In online marketing customers may view advertisements and visit a company website many times before purchasing an item (referred to as a conversion). Publishers that published the company's advertisement are, after the conversion, eligible to receive a fee. This is the main income source of many websites. Currently, the following simple payment model is used most of the time: “last click counts”, which awards 100% of the fee to the last advertisement which led to the website visit that resulted in the conversion. However, many advertisements, online and offline, advertising a product or just creating brand awareness, may step-by-step lead the customer to the conversion. How to attribute the right value of each of the steps of the funnel?
My PhD project focuses on developing conversion attribution algorithms. An accurate attribution algorithm would allow marketers to identify the most efficient advertising channels for their product thus optimizing their overall marketing spending. It also allows online advertisement delivery services to deliver the right advertisement to the right person at the right time.