Beating Amazon: Big Data for Retail Analytics

Written by

Team Egen

Published on

Apr 02, 2019

Reading time

3 min read

  • Data & Analytics
  • Business Strategy

Retail executives face an existential threat… how do they counter global behemoths like Amazon, Alibaba, and Walmart with their own initiatives to increase sales, improve customer service, and decrease expenses?

It’s harder than ever to truly make an impact and “wow” customers… according to a 2018 Forrester report on customer experience, despite numerous objective improvements, customer evaluations of major US brand experiences have stagnated. In other words, customer expectations are rising at least as fast as retailers’ ability to keep up.

On the other side of the coin, the ever-increasing economies of scale of global players and their willingness to cut margins mean that retailers have to act boldly to streamline their own operations and cut costs wherever possible… while simultaneously increasing their ability to nimbly anticipate customer needs.

Retail businesses are competing against global giants like Amazon.

That’s a Big Challenge. What’s the Answer?

The answer lies in a combination of clever technology, specialization, and relentless customer-focus.

We’re a technology firm so we’ll leave the specialization and customer-focus to you, but here’s our take on how retailers can leverage their relatively modest technology budgets to effectively compete, and beat, the global giants in their sector.

Big Data for Retail is Here. Now What?

Big data, machine learning, and artificial intelligence have all moved out of the lab and into production in organizations across the globe. Just last month, Gartner announced that there has been a 270% increase in the use of artificial intelligence in enterprise-sized organizations over the past four years with 37% reporting at least one project in production. CapGemni chipped in their own report in December of 2018 that showed over a quarter of all retailers have also deployed AI in some format.

big data retail analytics - shop

Over 25% of all retailers have deployed AI to increase sales, reduce customer complaints, and lower costs

AI in Retail, CapGemni, December 2018.

Customer Experience

So what’s all this technology good for? The most promising use cases fall into two broad categories:

CapGemni found that respondents who had implemented customer-facing AI expected increases in sales of 15% while simultaneously decreasing customer complains by 15%. Zero respondents were unable to quantify AI’s benefit to their organizations.

Some of the most common initiatives include:

  • Recommendation engines
    • Customization like more accurate sizing and color recommendations
    • Chatbots
    • Pricing optimization

Operations and Back-End Administrative

Interestingly, CapGemni found that operations-focused deployments of AI had the highest average ROI. Although not as “sexy” as customer-facing deployments, operations offer an incredible opportunity in areas such as:

  • Inventory management / assortment optimization
    • Procurement
    • Fraud prevention
    • Image recognition to decrease shrinkage/theft in physical stores
    • Optimized supply chain management

You Don’t have to Outspend Amazon

Given Amazon’s market cap, it’s probably a relief to hear that you don’t need to go toe-to-toe in a technology budgetary war. Here’s why: the best tools in the space are open source and/or available as pay-for-what-you-need services on commoditized cloud infrastructure like AWS, Google Cloud, and Azure.

big data retail analytics - money

In fact, the incredible increase in power and availability in big data and artificial intelligence for retail puts Moore’s law to shame. The most bleeding edge, proprietary, and highly experimental algorithms of two years ago are totally outclassed by the freely available and highly stable open source tools ready to run on the cloud infrastructure of your choice.

What will You Tackle First?

Analysts agree that the most successful deployments of big data and artificial intelligence are the ones that are most simple and highly focused. Leave the moon-shot projects to Amazon and Alibaba. After they’re successful, you can build a more stable version at 1% of the cost.

The first step is to gather and store as much of your structured and unstructured data into a data lake. Then, you can build analytics on top of that to tackle specific tasks.

Set up a conversation with one of our data architects to unearth the low-hanging fruit in your organization and get on the fast track to ringing-up your first big data win.

You might also like

All Insights

Looking to build something amazing?

Let's Talk