Dr. Kenneth Sanford, U.S. lead analytics architect with data science company Dataiku, believes that the better loyalty marketers understand their customers, the better chance they have of retaining them.
Sanford told Loyalty360 that loyalty marketers can use predictive analytics to minimize non-subscription churn (attrition) and maximize revenue.
Non-subscription churn occurs when customers leave your website or sales funnel. These types of customers may gradually reduce their purchase frequency over time, or they may never buy again. To combat non-subscription churn, companies can deploy predictive analytics to identify potential ‘churners’ and then take steps with short-term marketing campaigns to re-engage with them
Dataiku published a free white paper detailing how organizations can use predictive analytics to combat non-subscription churn titled How To: Address Churn with Predictive Analytics.
“Traditional marketing companies have a pretty good view of churn,” Sanford explained. “It’s sort of everyone else that struggles with non-subscription churn. In essence, it’s a trial and error problem. How long has it been since I saw certain groups of customers?”
When it comes to loyalty marketers’ biggest barriers to pursuing and properly executing predictive analytics to combat non-subscription churn, Sanford doesn’t hesitate; data silos.
“It’s the siloing of data and the difficulty in information integration,” he said.
Measuring churn is a “giant problem,” Sanford noted.
“You have to create metrics on what non-subscription churn is,” he said. “You have to cluster customers together. For example, one-month gap customers or three-month gap customers. Build that through trial and error. Successful companies in the analytics space that align themselves as data analytics first are the ones leading the pack.”
Predictive analytics can benefit loyalty marketers through effective customer segmentation, Sanford said, and tailored and triggered messaging can positively impact the customer experience and future brand advocacy.
“Data science gives us a lot of that,” Sanford said. “You can target email content and text messages.”
Sanford, however, pointed out that companies can’t expect to see immediate results.
“Trial and error is an inherent part of data science,” Sanford said. “Executives expect to see immediate results, but that is unrealistic. The only way to expect results is making a long-term investment having a data-first company with everyone in the company thinking that way. You have to be patient and expect some wins and losses. The better you understand your customers, the more likely you’ll be able to retain them.”
Sanford offered seven proven steps where predictive analytics can combat non-subscription churn:
1. Understand the Expected Time Between Purchases
It’s a good idea to conduct basic descriptive statistical analysis upfront (unsupervised/clustering) to decide which users should even be considered in the churn analysis. For example, if someone used the product or service only one time, are they considered a churner after that? Or is there some minimum threshold after which a user should be considered and included in churn analysis?
How will your specific business define churn? This step is crucial. Defining a churn period that is too long risks creating predictive models with artificially low churn rates, not capturing enough people and defeating the purpose of predictive modeling. But defining a churn period that is too short makes it difficult for marketing teams to evaluate churn prevention campaigns because they ultimately can’t distinguish between organic actions (users or customers who would have come back anyway without intervention) and effective campaigns.
2. Get Your Data
The minimum data required to predict churn is simply some form of customer identification and a date/time of that customer’s first and last interaction. This data, though not incredibly detailed, would allow you to build models to analyze and predict churn at a basic level.
3. Explore and Prepare Data
This step of the process can account for up to 80 percent of the total time spent on the project, so don’t be discouraged as you get your data into a useable format. Take time to ensure you understand what all the different variables in your data mean before moving on to cleaning up different spellings or possibly missing data to ensure everything is homogeneous. Thoroughly exploring and cleaning will save time in subsequent steps, particularly when it comes time for prediction.
4. Enrich Your Data
If you’re working with a more advanced data set than simply customer identification and date/time of last interaction (which is, as mentioned, highly recommended for better prediction), this is the time to enrich that data and join it to get down to the essentials. For example, if you have one data set with customer identification and date/time of last interaction and another with customer identification and demographic information, you’ll want to join these into one set of data.
5. Unleash a Machine Learning Algorithm
When building a predictive model, you need to make sure that it will actually learn what you want. For instance, one of the common pitfalls for a churn modeling project is to train your model on both past and future events. To avoid this common mistake, you need to put yourself in the position you’ll be in when your model will be deployed into production: What data will be available to you? When would you like your prediction to be: for next week, next month?
Visualization is an important step in the process because it allows a way for end users to consume the data quickly and easily.
7. Iterate and Deploy
This is where the interplay between data science and business is strongest. Work together to determine if the model is effective. Ensure models are sufficiently generic, which means using training, validation, and testing sets that are not specific to a certain time period or to a certain type of customer. For example, you would not want to train or test based on a data set from a time period where there was perhaps a pricing change or some other factor that caused churn rates to be different than usual.
Once you have a good churn prediction model in place, the job is only half complete. The final (and perhaps most important) step is to take actions based on your predictions. Many businesses make the mistake of taking those who scored the highest (i.e., are most likely to churn) and targeting them, but often it’s the customers with lower scores who can be saved from the ‘churning out.’ Often, short term, marketing campaigns (particularly those offering special deals or discounts) are the most effective means of re-engaging predicted churners.