$0.0025 Per Product: E-Commerce Data Has a Floor
Bright Data sells 100,000 product records for $250 — a quarter of a cent per SKU. With monthly refresh subscriptions discounted up to 80%, the per-record floor is now well below $0.001. Margin moves to the orchestration and refresh-cadence layer.
The e-commerce dataset market has a public price now. Bright Data’s eCommerce Datasets page lists 100,000 product records for $250 — a quarter of a cent per record on a one-time pull. Tiered up: 500K, 1M, 5M, and 20M+ packages, with monthly-refresh subscriptions discounted up to 80% of the one-time rate.
That is the unit price for raw e-commerce data in 2026. It is published. Anyone can buy at that rate. And it sets a floor that the rest of the market has to negotiate against.
Datafiniti keeps prices private and pitches “scale by volume not feature-gating” — its G2 listing shows no public price. Crawlbase quotes per-engagement. Both are competing against a published, transparent, sub-cent benchmark from the largest player in the segment, and neither has matched the move to public pricing.
That asymmetry is itself the story.
What the rate card actually says
Bright Data’s catalog claims more than 9 billion product records across the major e-commerce surfaces — Amazon, Walmart, Target, eBay, Shopify storefronts, AliExpress. The pricing tiers are linear in volume: 100K at $250, scaling proportionally up to 20M+ at custom enterprise pricing.
The discount structure is the load-bearing half. Refresh cadence is the real pricing axis:
- One-time pull: full price
- Bi-annual refresh: 25% off
- Quarterly refresh: 50% off
- Monthly refresh: up to 80% off
A buyer who commits to monthly refresh on a 1M-record dataset is paying roughly $0.0005 per record per refresh, which is below the marginal cost of running a self-hosted scraper at any reasonable engineering cost. The math says: at this price point, building your own e-commerce scraping infrastructure is a worse deal than buying.
That is the Bright Data thesis on this segment. They have decided the data is commodity, the only meaningful product is refresh cadence and reliability, and the price should be set low enough that build-vs-buy decisively favors buy.
The Filter API: paying for selection, not collection
The other pricing innovation worth flagging is Bright Data’s Filter API, which charges only on the filtered slice rather than the full corpus. A buyer who wants all Walmart skin-care products updated weekly pays for that slice, not for the entire Walmart catalog.
The architectural implication is that the cost is moving from collection to selection. Collection is now a fixed-cost utility. Selection — geo, category, price band, refresh cadence — is the metered surface. Vendors competing in this space have to decide whether to sell the corpus or sell the slice. Bright Data is selling both, with the slice as the higher-margin product.
That model puts pressure on every smaller vendor in the space. If you cannot match Bright Data’s per-record floor on the bulk corpus, you have to win on slice quality, latency, or coverage of long-tail targets. The smaller players (Datafiniti, Crawlbase, the various dropshipping-data vendors) are all trying some combination of those moves.
What the Apify Store data implies
The Apify Store ecosystem hosts a long tail of e-commerce scrapers — Amazon product listings, Walmart, eBay, AliExpress, Shopify storefronts. Per-event pricing on these actors typically ranges from $0.0015 to $0.005 per product record.
That range straddles Bright Data’s published rate. A high-quality Apify Actor priced at $0.003 per product is roughly competitive with Bright Data’s bi-annual refresh tier, and significantly more expensive than the monthly refresh tier. For buyers who need fresh data more often than monthly, Bright Data wins on price. For buyers who need ad-hoc pulls of specific slices that Bright Data does not pre-package, Apify Actors win on flexibility.
The remaining niche is the long tail — Shopify stores, regional marketplaces, vertical-specific catalogs — that Bright Data does not maintain as a packaged dataset. There is real margin in being the publisher of the only available scraper for a niche e-commerce target, and the Apify Store has those publishers. Whether they hold that position when Bright Data eventually packages those targets themselves is the next twelve months’ question.
What this means for buyer strategy
For a buyer evaluating e-commerce data sources in 2026, the read is clear:
- Bulk corpus, common targets: Bright Data wins on price. The published rate card is hard to undercut at scale.
- Niche targets, custom slicing: Apify Actors or specialty vendors win on flexibility.
- Real-time / near-real-time: nobody wins on price. This is custom infrastructure.
- Compliance-sensitive (EU, sensitive personal data): pricing is secondary to vendor compliance posture; the EU AI Act and Article 5 enforcement reshape what’s even legal to collect.
The macro pattern across all four buckets is the same as in lead data: collection is commoditizing, margin is moving to orchestration and selection. The vendors who win the next phase are not the ones with the largest catalog. They are the ones whose product surface is closest to what the buyer actually wants to buy — which is increasingly a filtered slice with a guaranteed refresh cadence, not a raw dump.
Bright Data is the first in this segment to price that explicitly. The rest of the market will follow, or not exist.
Sources