Kirti Agarwal

Helped business design marketing strategy to optimize the Retention efforts which led to more than $1M savings. Built multiple Machine Learning Prediction models using RF and XGBoost to identify High Risk Members who are most likely to Voluntary Lapse at any point in their journey. Classified Members into segments using K-means based on their lapse risk, channel preference, cost journey and other factors to drive targeted retention outreach tactics. Additionally, developed a Rapid Disenrollment model to predict new member's likelihood to Lapse within 3 months of joining and thus devise strategic campaigns to reduce churn. Created an attribution logic to measure individual risk components to drive Personalized member Retention efforts. Optimize Marketing efforts across different channels to improve engagement and promote a positive member experience. Identified the right channel and right time to outreach member to achieve maximum conversion by building Channel affinity Models. Multi-Touch Attribution using a fusion approach using LSTM with Attention Mechanism and ANN to identify drivers of campaign engagement. Cost saving by reduction in Campaigns sent through Omni channels and migrating only to member's preferred channel. Segmentation using semi-supervised clustering algorithm to increase campaign participation. Applied an Uplift Modelling approach to Prioritize the 'Persuadable' population and suppress the 'Lost causes' to achieve the higher increment saves in Retention Campaign efforts. Divided the Population into four segments - Sure Things, Lost causes, Sleeping dogs and Persuadable by applying a 4 Quadrant Multi-Classification approach. Gain received by measuring the impact a marketing action brings about in conversion. Measured value by quantifying the higher uplift and higher incremental saves gained due to precise targeting. Built Multi Touch Attribution Models and analyzed the touchpoint journey of a member which led to higher campaign closure and reduced