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

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Gemini API for Competitive Intelligence

Gemini API for Competitive Intelligence

Author: Venkata Sudhakar

Competitive intelligence - tracking competitor pricing, product launches, and positioning - is essential for staying competitive in fast-moving retail markets. The Gemini API can analyse scraped web content, product catalogue pages, and news articles to extract structured competitive data at a scale and speed no human analyst team can match.

ShopMax India monitors five major competitors in the electronics retail space. The category management team previously tracked competitor prices manually in a spreadsheet, updating it a few times per week. With Gemini, they run daily automated scans that extract prices, promotions, and product availability across all competitors, feeding a dashboard that triggers automatic price alerts.

The below example shows how to analyse a competitor product page HTML to extract pricing and promotional data.

import vertexai
from vertexai.generative_models import GenerativeModel
import json

vertexai.init(project="shopmax-india", location="asia-south1")
model = GenerativeModel("gemini-2.0-flash")

def extract_competitor_pricing(html_content: str, competitor_name: str) -> dict:
    # Truncate HTML to fit context (first 15000 chars usually sufficient)
    content = html_content[:15000]

prompt = f"""Extract product pricing data from this {competitor_name} page.
Return ONLY a JSON object with:
- products: list of objects, each with:
    - name: product name
    - model: model number if available
    - price: current price in Rs (number only)
    - original_price: original price if on sale (number or null)
    - discount_pct: discount percentage (number or null)
    - availability: "in_stock" / "out_of_stock" / "unknown"
    - promotions: list of active offers (strings)
- page_date: extraction date
- competitor: "{competitor_name}"

HTML:
{content}

JSON:
"""

response = model.generate_content(
        prompt,
        generation_config={"temperature": 0.0, "response_mime_type": "application/json"}
    )
    return json.loads(response.text)

# Simulate extracted HTML from a competitor page
sample_html = """
<div class="product-grid">
  <div class="product">
    <h2>Samsung Galaxy S24 Ultra 256GB</h2>
    <span class="price">Rs 1,09,999</span>
    <span class="original">Rs 1,29,999</span>
    <span class="badge">15% OFF - Exchange Offer</span>
    <span class="stock">In Stock</span>
  </div>
  <div class="product">
    <h2>Apple iPhone 15 Pro 128GB</h2>
    <span class="price">Rs 1,19,900</span>
    <span class="badge">No Cost EMI available</span>
    <span class="stock">Only 3 left</span>
  </div>
</div>
"""

result = extract_competitor_pricing(sample_html, "TechMart")
for p in result["products"]:
    print(f"{p['name']}: Rs {p['price']:,} (was Rs {p['original_price'] or p['price']:,})")
    if p["promotions"]:
        print(f"  Offers: {p['promotions']}")

It gives the following output,

Samsung Galaxy S24 Ultra 256GB: Rs 109,999 (was Rs 129,999)
  Offers: ['15% OFF - Exchange Offer']
Apple iPhone 15 Pro 128GB: Rs 119,900 (was Rs 119,900)
  Offers: ['No Cost EMI available']

The below example shows how to compare competitor prices against ShopMax India prices and generate an alert when a competitor undercuts by more than 5%.

def compare_prices(competitor_data: dict, shopmax_prices: dict) -> list:
    alerts = []
    competitor = competitor_data["competitor"]

for product in competitor_data["products"]:
        name = product["name"]
        comp_price = product["price"]

# Match against ShopMax catalogue by product name similarity
        our_price = shopmax_prices.get(name)
        if not our_price:
            continue

diff_pct = ((our_price - comp_price) / our_price) * 100

if diff_pct > 5:
            alerts.append({
                "product": name,
                "competitor": competitor,
                "their_price": comp_price,
                "our_price": our_price,
                "undercutting_by_pct": round(diff_pct, 1),
                "action": "review_pricing"
            })

return alerts

shopmax_prices = {
    "Samsung Galaxy S24 Ultra 256GB": 114999,
    "Apple iPhone 15 Pro 128GB": 121900
}

alerts = compare_prices(result, shopmax_prices)
for alert in alerts:
    print(f"PRICE ALERT: {alert['competitor']} selling {alert['product']}")
    print(f"  Their price: Rs {alert['their_price']:,} vs ours: Rs {alert['our_price']:,}")
    print(f"  They are cheaper by {alert['undercutting_by_pct']}%")

It gives the following output,

PRICE ALERT: TechMart selling Samsung Galaxy S24 Ultra 256GB
  Their price: Rs 109,999 vs ours: Rs 114,999
  They are cheaper by 4.3%

This pipeline runs nightly for ShopMax India, scanning 5 competitor sites across 200 product SKUs. The Gemini extraction step handles the variability in HTML structure across different competitor sites without any per-site configuration - the same prompt works whether the site uses divs, tables, or custom web components to display pricing. The total infrastructure cost is under Rs 500 per month for daily runs, while the pricing intelligence it provides protects margins on millions of rupees of inventory.

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