Preventing Churn Using Predictive Modeling Alex Herbert Sales Manager James Cousins Sr. Statistical Analyst March 26 th , 2019
Agenda • Who We Are • Implications of Churn • Ways to Prevent Churn • How we can help • Q & A
About Rapid Insight Predictive analytics and data preparation software company empowering professionals of all skill levels to turn raw data into actionable insights Serving hundreds of Code free Founded in 2002 customers worldwide, (but code friendly) and headquartered ranging from healthcare self-service in Conway, NH to higher education analytics platform The Veera platform enables users to easily build predictive models, perform advanced data analysis, and share insights
The Veera Platform Veera Construct enables everyone from citizen data scientists to PhD statisticians to turn any data into actionable information Veera Predict enables users of any skill level to analyze data and build predictive models in a fraction of time required by other tools Veera Bridge empowers organizations by democratizing data with its cloud-based collaboration platform
Preventing Churn Using Predictive Modeling
Our Simple Argument Against Churn It’s critical to your business There are several ways We offer a very convenient to be on top of churn to approach churn solution to prevent churn
Customer Churn a.k.a. Customer Attrition, Turnover, Defection,… • Losing Customers (voluntarily or involuntarily)
Customer Retention (avoiding churn) • Safeguards Revenue Reputation Morale
Customer Churn • Acquiring a new customer is between 5 and 25 times more expensive than retaining one 1 • Increasing customer retention by 5% can increase profits by 25-95% 2 1. www.hbr.org/2014/10/the-value-of-keeping-the-right-customers 2. www.bellycard.com/resources/customer-retention-marketing-insights/
Analytics in Customer Churn Descriptive Predictive Proactive
Descriptive Analytics in Customer Churn • What is your churn Descriptive rate? • Where is churn Predictive occurring? • Who is churning? Proactive
Predictive Analytics in Customer Churn • What factors predict Descriptive churn? • How much churn are Predictive you expecting Proactive
Proactive Analytics in Customer Churn • Reach out to those Descriptive likely to churn • Make changes to Predictive mitigate churn Proactive
Prevention Strategies 1 Personalization "Surprise reciprocity" (providing unexpected benefits Learn from Customers to build loyalty) (surveys, feedback) Thanking Customers Gamification & Referral (Zappos, Chewy) programs Develop towards making their Subscriptions lives more convenient Social Proof (Starbucks, apps) Have a "mission" Education- materials to (Toms, Warby Parker) support end-users' usage 1. blog.hubspot.com/service/customer-retention-strategies
Churn Use-Case: Telecommunications
Descriptive Analytics in Customer Churn • What is your churn Descriptive rate? • Where is churn Predictive occurring? • Who is churning? Proactive
Connect Integrate data in any format, from virtually any source Prepare Create step-by-step processes using easy, drag-and-drop visual workflows with no coding required Analyze Build and schedule jobs to run automatically, or run on-demand analyses Share Write back to databases, create and disseminate reports, publish dashboards to visual analytics tools such as Tableau, or output datasets for predictive modeling
Predictive Analytics in Customer Churn • What factors predict Descriptive churn? • How much churn are Predictive you expecting Proactive
What is Predictive Analytics? = Churned = Retained Churn Likelihood % of Plan Utilized
Predictors of Churn Outcome: Customer Churn Predictor Not a Predictor • Has children • Weekend vs Weekday ratio • Marital status • Bill delivery method • % of plan used • Gender
Customer Churn Predictions Plan Type # of Customers Predicted Churn Rate Predicted Loss Basic 1,350 17.27% 233 Family Minutes 14,956 2.49% 373 Starter Package, Basic 2,611 10.29% 269 Starter Package, Full 1,083 8.71% 94
Fully Transparent All methods and calculations performed are easily viewable Automated Data Mining Identify statistically significant data variables without manual effort Easy to Use New users can build expert models within two hours of training No Programming Needed Use simple pick lists instead of programming or custom command languages Automated Modeling Process Brings complex data analyses and predictive model building to users of any skill level
Proactive Analytics in Customer Churn • Reach out to those Descriptive likely to churn • Make changes to Predictive mitigate churn Proactive
Prevention Strategies 1 Personalization "Surprise reciprocity" (providing unexpected benefits Learn from Customers to build loyalty) (surveys, feedback) Thanking Customers Gamification & Referral (Zappos, Chewy) programs Develop towards making their Subscriptions lives more convenient Social Proof (Starbucks, apps) Have a "mission" Education- materials to (Toms, Warby Parker) support end-users' usage 1. blog.hubspot.com/service/customer-retention-strategies
Inspire Curiosity Democratize data by allowing users to seamlessly view and explore data Power Exploration Allow anyone, in any department, with any skill level, be inspired by data One Stop for All Insights A single hub allows you to streamline data and report publishing Self Service Reporting By supplying reports or datasets through Veera Bridge, have peace of mind that your users can now answer all questions dynamically
Personalized Demo data prep reporting predictive modeling advanced analytics cloud-based reporting data exploration To schedule your personalized demo contact sales@rapidinsight.com
Questions?
Thank You! James.Cousins@rapidinsight.com Alex.Herbert@rapidinsight.com information@rapidinsight.com For more information, visit www.rapidinsight.com
Recommend
More recommend