lagged metrics

In this project we want to investigate main differences between the metrics on the call and lagged metrics

difference

The lagged metrics considers a time span of 25 days when the customer could call again and turn around. The lagged metric loses around 10% customers

lagged_series time series of on-call and lagged metrics

We have different metrics to predict

lagged_metric metrics to predict, on call and lagged

We clean the features

lagged_boxplot normalized feature distribution

We check the feature independency

lagged_corr correlation between features

Effect of afiniti agent on feature distribution

lagged_overlay_on overlay of feature distribution

Overlay distribution of features on saved customers

lagged_overlay overlay of feature distribution

Overlay distribution of featured of saved/lost customers

lagged_corr Overlay distribution of features: joyplot

We build a predictive model for the on-call metric

lagged_confMatrix confusion matrix on prediction: on-call

We use the prediction of the previous model to predict the new metric

lagged_confMat confusion matrix on prediction: lagged

We knock-out features to calculate the relative importance of that feature

lagged_knockCall knock-out of features from trained model: on-call

lagged_knockLagged knock-out of features from trained model: lagged

We finally check the relative difference between on-call and lagged

lagged_featDiff relative difference between feature importance in lagged metrics

We can study the relative feature importance

lagged_featImp feature importance (week overlay)

and see how it evolves week by week

lagged_weekImp relative importance week by week

We study over week how the relative importance of that metric dropw over time

lagged_featImp importance difference relative to lagged metrics

We calculate the lift depending on categorical metrics


lift = SRon − SRoff

lagged_featImp lift relative to categorical metrics

cohort reshuffling

We look month per month how the association of agents per area varies over time

cohort_timeSer time series volume changes in cohorts

Shuffling among cohorts

lagged_sankey reshuffling of agents cohorts