The Same Product, Listed Twice
Catalogs collect duplicates: a product re-created with a new SKU, a variant entered as its own item, a supplier sending the same product under a new code. Each one splits stock, reviews and sales, and none of them shows up in an exact-match check.
Three listings for one product. Shoppers see different prices and split reviews, and the warehouse counts three items.
Duplicates Hide In Plain Sight
Duplicate SKUs rarely look like duplicates. One says “1L”, another “1 Litre”, a third “33.8 fl oz”. One puts the brand in capitals, another abbreviates the product type. One was created by a supplier upload, another by hand. An exact match on title or code finds none of them.
The cost is spread across the business. Stock is split, so one listing sells out while its twin sits unsold. Reviews and ratings are divided, so a strong product looks unproven. Search shows the same item several times at different prices, and every report counts the product more than once.
Manual cleanup is slow and it never ends. Catalog teams compare listings by eye, a few hundred at a time, while new duplicates arrive with every supplier feed.
Split Stock
One listing sells out while its duplicate holds the inventory.
Divided Reviews
Ratings are spread across copies, so good products look untested.
Cluttered Search
Shoppers see the same product several times at different prices.
Find Duplicates By Meaning, Not Spelling
Compare listings on what they describe, not how they are written. Normalise units and formats, ignore filler words, and match on the combination of brand, product type, size, variant and any codes that are present.
Group duplicates under one ID and decide once which listing survives. Then check every new listing and supplier feed against the existing catalog before it goes live, so duplicates stop arriving.
1 L and 33.8 fl oz are the same size. The match does not depend on how the unit is written.
Normalise First
Units, case and formats are standardised before listings are compared.
Group Under One ID
All copies of a product share one ID, so merging is a single decision.
Stop New Duplicates
Incoming listings are checked against the catalog before they go live.
The Identity Layer Under Product 360
Zingg runs as a workload inside your data platform and gives every entity one persistent ZINGG_ID. Everything above it reads the same resolved entities.
The same layer powersCustomer Analytics & MarketingFinancial Crime & RiskPrivacy & ComplianceSupply Chain & ProcurementMaster Data ManagementAI & Data Foundations
Deduplicate The Catalog With Zingg
Trained On Your Catalog
Label a few pairs of your own listings and Zingg learns what a duplicate means for your products. How Zingg matches
Stopwords And Standardization
Remove filler words and standardize fields so formatting differences do not hide duplicates.
Check Every New Listing
The incremental flow matches new and updated items against existing products. Incremental flow
See Why Items Matched
Review the evidence for each group before merging. Explainability
Catalog Deduplication Questions
Why don’t exact matches find duplicate SKUs?
Duplicates usually differ in how the title, units or brand are written, and codes are often missing or different. Exact matching only finds copies that are identical.
Does Zingg merge or delete listings for me?
No. Zingg groups duplicates under one ZINGG_ID and writes the result to your data platform. You decide which listing survives and apply it in your catalog.
Can it handle variants like sizes and colours?
Yes. You label examples of what counts as the same product and what counts as a different variant, and Zingg learns that distinction from your data.
How large a catalog can it handle?
Millions of items. Zingg learns which records are worth comparing, so it does not compare every listing with every other.
A Catalog With No Duplicates
Find duplicate listings across your catalog, in your own data platform.