Amazon’s World Product Graph Could Shift Product Research to Its Store

| 5 min read

A catalog built beyond Amazon’s inventory

Amazon’s reported World Product Graph targets a weak point in its shopping funnel: product discovery. The company already draws shoppers who know what they want, but many people begin researching elsewhere, before they’ve settled on a brand or model. A broad product database could help Amazon compete for those earlier decisions and keep more of the shopping journey on its platform. The aim, as described, is to make Amazon useful at the moment a shopper is still defining the question.

A Reuters report, citing an internal Amazon document, says the project would collect information on products sold across the internet, including items Amazon doesn’t carry. The proposed database could let shoppers compare models, review features and find alternatives without switching among Google, brand sites, review platforms and rival retailers. That would make the catalog broader in purpose than a list of goods available for purchase on Amazon: it could serve as a reference layer for researching a category, even when the eventual purchase happens elsewhere.

That distinction matters. A retailer’s own catalog naturally reflects what it sells, while a wider product database could help a shopper understand how its listings relate to products outside the store. The harder task would be keeping those relationships and product details useful as offerings vary across sellers and sites. The document describes an ambition, though; it doesn’t establish how the company would gather, check or update information at that scale.

That reach could matter as shopping tools evolve. AI agents need dependable product details to compare options and make recommendations, and incomplete or inconsistent information can weaken those comparisons. A catalog spanning products beyond Amazon’s own listings could give the company a larger role in those decisions, while making its site more useful even when a customer hasn’t settled on a specific item. Here’s the thing: being present in a recommendation is valuable only if shoppers trust the comparison behind it.

Amazon’s internal document framed the problem directly: “Amazon cannot be the go-destination for every shopping mission if customers must leave to research product relationships on Google, brand websites, and review sites.” It added that shoppers have had to do “painstaking research” across those destinations to find answers. That language points to a familiar friction in online shopping: product information is spread across sources, and buyers have to assemble the comparison themselves. A single starting point could reduce that effort. Whether Amazon could make it feel neutral and complete is another question (and this is the part most people overlook).

The discovery gap in the numbers

Research cited by Reuters points to a clear split between early browsing and product-led searches. A 2025 Morgan Stanley report found that 57% of online shoppers started product research on Google, versus 16% on Amazon. Those figures suggest that Amazon’s reach as a retailer doesn’t automatically make it the first stop when shoppers are still exploring a category. The gap is about where people begin their research, not necessarily where they ultimately buy.

When shoppers already had a particular product in mind, Amazon led: 41% began their search there, compared with 23% on Google and 18% on Walmart. The figures suggest Amazon is stronger at converting defined intent than shaping it. World Product Graph appears designed to narrow that gap by bringing comparison and discovery tools closer to the point of purchase. The numbers here are underwhelming for a company seeking to own more of the shopping journey, but they also explain why a broader catalog might appeal: discovery is where Amazon has room to gain.

The distinction between researching a category and searching for a known product is commercially important. A shopper who already knows the item can move quickly toward price, availability and purchase. Someone still comparing options may spend more time reading, checking features and considering alternatives across several sites. If you’re working in this space, the strategic point is that product discovery can shape which brands and sellers make the shortlist before a transaction is even close.

Plans remain unconfirmed

The document reportedly presents the initiative as a future press announcement and gives October 1, 2027, as a target release date. Reuters couldn’t verify that timeline or whether Amazon will proceed. A date in a draft can indicate an internal planning marker, but by itself it doesn’t show that a project is funded, technically ready or on track to reach customers.

Amazon told Reuters it is “constantly working on new ideas and technologies,” while cautioning that “draft, outdated documents often paint an incomplete picture” and that a brainstorming document doesn’t necessarily reflect current plans. That caveat deserves weight. The reported concept is specific enough to illuminate a business problem, but the company’s response leaves its status unsettled. The available account doesn’t confirm what features would ship, how the database would work, or whether the stated timing remains relevant.

What the proposal could mean

For now, the project is a reported proposal rather than a confirmed product launch. Its commercial logic is clear: if Amazon can help shoppers decide what to buy, as well as where to buy it, it may capture more of the research that currently starts on Google and other sites. That would extend the company’s role earlier in the decision process, where shoppers are weighing needs and alternatives rather than simply locating a known item.

This is more significant than it looks, but the value would depend on execution. A broader catalog could make comparison easier and give Amazon more chances to influence consideration; it could also disappoint if product details are hard to compare or the selection feels incomplete. The reporting doesn’t settle those questions. And yet, the strategic signal is plain: Amazon sees product research beyond its own listings as part of the competition for shoppers’ attention.

What this means for you as a reader is simple: watch for evidence of a working product, not just a stated ambition. Until Amazon confirms the initiative and explains what shoppers can actually do with it, the reported database remains a possible response to a real discovery gap. A proposal with a target date. Still a proposal.

Source: Sherin Shibu · www.entrepreneur.com