Business leaders are understandably overwhelmed by artificial intelligence due to its complexity and disruptive workplace implications. My latest research “Will People or Machines Rule Algorithmic Retailing?” explores the topic in a practical way, building framework for retail CIOs to enable active leadership in setting the course for this critical technology as part of an overall retail digital business strategy.
People and machines will collaborate in support of algorithmic retailing
Some estimates suggest that 50 per cent of jobs in existence today, regardless of industry, will vanish over the next 10 years. The job of a retail sales associate has been given a greater than 90 per cent chance of being automated. And why not? As a business leader, the appeal to leverage machines is obvious. They can work 24 hours a day, seven days a week, 365 days a year. They don’t require lunch breaks or vacations, and they won’t ask for a raise in pay. Machines can seemingly do everything better. Increasingly, we trust machines with more complex and personal tasks, and soon we will be living in the world of autonomous vehicles and drone deliveries. Of course, people are confused and concerned, wondering if they can be replaced or if they will have a job.
Gartner predicts 2020 will be a pivotal year in AI employment dynamics, during which AI-related job creation will cross into positive territory, reaching two million net-new jobs in 2025. Although AI will eliminate jobs, it will improve the productivity of many jobs, eliminating millions of middle- and low-level positions, but also creating millions more highly skilled management jobs and even entry-level and low-skill jobs. There are three choices:
- People only
- Machines only
- Both working in conjunction
The first choice is a nonstarter, as we have a long history showing that, when it comes to transformational technology, technology/machines always win. However, machines cannot usurp all interactions, as people still do — and will — provide tremendous value in many business processes in the future. The future of business, and certainly retail, is a combined machine and human environment that ensures each contributes value in a newly engineered organisation.
Labour is one of the highest costs on the traditional multichannel retailer’s balance sheet, and with the competitive pressure from an industry in the midst of a seismic transformation, humans will be impacted by job loss. Retailers will need to redeem savings by eliminating and revamping old, outdated processes. However, it is highly misguided to assume that people can be cut entirely from most processes, so it’s critical to understand how CIOs can approach this subject with business leaders.
Algorithms and artificial intelligence (AI) are converging to transform retail headquarters merchandising and marketing operations
Much of the debate on the impact of AI and intelligent automation services (IAS) on the retail workforce has focused on certain aspects of operational tasks in stores and distribution centers. There is no doubt that operational jobs such as cashiering will be affected by automation.
However, retail headquarters staff will also be disruptively impacted. With large numbers of relatively highly paid associates, who are responsible for making big-dollar decisions that affect every aspect of the business, headquarters are also prime targets for reduction and elimination through IAS. This is long overdue, as many of the existing activities are not optimised, with some surviving as leftovers from historically necessary processes that simply are now irrelevant. Others are the result of management turnover, failed technology implementations or previous job cuts done without a thoughtful organisation realignment.
In November 2015, Gartner published a prediction that, by 2020, merchant leaders will become algorithms, prompting the top 10 retailers to cut up to one-third of headquarters merchandising staff. This is starting to become reality, as retailers JC Penney and Macy’s 1 both recently announced elimination of positions and streamlining of merchandising organizations. At JC Penney, the position of chief merchandising officer was eliminated. In both cases, these retailers cited better use of information and the need for speed of decision making as major reasons. Several recent client interactions confirm that major retailers are moving along this same track.
Delivering customer-centric experiences will not be possible without AI-enabled intelligent automation services (IAS)
The old saying “product is king” is no longer true for retail. Experience is king, and product is a critical component of experience, but consumers increasingly determine the value provided by the associated experience. This is seen by the transition toward click and collect, home delivery, autoreplenishment, in-store entertainment and the growing use of subscription services. Retailers must find use cases for each of these key areas:
- Customer-Centric Marketing relies on analysis of extensive multichannel customer data, both structured and unstructured, to develop behavioural segments, content and delivery of marketing messages. For example, AI firm Sentient showed how providing 256 real-time website design variations for consumers for Swedish flower delivery chain Euroflorist can create personalised experiences.
- Customer-Centric Merchandising uses the same complex customer behavioural segments as marketing to develop the appropriate assortments by channel and geographic location. Both work together to build more accurate and targeted pricing and promotions that will influence customer behavior (see Figure 3). For example, AI firm Blue Yonder forecasts demand by size and colour daily to help German retailer Otto Group maximise its use of returns to fulfill customer orders.
- Customer-Facing Bots and Virtual Assistants using natural-language processes and linguistics, a field within AI, have hastened the emergence of bot platforms and virtual assistants. This has made emerging conversational AI platforms ideally placed to increase and enhance engagement with customers. In one example, Levi’s created its virtual stylist chatbot 2 working with Mode.ai.
AI will enable human creativity that fuels innovation
Retailers will be able to generate labor savings by eliminating highly repetitive and transactional jobs, but will need to reinvest some of those savings into training associates who can enhance the customer experience. Customer insights supporting an adaptive journey can be found through natural-language processing and machine learning. AI systems can learn from the huge data created by customers, which generates behavioural/usage insights and provides direction for product owners and retailers. This helps those product owners and retailers gain a better understanding of their consumers so they can customise their products, designs and shopping experiences around unique user needs.
Gartner clients can read the full research Will People or Machines Rule Algorithmic Retailing? to gain greater insights including many more real life examples of AI in retail.