it truthfully, if you strike up a level of rapport with them.

“With every question you draft, ask yourself if it still makes sense if you were asking a colleague the same question in real life.”

5. A warm up

“If you were interviewing consumers on the street, would anyone stop to talk to you if you jumped in with “What’s your favourite brand of cereal?”? Probably not. It’s just the same with surveys, it takes a few questions to get respondents into the right frame of mind. You could start by asking “Did you have breakfast this morning?” or “What did you have?”, these questions are much easier to answer and help warm the respondent up to the topic of the survey”.

Add in warm-up questions at the start of your survey. This will ensure that by the time the respondent reaches the questions you care most about, they’re ready to give you thoughtful information.

6. A sensible flow

“If your survey was a person, would you think they were scatty?”

If your survey jumps around from one topic to another, with no message cards signposting the way or guiding images to indicate a new topic, it’ll be much more difficult for the respondent to answer.

If they were having that conversation in real life, they’d be left confused, unable to keep up with the pace of change.

Make sure your survey follows a sensible flow from the broadest topics at the start to the narrower discussions in the middle and end, ensuring your respondents understand what’s being asked of them in each question.

7. A broad sample

A sample size of 400 is robust for the UK and any bigger population. But if you want to pay particular attention to one gender, age band or other demographic, then a bigger overall sample size will increase the robustness of the data from each group.

“The broader the sample you use, the better. Unless there are consumers for whom your product is totally irrelevant… but even then there are so-called “out of the box consumers”: husbands who buy female sanitary products for their wives, or grandparents who buy nappies for their grandchildren, for example.”

Keeping your survey sample broad means that you can still narrow down to the consumers that matter most to you, while only surveying the most important consumers will alienate those who fall outside of this definition who might have important insight to contribute.

8. Limited bias

A great survey achieves its business goals by gathering accurate data. In order to be accurate – representing the real opinions of consumers – the person writing the survey must keep track of and correct their own biases in the questions posed.

“There are lots of ways to make your survey as neutral as it can be: always make sure there’s an option for the respondent to disagree, include an equal weighting of positive and negative responses, avoid yes/no questions…”

9. Interesting parts

Another way to dial up the accuracy of your data is to keep your respondents engaged throughout. A bored respondent will provide inaccurate data – which is bad news.

“Pepper your survey with message cards and media (images, audio and video clips) to keep the respondent interested throughout. Thank respondents for their help at the start, let them know how long it should take to complete, and set the pace by telling them when they’re half way through. All of which encourages them to be honest with their answers, just as you’ve been honest with them.”

10. Consideration for the country you’re targeting

“You must understand the colloquialisms and sensitivities in the countries you’re targeting. Without this, your data will suffer because the respondent (at best) doesn’t understand the question or (at worst) is offended by it.”

A great survey makes sense to the respondents taking it, so invest in your translations. But also be wary that some topics may not be considered appropriate in other cultures, so surveys aren’t one size fits all countries. Write each survey for the audience receiving it to make sure they work to answer the overarching questions you have.

The next steps…

It’s possible for even a survey novice to script a great survey. There’s no need for trial and error, especially when the ACE team are at the end of the phone and itching to help. We’re firm believers that everyone should have access to the data they need to make their important business decisions, and a great survey is the perfect vehicle to deliver the answers.

Tick off these 10 elements while you’re scripting your survey and over time you’ll become a veritable pro! Meanwhile, if you’re new to research, or just want an objective pair of eyes give the ACE team a call. Get in contact with us today.


Content Team 

Our in-house marketing team is always scouring the market for the next big thing. This piece has been lovingly crafted by one of our team members. Attest's platform makes gathering consumer data as simple and actionable as possible.

See all articles by Attest