3.8 KiB

Raw Blame History

📚 Maven Toys Dataset Schema Relationship Guide

This document outlines the schema relationships and foreign key connections between all CSV files in this directory, suggesting how they can be joined for comprehensive data analysis.

🧩 Entity/Dimension Tables (The "Who" and "What")

These tables define core entities and are typically used as lookup tables.

stores: Information about the physical retail locations.
- Primary Key (PK): Store_ID
products: Master list of all items sold.
- Primary Key (PK): Product_ID
calendar: Time dimension data for the business.
- Primary Key (PK): Date (Assuming unique dates are recorded)

📊 Fact/Snapshot Tables (The "When" and "How Much")

These tables record events, measurements, or snapshots in time that link the dimensions together.

sales: The core transaction log. This is the most frequently joined table.
- Foreign Keys (FKs): Store_ID (references stores), Product_ID (references products).
inventory: Snapshot of stock levels at a point in time.
- Composite Key/FKs: (Store_ID, Product_ID) \to Links to both stores and products.
data_dictionary: Metadata describing the other fields (Not used for joins, but crucial for understanding column definitions).

🗓️ Time Dimension

The calendar table provides temporal context, which can be joined with sales records to analyze performance around holidays or specific periods.

🔗 Relationship Map and Join Paths

The following sections show the explicit paths you can use for joining data in SQL or Python (Pandas/DuckDB).

1. Sales Analysis Path

Goal: Analyzing a transaction's details, location, and item description.
Join Chain: sales \to (stores, products)
Example Join: FROM sales s JOIN stores st ON s.Store_ID = st.Store_ID JOIN products p ON s.Product_ID = p.Product_ID;

2. Inventory Valuation Path

Goal: Calculating the total value of current stock across all stores.
Join Chain: inventory \to (stores, products)
Example Join: FROM inventory i JOIN stores st ON i.Store_ID = st.Store_ID JOIN products p ON i.Product_ID = p.Product_ID;

3. Comprehensive Performance Path (The Full Picture)

Goal: Linking sales performance to store location details and calendar dates.
Join Chain: sales \to (stores, products, calendar)
Notes: You can join on the date field from both sales and calendar.

💡 Example Queries (Ready for Use)

These queries demonstrate how to combine the tables.

1. Total Revenue Over Time

Calculate the total revenue generated month-by-month, showing store performance over time.

SELECT
    strftime('%Y-%m', s.Date) AS sales_month, -- Grouping by Year and Month
    st.Store_Name,
    COUNT(DISTINCT p.Product_ID) AS distinct_products_sold,
    SUM(s.Units * p.Product_Price) AS total_monthly_revenue
FROM sales s
JOIN stores st ON s.Store_ID = st.Store_ID
JOIN products p ON s.Product_ID = p.Product_ID
GROUP BY 1, 2
ORDER BY 1 DESC, total_monthly_revenue DESC;

2. Top Performing Product/Category Analysis

Identify the best-selling categories and the top 5 specific products by units sold.

SELECT
    p.Product_Name,
    p.Product_Category,
    SUM(s.Units) AS total_units_sold
FROM sales s
JOIN products p ON s.Product_ID = p.Product_ID
GROUP BY 1, 2
ORDER BY total_units_sold DESC
LIMIT 5;

3. Low Stock Alerts (Inventory Management)

List all stores and products where the current stock is below a specified threshold (e.g., < 50 units).

SELECT
    st.Store_Name,
    p.Product_Name,
    i.Stock_On_Hand
FROM inventory i
JOIN stores st ON i.Store_ID = st.Store_ID
JOIN products p ON i.Product_ID = p.Product_ID
WHERE i.Stock_On_Hand < 50;

3.8 KiB Raw Blame History