THE IMPACT OF THE CALL TO TEXT CAMPAIGN ON RETURN VOLUME
In an effort to reduce product returns, a call center Agent contacts the customer to confirm date and time as well as measurements and other specifications for their appliances before delivery. It is important to know if this initiative is effective in reducing losses.
REFRAMED AS ANALYTICS PROBLEM
Historical data was used to determine if calls to the consumer is an important predictor of Return Volume, and to examine how this and other factors influence increases and decreases in return volume. Data from SalesPortal POS database was used in a Regression Analysis which selected an appropriate model. The data included factors such as Sales, Cancels, Shipped Units, Calls made to the consumer, and Returns.
Multivariate regression is a set of advanced statistical techniques that estimates a single regression model with more than one outcome variable. Multivariate regression was used to determine if calls to the consumer is an important predictor of a return, and to examine how this and other factors influence increases and decreases in a return.
Continuing the initiative when it is not effective inflates costs and creates an unnecessary touchpoint for the customer.
The resulting model was found to be statistically significant (p < 0.001) and explained 80% (R-sq = 0.8196) of the variation in a return and it was concluded with 95% certainty that holding all other variables constant, for each call made to the consumer, you can expect a return to decrease on average by 0.04 Units. In other words, you can expect for each 25 calls, a return will decrease on average by 1 Unit.
The results were similar for Scorp, but not for Tcorp.
Continue the effective initiative for Jcorp and Scorp.
Returns are expected to be reduced by 4 for every 100 calls.
Discontinue the ineffective initiative for Tcorp.
This will not affect the volume of returns.
Gains: Returns will be reduced by 4 for every 100 calls.
Reduce budget by 70% of the original budget.