Stock & Price Tracking Methodology
Discover how we compile price points, fetch stock indicators, and maintain accuracy across retailers.
1. Real-time Polling & Scraping Engine
Data fresh from the source is essential for competitive console tracking. Our backend uses a decentralized network of Cloudflare Workers that poll our retail partners continuously.
Each retailer has a custom-written crawler script configured with exact request variables:
- Amazon India: We inspect product listing pages and select JSON data structures returned by the Amazon Product Advertising API (PA-API) to fetch current prices, seller info, and availability flags.
- Flipkart: We target product schemas and API endpoints used by Flipkart's shopping application, detecting the presence of the "Add to Cart" button and the active merchant profile.
- Croma: Our engine directly calls Croma's inventory status endpoint using postcode validation parameters to determine localized stock.
- ShopAtSC: We monitor the Shopify inventory status database attached to Sony India's official direct showroom portal.
2. Postcode Validation & Localized Stock Checks
In India, stock availability is not universal. E-commerce companies utilize localized warehouse shipping hubs. A PS5 Slim might show "In Stock" in Mumbai, but show "Out of Stock" if checked from a postcode in Kolkata.
To manage this variance, our polling engine runs queries using standard representative postal codes across different Indian zones:
| Zone | Pincode Check | City Focus | Fulfillment Hubs Checked |
|---|---|---|---|
| West | 400001 | Mumbai | Bhiwandi / Thane Central Hubs |
| North | 110001 | Delhi (NCR) | Gurugram / Noida Logistics parks |
| South | 560001 | Bengaluru | Nelamangala / Kolar Warehouses |
| East | 700001 | Kolkata | Dankuni Fulfillment Warehouses |
If a product is in stock in any major zone, we mark it as "IN_STOCK" on the main list but surface city-specific tags in the detailed views.
3. Data Confidence Scores & Data Tiers
Not all stock sightings have the same level of accuracy. To give you the best picture, we assign every sighting a confidence tier based on its source:
🟢 High Confidence (95% - 100%) Direct API
Data retrieved directly from official retailer catalog APIs or verified affiliate data feeds. This is automatically confirmed by our servers and requires no manual approval.
🔵 Medium Confidence (80% - 94%) Community Verified
A stock drop reported by a site user that has received 5 or more upvotes from verified IP addresses within the community.
🟡 Low/Unverified Confidence (0% - 79%) New Report
User-submitted stock reports that have not yet passed voting verification or older reports that may have gone stale.
4. Fallback Safeguards & Rate Limits
E-commerce storefronts periodically introduce script changes, change HTML structure, or temporarily block checking agents. If our scraper encounters a network failure or fails to parse a page, we initiate our fallback safeguard:
- Automatic Retry: The engine retries the request using a secondary IP address from a different routing node.
- Soft-State Maintenance: If the retry also fails, the status flag stays set to the last known verified state (instead of showing out of stock) for up to 30 minutes, preventing false alerts.
- Error Notification: If a crawler fails for more than 5 consecutive cycles, our system triggers an internal alert to our dev team for immediate script adjustment.