When running a market research survey, there are many things to consider. In fact, so many one could write a pretty thick book about it.
These days, time is scarce and we usually have weeks if not days to plan and run a survey, while still needing to make sure we don’t make any obvious mistakes. This is where our list of survey tips comes in handy.
And if you are thinking about running a quantitative survey, especially a conjoint survey – don’t hesitate to get in touch with us – our seasoned survey and pricing experts will guide you through the whole process.
Market research survey planning stage
Planning the survey is like setting the foundation. If you don’t do it right the whole building (survey) will collapse.
1) Choose the right methodology
The first survey tip comes before you start creating a survey, and it includes a choice!
Before you start sending out the surveys, you need to make a very important decision: which methodology will you use? Qualitative or quantitative? Will MaxDiff do the trick, or you need a Conjoint Analysis?
There are many choices available, but if you know what you want to find out, the decision should become pretty obvious!
For example, if you want to know how much importance customers attribute to each feature of your product or service, we suggest using a MaxDiff Survey.
On the other hand, if you already know which features customers value most, and are keen to understand the price they are willing to pay for those features, we recommend conducting a Rapid Conjoint Survey.
2) More than one survey for higher quality
This tip might not be an obvious one, but if you want to have high quality insights, you need to plan for several surveys from the very start.
One survey will likely give you interesting & useful answers, but running more than one will ensure that you have higher-quality data, through analyzing either different market segments or changes happening to the same one.
Just look at how COVID-19 changes things on a weekly if not daily basis. If you took actions based on data from before the pandemic — or from just a month ago, or from earlier this week — how useful would it really be?
3) Clearly define the purpose
If you want your survey respondents to provide quality answers, they need to understand what the research is all about.
This is why you need to make sure to create a clear and thorough explanation of the survey and its goals before you start putting things together. This should include as many details as needed about the product or service and its features, benefits, and specs.
As a bonus, this will help you explain the survey to your team as well!
Market research survey design stage
Survey design is critical for success. If you do it right – you will get high quality data to analyze and make decisions. If you do it wrong – you will get the data to analyze, but the decisions you make might be wrong.
4) Logical ordering of questions
When you start putting together the questions for your market research survey, make sure they follow a natural and logical flow. Always start with the broadest questions and then move on to the ones that go deeper.
This will make it easier for respondents to answer, and their thought process will be beneficial for your survey.
Respondents are more likely to provide quality answers to well-ordered questions than to a chaotic mess, where they will try to click away as soon as possible.
5) Short and focused
When talking about the length of a survey, it should be as short as possible while still providing answers to all the questions you have for a single objective.
If you have more than one objective, then you need more than one survey. And with modern tools and cost-effective platforms like EPIC Conjoint (shameless plug), you can do just that!
Also, if your survey becomes too long, you increase the chance of respondent fatigue and this will in turn lower the quality of their answers, meaning less useful data as a result…
6) Prepare simple questions
A very simple tip – to collect higher quality data you need to ensure survey questions are clear, and without any kind of ambiguity.
Making it easier for respondents to understand what you’re asking will significantly improve the results you will get!
And this will then make sure decisions you take based on survey results are of higher quality…
7) Avoid open-ended questions
Open-ended questions are great to give context to your respondents’ answers, but they are a nightmare to analyze, so you should use them sparingly. Instead, go for the closed-ended questions whenever you need exact data to analyze further.
In many cases, yes/no questions are excellent – respondents don’t have to think much, just make a quick decision about the proposition you are presenting.
If you need them to choose an option from a wider set of possibilities – use multiple choice questions.
And there are rating and ranking scale questions that enable respondents to assign values to individual statements.
Whatever the combination of questions you end up using, remember our previous tip #6 to improve the quality of the survey and increase the chances of capturing relevant data.
8) Use consistent rating scales
How many times have you been asked to respond to a survey and then got frustrated by the rating scale questions? Researchers tend to use them a lot, to quickly collect vast volumes of numerical data that is easy to analyze, but too often they don’t keep things consistent.
Sometimes they reverse the order of values on the rating scale – higher values on the right, then on the left, then back again… Sometimes they keep changing the number of options to choose from – one set of questions has 5 values, the other set 10… And then they change what the higher value represents – sometimes it is a good thing, for other questions it is a bad thing…
All of this leads to respondent frustration and wishing they hadn’t started completing the survey. Often they will either stop it or answer quickly, without too much thinking. Or choose wrong values being misled by what they thought values represent.
So, keep the rating scales consistent throughout the survey. Make sure you always use the same set of values and that values always have the same meaning. That is the only way to ensure data quality!
9) Count (down) the questions
You have surely been in a position to respond to a survey, and 10 minutes in you are not sure how much longer it will last. Yes, it said it would take about 15 minutes, but was it overly optimistic? And suddenly it seems like you’ve been working hard on that survey for hours…
Nobody wants to feel like that when answering market research survey questions, so don’t make your respondents do so. Besides stating transparently how much time it should take them to respond to a survey, count (down) the questions. It will help them feel in control and satisfied that they will finish the task soon.
In turn, they will patiently go through every question and give high quality answers – just what you need to ensure you’re making the right decisions.
10) Single question at a time
This tip should be an obvious one, but too often we find ourselves reading through the question several times only to get more confused by what we should provide as an answer.
To make sure your survey respondents are not confused – prepare questions that are as simple and clear as possible, so there is nothing left open for interpretation.
11) Ask challenging questions
To uncover truly meaningful insights, you shouldn’t only ask the usual, comfortable questions. Dare to challenge the respondents to go deeper into their thoughts and even feelings about brands and products.
This is the part of the survey that might not be quantifiable but will elevate the quality of your customer insights as a whole. And you are surely to be surprised by what you end up reading about you and/or your competitors.
12) Avoid data overload
Quantitative studies are great at generating data. This is one of their greatest benefits and biggest obstacles at the same time. Too much data can lead to too much analysis and difficulties in getting to relevant insights quickly.
This is why collecting data “just in case” should be avoided. Focus only on those data points you will actually need in your analysis and make sure to choose a research partner and survey tools that can help you cut through the data noise.
To help our customers analyze the data collected during conjoint surveys, we’ve built tools like Interactive Price Optimizer and Market Simulator with Business Case Parameters. They allow you to quickly see how changing different parameters influences your bottom line. To find out more, read our February updates announcement.
13) Mirror questions for improved quality
We’re sure that you will not base all your decisions on just one survey. That is only the starting point.
As your sales & marketing strategies are implemented, you will need to check them against fluid customer preferences and ensure you are on the right track.
To increase chances of success, make sure to ask the most important questions across several surveys. Using mirror questions will give you additional confidence for the decisions you are making…
Market research survey distribution stage
Distribution is all about delivering your survey to the right people with laser precision at the right time.
14) Pre–test the survey
Before you start sending the market research survey to respondents, make sure it works as intended.
Like with many other things, test your survey thoroughly and don’t just rely on a couple of tests done by yourself!
Include as many colleagues as you can, so you don’t only test if the survey will run successfully, but also if data is collected and presented as expected.
This will help you iron out any bugs and prevent you from having to run repeat surveys, which can be costly and can lead to lower quality results!
15) Precisely target your audience
The quality of your survey depends on the responses you get. Yes, you can blast it out to the world and hope for the best, but the results will most likely be inconclusive and place you back at the drawing board.
To get valuable insights, you need to ensure that respondents come from the target group for which you have designed your product, service, or offer. This includes not only reaching out to people you think are part of that group but making sure they are part of the target group by asking them screening questions!
To ensure not only speed but also panel quality – we recommend working with panel providers. Our tool seamlessly integrates with the leading global panel provider CINT, enabling clients to conduct surveys quickly and seamlessly in 130+ international markets in a click of a button!
16) Ensure statistical significance
To ensure your survey provides actionable customer insights, it is not enough to send it to the right audience. You also need to collect a sufficient number of qualified responses to ensure the survey results will lead to quality decisions.
This is why our Rapid Conjoint Analysis platform tells you, before distributing the survey to the target audience, how many responses you need to ensure statistical significance.
17) Offer an incentive
If your respondent database is too small to be sure you will get enough responses, offer an incentive. It can be a shopping voucher, product discount, or something similar – you just need to make sure it is something meaningful for your target audience.
Data shows this can lead up to a 50% higher response rate, which means you can get enough results even with a smaller respondent database.
18) Choose your timing
To ensure your respondents pay attention to your survey and participate in it, choose carefully when you send it.
Research shows that Monday, Tuesday, and Thursday are the best days. But this, as well as the time of day, will depend on your target audience and their habits. Don’t take our word for it, HubSpot aggregated the data from several studies.
We hope these tips will help you successfully plan, design and run your market research survey with confidence.
If you are looking for the easiest way to run an accurate conjoint analysis – get in touch with our team and let us show you how EPIC Conjoint enables you to gather decision-making results within 3 business days!
Want to see EPIC Conjoint in action and find out how it can help you uncover customer insights, make pricing decisions and drive growth?