Amazon is building a database of every product sold online

Amazon is building a database to log every product sold online—including those it doesn’t sell. Called the World Product Graph, the project aims to feed Amazon’s AI shopping tools and could make Google or Reddit searches redundant for millions of shoppers. An internal document points to a 2027 launch, but Amazon says the timeline is still aspirational.

Tarun Mottlia

Editor, Lazyfounder

Published Nov 16, 20263 min read

Amazon is building a database of every product sold online — Image: Amazon, Crate & Barrel) via source

30 SEC SUMMARY

Amazon is building a database—World Product Graph—to catalog every product sold online, including rivals’ items, aiming to strengthen its AI shopping tools.
The project could cut shoppers’ reliance on Google or Reddit for product details by centralizing reviews and specs in one place.
Manufacturers like Coca-Cola and Bayer are already feeding data into the system.
An internal document flags October 1, 2027 as a target date, though Amazon calls the timeline aspirational.

KEY HIGHLIGHTS

World Product Graph aims to catalog every product sold online, including rivals’ listings.
Database will include customer reviews and specs for items Amazon doesn’t sell.
Manufacturers like Coca-Cola and Bayer are already supplying product data.
Amazon’s AI shopping tools, including Alexa, would tap the database for better recommendations.
Internal document targets October 1, 2027, but Amazon says the timeline is aspirational.

Amazon’s plan to catalog every product sold online

Amazon wants to catalog every product sold online—even those it doesn’t stock. Its new database, the World Product Graph, surfaced in an internal document reviewed by Reuters and confirmed by sources who spoke with Engadget. The system is designed to log everything from customer reviews to technical specs for items sold anywhere on the web.

That includes competitors’ listings. By pulling in reviews and details from outside its own marketplace, Amazon hopes to become the only stop shoppers need for product searches. Instead of hopping between Google, Reddit, and Crate & Barrel, users might soon find all the answers inside Amazon.

Coca-Cola and Bayer are already on board, feeding product data into the graph. Amazon’s AI shopping assistant, Alexa for Shopping, will use the information to sharpen its recommendations and answers.

The company has set an internal target of October 1, 2027 for launch, but Amazon calls the date aspirational. No public rollout is guaranteed.

Why Amazon’s database could dethrone Google search

The World Product Graph isn’t Amazon’s first shot at owning product search. Earlier LazyFounder reporting showed Amazon quietly assembling a similar database ahead of Black Friday, aiming to pull shoppers away from Google.

Where Google scrapes public listings, Amazon is cutting deals directly with manufacturers. That could give Amazon cleaner, more detailed product data than search engines or rivals can offer. If shoppers start searches on Amazon instead of Google, ad dollars and affiliate revenue could follow.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

If Amazon succeeds, shoppers may start searches on Amazon instead of Google—shifting ad dollars and affiliate traffic away from search engines and review sites.

Key takeaways

Amazon’s World Product Graph could become the default product search hub, bypassing Google for millions of shoppers.
Manufacturers supplying data early may gain better placement in Amazon’s AI-driven recommendations.
A 2027 launch is possible but not certain; delays or scale-backs could slow Amazon’s search ambitions.

Sources

TechRadar · 2026-11-13
No gimmicks, just genius — I’ve rounded up the best genuinely useful kitchen gadgets under $50 to gift your favorite home cook
Engadget · 2026-11-15
Amazon Is Reportedly Working On A Database For 'Virtually Every Item For Sale On Earth'

This story is an original summary drafted with AI by Lazyfounder from the reporting listed above and checked by automated validation. Facts are attributed to their original publishers; sections marked as analysis are Lazyfounder's. Where a source is in another language, facts were machine-translated and quotations are reported, not reproduced. Read the original coverage via the links, and see our AI policy and corrections policy.

About the author