Shopping for databases: finding the right fit for retailers and eCommerce companies

Send to friend

By Luke Whitehead, head of European marketing, Couchbase.

In today's Digital Economy, retail and eCommerce companies are competing based not only on their individual offerings but on how they deliver customer experience and operational efficiency.

Their customers expect personalised interactions, relevant information and instant gratification, but these expectations must be met whilst the business reduces operational costs and time to market. Those retail and eCommerce companies that will succeed are those who create loyal customers who spend more, but cost less.

As well as facing competition from within their certain industries – for example clothing, automotive, electronics, office supplies etc – retail and eCommerce companies also experience wider competition from internet companies like Amazon and eBay. These digitally born, tech-savvy companies are renowned for how they evolve their eCommerce applications, and for the speed and agility with which they release new features.

When it comes to customer experience, performance and availability are critical factors of success. Whether customers know it or not, they're interacting with a database. They're accessing product data, customer data, engagement data – and if the data is not readily available, the customer's experience suffers.

Shoppers expect every request, whether it's finding and viewing a product, adding it to the shopping cart, or proceeding to checkout, to be handled immediately – not within seconds, but milliseconds. It doesn't matter what time it is, or what time zone they're in. It doesn't matter if it's Black Friday, Cyber Monday, or Super Sunday – shoppers expect eCommerce applications, web and mobile, to be available 24 hours a day, 365 days a year, anywhere in the world and on any device. Retail and eCommerce companies are increasingly turning to NoSQL technology to meet their demands for a higher level of performance and increased availability.

NoSQL is a modern database technology developed about a decade ago by leading Internet companies including Google, Facebook, Amazon and LinkedIn. It was created to overcome the limits of relational databases that are far more structured and less flexible in meeting the requirements of modern web, mobile and IoT applications.

Typically, relational databases run on a single server, so the resources available to the database – processing, storage, and memory – are not only limited, they're fixed. This was not a problem when relational databases powered internal-facing applications with a limited number of users and predictable workloads. However, applications with thousands, if not millions, of web and mobile customers can overwhelm relational databases with increasing, and more importantly, volatile workloads.

NoSQL document databases are well suited to house many types of data – anything from product and pricing catalogues, to content management, to personalisation to customer profiles, and more – making them very effective as a general-purpose database for retail and eCommerce applications. They are ideal for product, click- stream, session, order, and for user or member data – all part of today's customer experience, whether online or in-store. However, that doesn't mean there isn't a place for relational databases. In particular, relational databases are well suited to legacy business management applications, such as enterprise resource planning (ERP) and supply chain management (SCM). Whereas NoSQL is better suited to meet the performance, scalability, availability, and agility requirements of interactive customer-facing applications.

The most innovative retail and eCommerce companies have already embraced NoSQL by successfully introducing it into their relational environments to support a growing number of use cases. Not only does NoSQL improve the customer experience with greater performance and better features but it also to enhances operational efficiency by reducing costs and enabling a faster time to market.

Comments (0)

Add a Comment

This thread has been closed from taking new comments.