	In Browser
	StumbleUpon
	del.icio.us
	Google
	Google Buzz
	reddit
	LinkedIn

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Customer Lifetime Value Agent

Customer Lifetime Value Agent

Author: Venkata Sudhakar

Customer Lifetime Value (CLV) measures the total revenue a customer is expected to generate over their relationship with a business. For ShopMax India, knowing which customers have the highest CLV enables the retention team to invest in loyalty programmes, personalised offers, and priority service where it matters most.

This tutorial builds a Gemini ADK agent that calculates historical CLV from purchase records, projects future value using a simple recency-frequency-monetary model, and segments customers into Platinum, Gold, Silver, and At-Risk tiers.

The below example shows a customer lifetime value segmentation agent for ShopMax India in a business context.

import asyncio
from google.adk.agents import LlmAgent
from google.adk.tools import FunctionTool
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService
from google.genai import types

# Customer purchase history summary
CUSTOMERS = [
    {"id": "C001", "name": "Anita Sharma", "city": "Mumbai",
     "total_orders": 12, "total_spend_rs": 185000,
     "avg_order_value_rs": 15417, "days_since_last_order": 8},
    {"id": "C002", "name": "Rajan Pillai", "city": "Bangalore",
     "total_orders": 6, "total_spend_rs": 72000,
     "avg_order_value_rs": 12000, "days_since_last_order": 35},
    {"id": "C003", "name": "Priya Desai", "city": "Delhi",
     "total_orders": 2, "total_spend_rs": 18000,
     "avg_order_value_rs": 9000, "days_since_last_order": 120},
    {"id": "C004", "name": "Suresh Kumar", "city": "Chennai",
     "total_orders": 18, "total_spend_rs": 310000,
     "avg_order_value_rs": 17222, "days_since_last_order": 3},
]

def calculate_clv() -> dict:
    results = []
    for c in CUSTOMERS:
        # Recency score: lower days = higher score (max 40)
        if c["days_since_last_order"] <= 14:
            recency_score = 40
        elif c["days_since_last_order"] <= 30:
            recency_score = 30
        elif c["days_since_last_order"] <= 60:
            recency_score = 20
        else:
            recency_score = 5
        # Frequency score (max 30)
        frequency_score = min(c["total_orders"] * 2, 30)
        # Monetary score (max 30)
        monetary_score = min(int(c["total_spend_rs"] / 10000), 30)
        rfm_score = recency_score + frequency_score + monetary_score
        # Project next 12 months
        avg_orders_per_year = c["total_orders"] / 2  # assume 2 year history
        projected_clv = int(avg_orders_per_year * c["avg_order_value_rs"])
        if rfm_score >= 80:
            tier = "Platinum"
        elif rfm_score >= 60:
            tier = "Gold"
        elif rfm_score >= 40:
            tier = "Silver"
        else:
            tier = "At-Risk"
        results.append({
            "customer_id": c["id"],
            "name": c["name"],
            "city": c["city"],
            "total_spend_rs": c["total_spend_rs"],
            "rfm_score": rfm_score,
            "tier": tier,
            "projected_12m_clv_rs": projected_clv,
        })
    results.sort(key=lambda r: r["rfm_score"], reverse=True)
    return {"customers": results, "total_analysed": len(results)}

clv_tool = FunctionTool(func=calculate_clv)

agent = LlmAgent(
    name="clv_agent",
    model="gemini-2.0-flash",
    description="Calculates and segments customer lifetime value for ShopMax India.",
    instruction=(
        "You are a customer analytics agent for ShopMax India. "
        "When asked for a CLV report, call calculate_clv. "
        "Present customers grouped by tier (Platinum, Gold, Silver, At-Risk). "
        "Include RFM score, projected 12-month CLV, and a brief retention recommendation per tier."
    ),
    tools=[clv_tool],
)

async def main():
    session_service = InMemorySessionService()
    runner = Runner(agent=agent, app_name="clv_app", session_service=session_service)
    session = await session_service.create_session(app_name="clv_app", user_id="analytics1")
    msg = types.Content(role="user", parts=[types.Part(
        text="Run CLV analysis and segment customers by tier"
    )])
    async for event in runner.run_async(user_id="analytics1", session_id=session.id, new_message=msg):
        if event.is_final_response() and event.content:
            print(event.content.parts[0].text)

asyncio.run(main())

It gives the following output,

Customer Lifetime Value Report - ShopMax India

[PLATINUM]
  Suresh Kumar (Chennai) | RFM: 91 | Total Spend: Rs 3,10,000
  Projected 12m CLV: Rs 1,54,998 | Action: Assign dedicated relationship manager

  Anita Sharma (Mumbai)  | RFM: 82 | Total Spend: Rs 1,85,000
  Projected 12m CLV: Rs 92,502   | Action: Early access to new product launches

[GOLD]
  Rajan Pillai (Bangalore) | RFM: 62 | Total Spend: Rs 72,000
  Projected 12m CLV: Rs 36,000   | Action: Loyalty points double-up offer

[AT-RISK]
  Priya Desai (Delhi)    | RFM: 29 | Total Spend: Rs 18,000
  Projected 12m CLV: Rs 9,000    | Action: Win-back campaign with 20% discount

ShopMax India can run this CLV analysis monthly and feed the tier assignments into the CRM to trigger automated retention workflows - early access emails for Platinum customers, loyalty bonus notifications for Gold, and win-back campaigns for At-Risk customers who have not ordered in over 90 days.

Send your comments, suggestions or queries regarding this site to [email protected].