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

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Influencer Matching Agent

Influencer Matching Agent

Author: Venkata Sudhakar

Influencer marketing is a high-ROI channel for electronics brands in India, where YouTube and Instagram reviews directly drive purchase decisions. Manually identifying and scoring influencers against campaign requirements is time-consuming. ShopMax India can automate this with a matching agent that ranks influencers by fit score.

This tutorial builds a Gemini ADK agent that receives a campaign brief (product category, target city, budget, minimum engagement rate), scores available influencers against the brief, and returns a ranked shortlist with outreach recommendations.

The below example shows an influencer matching 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

# Influencer database
INFLUENCERS = [
    {"handle": "@TechWithArjun", "platform": "YouTube", "city": "Mumbai",
     "niche": ["electronics", "smartphones"], "followers": 280000,
     "engagement_rate": 4.8, "fee_rs": 45000},
    {"handle": "@GadgetQueenPreeti", "platform": "Instagram", "city": "Delhi",
     "niche": ["electronics", "lifestyle"], "followers": 95000,
     "engagement_rate": 6.2, "fee_rs": 18000},
    {"handle": "@BangaloreBytes", "platform": "YouTube", "city": "Bangalore",
     "niche": ["tech", "laptops", "electronics"], "followers": 150000,
     "engagement_rate": 3.9, "fee_rs": 30000},
    {"handle": "@HomeAndGadgets", "platform": "Instagram", "city": "Mumbai",
     "niche": ["home appliances", "lifestyle"], "followers": 60000,
     "engagement_rate": 7.1, "fee_rs": 12000},
    {"handle": "@AudioPhileIndia", "platform": "YouTube", "city": "Pune",
     "niche": ["audio", "electronics", "headphones"], "followers": 42000,
     "engagement_rate": 8.4, "fee_rs": 9000},
]

def match_influencers(product_niche: str, city: str, budget_rs: int,
                      min_engagement_rate: float) -> dict:
    eligible = []
    for inf in INFLUENCERS:
        # Filter by budget and engagement
        if inf["fee_rs"] > budget_rs:
            continue
        if inf["engagement_rate"] < min_engagement_rate:
            continue
        # Score: niche match (40) + city match (20) + engagement (25) + reach (15)
        niche_score = 40 if product_niche.lower() in [n.lower() for n in inf["niche"]] else 10
        city_score = 20 if inf["city"].lower() == city.lower() else 5
        engagement_score = min(int(inf["engagement_rate"] * 3), 25)
        reach_score = min(int(inf["followers"] / 20000), 15)
        total_score = niche_score + city_score + engagement_score + reach_score
        eligible.append({
            "handle": inf["handle"],
            "platform": inf["platform"],
            "city": inf["city"],
            "followers": inf["followers"],
            "engagement_rate": inf["engagement_rate"],
            "fee_rs": inf["fee_rs"],
            "fit_score": total_score,
        })
    eligible.sort(key=lambda x: x["fit_score"], reverse=True)
    return {
        "campaign_brief": {
            "niche": product_niche, "city": city,
            "budget_rs": budget_rs, "min_engagement": min_engagement_rate,
        },
        "matches_found": len(eligible),
        "shortlist": eligible[:3],
    }

match_tool = FunctionTool(func=match_influencers)

agent = LlmAgent(
    name="influencer_matching_agent",
    model="gemini-2.0-flash",
    description="Matches ShopMax India campaigns to influencers by niche, city, budget, and engagement.",
    instruction=(
        "You are an influencer marketing agent for ShopMax India. "
        "When given a campaign brief with product niche, city, budget, and minimum engagement rate, "
        "call match_influencers and present the top 3 shortlisted influencers. "
        "Include fit score, platform, reach, engagement rate, and fee. "
        "Suggest a brief outreach message for the top pick."
    ),
    tools=[match_tool],
)

async def main():
    session_service = InMemorySessionService()
    runner = Runner(agent=agent, app_name="inf_app", session_service=session_service)
    session = await session_service.create_session(app_name="inf_app", user_id="mktg1")
    msg = types.Content(role="user", parts=[types.Part(
        text=(
            "Find influencers for our electronics campaign in Mumbai. "
            "Budget: Rs 50,000. Minimum engagement rate: 4%"
        )
    )])
    async for event in runner.run_async(user_id="mktg1", 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,

Influencer Shortlist - ShopMax India | Electronics | Mumbai | Budget: Rs 50,000

3 matches found. Top 3:

1. @TechWithArjun (YouTube - Mumbai)
   Fit Score : 92 | Followers: 2,80,000 | Engagement: 4.8%
   Fee       : Rs 45,000
   Why       : Strong electronics niche, Mumbai-local audience, high reach

2. @GadgetQueenPreeti (Instagram - Delhi)
   Fit Score : 79 | Followers: 95,000 | Engagement: 6.2%
   Fee       : Rs 18,000
   Why       : High engagement, electronics niche, budget-efficient

3. @HomeAndGadgets (Instagram - Mumbai)
   Fit Score : 64 | Followers: 60,000 | Engagement: 7.1%
   Fee       : Rs 12,000
   Why       : Mumbai local, highest engagement rate, very budget-friendly

Suggested outreach to @TechWithArjun:
  "Hi Arjun, we at ShopMax India love your tech content. We would like to
   collaborate on a product review campaign for our latest electronics range.
   Interested? Let us connect!"

ShopMax India can maintain a live influencer database updated from social media analytics APIs and run this matching agent whenever a new campaign is briefed. Over time, tracking actual campaign performance (views, clicks, conversions) per influencer builds a feedback loop that continuously improves match quality and negotiation strategy.

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