As professional market researchers, we know that DIY research is sometimes the best solution - there are times you don’t have the budget or time to engage with professionals. 

Rather than leave you high and dry, our team is here to help you do the best DIY research possible. We’ve seen plenty of common mistakes that, if corrected, would greatly increase the chances of useful DIY research results.

Mistake #1: Improperly Defining your audience

When conducting online quantitative research, you have access to all sorts of people - people that could fit almost any profile! This in and of itself is a wonderful thing, but not everyone on the internet is who you are looking to complete your survey.

From the get-go, you need to have a firm definition of your audience. This will depend on what research you are doing and what are the objectives of your research.

BEST PRACTICE: Define all critical demographics / psychographics / behaviors that make up your audience

Do they need to meet certain demographic criteria?

• Are there specific behaviors or attitudes they need to have to be a useful respondent to you?

• Are there quotas you need to consider?

DEVELOP YOUR SCREENER

Once you have your definition, you need to make sure the screener of your questionnaire is properly set up so you can efficiently narrow down the vast sea of respondents appropriately to reach exactly who you need.

MISTAKE #2: DRAFTING QUeSTIONS WITH POOR WORDING, BIASES, oR FLOW

By this point you have at least a general idea of what types of questions you want to ask of your audience.

As you begin crafting your questionnaire you need to be mindful of the wording and options you present to respondents.

It’s important to reduce biased wording as much as possible and provide comprehensive answer options. It is also important to think what type of question is more appropriate - single select? Multiple select? Ranking?

Think of the question order. Does a question you asked previously bias the answers respondents may provide to subsequent questions?

Thinking about all these elements ahead of time will result in easier execution, a more accurate picture of respondent sentiments, and increases the validity of your data.

BEST PRACTICE: Develop clear, unbiased questions

MISTAKE #3: Creating questions incompatible with survey software

While creating your questionnaire, it is important to be familiar with the survey platform you’re planning to use (as well as its limitations).

Too often we have clients drafting questions in a way that can’t be executed by the software they have access to. 

Especially if you’re using a basic level or free version of a platform, there are even more limits to the types of questions and logic that can be programmed. 

Knowing these limitations ahead of time will save hours of rework when it comes time to programing.

BEST PRACTICE: Align questions with your survey platform

MISTAKE #4: Accepting poor quality responses

The time has come and you’re fielding your DIY study. The data is coming in but things seem off.

Most often this is because you haven’t cleaned the data properly. There are automatic measures to help improve data quality like IP duplication checks or eliminating speeders but these don’t catch every poor response.

With the rise of generative-AI, we’ve seen more and more seemingly human responses that, upon further investigation, are actually bots. 

We recommend adding in open ended questions that require a respondent to type a response to a question. Reviewing these answers manually is an easy way to sift out poor responses from both humans and bots alike.

A good rule of thumb is that you can expect to remove upwards of 25% of survey responses due to poor quality data - so be sure to overfield so you aren’t left with fewer of your target audience.

BEST PRACTICE: Implement various data quality checks

MISTAKE #5: LIMITED ANALYSIS FOR COMPREHENSIVE INSIGHTS

The questionnaire has been written, survey programmed, responses collected, data cleaned and now it’s finally time to analyze results!

When doing DIY research you may be limited in the outputs provided by the survey platform but this shouldn’t stop you from digging into your data fully.

Often DIYers limit themselves by only looking at their data among total respondents.

While this is obviously a useful lens, it shouldn’t be the only one used. Looking at data among subgroups often paints a fuller, more detailed picture of the study results.

BEST PRACTICE: Conduct detailed subgroup analysis

Avoiding these mistakes should make the most of your DIY research endeavors but if all this seems like too much, consider reaching out to our team at Lab42.

All our research is fully customizable, and we are often able to work with quick timelines and more limited budgets. Our team of experienced researchers is happy to take the load off you and your team if you’d rather hand off your DIY and leave it to the pros.