Attribution Modeling and Multi-Channel Marketing
In this webinar, participants will be introduced to Attribution Modeling, including Markov, Hidden Markov, and Survival Modeling. All of these methods attempt to quantify the importance of each touch-point in the customer’s journey towards purchase. Data requirements in today's multi-channel environment will also be discussed.
The term “attribution modeling” is commonly used with two different meanings. On the one hand, the term is often used to refer to how much unit change in sales is attributable to a given unit change in a type of media activity. For example, an increase of 10,000 unit sales is attributable to one Gross Rating Point of TV advertising. The first type of attribution modeling is very much the same thing as traditional Media Mix Modeling
On the other hand, the term “attribution modeling” refers to a technique used to measure the relative impact of each touch point in a customer’s journey in generating the final sale. For example, buyers may explore various websites to decide whether to buy a book, ultimately buying the book through Amazon. Some of the credit for the final Amazon sale is attributable to exposure to the influential content on websites visited earlier in the customer’s journey. The second type of “attribution modeling delivers a unique insight: The relative degree to which final sales is attributable to each touch point in a typical customer’s online journey towards purchase.
Customers today use many channels for shopping and buying products or services. For example, one might search for product information on a cell phone while in a brick-and-mortar store, visit websites of multiple brands, hear a advertisement on the radio, see signage in the store, discuss products on social media, and so on. Attribution modeling most often focuses on only the digital portion of the purchase journey, as depicted in the simple diagram where 60% of shoppers start with website A and 40% with website B. From each website, a shopper can branch off in multiple directions. For example, 50% move from website A to website B and 50% move from website A to website D. As seen in the diagram, 100% of those on website B move to website C, where 50% are converted into a purchase. Conversion may be defined in many ways, such as registering as a member or purchasing a product.
Expanding our diagram to non-digital channels, the consumer might visit a website but then branch off into social media, check email for promotions, drive past a billboard to receive brand reinforcement, stop at a brick-and-mortar location to handle the product, receive a promotion in a mobile app, and receive a promotional call on their phone. In other words, the digital shopping journey is only a portion of the actual journey, and multi-channel attribution modeling should include non-digital channels to be completely accurate in measuring attribution.
Presenters
Elizabeth Horn
Senior VP, Advanced Analytics
Beth has provided expertise and high-end analytics for Decision Analyst for over 25 years. She is responsible for design, analyses, and insights derived from discrete choice models; MaxDiff analysis; volumetric forecasting; predictive modeling; GIS analysis; and market segmentation. She regularly consults with clients regarding best practices in research methodology. Beth earned a Ph.D. and a Master of Science in Experimental Psychology with emphasis on psychological principles, research methods, and statistics from Texas Christian University in Fort Worth, TX.
John Colias, Ph.D.
Senior VP Research & Development
As a leader with both university teaching and business consulting experience, John focuses on predictive modeling, prescriptive analytics, and artificial intelligence. As Senior Vice President, Research & Development, at Decision Analyst, John combines academic and business interests to help analytics professionals by offering cutting-edge analytic solutions tempered by business realism. He holds a doctorate in economics from The University of Texas at Austin, with specializations in econometrics and mathematical modeling methods.