Twitter
Facebook
Linked In
RSS
Login or Register
New to InternetRetailing?
Register Now
Internet Retailing
IREU Top500 The Customer Report: 2018

IREU Top500 The Customer Report: 2018

You are in: > Home > Magazine > Magazine Articles

This is your 1 complimentary article for this month

Become a member for unlimited and immediate access.


Register
Already a member? Log in here

AI: human-guided automation is here

Linked InTwitterFacebookeCard

André Brown, Group CEO of Attraqt discusses the rise of AI and automation and what’s ahead for retailing.



NASA’s mission to Mars relies on complex, advanced technologies created by some of the finest human minds. But according to former chief scientist Dr Ellen Stofan, whatever is waiting to be discovered will be beamed back to Earth by human beings rather than robots.

Stofan argues: “It’s much quicker to send people to Mars to do the research required. Humans are creative, flexible and mobile – we can cover a lot of ground in a short amount of time and we just don’t have the robotic ability to replace a human. To get the work that we need done, we don’t have the artificial intelligence. Maybe many years from now we will, but we’re going to get the information a lot sooner by sending a person.”

And so, it appears that expected signs of life – fossil microbes or extant life under the surface of the planet, rather than little green men – will be revealed by human hands, assisted by machines. Interplanetary space travel and online retail may be miles apart, but the relationship between man and machine is proving to be just as experimental – and controversial – here on earth. Research carried out in March last year by PwC predicts that robots and AI will likely take over almost a third of all UK jobs by the early 2030s – with the retail industry one of the high-risk sectors, alongside wholesale, manufacturing and others. According to the findings, there are 2.25 million jobs under threat of automation in retail and wholesale alone.

Robots stealing jobs is an oft-quoted line. It’s true that AI can offer the retail industry a fresh take on productivity and efficiency, but this means eliminating monotony. Some jobs will go, while others, such as those with a creative focus will be born. Peel away the alarmist headlines about internet retailing roles being defined as man or machine and ask the question: “does it have to be one or the other?” and this offers a refreshing perspective. Self-learning automation and human guidance working in tandem is a steady trend that will continue to permeate ecommerce in 2018 – accelerating efficiency and giving retail teams nimble and immediate operational and creative control of a brand’s website at any given moment.

WHY SHOULD RETAILERS CARE?



Let a machine do everything and it will become a self-fulfilling prophecy. You’d likely end up selling the same products over and over again, while optimisation of operations towards one goal might mean other goals will be neglected, despite the ability to set sophisticated technology to multi-task. On the other hand, it would obviously be impossible to run an ecommerce business without technology, as there are too many variables – huge product volumes, millions of different customers and so much data.

Styles, products and seasons are ever changing, so a combination of automation and manual becomes critical. Rules will do the heavy lifting within technology solutions and retailers will apply strategies, with presentation finessed according to conditions at any time. It’s a controlled experience, with the needs of the brand at its core, hence the term human guided automation. Luxury retailers will often showcase aspirational, expensive products in line with brand and customer expectations, even though their bestsellers might be wallets and socks. A machine would push the latter, but the human element, the brand’s strategy, would layer on top of that. People are behind the strategy and look, with machines populating websites following those guidelines.

If you look at Harvey Nichols’ website and go to the ‘New In’ section, the retailer is using row-based merchandising to curate these pages on a granular level to make sure they are pushing appropriate products. On each row, you’ll see four items from a particular brand or from the same category – evening dresses, say, or handbags, which complement each other visually and stylistically. These aspirational products and hottest design trends are carefully selected both to embody Harvey Nichols’ brand image while reflecting its business requirements, such as stock levels and current campaigns. This is an example of human-assisted merchandising according to trends. The same situation arises with onsite search. Technology will guarantee results, but it will take human interaction manipulating automated search rules to ensure all relevant product matches are found. An example is a search for the term ‘rucksack’ on Amazon . Search algorithms wouldn’t automatically relate ‘rucksack’ to ‘backpack’ as the terms are too different, but a person knows they are the same thing and can ensure results are returned for products called by either name, no matter which term the shopper enters. Machines are pretty smart, but they still can’t take over some parts of your intellectual property – that is, what you know about your own business.

We have different, complementary talents. Machines excel in identifying patterns in enormous amounts of data and drawing conclusions. Humans are good at understanding strategy, brand image and aesthetics under fast-changing circumstances. Used together, both skillsets can set trends and inspire shoppers.Consumers are also able to help retailers refine communication, specifically when they are presented with content. Facebook’s ‘see fewer posts like this’ function allows people to control the results they wish to see in their timeline, or more precisely, things they don’t want to see. In Facebook adverts, this function can be used to make sure retailers are showing the right products to shoppers according to their preferences. They can select multiple options to explain why they don’t want to see certain products, or the complete advert, allowing Facebook to learn. This guidance then controls all future recommendations and posts that automatically appear in their timelines.

What we are seeing is the relationship between humans and technology becoming more powerful across the ecommerce journey – from merchandising, personalisation and digital marketing to virtual assistants, warehousing and logistics. A magical device capable of reducing operational efforts and human intervention to a bare minimum first resonated with the market a couple of years ago, but the expected new era of ecommerce, at least for now, hasn’t dawned. A harmonious combination of human and machine is the way ahead for the retail sector. Space exploration might not be such a distant connection after all. Although it’s worth remembering – it took three days from earth for mankind to make one giant leap on the moon, but we’re looking at seven to eight months before stepping foot on Mars.
Linked InTwitterFacebookeCard

Become a Member

Create your own public-facing profile
Gain access to all Top500 research
Personalise your experience on IR.net
Internet Retailing
We are the magazine, portal and research source for European ecommerce and multichannel retail, hosting the board-level conversation for retailers, pureplays and brands across all of our platforms. Join the conversation.

© InternetRetailing Media

Latest Tweet

Internet Retailing
Tamebay
eDelivery
Twitter
Facebook
Linked In
Youtube
RSS
RSS
Youtube
Google
Linked In
Facebook
Twitter