Big Data and E-Commerce: How Global Retailers Are Adapting.

Are offline and online retail channels different from each other? Both channels aim to optimize the customer experience, drive traffic & frequency, increase basket sizes and improve sales & margin. Yes, the goals of each channel are the same, but the access to data in the e-commerce world is a major difference. It ultimately improves the channel’s ability to deliver the aforementioned objectives.

I’ve had a front row seat to some jaw dropping big data and machine learning projects. Luckily, the fixes to retail’s big data and e-commerce challenges are not too difficult, when you have the right tools. Below are four ways retailers can adapt to the data-rich e-commerce world to sell more products. 

1. Prioritize Site Search

Many e-commerce sites have become too difficult to navigate. When was the last time you purchased a product online by clicking through the navigation bar’s links? I cannot remember the last time I completed an online purchase this way.

Site complexity is precisely why it is critical to have a prominent, well functioning, search capability on your e-commerce site. Site search is a key interaction point for customers and can make or break a shopping experience. It appeals to visitors with strong purchase intent. They know what they want and are on a mission to find and purchase these items.

In client engagements, I have found that on average, 33% of all e-commerce sessions use the search function. Almost 66% of those search sessions result in a purchase. Clearly, a better search can lead to higher conversion rates.

One way to improve site searches is to identify distinct search behaviour trends for each product category. Perhaps now is a good time to revisit your site’s search capabilities to determine if it best suits your customers’ needs.

2. Optimize Customer Journeys

A great deal of planning is put into creating the ideal customer journey before a new or updated e-commerce site is launched. However, over time, the site changes. New categories or products are added, promotional campaigns are initiated, and before you know it, the e-commerce site is supporting unintended customer journeys that do not enable easy conversions.

Since 88% of US consumers research products online before buying, neglecting to manage customer journeys on your site can seriously hurt your conversion rate. As an e-Commerce leader, you must know your statistics. What are the typical customer journeys on your site?   Which ones yield the highest and lowest sales? When do journeys result in bottlenecks or worse yet, an abandoned cart?  

To get to the root of the vast array of customer journeys, you'll need to conduct a formal journey path analysis. Start by understanding the most common paths that result in a sale, an abandoned cart, or even a return. You'll quickly learn how your site and products are influencing the path to purchase.

With this information in hand, e-commerce retailers are able to identify and eliminate non-converting pages, dead links, and design new paths that lead to increased conversions.

3. Know The Preferences of Desktop and Mobile Consumers

Mobile technology is changing traditional retail, but perhaps not in the way that you’d think. Although mobile-based transactions are on the rise, these purchases only account for a small share of US retail sales. According to IBM, “smart phones browse and tablets buy.” Yet, desktops are still king when it comes to online spending. The 2014 holiday season numbers indicate that desktop traffic represented 55% of online traffic and 77% of online sales. Research from Mobify shows that conversion rates in 2014 on smartphones range from .63%-1.37% and conversions on tablets were between 1.67%-3.65%. These stats mirror our internal findings quite closely.  

So how are consumers using their mobile devices? In my experience it depends on retail vertical and region. For some multi-channel retailers, customers are primarily browsing and researching products on mobile devices (phones & tablets). Perhaps customers do this while waiting in a queue or on a break.  These consumers also tend to look at higher value products on their mobile device versus on their desktop.

 Clearly, consumer behaviours are different between channels and retailers must understand these differences before investing heavily in technology.

 4. Understand Your Customer Data

Many multi-channel retailers are challenged to bring their data together from across a variety of online, offline, and mobile sources. A recent Forester and Retail Council of Canada survey found that nearly half of retailers struggled to connect data across devices/channels. As a result, e-commerce teams spend countless hours manually collecting data to analyze using Microsoft Excel, in the hopes of finding a powerful consumer insight to improve the business.

This is an ineffective process that is based on a 25-year-old retail paradigm. According to an Econsultancy survey, only a third of client side marketers reported that they were doing a good or excellent job at taking actions based on insights derived from customer data. Many are taking actions on inaccurate or poor data inputs.

Stop making retail decisions by gut feel and start a weekly routine of reviewing the key data points that drive your business. Begin using tools that speed up the process. The key is to bring in multiple sources of information in order to form a 360 degree view of how customers are behaving and interacting across all retail channels and touch points.  You can start small, but start today!  

Online Retailers Share Common Challenges

So, there you have it. Four ways eCommerce retailers are adapting to a data rich world.  Multi-channel retail organizations store a tonne of data. Data science and machine learning tools are a practical and cost-effective way to make sense of this data.

Did you find yourself nodding in agreement at some of the examples above? If so, I’d love to hear your stories or other ways you’ve been able to improve conversion rates at a large multi-channel retailer by leveraging data. Let me know by leaving some feedback below, or by sending me an email at [email protected]. If you liked this article, please give it a ‘thumbs’ up and share it or my slideshare version with your network.