How can Analytics Drive Customer Values?
3 likes1,200 views
The document discusses how analytics can be used to drive customer value. It provides examples of using analytics for customer churn prediction, targeted marketing, fraud detection, cash demand forecasting for ATMs, and optimizing customer offers. Social media analytics and text mining are also discussed for understanding customer sentiment and preferences.
How can Analytics Drive Customer Values?
- 1. How can Analytics Drive Customer Values? Franklin So Regional Analytic Practice, SAS AP June 28, 2011 Copyright © 2010 SAS Institute Inc. All rights reserved.
- 2. 2 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 3. My new iPhone 4… FREE!!! Guess from whom???? 3 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 4. 4 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 5. 1 week later She moved to another telco operator in early June... 5 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 6. SORRY THIS ATM IS TEMPORARY OUT FO CASH 6 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 7. Unfortunately there was long queue in the bank 7 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 8. Leave my comments in Facebook for these promotion offer… 8 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 9. = All these involved Analytics … 9 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 10. Analytics to manage CHURN customers 10 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 11. 11 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 12. Churn Scores for prediction… 12 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 13. Social network for viral marketing Linking individuals and measuring the strength of their relationships. •Identify communities based on behavioral relationships between customers •Measure and segment customers based on social influence (e.g. “leaders”, “followers”, “marginals” and “outliers”) •Target customers based on community status and behavioral changes within communities (e.g. when a community “leader” changes, target his/her “followers”) Leader Follower 13 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 14. SORRY THIS ATM IS TEMPORARY OUT FO CASH Analytics to forecast demand e.g. cash demand in ATM machines 14 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 15. ATM Replenishment Forecasting and Optimization SAS Forecast Server SAS/OR Bank ATM system transaction log Demand Forecasting for each ATM SAS BI Optimization of the Replenish schedule by denominations Executive report on ATM Replenishment Status and Manual Overrides 15 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 16. Business Forecasting and scenario (“What-if analysis”)… Red line represents the forecast of transactions of the ATM. White dots represent the history of ATM transaction. 16 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 17. Optimized replenishment schedule tells you which ATM to replenish at what time • Reduced ATM down-time due to cash out • Optimized no. of ATM replenishment trips • Improved operational efficiency of maintaining ATM Network • Increase customer satisfaction for reducing the cash out events 17 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 18. Simulate the operational process and time required for queuing the teller… How many queues are required? One queue for one teller or one single queue for all tellers 18 1 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 19. Analytics to identify customer strategy who to sell what, on when, and how… 19 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 20. Who should be targeted on what, when, and what’s NEXT… Deposits Bank Knows savings balance and some demographic details Credit Card Knows income , purchasing & payment behavior Insurance Life stage and some Customer asset and liability details Investments Knows investments Channels 20 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 21. Segmenting customer based on behavioral attributes results in segments which are: Identifiable – customer behavioral data are much richer. Actionable – behavioral attributes tell us more about the needs of a customer. relationship product holdings customer tenure, number of what products does the customer accounts/products, recency of have with the bank? account, branch, etc. financials transaction pattern Customer average, trend, variance, what type of transactions, how often, etc. for balances, amount per transactions, etc. interest, fees, etc. channel preference & usage what channels does the customer own, access and how often, etc. 21 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 22. Customer Value 22 Loyalty Score Copyright © 2010, SAS Institute Inc. All rights reserved.
- 23. Acquisition Retention & Retention Customer Value Cross selling 23 Loyalty Score Copyright © 2010, SAS Institute Inc. All rights reserved.
- 24. Analytics to identify the NEXT BEST and Optimize Offers 24 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 25. Personalize Offer Optimization 25 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 26. Towards Offer Optimization 26 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 27. And finally Offer Optimization 27 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 28. Analytics to analyze Social Media and Unstructured Data 28 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 29. WHAT’s Important to Overall Brand Your Business What’s the ►Service, Price, Network Sentiment? Topics Relevant to Your Organization Performance, Marketing Program Media Types WHERE You’re Being Talked About Positive Negative • Internal: Portal, Call Centre Neutral Media Sources External: What’s the Bloggers Bloggers WHO’s Talking about You and What Value? They’re Saying i.e. Harvey West ► the top Twitter influencer Commentary with 300,000 followers ….What’s the Trend…. Time…. Historical Forecasted Future 29 HP Confidential. Commercially sensitive. Distribution prohibited. Copyright © 2010, SAS Institute Inc. All rights reserved.
- 30. 30 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 31. Statistical Process Control Analysis of Variance Categorical Data Analysis Social Network Analysis Spectral Analysis Statistical Analysis R Integration Process Capability Analysis Forecasting Scheduling Reliability Analysis Nonlinear Programming Design of Experiments Cluster Analysis Sentiment Analysis Linear Programming Data Visualization Network Flow Models Predictive Modeling Econometrics Vector Autoregressive Models Discrete Event Simulation Exploratory Data Analysis Mixed-Integer Programming Sample Size Computations Nonparametric Analysis Interactive Matrix Programming ARIMA Models Matrix Programming X11 & X12 Models Neural Networks Scoring Acceleration Ensemble Models Bayesian Data Mining Survival Analysis D-Optimal High Performance Forecasting Text Analytics Decision Trees SAS leads advanced analytics market by wide margin (IDC, June 2011) Psychometric Analysis Information Theory Statistics Descriptive Modeling Mixed Models Multivariate Analysis Quality Improvement Text Mining Multinomical Discrete Choice Study Planning Gradient Boosting Machines Predictive Analytics Analysis of Means Interior-Point Models Random Forrests Survey Data Analysis Content Categorization Genetic Algorithms Operations Research Discrete Event Simulation Content Categorization Automated Scoring Simulation Ontology Management Time Series Analysis Model Management Association & Sequence Analysis Constraint Programming Fractional Factorial Ontology Management Large-Scale Forecasting Regression 32 Copyright © 2010, SAS Institute Inc. All rights reserved.
- 32. Thank You Copyright © 2010 SAS Institute Inc. All rights reserved.