Editor’s note: Juliet Bakhmut is senior content strategist, CoolTool, a San Francisco-based research platform. This is an edited version of a post that originally appeared under the title, “How to run an implicit test – simple tips to get great results.”

Recently, one of our new clients – a small company with a modest research budget – told us that in order to save money they make the same respondents participate in one implicit test (IPT) two or more times. In case you hadn't realized – this is wrong. Don’t do that.  

Imagine that you are in the bar and you are approached by a someone who you don’t really want to talk to. They ask, “Can I buy you a drink?” You reply with, “No” (or “Yes,” it doesn’t really matter). After that, they turn around and ask you the same question again, again and again. How would you feel? Embarrassed and frustrated to say the least. The same happens when you approach a person with your product and ask, for example, “Do you like the packaging?” The person replies to you but you carry on and ask them the same question again. Seems a little bit weird, doesn't it? So does the repeated participation in IPTs. 

Implicit tests are effective and provide comprehensive data about customers only if several simple rules and tips are observed. The first one you already know: Don’t ask the same respondents to participate in the same survey several times. 

Let’s see what the other rules are:  

Consider the representative sample size: The minimal representative sample of respondents in the implicit priming test should not be less than 70-100 individuals depending on the scope and aim of the research. The larger the sampling, the more credible information on the diversity of customers’ attitudes toward products (and other tested materials) you will receive. 

Don’t disturb your respondents: It’s crucial to let respondents fully concentrate on the test, as any disturbance can influence their emotions and feelings and thus compromise the quality of the data you will collect about the tested object. Also, don’t forget that within implicit priming tests you are capturing the unconscious reactions of respondents. That’s why when conducting an IPT, make sure that questions you ask are clear and simple, so respondents can answer quickly without thinking too much. Over-thinking because the question is too complex will not give you clear IPT results.  

Tested objects should be familiar to respondents: Respondents should have at least a basic concept of what you are testing in order to swiftly form an attitude toward it. For instance, if you are planning to test a new name for the brand, make sure that it doesn’t contain any unfamiliar or invented words. Unknown or complicated concepts are hard to grasp for respondents and therefore aren’t easily associated with objects or products – something you want to avoid while doing IPT. 

These simple tips will help you avoid common mistakes in selecting and managing respondents’ groups.