Managing data science engagement, which delivers insights to reduce customer contacts

Marketing Mix Modeling -o Managed a team of five data scientists/engineers to deliver marketing mix models for a leading food delivery organization in APAC region. Delivered an optimization tool to assist business in marketing spend decisionso Designed market mix frameworks for multiple geographies of a leading Travel and Hospitality client

Forecasting key metrics - Created forecasting models (Visits, Calls and Units) for the marketing team of a leading technological company using marketing spends as inputs. Forecasts were instrumental in resource planning for call centers and assisting marketing managers in meeting their plans. 6% improvement in forecast accuracyMarketing insights - Utilized Adobe Omniture to deep dive into the performance of different marketing metrics (site visits, units, revenue and margin). Lift analysis done for Black Friday, Cyber Monday and New Year’s day performanceCustomer Loyalty Driver Analysis - Designed a loyalty framework that involved the following: a) creating definitions for different loyalty behaviors (expansion, continuity and advocacy) b) Identifying drivers for different loyalty behaviors c) Sizing the opportunity and creating scorecard for customers. This project helped the business to rank existing customers based on their loyalty scores and start programs to improve customer loyalty scores. A 7% increase observed on an average in the customer loyalty after the implementation of this projectPropensity Modeling - Worked with the marketing team of a leading apparel company to identify target customers for their cross-sell brand. Customers were identified based on their propensity to buy the product. Customer universe was scored using Logistic regression model. Gains of 3 and 2.4 were observed in the top 2 deciles

a) Call Volume forecasting - Time Series Analysis (ARIMA, Winter Holt’s) - to assist the operations team in resource and capacity planning. The average waiting time for a customer on an average reduced by 5%b) Audience Insights - Text analytics (Boolean Query, Sentiment Analysis) analyzed the customer feedback and reviews & recommending changes in the existing processes to improve customer experienceRoles and Responsibilities in Professional Careera) End to end analytical solutions to clients’ business problems. Understanding the customer problem, data and creating an analytical solution using statistical techniques in SPSS or R and providing recommendations to client based on the sameb) Creating business proposals for prospective clients to showcase the use of analytics in data driven decision makingc) Call volume forecasting for call center datad) Analyzing customer survey data to enable better targeting of customers and understanding the concerns raised by the consumers using text analytics

a) Satisfaction Loyalty Analysis - Pilot project to identify major factors driving customer loyalty. Proposed a metric based framework for better targeting and improved customer retention. Owing to success of the pilot an annual retainer was signed for USD 1Mnb) Identifying drivers of Overall Quality of Service and Loyalty - A comparative study of client’s business with competition and key focus areas identified (Linear Regression model)c) Customer acquisition - Propensity Modeling for one of the leading cloud computing platform to identify positive and negative drivers leading to conversion from a trial customer to a paid customer (Logistic Regression Model)Roles and Responsibilities:a) Understanding client requirements and client data. Conduct research to understand which statistical techniques and analysis can be conducted on the datab) Present insights and recommend measures to improve the clients’ businessc) Mentored new recruits and hosted knowledge sharing sessions which allowed the new recruits to get up to speed on the projects and culture of the companyd) Documentation of the best practices from the work experience and learnings from each project

a) Customer Segmentation - Clustering based on customers’ usage patterns (K-means clustering). Helped improve offer customization & targeting of customers, reducing budget expenses by $150K/quarter