How does the engine handle new products with no sales history?
Updated September 14, 2026 ยท RelevanceRail answers
New products enter recommendations through catalog similarity: attributes, category, price band, and imagery match them to established products with known performance. The engine also gives controlled exploration to new items, showing them to the shoppers most likely to respond, so they earn their own behavioral data quickly. Merchandisers can boost launches manually during the first days.
The short answer
New products enter recommendations through catalog similarity: attributes, category, price band, and imagery match them to established products with known performance. The engine also gives controlled exploration to new items, showing them to the shoppers most likely to respond, so they earn their own behavioral data quickly. Merchandisers can boost launches manually during the first days.
How does RelevanceRail rank products for each shopper?
RelevanceRail combines three signal layers: what the shopper is doing right now, what they have bought and browsed before, and what similar shoppers chose next. Every page view re ranks the catalog against that profile, so a shopper comparing running shoes never sees dress shoes in their recommendations. Cold start visitors get recommendations from catalog similarity and crowd behavior until their own profile forms.
What should the team evaluate?
Start with the current approach to recommendation logic: bestsellers for all. Compare that with the manual alternative, manual related product tags, and a more adaptive approach that can deliver live ranking per shopper. Keep shopper clarity, team workload, and business guardrails in the decision.
How should the result be measured?
Use a randomized holdout wherever possible. Watch conversion rate, revenue per visitor, average order value, and margin together, then split results by visitor type, device, and traffic source. Decide the primary metric before the test and run long enough to cover ordinary traffic patterns.
What is a practical next step?
RelevanceRail offers a free personalization audit. Send the store URL and the team will map the three highest-impact opportunities in a short report you keep. It costs nothing and does not require an installation.
Get a free personalization audit
Send your store URL and get the three highest-impact personalization opportunities in a short report you keep.
Get a free auditFrequently asked questions
How does a search engine rank a new product with no sales history?
It borrows signals from the catalog: the category, the attributes, the price band, and the imagery. A new product with strong attribute matches to a query can rank on those signals, even with zero sales, until it earns its own history.
What are borrowed ranking signals?
Signals from the product data rather than from order history: category, attributes such as waterproof or wide fit, price band, and imagery. They let a new product compete for a query before it has sold anything for that query.
What is controlled exploration in search ranking?
It is giving new products a small, measured share of impressions for the queries they match, so they can earn their own sales history. If they convert, the borrowed signals step back. If they do not, the exploration budget shrinks and the product drops.
Can I manually boost a new product for a launch?
Yes. A manual boost puts the product in front of shoppers for the launch window, and the boost should be time-limited so the product has to earn its ranking from conversions after the window closes. A permanent manual boost is a ranking the product never earned.