BDM Capstone · Jan 2024 – Jan 2025
Jai Maa
Jhandewali
Store
Optimizing Inventory & Operations for Sustainable Growth in a Family-Run Grocery Business — VP Block, Pitampura, Delhi.
398
Days of Sales Data
13
SKUs Tracked
5.2K
Data Records
15K
Milk Units Sold
Business Context
Store Overview & Challenges
Jai Maa Jhandewali Store is a family-run B2C grocery established ~2 years ago with ₹7 Lakh investment, breaking even in 1.5 years. It operates 9 AM–9 PM and sells across Dairy, FMCG, and Groceries.
Store Profile
Store Name
Jai Maa Jhandewali Store
Location
VP Block, Pitampura, Delhi
Owner
Mr. Satyam Prakash
Store Type
Family-run B2C Grocery
Operating Hours
9 AM – 9 PM Daily
Established
~2 years ago · ₹7L invested
Break-even
1.5 years · Now profitable
Study Period
Jan 2024 – Jan 2025
Product Categories
DAIRY — Category A
Milk (top SKU — 25.85% of all sales), Butter. Highly perishable; demand varies 4–108 units/day for milk alone.
FMCG — Categories A & B
Bread, Chips, Biscuits, Soft Drinks, Toiletries. High velocity — mirrors Dairy seasonality, peaks Mar/Jun/Oct/Dec.
GROCERIES — Categories B & C
Rice, Wheat, Pulses, Sugar, Salt, Spices. Steady year-round demand. Pulse–Wheat correlation 0.729 — high bundling potential.
Inventory Management
Highly volatile demand makes stocking unreliable without data-backed systems.
- Milk demand: 4–108 units per day
- Overstocking perishables causes regular wastage
- No reorder-point system — intuition-based ordering
- Seasonal spikes go unplanned and understocked
Limited Business Growth
Manual operations constrain scale and leave the store vulnerable to quick-commerce.
- No POS — all tracking done manually on paper
- Owner has no bandwidth for strategic decisions
- Competition from Blinkit, Zepto, Instamart growing
- No home delivery despite nearby customer demand
Space Constraints
Physical limitations cause poor SKU placement and spillover storage at home.
- Limited retail floor area for 13+ SKU categories
- Category C items occupy prime shelf space
- No vertical shelving — floor space under-utilized
- Some stock stored at home — causing access delays
Data Collection
Data SourcePrimary · Handwritten bills
Study DurationJan 2024 → Jan 2025
Total Records398 days × 13 SKUs
VariablesDate, SKU, Qty, Category
DigitizationCleaned Excel sheets
Collection methodOwner logs & interviews
Tech Stack
HTML5 / CSS3
Chart.js
Excel / Python
Google Sheets
Linear Regression
ABC Analysis
AUTHOR
Sumit Kumar
22f2000848@ds.study.iitm.ac.in
BDM Capstone · IIT Madras · 2025
Visual Insights
Sales Analytics
13 months of granular SKU data across three categories — Dairy, FMCG, and Groceries — revealing seasonality, distribution patterns, and cross-product correlations.
Monthly Sales by Category
Dairy · Groceries · FMCG — actual unit totals per month
Monthly Sales Distribution
Average daily sales and Q1–Q3 range across all SKUs per month
Category Sales Trend — Jan 2024 to Dec 2024
Monthly unit totals per category showing seasonal patterns
ABC Classification — All 13 SKU Total Sales (Jan 2024 – Jan 2025)
Cat A = top 70% volume (red) · Cat B = next 20% (amber) · Cat C = bottom 10% (green)
Correlation Heatmap
Pearson correlation — category level · Dairy / Groceries / FMCG
Dairy
Groceries
FMCG
Dairy
1.00
0.43
0.61
Groceries
0.43
1.00
0.65
FMCG
0.61
0.65
1.00
Top SKU-level correlated pairs
Pulse & Wheat0.729 — strong co-purchase
Pulse & Rice0.691 — grocery staple bundle
Bread & Sugar0.550 — FMCG–Grocery cross
Chips & Biscuits0.499 — snack bundle
Milk & Spices0.119 — no dependency
SKU Sales Breakdown
Total units · Jan 2024 – Jan 2025 · 13 SKUs
| Product | Class | Units | Share |
|---|
Milk Sales Forecast — Linear Regression
Actual monthly avg vs predicted · Features: lag, month, weekday, category
R² ≈ 0.78
ML FORECAST
0.78
R² Score
Mar · Jun
Oct · Dec
Oct · Dec
Peak Months
4–108
Daily Range (units)
Methodology
Analytical Framework
Four complementary analytical techniques applied to the 13-month primary dataset to extract actionable operational insights.
01
Trend Analysis
Sales Trend & Seasonality Detection
Monthly sales were visualized to reveal seasonal spikes. Milk peaks in March (summer onset), June, October (festive season), and December. Time-series decomposition using moving averages separated trend from cyclic noise, confirming double seasonality in Dairy and FMCG. Groceries showed comparatively stable year-round demand with minor festive bumps.
02
Sales Forecasting · ML
Linear Regression — Milk Demand Prediction
A linear regression model was built for Milk using features including lag-1 sales, month-of-year, weekday indicator, and category. The model achieved R² ≈ 0.78, enabling 3-month forward forecasting to support proactive procurement and reduce stockout risk for the store's single largest SKU.
R² ≈ 0.78 · RMSE = Low
03
ABC Analysis
Inventory Classification by Volume Contribution
All 13 SKUs were ranked by cumulative sales using the Pareto principle. Category A (Milk, Bread, Chips, Biscuits, Toiletries) accounts for 77.6% of total volume. Category B (Pulse, Sugar, Soft Drinks, Butter, Wheat) contributes 18.2%. Category C (Rice, Spices, Salt) contributes just 4.1% — prime candidates for space reallocation or discounting.
Cat A · 77.6%
Cat B · 18.2%
Cat C · 4.1%
04
Correlation Analysis
Pearson Correlation — Cross-SKU Purchase Patterns
Pearson coefficients computed across all 13 SKU pairs identified significant co-purchase relationships. Pulse–Wheat (0.729) and Pulse–Rice (0.691) suggest grocery staple bundling. Chips–Biscuits (0.499) validates snack combo potential. Milk–Spices (0.119) confirms independence — no joint promotion needed. These findings directly inform shelf layout and promotional bundle design.
Recommendations
Strategic Action Plan
Data-backed recommendations across four pillars — Inventory, Space, Automation, and Growth — designed to be implementable with minimal capital outlay.
Inventory Optimization
Apply data-driven ordering to eliminate perishable waste and prevent stockouts of high-demand SKUs.
- Use ML forecasting model for Milk ordering — anticipates peaks, reduces overstock
- Prioritize Category A SKUs in procurement budget allocation
- Apply Just-In-Time (JIT) ordering for Milk and Butter to minimize wastage
- Set safety stock levels based on 398-day demand distribution (±1σ buffer)
- Implement weekly reorder cycle aligned with seasonal demand patterns
Space Utilization
Redesign store layout based on ABC classification and purchase correlation data.
- Front-load Category A SKUs — prime eye-level placement for Milk, Bread, Chips
- Invest in vertical shelving — doubles usable storage without expanding floor area
- Co-locate Pulse+Wheat+Rice (high correlation 0.69–0.73) for intuitive aisle flow
- Discount or phase out low-performing Category C items to free shelf space
- Eliminate home storage overflow — all stock should remain on-premises
Process Automation
Replace manual record-keeping with lightweight digital tools for real-time decision-making.
- Adopt basic POS (e.g., Vyapar, OkCredit) for daily sales tracking
- Build Excel inventory tracker with automatic reorder alerts
- Log daily SKU sales digitally — eliminates handwritten bill dependency
- Use Google Sheets dashboard for weekly owner review — zero cost, no friction
Business Growth
Expand reach and revenue through low-cost local engagement strategies.
- Launch WhatsApp-based home delivery for nearby customers — no app needed
- Create festive combo bundles: Chips+Biscuits, Pulse+Rice+Wheat (correlation-backed)
- Hire temporary help Oct–Dec during confirmed high-demand festive season
- Leverage seasonal peaks (Mar, Jun, Oct, Dec) with targeted promotions
- Introduce loyalty cards for repeat customers to compete with quick-commerce apps
Projected Impact
Expected outcomes across financial, operational, and strategic dimensions
Financial Gains
Increased net revenue through reduced perishable wastage. Better cash flow via optimized procurement. Lower holding costs from leaner inventory.
Operational Gains
Fewer stockouts of Category A items. Optimized floor space allocation. Reduced manual workload allowing owner to focus on growth.
Strategic Benefits
Improved scalability and competitive positioning vs quick-commerce. Higher customer satisfaction. Foundation for data-driven management culture.