Rahul K Email & Phone Number

London, GB

London, England, United Kingdom

Global APAC lead- RGM transformation

London, England, United Kingdom

• Developed E2E scalable framework for Revenue Growth Management from model building to ROI estimation, simulation and optimization across 14 markets for a fortune 200 Major
• Managing Product Roadmap for inhouse Consumer Analytics product

• Incubated PoC concept into a scalable architecture & product and received US patent grant for the underlying NLP based technology framework
• Leading a team 10+ through agile scrum methodologiesConsulted for a Fortune 200 CPG major in 3 year brand strategy planning exercise on ecommerce strategy onsite at client headquarters

Bengaluru, Karnataka, India

• Played a vital role in Ecommerce project on Consumer Purchase Journey tracking for a Fortune 200 major that gave top results for the client business.
• Achieved +20% versus baseline (+30% vs PY) Net revenue
• Gain in market share around 15% in tested category
• High single digit gross margin expansion through better segmentation and channel shifts
• Generated the best ever month, including Prime days month, in history of client owned products sale in Amazon.
• Developed end to end R shiny based solution from design to deployment for unstructured NLP data processing and insight generation.

Bengaluru Area, India

• Developed end to end solution for identifying the right SKUs that can be added or removed from a store and create weekly orders with optimised volume at store X SKU level
• Proposed and implemented PAM clustering technique to segment stores to set benchmarks and estimate opportunities.
• Implemented parallel processing to achieve massive performance gains and reduced processing time.
• Deployed the solution to run automated and at scale ( ~2 lakh stores ) with weekly order refresh, removing any need for user intervention.
• Innovated a one stop solution for Bayesian regression modelling, visualization and simulation using R shiny.
• Created an easy to use GUI for adding variables to model with the help of residual correlation tables aided with actual vs predicted charts.