Cross-platform pricing
Coming soon
The Cross-Platform sub-score is one of the six inputs to the SellerScore. Today it’s a neutral placeholder (always around 50) because the upstream data source — Crawlee, Hilal’s web-scraping microservice — is stubbed.
Cross-platform pricing flips that sub-score from placeholder to real signal.
What this will let you do
- See AliExpress / Alibaba pricing for the product in the same view, alongside its Amazon price. The gap between the two is your sourcing leverage.
- See TikTok Shop pricing for products that exist there. TikTok Shop is increasingly Amazon-adjacent for impulse / younger-skewing categories.
- See Google Shopping competitive listings beyond Amazon. Useful for understanding the category outside the Amazon bubble.
- Filter and weight by cross-platform leverage in the search form. Find products specifically where the Amazon price is high relative to AliExpress.
- Make sourcing decisions faster. Instead of opening a search tab to AliExpress, the data is in the same view.
How it’ll work (preview)
A separate Crawlee microservice (Run 3 in the engineering plan) will be deployed alongside Hilal’s main backend. When you run a search:
- Standard scoring runs first — the SellerScore is computed with five real sub-scores and the placeholder for cross-platform.
- Crawlee fans out in parallel — for the top results, hits AliExpress, TikTok Shop, and Google Shopping for the equivalent product or category.
- Cross-platform sub-score is computed from the gathered data — a higher score for products with bigger price gaps (more sourcing leverage).
- The composite SellerScore is recomputed with the real cross-platform data.
The whole thing should add 5–15 seconds to a search depending on how aggressive the Crawlee fan-out is.
Why scraping is hard
Crawlee is gated behind a feature flag (CRAWLEE_ALIEXPRESS_ENABLED) and is not on for any user today. Reasons:
- Proxy pool sizing. Reliable scraping at scale needs a robust proxy pool; we’re sizing this carefully.
- Anti-scraping countermeasures. AliExpress, TikTok Shop, and Google Shopping all push back on automated access. Our crawlers need to handle the countermeasures gracefully.
- Data quality. Scraped data is messier than first-party APIs. We’re tuning the parsers and confidence scoring.
- Cost. Each Crawlee request has a cost; we’re modeling how the credit price needs to reflect this.
How the search experience will change
You’ll see:
- Cross-platform sub-score chip turn from grey placeholder to a real 0–100 number.
- A new section on the product detail page showing the lowest AliExpress / TikTok Shop / Google Shopping price next to the Amazon price.
- A “sourcing leverage” indicator — the gap, expressed as a multiplier (e.g., “Amazon 3.5× AliExpress”).
- AI brief mentions cross-platform in the sourcing notes section when the gap is significant.
- Filter / weight options for cross-platform-driven research.
Status
- Crawlee microservice scaffold: ready, deployed to staging.
- AliExpress integration: active development, behind feature flag.
- TikTok Shop integration: planned in the same Run.
- Google Shopping integration: planned in the same Run.
- Frontend wiring: part of Run 4.
- Beta: rollout planned alongside the rest of Run 3, behind per-org feature flag at first.
What you can do today to prepare
- Don’t over-trust the cross-platform sub-score in your weights. A weight of 25%+ on cross-platform makes today’s SellerScore unreliable since the sub-score is a placeholder. Until Crawlee ships, leave the cross-platform weight at the default 10% or set it to zero.
- For sourcing today, do the AliExpress lookup manually and keep notes in your watchlist. Once Crawlee ships, you can compare your manual data against Hilal’s automated data.