Discrimination is - in a market research context - recognizing one thing from another. This is exactly what you do when defining your target audience and the basis for your screening of respondents in a survey: you want to define whether it’s purely based on socio-demographic background data or behavioral or perception based as well. Definitions should be sharp, yet ensure volume in your potential customer segment to be commercially viable.
Screening your target audience precisely, and defining which problem you want to solve is not only of high value, it is crucial if you want to position your brand well and communicate effectively.
First step is, thus, to define your target audience well; if you are precise you will have a much easier time taking action on your results. Focus can be on understanding a broader prospect base in a market, current clients or people with certain characteristics, (purchase) behaviour or perceptions. Maybe you have a persona created that you imagine seeing or being when receiving the story of your data.
Regardless of the purpose, a good screening takes the minimum viable approach to not tire respondents. We want to remember that screening has to have some kind of business case and hence a volume in the general population that can justify your company's strategic focus.
Screening questions are often a major source of frustration but also a necessity for high quality in a piece of market research. Hence, it’s relevant to understand the purpose and considerations involved in high quality screening. Often the optimal respondent screening is a combination of background data excluding people with certain socio-demographic characteristics and a minimum viable set of purchase and/or behavioral requirements to be able to create a funnel throughout the questionnaire.
In theory there’s no limit to the number or type of screening questions you can pose but three things are important: First, you want to make them as orthogonal (uncorrelated) as possible. Second, you want to keep it short to not tire respondents. Finally, you want to keep the level of complexity low for the respondents.
When asking screening questions, it’s important what is screened out and what is registered and categorized. Unfortunately, I often see a tendency to screen a bit harder than needed. Often because they are defined from ideas rather than a fact-based understanding of what the actual target audience is for a brand, product or service. Thus brands risk losing insights from parts of their potential customer segments.
To minimize this, you should think of potential; challenge your beliefs (especially if they are not fact-based) when defining your screening criteria. That will minimize your risk of targeting a too limited audience for the evaluation of your products and/or services and probably improve your target audience definitions.
In short, when screening your target audience in market research, here are my recommendations: