Retail Ultraplan

# 🏢 Almond House — Retail Intelligence Platform (UltraPlan v3) ## What This Is Complete analysis of **Almond House** POS data — 11 retail stores, 728 items, 38K+ receipts across 10 days (March 2026). Built from the **actual data** in `AI_Forecasting_Main.xlsx`. Every number below comes from your real transactions. ## Key Findings | Metric | Value | |--------|-------| | **Total Revenue** | ₹3.77 Cr (10 days) | | **Daily Average** | ₹37.7 L/day | | **Top Store** | KPHB (₹60.8L, 6,854 receipts) | | **Bottom Store** | BMT1 (₹16.2L, 1,950 receipts) | | **Top Item** | Kaju Burfi (₹23.5L alone!) | | **Top Category** | SWEET (51.6% of all revenue) | | **Avg Basket** | ₹980, 2.5 items | | **Peak Hour** | 6:00 PM (9.9% of daily revenue) | | **Best Day** | Saturday (₹47.5L) | | **Cross-sell Rules** | 2,321 item pairs discovered | ## Critical Data Findings (from audit) 1. **Net Amount is NEGATIVE** (ERP convention) → Revenue = -Net Amount 2. **Price == Net Price 100%** → NO discounts in data → promo analysis not possible 3. **Gross Amount is RECEIPT-level**, not line-level 4. **9 warehouse stores** separated from 11 retail stores 5. **Feb data is sparse** (83 rows) → only March used 6. **47% revenue from KGS items** (sold by weight: 0.25kg, 0.5kg) ## Output Files (`output/` folder) | File | What | Rows | |------|------|------| | `00_cleaned_sales.csv` | Cleaned transaction data | 94,413 | | `00_returns.csv` | Return transactions separated | 303 | | `00_warehouse_transfers.csv` | Warehouse/production data | 5,876 | | `01_store_performance.csv` | Store ranking & KPIs | 11 stores | | `02_daily_sales.csv` | Daily revenue trend | 10 days | | `02_hourly_pattern.csv` | Hourly sales pattern | 17 hours | | `03_item_performance.csv` | All items ranked by revenue | 671 items | | `04_category_performance.csv` | Category breakdown | 19 categories | | `04_brand_performance.csv` | Brand breakdown | 8 brands | | `04_product_line_performance.csv` | Product line breakdown | 39 lines | | `05_baskets_detail.csv` | Every receipt with items | 38,478 baskets | | `05_crosssell_rules.csv` | "Bought together" pairs | 2,321 rules | | `05_category_copurchase.csv` | Category pairs | 78 pairs | | `06_inventory_abc_xyz.csv` | Item classification + safety stock | 725 items | | `07_store_item_matrix.csv` | Store × Item sales | 5,370 combos | | `07_item_store_coverage.csv` | Which items at which stores | 725 items | | `07_store_category_mix.csv` | Category mix per store | 11 × 19 | | `08_executive_kpis.csv` | Summary KPIs | 22 metrics | ## How to Use ```bash # Open any CSV in Excel/Google Sheets/Power BI # Or re-run analysis: pip install pandas numpy scikit-learn scipy openpyxl python ultraplan_v3.py --data AI_Forecasting_Main.xlsx --outdir my_results ```