Create a Report based on multiple Data Store

Build reports on Retail Data in BDB using Datastores, standard KPIs, visualizations, and ML features (Segmentation, Anomaly Detection, and Forecasting).

Purpose

This workflow explains how to create a Self-Service Report on the BDB Platform using Retail Store Purchase and Sales Data. It demonstrates how to use the Data Sandbox, build Datastores, and visualize insights in the Report Module using standard KPIs, charts, and Machine Learning (ML) capabilities such as Customer Segmentation, Anomaly Detection, and Time Series Forecasting.

Business Context

Retail organizations often deal with large amounts of purchase and sales data that require real-time insights to optimize procurement, pricing, and customer engagement strategies. This workflow empowers business analysts to:

• Visualize Procurement Trends through KPI-driven dashboards.

• Identify Top Performing Brands and Products.

• Monitor Sales Patterns and Category Performance.

• Leverage Machine Learning (ML) tasks for forecasting, segmentation, and anomaly detection — all within a single self-service reporting environment.

Key Highlights

Category	Description
Goal	Build a comprehensive Self-Service Report for retail purchase, sales, and insights.
Data Source	Retail Store Purchase and Sales Data uploaded to the Data Sandbox.
Modules Used	Data Center and Report modules.
Key Features	KPIs, interactive charts, ML View (Segmentation, Time Series, Anomaly Detection).
Outcome	A fully interactive, analytics-driven reporting workspace with intelligent forecasting.

Workflow Overview

Step	Module	Description
1	Data Center	Upload Retail Purchase and Sales data to the Data Sandbox.
2	Data Center	Create Datastores for Purchase and Sales data.
3	Report Module	Create a Self-Service Report for Purchase data.
4	Report Module	Create a Self-Service Report for Sales data.
5	Report Module	Explore Machine Learning (ML) features for deeper insights.

Step 1 – Upload Data to the Data Sandbox

1. Navigate to the App Menu → Data Center.

2. Click Create (+) → select Data Sandbox.

3. Provide the following details:

   • Sandbox Name: Retail_Store_Data

   • Description: Contains retail store purchase and sales records

4. Upload your Excel or CSV files:

   • Retail_Store_Purchase_Data.xlsx

   • Retail_Store_Sales_Data.xlsx

5. Click Upload to complete.

6. Once uploaded, both datasets will appear in the Sandbox list.

Step 2 – Create Datastores for Purchase and Sales Data

1. In the Data Center, click the three dots (⋮) beside the Retail Sandbox.

2. Select Create Data Store.

3. Create two separate Datastores:

   • Retail_Purchase_Store

   • Retail_Sales_Store

4. Verify that both Datastores are active and available under the Data Store list.

• Result: Two Datastores are now available for report creation — one for purchases, one for sales.

Step 3 – Create a Self-Service Report for Purchase Data

1. Navigate to the App Menu → Report Module.

2. Click Create Report.

3. Provide:

   • Report Name: Product Procurement Overview

   • Description: (Optional) A report visualizing procurement trends and top-purchased products.

4. Select Existing Data Store as the data source.

5. Choose the Retail_Purchase_Store and click Create Report.

3.1 Create KPI Tiles

1. Search for Purchase Amount and plot a KPI tile showing the Total Purchase Amount.

2. Search for Quantities and plot a KPI tile showing the Total Quantity Purchased.

3. Use Chart Properties to adjust:

   • Color Theme (e.g., gradient green or blue)

   • Font Size and KPI Format (e.g., Currency, Numeric).

4. Save the KPI tiles.

3.2 Create Brand and Product Insights Charts

1. Search for Brand Name and Purchase Amount.

   • Create a Bar Chart showing the Most Preferred Brand by purchase amount.

   • Customize chart colors and label rotation under Chart Properties.

2. Search for Product Name and Purchase Amount.

   • Create a Column Chart showing Top Purchased Products.

   • Apply data sorting to show products in descending order of purchase value.

3. Add more KPIs as needed, such as:

   • Average Purchase Value

   • Total Unique Products Purchased.

4. Once complete, save the report as “Product Procurement Overview.”

Step 4 – Create a Self-Service Report for Sales Data

1. In the Report Module, click Create Report again.

2. Provide:

   • Report Name: Product Sale Overview

   • Description: (Optional) Visualizing sales trends and category performance.

3. Select Existing Data Store → Retail_Sales_Store.

4. Click Create Report.

4.1 Create Sales KPIs

1. Search for Billing Date and Net Sale Value.

   • Plot a KPI tile showing the Net Value of Goods Sold.

2. Search for Billing Date and Quantities.

   • Plot a KPI tile showing the Total Quantities Sold

4.2 Create Category Analysis Charts

1. Search for Product Categories and Net Sale Value.

   • Plot a Bar Chart titled “Most Valued Categories.”

   • Use Charting Properties → Sort → Descending to show top performers.

2. Create another chart using the same fields.

   • Change Sort → Ascending and title it “Least Valued Categories.”

3. Apply consistent visual themes and axis formatting across all charts.

4. Save the completed report as “Product Sale Overview.”

Step 5 – Create the ML View for Insights

The Machine Learning (ML) View enables users to apply predictive analytics directly within the report — without writing code. It supports four ML tasks:

• Customer Segmentation

• Anomaly Detection

• Time Series Analysis

• Sentiment Analysis

5.1 Customer Segmentation

1. Search for the following fields:

   • City, Net Sale Value, Basic Cost, Tax Amount, Discount.

2. Click the ML Tab → Customer Segmentation task.

3. Configure parameters:

   • Feature Columns: Select all the fields listed above.

   • Cluster Count: 3 (or as per requirement).

   • Operator: K-Means Clustering.

4. Click Predict to visualize customer segments.

• Result: The chart groups customers into clusters based on purchase patterns.

5.2 Time Series Forecasting

1. Search for Billing Date and Net Sale Value.

2. Create a chart visualizing sales trends over time.

3. Click ML Tab → Time Series Analysis.

4. Configure:

   • Feature Column: Net Sale Value

   • Prediction Count: 10 (Next 10 Days)

5. Click Predict.

• Result: The model forecasts sales for the next 10 days based on historical data.

5.3 Anomaly Detection

1. Search for Store List and Quantities.

2. Create a chart displaying quantity distribution per store.

3. Click ML Tab → Anomaly Detection.

4. Configure:

   • Feature Column: Quantities

   • Store Filter: Wonders

5. Click Predict.

• Result: The system highlights stores with abnormally high quantities as anomalies.

5.4 Sentiment Analysis (Optional)

1. Upload a Customer Review Dataset (if available).

2. Select Review Text as the feature column.

3. Run Sentiment Analysis under the ML tab.

4. Review polarity outputs (Positive, Neutral, Negative) as insights.

Step 6 – Finalize and Save the Insights Report

1. Review all KPIs, charts, and ML results.

2. Click Theme Settings and apply a visual theme for branding consistency (e.g., Retail Blue or Dark Mode).

3. Provide:

   • Final Report Name: Retail Insights Dashboard

   • Description: (Optional) Integrated overview of Purchase, Sales, and ML insights.

4. Click Save Report.

• Final Output: Three separate but linked reports:

  • Product Procurement Overview

  • Product Sale Overview

  • Retail Insights (ML View)

Outcome

• Two self-service reports created using Retail Purchase and Sales Datastores.

• Insights visualized through KPI Tiles, Bar, and Column Charts.

• Advanced analytics achieved using ML View (Segmentation, Forecasting, Anomaly Detection).

• Unified Retail Insights Dashboard enabling data-driven decision-making.

Best Practices

• Use consistent column names across purchase and sales data.

• Regularly update Sandbox data to reflect recent transactions.

• For forecasting, ensure at least 30 days of historical data.

• Save custom chart views for reuse across reports.

• Apply color-coded themes for better visual distinction (e.g., blue for sales, green for purchase).

• Validate all KPI metrics before publishing reports.

Business Value

This workflow helps retail teams:

• Monitor procurement and sales KPIs in one consolidated view.

• Predict future sales and detect operational anomalies.

• Segment customers for targeted marketing and loyalty programs.

• Enable self-service analytics without technical dependencies.

• Enhance operational agility with real-time data insights.