In marketing, the key question is how to measure the effectiveness of your campaigns. But how can you get your attribution model to tell a story? By asking this question you are already thinking far ahead of many digital marketers. Most of the digital advertisers use the last click attribution model to evaluate the success of campaigns. In other words, during the customers’ journey and among different touch points or ads which impressed the customers, only the very last event is considered as significant while the others are ignored. This is simple and easily implemented but, in most of the business cases it doesn’t help in making smart marketing decisions.
Use survival analysis for making an actionable attribution model
Survival analysis is a popular approach among biostatisticians for analyzing the expected duration of time until one or more events happen, such as death in biological organisms. However, this statistical technique attracted scientists and technicians in other domains. Survival analysis is called reliability theory or reliability analysis in engineering, duration analysis in economics, and event history analysis in sociology. Two example questions that survival analysis attempts to answer are listed as follows:
In survival analysis, death or failure is considered an 'event'; traditionally only a single event occurs for each subject, after which the organism or mechanism is dead or broken. Chandler-Pepelnjak (2010) proposed to use survival analysis for marketing attribution analysis. He suggested that each event in a customer's’ journey gives an indication of whether the journey is still alive. And when the customer converts to purchase, the journey then 'dies'.
If you are into marketing do you want to know how long your customers are likely to stay with you; or whether you customers fall into a certain demographic profile, or whether those who enter your website via a certain channel or campaign tend to convert more quickly?
We found this survival analysis technique extremely valuable for attribution modelling. Using survival analysis, we measure time of conversion and probability of conversion based on the channels that customers used on their journey. The probability of being alive (not converting) and being dead (conversion) is used for the following example:
What is the visitors’ probability to convert after 50 days?
We used survival analysis technique to answer this question. Our target population is our visitors. Our treatment is our marketing channels. Events are conversion and time to death is equal to the time of conversion.
Attribution can be a problem or a solution, depending on how you use it. It can be a solution if your attribution tells you a story. The story says how your customers fell in love with your brand and how you broke up with them and how after all that time, they found you again and what made them committed to your brand. We found the best way to tell this story is using survival analysis. By using survival analysis you’ll have a better understanding about your customers' conversion trend. Probably what you are more interested in is understanding and analyzing the variables and channels that convince your customers to convert. Our approach does the trick for you and also calculates the statistical significance of each variable. In the next blog posts, we will discuss how survival analysis helps us in campaign optimization and attribution modelling. We state how our approach explains the role of each channel in the love story between you and your customers, and how you should use that information to make a smart decision about your advertising investment.
The approach above not only works for ECommerce website but for content based as well. Reach to us about our Enhanced Ecommerce for Content approach which provides you with deeper insights into how really your content is consumed. The Government Analytics solution helps federal and state government understand the deliverability of the content and know how website visitors are using it.