Creating The Intelligent Enterprise In Retail (Part 2): AI, RPA, And The Future Store

Marc Teerlink, Steve Mauchline and David Judge

Part 2 of the 3-part series “Retail and Intelligent Technology

While out shopping recently, a friend of ours signed up for a retail credit card in the store (mostly to get the additional discount). When the physical card arrived in the mail a few weeks later, he was disturbed to find that his last name was spelled inaccurately! Someone had mistyped his name by one letter. This is just one example of the manual efforts that connect two systems – they take time and still can be incorrectly executed.

After the challenge of having his name fixed on the new credit card, our friend chose to redeem his “sorry we messed up” discount and bought some shirts online. Unfortunately, one didn’t fit well, so he took it to a store to return it. To his dismay, the store refused to accept the shirt, as the store systems didn’t match the online digital inventory, and he had to return it by mail. Such stories are a far cry from the “seamless customer experience” or the digitally enhanced future of retailing.

From checkout clerks to switchboard operators, people (i.e., human labor) have long supported commerce. Technology advances have facilitated improvements in service and delivery time – when companies deploy them successfully. While self-checkout and scanning devices come to mind, automation taken further can drastically reshape and advance accuracy and service for companies willing to take the risk.

Currently, the average retailer or consumer products company has automated around 30% of its back-office processes. This means 70% of processes are not automated. As macroeconomic conditions continue to put pressure on profit margins across sectors and especially on retail, cost productivity and unlocking new value are back at the top of the senior-management agenda.

The question is, what else can be done? If you can automate a majority of mundane and recurring activities, you can create large-scale efficiencies. We also know that if you can automate to that level, it’s not just about cost, it’s also about quality. Automation can perform this manual work to a high degree of precision and speed. That means faster, better, cheaper processes and more time spent improving the customer experience.

From automation to artificial intelligence, a business strategy

The business case for automation of repetitive retail processes shouldn’t be hard to imagine. Automation and technology support myriad topics, including the use of online channels, curating offers, personalizing products, automating supply chain and payment processes, and adding experiences that cater to increasingly niche interests.

These advances, however, can only be delivered through robust digital and analytical capabilities. Many retailers that we interviewed referred to digitization as the process of using tools and technology to create more intelligence in their processes for the sake of cost advantages and competitive differentiation, becoming an intelligent enterprise as the result of a successful digital transformation. The key concept is the ability to “replace repetitive tasks” and still have the flexibility to adapt to change.

This is where intelligent technologies come in. Artificial intelligence (AI), powered by data science and machine learning (ML) techniques, can go further than ever before, as multiple layers of decisions can be combined to have a machine make decisions or provide strong recommendations for actions. This is why we can safely estimate that a fair portion of business processes (around 50%) will be fully digitized and automated in the next three years.

For the record, AI is not “one” technology: it’s process knowledge or a combination of different techniques, technologies, tools, and sets of training data. Think ML (and more traditional predictive analytics approaches), blockchain, IoT, process automation, and conversational user interface flows. These are the pragmatic approaches to making the sci-fi AI real and ready to deliver value today.

A good definition of AI is “the orchestration of a set of knowledge-embedded technologies that, when used together, augment or automate a complete task of a specialist, an expert, or a professional knowledge worker.”

Once existing processes are automated, new processes will be created as we move from a process-driven world to a data-driven world. Based on the insights from that data, businesses will create new processes, reshuffle existing ones, and digitize the process experience.

Harness the value of intelligent process automation

One of the most powerful sets of knowledge-embedded (intelligent) technology components within AI is robotic process automation (RPA). Intelligent RPA can be thought of as “the orchestrator,” aka the organization or the meta-layer. Intelligent RPA brings together a set of tasks, consolidates them, and orchestrates multiple systems that replace the repetitive human intervention processes that involve aggregating data from multiple systems (or taking a piece of information from a written document and entering it as a standardized data input).

Hopefully, when reading this, the example of the manual processes that duct-taped two systems together to issue a retail credit card springs to mind as a candidate for intelligent robotic process automation!

Our credit card example offers a view into where intelligent process automation can enter. Looking forward, we believe intelligent automation will be a core part of companies’ next-generation operating models. Some early results from companies employing intelligent process automation are shown in the graph below:

To further support automation going forward, many industry analysts and strategic consultancy firms support the notion of intelligent RPA as the future of operations planning. In short, while intelligent RPA and other intelligent automation are unfolding, there are certainly wins to be achieved when carefully considered and intentionally tested and deployed.

A case study and real-life application

Which use cases of AI are right for your business? It’s a combination of aligning your business strategy with the capabilities of your subject matter experts. You may have a compelling use case, but no true experts to help train the solution, or you may be light on use cases but have deep experts looking to help you. The sweet spot is when you have both!

The adoption of a “crawl, walk, run” approach is very appealing and absolutely sensible to manage IT and business team expectations and risk. Once you have a solid foundation, you can vary the speed of implementation.

The belief is that this type of intelligent automation will enable retailers and consumer product companies to augment existing jobs – and transition to those 2020 and beyond jobs that have yet to be created. In short: process automation and machine learning, as AI’s most important components, have the vast potential to drive the first big change in retail since e-commerce.

If you dislike change, you’re going to dislike irrelevance even more” – Four-star general Eric Shinseki in “Marines Turned Soldiers.”

You may be reflecting on what all this means to you. We suggest that you think about a real-life use case that is either directly applicable or might be the catalyst you need to create your own.

With today’s technology, if we use robots to roam a store and survey and replenish stock (RPA to input stock depletion, reorders, and replenishment), a supervisor will be freed to better manage employees and service consumers. We can reduce processes while helping customers buy!

In future blogs, we will look at how sensors and tags will impact retail and consumer industries.

Coming to the AI and Big Data global expo in London on April 25–26? Marc Teerlink and David Judge would love to hear your perspective following their keynote session focused on how to make money from AI and ML.

SAP worked with more than a dozen industry experts to uncover five trends that will determine the customer experience over the next decade. The Future Customer Experience: 5 Essential Trends report, examines each of these trends and offers recommendations for how brands should respond now to prepare. Download the report.


Marc Teerlink

About Marc Teerlink

Marc Teerlink is Global Vice President of Intelligent Enterprise Solutions at SAP. He drives the strategy, vision, and production of AI and machine learning technologies delivered through SAP Leonardo. Prior to his current role, Marc was IBM Watson’s Chief Business Strategist, where he oversaw the new offerings portfolio for the Watson platform during IBM's formative years of artificial intelligence. During his time at IBM, Marc executed a number of successful transformational projects and created and delivered cognitive computing solutions and services offerings. Before IBM, he built expertise as a banker, consumer products business manager, consultant, and change leader within nine countries across three continents.

Steve Mauchline

About Steve Mauchline

Steve Mauchline is a Business Architect in SAP North America's Presales unit. He helps customers rationalize business capabilities to ensure clear understanding of their customer vision & strategy and provides business capability recommendations to operationalize the future business and operational model. Prior to his current role, Steve spent 10 years with IBM, focused on software and services solutions in Retail, Consumer Products and Transport & Travel industries. He spent his formative years working for a large UK Retailer, building his expertise across stores, supply chain, merchandising and business analytics.

David Judge

About David Judge

David Judge is Vice President of Intelligent Enterprise Solutions at SAP, and drives solution strategy for the SAP Intelligent Robotic Process Automation service. As an SAP executive focused on innovation, David guides product strategy and market awareness for multiple topics including machine learning, automation, and conversational AI. Prior to joining SAP, David was Practice Leader for Artificial Intelligence in the Emerging Business Accelerator with Cognizant, and Executive Director for Artificial Intelligence market development at IPsoft.