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

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Price Negotiation Agent

Price Negotiation Agent

Author: Venkata Sudhakar

Price negotiation is common in Indian e-commerce, especially for bulk buyers and loyal customers. A ShopMax India price negotiation agent evaluates each discount request against product margin thresholds, the customer loyalty tier, competitor pricing, and current stock levels - then either approves the discount, issues a counter-offer, or declines with a reason. This removes the need for manual approval on routine negotiation requests.

The agent uses four tools: fetch product margin data, get customer loyalty tier, check competitor prices, and compute the negotiation decision. The model applies business rules and generates a natural, respectful response whether approving or declining.

from google.adk.agents import LlmAgent
from google.adk.tools import FunctionTool
from google.adk.sessions import InMemorySessionService
from google.adk.runners import Runner
from google.genai.types import Content, Part

def get_product_margin(sku: str) -> dict:
    """Get cost price and margin data for a ShopMax India product."""
    products = {
        "SKU-S24":  {"name": "Samsung Galaxy S24", "mrp_rs": 79999, "cost_rs": 67000,
                     "list_price_rs": 74999, "min_margin_pct": 8},
        "SKU-OP12": {"name": "OnePlus 12",         "mrp_rs": 69999, "cost_rs": 57500,
                     "list_price_rs": 64999, "min_margin_pct": 10},
    }
    return products.get(sku, {"error": "Product not found"})

def get_customer_loyalty_tier(user_id: str) -> dict:
    """Get ShopMax India loyalty tier and max discount authority for a customer."""
    tiers = {
        "u_gold":     {"tier": "Gold",     "lifetime_spend_rs": 250000, "max_discount_pct": 8},
        "u_silver":   {"tier": "Silver",   "lifetime_spend_rs": 80000,  "max_discount_pct": 5},
        "u_new":      {"tier": "Standard", "lifetime_spend_rs": 5000,   "max_discount_pct": 2},
    }
    return tiers.get(user_id, {"tier": "Standard", "max_discount_pct": 2})

def get_competitor_price(sku: str) -> dict:
    """Get current competitor prices for a ShopMax India product."""
    comp = {
        "SKU-S24":  {"amazon_rs": 73499, "flipkart_rs": 74000, "croma_rs": 75999},
        "SKU-OP12": {"amazon_rs": 63999, "flipkart_rs": 64499, "croma_rs": 65000},
    }
    return comp.get(sku, {})

def evaluate_discount_request(sku: str, requested_price_rs: int, list_price_rs: int,
                               cost_rs: int, min_margin_pct: float,
                               max_discount_pct: float, competitor_min_rs: int) -> dict:
    """Evaluate a discount request and return approval, counter-offer, or decline for ShopMax India."""
    requested_discount_pct = round((list_price_rs - requested_price_rs) / list_price_rs * 100, 1)
    min_price_rs = round(cost_rs * (1 + min_margin_pct / 100))
    loyalty_floor_rs = round(list_price_rs * (1 - max_discount_pct / 100))
    # Best we can offer: highest of min margin price and loyalty-gated floor
    best_offer_rs = max(min_price_rs, loyalty_floor_rs)

if requested_price_rs >= list_price_rs:
        return {"decision": "NO_DISCOUNT_NEEDED", "offered_price_rs": list_price_rs}
    elif requested_price_rs >= best_offer_rs:
        return {"decision": "APPROVED", "offered_price_rs": requested_price_rs,
                "discount_pct": requested_discount_pct}
    elif best_offer_rs <= competitor_min_rs:
        return {"decision": "COUNTER_OFFER", "offered_price_rs": best_offer_rs,
                "reason": "Below our minimum margin at requested price"}
    else:
        return {"decision": "DECLINED",
                "reason": "Requested price is below our cost floor and loyalty tier limit",
                "list_price_rs": list_price_rs}

agent = LlmAgent(
    name="shopmax_negotiation_agent",
    model="gemini-2.0-flash",
    tools=[
        FunctionTool(get_product_margin),
        FunctionTool(get_customer_loyalty_tier),
        FunctionTool(get_competitor_price),
        FunctionTool(evaluate_discount_request),
    ],
    instruction=(
        "You are a ShopMax India price negotiation agent. "
        "When a customer requests a discount: get product margin, get customer loyalty tier, "
        "get competitor prices, then evaluate the request. "
        "Be polite and explain the outcome. For counter-offers, highlight the value offered."
    )
)

session_service = InMemorySessionService()
runner = Runner(agent=agent, session_service=session_service, app_name="shopmax")

requests = [
    ("u_gold",   "SKU-S24",  70000, "Can I get the Samsung S24 for Rs 70,000?"),
    ("u_silver", "SKU-OP12", 58000, "Can I get the OnePlus 12 for Rs 58,000?"),
    ("u_new",    "SKU-S24",  60000, "I want the Samsung S24 for Rs 60,000"),
]

for uid, sku, req_price, query in requests:
    session = session_service.create_session(app_name="shopmax", user_id=uid)
    events = list(runner.run(
        user_id=uid,
        session_id=session.id,
        new_message=Content(parts=[Part(text=query)])
    ))
    for event in events:
        if hasattr(event, "content") and event.content:
            for part in event.content.parts:
                if hasattr(part, "text") and part.text:
                    print(f"[{uid}] {part.text.strip()[:200]}")
    print()

It gives the following output,

[u_gold] Great news! As a Gold loyalty member, we can approve Rs 70,000 for
the Samsung Galaxy S24 - that is a 6.7% discount on our list price of Rs 74,999.
This is also lower than Amazon (Rs 73,499) and Flipkart (Rs 74,000). Your price
is confirmed at Rs 70,000.

[u_silver] We are unable to match Rs 58,000 for the OnePlus 12 (our list price
is Rs 64,999). As a Silver member, the best we can offer is Rs 61,749 - a 5%
discount. This beats Amazon at Rs 63,999. Would you like to proceed at Rs 61,749?

[u_new] Thank you for your interest in the Samsung Galaxy S24. Unfortunately,
Rs 60,000 is significantly below our cost floor. As a new customer, we can offer
up to 2% off: Rs 73,499, which matches the best Amazon price today.

The Gold tier customer got their price approved outright (within loyalty limits and margin floor). The Silver customer received a counter-offer at the loyalty tier maximum. The new customer was politely declined with an explanation and a best-available offer. All three responses reference competitor prices to reinforce value.

For ShopMax India production, store negotiation outcomes in BigQuery with columns for user_id, sku, requested_price, offered_price, decision, and whether the customer accepted. Run weekly analysis to find which SKUs have the highest negotiation rates - these may need a price repositioning review. Integrate with the CRM so Gold and Platinum customers who negotiate frequently are flagged for proactive outreach with pre-approved special pricing before they need to ask.

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