Click to learn more about author Dave Kelly.
Human beings are imperfectly dynamic creatures shaped by experiences, biases, and cognitive processes, and this combination of behavior-shaping factors is unique from individual to individual and can even change from day to day. In the always-on digital world we live in, where competition for consumer attention is fiercer than ever while actual attention spans shrink, marketers looking to personalize communications and data scientists hoping to build better models may find it nearly impossible to do so when relying on traditional data like simple demographics while ignoring – or unable to quantify – internal motivations.
How can organizations gain insight into the hearts and minds of individuals in order to understand the why driving their decisions? After all, we’re so much more than age and gender, right?
Lucky for marketers and data scientists hoping to learn who individuals are, what they do, and why they do it, there is an entire scientific discipline dedicated to understanding the human mind and the way it works known as cognitive sciences. And when attempting to go beyond known demographics and firmographics, pairing best practices in cognitive science with Data Science and analytics is a proven way to gain insight into the “unknown” and vastly improve results.
Taking a Scientific Approach to Surveys
When marketers and data scientists find themselves in need of data that is not readily available from traditional sources or simply does not exist, they often turn to surveys. In theory, this sounds like a great approach – just ask your subjects directly for their response to a few questions. However, as the imperfectly dynamic creatures that we are, self-reported survey results are not always trustworthy.
This is where cognitive science can make all the difference, as not all surveys are created equal.
Frequently when responding to a survey, people are not the best barometers of their innermost motivations and characteristics. There is also no guarantee that every respondent interprets and answers a particular question as intended. For example, survey respondents cannot be asked bluntly if they believe they possess a certain trait. Every respondent will answer the question based on a slightly different interpretation of the trait, and whether they perceive it as positive or negative.
Taking an expert, psychological approach to capturing and analyzing this data can be a game changer. Cognitive psychologists are armed with the skills to overcome these initial hurdles and produce dependable survey results that can fuel strong predictive models.
Instead of asking respondents if they possess a certain trait, a better survey may ask if respondents have previously engaged in explicit behaviors that speak to the factor being explored in multiple different ways. Having a collection of responses that can only be answered in a definitive manner gives data analysts a much more reliable and predictive dataset to model for marketing analytics.
But because the type of thinking that comes from years of studying cognitive processes is the foundation of reliable survey-based research, it cannot be abandoned after survey construction.
Validation Is a Must
No matter how many measures are taken to remove the bias and corruption from survey responses through cognitively sound design, some respondents will simply not provide quality information.
This leads to incorrect, inconsistent, or missing data, which is a big problem for marketers and data scientists. Their job relies on sound data so they can “crunch the numbers” and reach their desired output. Any marketing data scientist will tell you that clean data is the key to a powerful, predictive model.
So how does cognitive science assist with this critical data cleansing?
By purposefully designing surveys based on a sound psychological foundation, researchers can validate accuracy through data comparisons. For example, if a respondent answers a question in one manner and then answers a similar question asked through the same cognitive lens with conflicting feedback, this is a sign their responses may be inaccurate. On the other hand, if every response related to the cognitive motivation being examined provided by an individual is consistent, the information is most likely valid. Think of this process as the “cognitive validation” that takes place before the “analytical validation,” which is done by distributing the same survey to another random survey panel.
Know Your Customers Like You Know Your Friends
Knowledge and understanding of the inner workings of the mind not only greatly enhances the quality of the survey and increases the chances of an accurate response but can also help in ensuring the final dataset is clean, valid, and reliable. By layering cognitive science on top of marketing data creation and Data Science, organizations can deepen the relationship with their customers and gain a better understanding of their behaviors and motivations.
Credit: Source link