Shopper Research Blog

My Path To Purchase

How would one develop a branding model using text data and deep learning methods?

Trying to specify consumers’ purchase paths (the behaviors and inputs that impact shoppers’ decisions) as a marketing researcher is often very challenging. The reason being that, in most cases, purchase decisions are unique to each individual and are composed of numerous decisions that take place subconsciously. These subconscious decisions are also influenced by outside factors, ingrained habits, and biased perspectives that consumers won’t admit to. So why do we try? We try because it’s important for brands to understand the process, even if it’s an incomplete picture.

My colleague, Clay Dethloff, wrote a blog not too long ago about understanding in-store behaviors in an omnichannel world. It’s a great blog and while I won’t revisit the ground he covered in it, it’s certainly worth repeating: today’s shopper lives in an omnichannel world, and mobile plays a huge part of that omnichannel experience. Often, initial qualitative research can be conducted to help identify many areas that additional quantitative research can target. That’s what I would like to focus on. I’m more of a quantitative researcher, after all.

For quantitative path-to-purchase research to be effective, it needs to be targeted. The objectives must be specific, and the survey instrument must be focused on answering those objectives. If my objective is to understand the purchase path for buying a home, for instance, that’s likely too broad of an objective to answer in a single quantitative survey. A home purchase is generally comprised of a number of complicated steps that would be better uncovered by breaking them down into manageable components, so be thoughtful about setting your research goals and make them specific to an outcome that the survey instrument can achieve.

A more specific objective might be to understand the shopping behavior associated with purchasing a major appliance. This objective can be achieved with an online survey about shopping behaviors, and includes an understanding of what product research was done and identifies the steps that were taken that led to the product purchase. The main point is that it can be accomplished in the context of a single survey.

Quantitative survey methodologies for path-to-purchase research can be simple or complex, as dictated by the category, needs, and budget. Complex path-to-purchase studies may require attribution modeling, as described in this brief video by my colleague, Dr. John Colias, where the contribution values of certain interactions are linked to the final purchase outcome.

But in many cases, the methodology doesn’t need to be that complicated. I often find that the final output will revolve around profiling the “steps” involved in making a purchase and determining the likelihood that general consumers will take those steps. These steps are often referred to as “decision trees.” Let’s go back to our example of a consumer buying a major appliance at a big-box store. Are they focused more on brand than price? Are delivery options of greater concern than warranties or online ratings are? As another example, what about consumers who buy spray cleaning products? Are they prioritizing brand, price, claims, functionality, or size? Is shopping convenience more important than the actual product? What impact does a consumers’ online product research play in the purchase path? Knowing the answers to these questions helps retailers, manufacturers, and service providers tailor their products to better serve consumers and, in the end, help themselves.

My wife and I recently purchased a new washer and dryer for our home. Even though we participated equally in the purchase decision, our individual decision trees were different. There was overlap, certainly, but the things she emphasized over the purchase journey were different than mine in key ways. She did a great deal more online research and read online reviews, while I visited physical stores to touch and feel the products. As we got closer to the purchase decision, I prioritized brand over price and features while she prioritized price and features above brand. This example illustrates the importance of understanding what compromises take place along the purchase path when there are multiple decision-makers in the household.

Quantitative path-to-purchase research, like shoppers’ purchase paths, are not one-size-fits-all studies. Methodologies and instruments should be customized to fit objectives and budgets. The good news is there are lots of options to choose from. How you decide to pursue your objective is your own unique research path to purchase!

Tom Allen

Senior Vice President

Tom has over 20 years of research experience spanning several industries, including retail, restaurants, consumer packaged goods, and financial services. He earned a Bachelor of Business Administration from The University of Texas at Austin and a Master of Business Administration in Marketing from The University of Texas at Arlington.

• 
•