Scaling Retail Planning at IKEA: Orchestrating Sales, Fulfillment and Capacity Assessment with Metaflow
summary
IKEA orchestrates sales, fulfillment and capacity forecasting into a single pipeline using Metaflow + Argo Workflows to run parallel jobs for 30+ countries, collapsing eight repositories down to four and cutting a long-running computational pipeline from seven hours to about one. The actual deliverable is a reproducible, namespaced pipeline plus a customizable dashboard that preserves local development, provides isolated cloud experiments, triggers cross-workflow dependencies, and materially speeds iteration while keeping planners able to adjust forecasts.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:13Retail planning overview
- 4:05Sales fulfillment & capacity planning
- 9:03Forecasts and data insights
- 11:09Data science team & challenges
- 14:18Metaflow orchestration
- 16:25Simpler workflows with Metaflow
- 18:13Metaflow development phases
- 20:01Argo Workflows with Metaflow
- 21:47Complexity reduction
- 22:50Isolation & collaboration
- 24:16Simplicity & scalability benefits
- 25:20Teamwork and final remarks
- 25:42Q&A
related talks
29mModern Development in Traditional Retail
44mHow Metaflow Became Netflix's Beloved Data Science Framework
36mHuman-Friendly, Production-Ready Data Science Stack with Metaflow & Kubernetes
21mBreaking the 1.5MB Barrier: Running Large Metaflow Flows with Argo for AI/ML Workloa...
35mBuilding Robust Reactive ML Systems In A Multiplayer Setting
32mHuman-Friendly, Production-Ready Data Science Stack With Metaflow & Kubernetes
40mOptimizing AI/ML Workloads
31mEvolving CI/CD at Flipkart