stkisengese/duck-bidco-retail-insights

A data engineering solution transforming raw retail sales data for Bidco Africa into actionable KPIs: Data Health Score, Promotional Uplift and Competitive Pricing Index. # Bidco Africa Sales Performance and Pricing Engine A data engineering solution transforming raw retail sales data for Bidco Africa into actionable KPIs: Data Health Score, Promotional Uplift and Competitive Pricing Index. ## Context & Executive Summary This analysis converts one week of raw supermarket transactions into three actionable deliverables for Bidco: 1. a data health report per store and supplier so you can trust numbers 2. promotion performance metrics that quantify uplift, discount depth and coverage 3. a pricing index that shows where Bidco sits versus peers by Section/Sub-Department and store. My approach prioritises business value first: identify high impact SKUs and stores where pricing or promo changeswill most quickly increase sales or margin. All assumptions, definations and calculations are documented below. ## Data Health: Key Findings ### What's Working - 99.8% data completeness across 30,000+ transactions - All stores show >95% data quality - Zero critical data gaps that would block decision-making ### What Needs Attention 1. **15 records missing supplier** (includes potential Bidco products) - Impact: Cannot include in competitive pricing analysis - Fix: Cross-reference with Bidco product catalog 2. **32 records missing RRP** (0.1% of the dataset) - Impact: Cannot detect promotions for these SKUs - Fix: Use category average as proxy for low-value items ### Recommended Actions - **For Bidco** Proceed with annalysis using 99.9% of the data - **For Retailer** Request RRP/Supplier updates for flagged records - **For Duck** This data quality is suficient for real-time decision making ### Business Confidence Level: HIGH This dataset can reliably inform: - Promotion strategy decisions - Pricing adjustments - Inventory planning ## Promotional Uplift ### Promo Detection Rule(Assumption): SKU is "on promo" when realized price is ≥10% below RRP for ≥2 days in the week Why this rule? - 10% threshold: Filters out normal price variations/roundin …