Radial powers retails sales with 10x higher performance on MongoDB Atlas

INTRODUCTION

The platform behind leading ecommerce experiences

When we’re shopping online, we want an experience so smooth we don’t have to think twice about it. Behind the scenes, retailers are partnering with companies like Radial to make sure they can fulfill every order quickly, from displaying their online listings, to managing orders and inventory, shipping, and logistics.

With more than 35 years of experience, Radial is a trusted expert known for delivering best-in-class customer experiences for the world’s biggest retail brands. It also knows where to invest to have the biggest impact. By swapping its on-premises relational database for MongoDB Atlas, Radial delivered 10x more performance for less money, and with minimal downtime to boot. It's not surprising, then, that Radial received a 2023 North America MongoDB Innovation Award for its MongoDB Atlas deployment.

Radial’s broad suite of specialist tools allows shoppers to browse with confidence all year round, which is particularly challenging for ecommerce providers to achieve during Q4 of the retail calendar, when they need to handle Black Friday and Cyber Monday sales as well as Christmas and other holiday peaks.

“As a retail platform, we look for every opportunity to expand into the cloud. It offers the scalability we need to accommodate those seasonal peaks without creating complexity or escalating costs,” explains Eric Lutts, Senior Director of Database Engineering at Radial.

THE CHALLENGE

Managing shopping peaks without latency

Radial Order Management helps retailers by orchestrating every step from inventory management to fulfillment and providing real-time business analytics. Data needs to be accurate to make sure the inventory is up to date so retailers don’t oversell their stock, and the system needs to be ready to handle high-volume, high-velocity transactions 365 days a year.

“We were using a relational on-premises system with row locking to manage inventory, but during peak times hot products fly off the shelves at high speeds across multiple channels, which risks latency issues and the database getting overloaded,” recounts Lutts. “It’s not an issue we could fix by simply buying more hardware for busy times; it’s not cost effective to have it plugged in but unused for most of the year.”

Radial began looking for a flexible, high-performing database platform that could handle high throughput, horizontal and vertical scaling, and was easy to develop in line with changing needs. “Retail is an ever-changing market, and there are different attributes that need to be added to data all the time. A hard data model would severely limit us, so finding an agile, future-proofed platform was critical for success,” Lutts added.

THE SOLUTION

Powering order and inventory management with MongoDB Atlas

Radial selected MongoDB Atlas on AWS, a managed cloud solution that provides the flexibility and high levels of performance required to support retailers throughout the year. MongoDB Professional Services worked closely with Radial to develop a migration plan to ensure there was minimal downtime during the migration.

“The support we get from MongoDB is second to none; amazing customer service is its DNA. I have experience with a lot of other database vendors, and what sets MongoDB apart is that they don’t just help, they educate so we can get the most from the platform in the future,” explained Lutts. “They have a support model I wish other companies would adopt. We’re confident we can develop the platform ourselves as our requirements evolve.”

MongoDB Atlas integrates with Radial Order Management and supports inventory management. Consistent real-time updates ensure customers are informed immediately if an item is out of stock and eliminate the risk of fulfillment issues if there’s latency between orders being placed and the volume of available stock being updated. This avoids a poor customer experience if more orders are accepted by the system than the retailer can provide.

“Shifting from a rigid relational system with lots of controls in place to a more fluid system like MongoDB can be a big change for developers, but it’s worth taking the time to learn and adapt,” said Lutts.

With MongoDB Atlas providing a more flexible, feature-rich, modern, multi-cloud database platform, Radial can offer more value to retailers. Atlas’s flexible model allows Radial to adopt and reengineer their solutions quickly, which is a substantial differentiator for other solution vendors. For example, it makes it quick and easy to spin up a load test environment, so the company doesn’t have to spend resources maintaining one on an ongoing basis. This means retailers can test how their front-end web store connects to backend systems via the API, which is vital for peace of mind for new clients or retailers who regularly refresh the look and feel of the web store and need to know it will work seamlessly when it goes live.

“It takes less than an hour to spin up a complete copy of our production dataset. That’s invaluable to staff and clients because we can run client tests on demand without impacting production,” added Lutts. “We can give retailers greater confidence that everything will work seamlessly throughout Q4 even if they’ve changed the front end of their web store.”

THE RESULTS

10x increase in performance at a lower cost

Since moving to MongoDB Atlas, Radial has improved performance and reduced latency on its order management platform while saving money. This allowed the company to rapidly expand strategic customer accounts and process high volumes of transactions quickly and seamlessly resulting in impressive impacts for Radial’s clients.

One retailer saw record-breaking volumes of sales from a one-day event that went off without a hitch, which was make or break for the retail company at the time. There were more than 10 million transactions processed during this short timeframe.

“We can process 10 times more throughput with MongoDB Atlas than we could on-premises, and we’ve got room to handle even higher capacity,” Lutts explained. “We can also seamlessly scale up and down as required, which reduces the risk of outages and means we don’t have to pay to have extra hardware on standby.”

Developers, now free from system maintenance, can focus on designing new features and functions to better serve retailers. And with a wealth of actionable data at their fingertips, they can identify the developments that will have the biggest impact on the shopper experience.

In the future, Radial is exploring how MongoDB Atlas Search will benefit its Extract, Transform, and Load (ETL) developers. It’s also considering zone-based sharding across geographic regions for retailers wanting to run targeted sales in specific locations, and looking at how Relational Migrator could help to simplify moving other legacy systems onto the platform.

“MongoDB regularly releases new products and upgrades, which has been a huge win for us,” says Lutts. “The retail market is always evolving. There are always new products and new ways of addressing a customers’ needs. The only thing that doesn’t change is that when someone orders something, they want to make sure there’s inventory, and that the order is handled properly. Both of those things are what MongoDB helps us with today. The market changes, and we address the market. MongoDB is a flexible platform — it will allow us to evolve as the market evolves.”