Databricks AI Segmentation → Salesforce Data Cloud

This repository contains a lightweight Databricks / PySpark notebook that demonstrates how customer segmentation models can be operationalized within Salesforce Data Cloud and Agentforce.

---

🎯 Scenario: Customer Segmentation for AI-Driven Engagement

A marketing team wants to identify high-value and at-risk customers using transaction data processed in Databricks.
The enriched dataset is exported to Salesforce Data Cloud for downstream activation in CRM Analytics and Agentforce.

---

🧱 Architecture

Databricks (PySpark DataFrames) → Feature Engineering → Segment CSV → Salesforce Data Cloud (DMO) → CRM Analytics → Agentforce Actions

---

🧩 Repository Contents

Path	Description
notebooks/segmentation_notebook.ipynb	Jupyter/Databricks notebook that computes customer RFM-style segments
data/sample_transactions.csv	Synthetic dataset for demo
images/architecture_diagram.png	Visualization of architecture
docs/demo_story.md	Full enablement context and storytelling guide

---

🧪 Key Demo Steps

1. Load sample transactions in Databricks or local PySpark.
2. Aggregate by customer and compute metrics (total spend, recency, frequency).
3. Label segments (High/Medium/Low value).
4. Export results as segments.csv.
5. Ingest into Salesforce Data Cloud and visualize in CRM Analytics.

---

💬 Demo Talking Points

• Shows integration of external AI/ML pipelines with Data Cloud.
• Demonstrates how DataFrames (PySpark/pandas) fit into Salesforce architecture.
• Storytelling bridge for SEs: external data → unified Data Cloud profile → AI-driven actions.

---

🧠 Intended Audience

Salesforce Solution Engineers, Data Cloud & AI practitioners, and community learners building Data Cloud–AI demo flows.

---

📚 Related Reading

• Dynamic Dashboard Filters in CRM Analytics + Data Cloud (LinkedIn Article)

---

📄 License

MIT © Lakshmi Achary
(Sample enablement demo — not affiliated with Salesforce.)