Programmatic Planning Using Predictive Data Analytics
Solix Case Study
Situation
One of the largest U.S. telecoms was operating a mature and established low cost broadband program and needed a way to strategically plan for increases in applications and eligibility reviews/determinations. The primary goal was to use historical data trends in order to forecast and plan budgets and better align resources in anticipation of expected surges in program activity. By utilizing predictive analytics prepared by Solix, the client was able to use fact based data to make both operational and financial programmatic decisions. Solix’s understanding of the customer lifecycle and leveraging its business intelligence capabilities demonstrated its ability to analyze raw data to enable predictive forecasting for critical business decision making.
Objectives
- Utilize predictive analytics to develop program budgets and forecast revenue.
- Develop and implement client self-service predictive data analytics and market analysis dashboards.
- Utilize U.S. census block data to drive planned community sign-up/marketing events based on socioeconomic information.
Challenges
- Obtaining census block data, income information, by area/zip code.
- Removing outlier data to help forecasting.
Solution
- Captured historical customer information to perform predictive analytics.
- Identified key performance indicators of applicant/customer lifecycle including application received date, reviewed date, decision date, decision outcome (approval/denial), purchase/provision date/count.
- Utilized historical programmatic KPI information to visually forecast six months ahead with an 80% level of confidence:
- Income application and review volumes.
- Incoming number of applications with a decision result of approved.
- Number of service purchase/take rate.
- Integrated census block income and socioeconomic data into self-serving analytics dashboard.
Results
- Delivered client-self service reporting dashboard providing data on demand with key performance indicators and visualizations as it related to applicant/customer lifecycle.
- Improved contact center staffing to meet actual needs of the business for providing operational support for the customer provisioning/order process.
- Created ability to use historical data to provide planning budgets and updated quickly using fact-based data.
- Enabled client marketing teams to target specific census blocks/zips codes that have a higher take rate probability; high customer acquisition required.
- Reduced customer acquisition cost.
- Produced more accurate budgets using real-time program data.