Data Science for Market Research


What people think of your product or company is secondary to how they feel.

While the two are obviously intertwined, people’s feelings will drive them to make decisions that are advantageous for your company.  We have a full slate of big data tools to understand how people think about a product, but most importantly, we know how to understand the deep feelings a consumer can have.  As evidence, in the past we have used sentiment analysis to examine what will predict purchase behavior.  In any market research survey, it all comes down to who is going to buy a product.  Asking people about their behavior around the product, however, is somewhat suspect because people can often say they do one thing but in reality do another.  I may think I am likely to buy a sports car in my life time, but that moment of decision-making where I actually buy a sports car is a long way from what I might report on a survey.  But we can look at other behaviors, we can look at how people talk about a product and use that to predict their actual brand-based behavior.  In this case, we were able to use short statements about a variety of products and use them to predict their overall brand behavior, including past purchases, at a greater than chance rate.  We are able to do so because we understand that feelings, thinking, and decision-making are intertwined and that emotion and feeling play a large role in the process.  We often want to resort to “rational” logic when we think we will predict how someone will make a decision.  Subtle cues in how people feel about a product are likely to be more predictive of purchase behavior than a pros-and-cons decision making process.  We look for these subtle cues, leverage them, and pass them on as actionable insight into how people are going to decide to choose a given product.

Leave a Reply

Your email address will not be published. Required fields are marked *