Amazon sells 4,000 products every minute, about 50% of which are presented to users by personalized recommendation engines.

When browsing the Amazon website, algorithms predict what you want at this moment and select a group of approximately 353 million products to push to you.

What drives personalized recommendations is Amazon’s ever-evolving procurement graph, which is the digital representation of the “physical elements” in reality—all store information such as customers, products, purchases, events, and store locations—and the relationships between these elements.

Amazon's purchase graph links purchase history with website browsing, Prime Video viewing, Amazon music listening and data from Alexa devices. The algorithm uses collaborative filtering to combine diversity (how different the recommended products are), unexpectedness (

Factors such as the surprising degree of recommended products) and novelty (freshness) generate the most complex recommendations in the world.

With its rich data and industry-leading personalized recommendations, Amazon now accounts for 40% of the U.S. e-commerce market, while its closest rival Walmart has a market share of only 7%.

In order to compete with Amazon, Google announced in April 2021 the launch of Shopping Graph, an AI model that recommends products when users search.

More than 1 billion people search for products on Google every day, and shopping images connect them to more than 24 billion product listings from millions of merchants across the web.

The foundation of this model is Google's unique Knowledge Graph, which captures information about entities and their relationships across the vast web, including from Android, voice and image search, Google Chrome extensions, Google Assistant, Google

Structured and unstructured data from Email, Google Photos, Google Maps, YouTube, Google Cloud and Google Pay.

Google Shopping Graph allows 1.7 million merchants to use simple but similar tools to display relevant products on Google, and Google can meet the challenge of Amazon.

Data graphs like Amazon and Google rely on product usage data (that is, behavioral data generated when users use a platform or product) to capture the connections and relationships between companies and their customers.

The concept of data graph originates from social network and graph theory. This theory defines social graph as the presentation of connections and relationships between people, such as friends, colleagues, bosses, etc. Each person is presented as a node, and the relationship is between points and

Connections between points.

This concept emerged from the work of social psychologist Stanley Milgram and over the past two decades has provided a practical lens for analyzing the structure and dynamics of organizations, industries, markets, and societies.

In 2007, Facebook launched the social platform of the same name, allowing developers to create applications integrated into website information flows and interpersonal connections, making digital social graphs popular.

Leading technology companies use data graphs to provide personalized recommendations, upgrade products, optimize advertising, and more.

The most successful examples, such as Amazon’s purchasing graph, Google’s search graph, Facebook’s social graph, Netflix’s movie graph, Spotify’s music graph, Airbnb’s travel graph, Uber’s travel graph and LinkedIn’s career graph, use

The continuous collection of user usage data, coupled with unique algorithms, has left competitors behind in all aspects from product development to user experience.

This article discusses how companies can learn from data mapping leaders to create new competitive advantages.

Data Network Effects To understand the data graph, you must first understand the data network effect, which is the effect in which the data generated when users use a product or service makes the product or service more valuable to other users.

Unlike direct network effects where the value increases as more users join (such as Facebook and LinkedIn), data network effects do not require an increase in the number of users to increase the value of the network. Instead, existing users continue to use it, resulting in more extensive and in-depth use.

Data enables algorithms to produce continuously improving results.

For example, Google's two trillion searches every year help Google enrich its knowledge graph, improve its search engine, and provide users with better search results.

And if users stop using the platform, improvements in platform service quality will stall and become less helpful.

Data graphs are not static and do not reflect data at a certain point in time, but what data scientists call dynamic data.

This is part of the reason why manually graphing your data is impossible.

Technology must be leveraged to collect and interpret in real time the millions of pieces of data generated by the use of a company's products by consumers around the world.

Data Graph Success Factors Data Graph leaders collect user behavior data and quickly use it to improve all aspects of their products and services.

These companies are constantly modifying the methods they use to classify and label product data, looking for relationships between entities so that algorithms can better categorize and provide personalized recommendations.