Big Data Analytics: Driving Sales for E-Commerce

More and more people now prefer online shopping carts and home deliveries over crowded parking lots and long check-out lines, and that’s no surprise! The convenience and stress-free shopping experience that e-commerce offers are just a few reasons why “internet malls” are growing exponentially -- with help from Big Data.

What is Big Data?

Big Data is a nearly immeasurable collection of information streamed and gathered from internet-enabled platforms such as social media, location-based applications and web browsers that reveals patterns in various activities of interest by internet users. This includes critical information about business transactions performed by consumers. However, proper analytics and algorithms are needed to transform these large data dumps into actionable information.

"Information is the oil of the 21st century, and analytics is the combustion engine."

- Peter Sondergaard, Gartner Research

Analytics: Transforming Big Dumps into Big Gold Mines

Big Data provides e-commerce with useful attributes such as a buyer’s browsing history, time spent on product pages, recent purchasing patterns, personal profile, and more — all of which can be used to create personalized sales and gain consumer loyalty. The less exploited yet more encompassing potential of Big Data is much more omnipotent - its ability to form accurate predictions based on past data. For e-commerce, this means fully realized Big Data can literally spoon-feed them with precise inventory information needed to supply upcoming consumer demands.

The Next-Wave of Big Data: Predictive Analytics

If we consider what Big Data offers e-commerce business today as gold, then the coming wave of predictive analytics will give businesses benefits beyond a diamond mine.

One way researchers and corporations are approaching predictive analytics is by using advanced machine learning to scavenge enormous sets of data for patterns, correlations and relationships. The analysis can give e-commerce a priceless tool to optimize inventory management, supply chain decisions and much more.

Visualizing Data-Driven Success in the Works

Amazon is an example of Big Data driven e-commerce success. Amazon’s homepage caters especially to the shopping needs of the repeat visitor. Items and brands they like are displayed on the center of the page under “Related to Items You’ve Viewed” and “More Items to Consider.” If visitors scroll down a bit further, they will see another section titled “What Other Customers are looking at Right Now.”

These personal shopping recommendations are, in Amazon’s words, “inspired by your browsing history.” Products generated for the above categories make use of real-time data such as visitors’ previous browsing history and other consumer behavioral information pulled from a central Big Data warehouse. These personalized modules are specifically designed to make consumers browse more and buy more.

On the inventory side of things, Amazon is using predictive analytics to optimize their inventory management, that is, they anticipate what customers are going to buy according to past trends and ship pre-packaged inventory to local sites even before the orders are placed. For Amazon, this means lower overhead and better inventory management, for consumers, this means expedited shipping due to local inventory supplies. In both cases, customers are happier, stickier and thus buying more - all of which results in increased sales for the e-commerce business.

So, how are revenues looking for Amazon? Aside from its unarguable market dominance, according to statistics, Amazon saw a sales increase of 15% for the quarter ending in March 2015 and much of that is the direct result of relevant product recommendations and optimized inventory delivery for repeat customers, which is only possible with Big Data.3

Apart from e-commerce giant Amazon, RedBox is another well-known company which leverages Big Data Analytics. Research reveals that the hot video rental company uses predictive analytics to understand same-store sales throughout the US, covering over 35,000 kiosks.4 The company then utilizes data visualization tools to determine the best locations for the next kiosk.

RedBox tracks their DVD transactions, including details like how many times each movie gets rented out and the movement of those disks. They say that 50% of their disks get returned at a different kiosk site, so determining the information on kiosk location, sales figures, and movement of inventory all requires Big Data analytics. Furthermore, by tracking the locations of the DVDs, they are able to draw relationships between different areas, which would not be possible without their Tableau analytical software.

Additionally, RedBox uses predictive analytics for their movie placement decisions as well. Matt James, senior director at the company, shares how their movie placement helps them. “We can determine how many copies of ‘Jack Ryan’ each kiosk gets and we can leverage that knowledge for the next movie that is similar to ‘Jack Ryan.”

Then of course, there is the e-commerce heavyweight eBay.5 Ebay uses Hadoop clusters and Teradata servers to allow search results to pop up in a tenth of a millisecond. They use predictive analytics to give merchant partners valuable information regarding predictions about customer segments, for instance “adventure seekers,” without disclosing private personal information about specific customers. This allows eBay and its merchants to apply real-time, targeted content, pricing, and marketing messages to future visitors to the site to incur visitor actions and drive their sales.

Another added benefit of employing Big Data and predictive modeling is that the model will train itself over time, with more data and behavior, without eBay having to go back and redesign that model. In other words, a feedback loop allows the prediction model to self-improve over time. According to Ashok Ramani, product lead of Big Data, the approach will let eBay scale their service to many merchants.

What’s Next?

Big Data and Analytics is a solid addition to the working of the modern e-commerce firms, allowing them to use clickstream information to predict consumer trends on which they can build better business models to drive up sales. But the full potential of Big Data and Analytics for e-commerce is yet to be realized. Not far into the future, e-commerce will leverage the vast amount of information to map our needs so precisely that all we have to do is consent to purchasing. If autonomous shopping does happen, it’ll be due to Big Data and Analytics.

Businesses looking to move forward with web analytics can leverage the current end-to-end comprehensive solutions including Big Data and Analytics and more, to speed up their time-to-market. For more information about analytical solutions, consult Aricent’s qualified analytics architects.




Leave a Reply

Your email address will not be published. Required fields are marked *