The biggest shopping season of the year is underway and a lot of it is being done online this year.  In 2020, US holiday sales will grow 33% year-over-year, according to a recent report by Adobe Analytics. This surge in online shopping and distanced gift-giving is expected to result in three billion packages being shipped in the US, an increase of over 800 million from 2019.  To support the growth and demands of eCommerce, many retailers have improved their systems and capacity in recent years – a cycle of innovation that has sped up to meet the challenges of 2020.

Many retailers rely on Azure Cosmos DB to support critical services and applications, and that has only increased this year. Regardless of size or specialty, the requirements are similar: dynamic (and often unpredictable) scale, changing data types and schemas, dispersed customers and operations, and the need to deliver fast and always-on customer experiences across channels. NoSQL databases are often chosen in these circumstances because they are built for flexibility and scale; Azure Cosmos DB goes a step further by guaranteeing low-latency and high-availability and with its ability to replicate data across any Azure region.

In this post, we’ll walk through real-world customer examples of top use cases and share code samples and other resources you can use to build your own.

Walmart supports hyper-growth with a cloud-native eCommerce platform

The global tech team at Walmart was tasked with building a new cloud-native platform to support 10x growth in the eCommerce business, and recently shared details of this journey on their blog. Online sales were already a huge part of Walmart’s business, delivering $5 billion annually and enjoying 30%+ year-over-year growth. Walmart saw an opportunity to grow the eCommerce business even more but realized their existing monolith software couldn’t support the scale or cost-effectiveness required. The answer was a custom-built platform using cloud-native services including Azure Kubernetes Service and Azure Cosmos DB. Read the full story.

Develop your own solution:

• Scalable order processing solution architecture
• Retail eCommerce product recommendations hands-on-lab and demo
• Real-time recommendation solution accelerator (this IoT telemetry architecture can be used for retail personalization!)

ASOS delivers personalization at scale

Online fashion retailer ASOS runs a microservices architecture, with NoSQL data in Azure Cosmos DB powering inventory, ordering, user profiles, and recommendations. With over 15 million customers worldwide and more than 85,000 items in its product catalogue, the team at ASOS needed a way to generate personalized recommendations in real-time and at an enormous scale. Plus, they needed to ensure their site was fast and responsive even during the peak holiday shopping season. Read the full story.

Develop your own solution:

• Retail personalization solution architecture
• eCommerce recommendation system hands-on lab and demo
• Order processing solution architecture

Jet.com real-time eCommerce pricing engine

Now part of Walmart, Jet.com built a microservices-based eCommerce platform using Azure Cosmos DB and other Azure cloud services. One of the key services introduced was a dynamic pricing engine that made real-time adjustments to encourage customers to both buy more items and items in the same distribution centre. The event-driven system behind this engine tracks multiple events related to each order, customer, product inventory, and pricing – in real-time and across geographies. Read the full story.

Develop your own solution:

• Scalable serverless web application solution architecture
• Event-driven retail solution lab
• Order processing solution architecture

Chipotle digital ordering platform

To create an integrated digital experience for individual and catering customers across 2,500 stores globally, Chipotle developed a new website from scratch using .NET and a variety of Azure services. Azure Cosmos DB plays a big role in the new architecture, along with Azure Functions, Azure Cache for Redis, and Azure Event Hubs. Part of this project included replacing some relational databases with NoSQL, and storing customer orders as single JSON documents in Azure Cosmos DB to gain scalability and high-availability. Read the full story.

Develop your own solution:

• Watch Mike Smith, Chipotle Lead Software Developer, share details of the solution
• Order processing solution architecture
• Scalable serverless web application solution architecture

Adobe uses customer insight for real-time personalization 

Used by many retailers, the Adobe Experience Platform builds up-to-date customer profiles using a variety of data which then power AI-driven personalized customer experience across digital channels. These unified profiles are 360-degree views of customers that integrate data from many disparate sources, solving one of the biggest problems in eCommerce and digital customer experience: knowing who the customer is and how best to engage them.  As detailed on the Adobe technical blog on Medium, Azure Cosmos DB supports Adobe Experience Platform Customer Profile (read the story) and is one of the databases used in Unified Profile (read the Unified Profile story).

Develop your own solution:

• Retail personalization solution architecture
• eCommerce recommendation system hands-on lab and demo

Get Started

Now that you’ve seen how leading retailers use Azure Cosmos DB to power their eCommerce solutions, you can build your own with the architectures and code samples above.  You can also:

• Discover the benefits and capabilities of Azure Cosmos DB and get started free at AzureCosmosDB.com.
• Find more Azure Cosmos DB use cases
• Learn about NoSQL (non-relational) databases and how they differ from relational databases
• Develop new skills with Microsoft step-by-step training modules for Azure Cosmos DB