predicting churn

PREDICTING CUSTOMER CHURN AT QWE INC. Group10: Richard Ely, Yuchen Luo, Xinyu(Frank) Meng, Yijia He, Simeng Yin

Trusted by over 1 million members

Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.





Agenda ■ Ex Exec ecu uti tive ve Su Summ mmar aryy ■ Met eth hodology – Multiple-variable Logistic Regression (MLR)







Executive Summary ■ Problem: how to estimate the probability that a given customer would leave and identify the drivers that contributed most to that customer’s decision ■ Decisions to make: - methodology - ident identify ify the the 3 most most influent influential ial variab variables les related related to probabi probability lity of churn churn







Relationship between Age and Churn does not align with Mr. Wall's belief Mr. Wall’s Belief of Age vs. Churn

■ Ag Age e 6 and and 14 are are no nott good good cut cutof offf poin points ts ■ On Only ly cus custom tomer erss age age > 35 les lesss like likely ly to to leav leave e

Customer Age (in month)

Likelihood to Churn

<6

Less Likely

6-14

Most Likely







Top 3 factors in Multiple-variable Logistic Regression - “CHI Score in Dec” , “Change in Login Recency”, “Change in CHI Score” ■ Be Best st fa fact ctor orss be beca caus use: e: Smaller p-value

Statistically Statistic ally significant

Variable

Standardized Coefficient

P-value

CHI Score in Dec.

-0.37

1.87e-07 ***







MLR with Five Variables Is Not Good at Predicting Churn Customers Methodology: Five variables with statistically significant coefficient CHI Score in Dec

∆Days since Last Login

(Dec-Nov)

∆CHI

Score (Dec-Nov)

Customer Age

∆ Views

(Dec-Nov)









Reasons:







Top 3 Predictors in Decision Tree - Change in Login Recency, Change in Login Frequency, Customer Age







Change in Login Recency < 18

Condition met Condition unmet







Decision Tree-An Extract of Predicted Churn Customers







Decision Tree Excellent in Avoiding















predicting churn

Recommend Documents