As we look back on a year of lockdown for most of the West, we can take stock of the extent to which the pandemic has changed business forever. It forcibly accelerated swathes of business transformation, subverting entire industries. Legacy businesses have had to modernise or face terminal decline, and many newer businesses, even those previously considered ‘digital-first’, have had to re-imagine their core proposition.
It’s put technology front and centre, thrusting into the spotlight those companies who were merely paying lip-service to embedding tech within everything they do. With customers – whether B2C, B2B or even D2C – nearly exclusively coming through digital channels during the pandemic, this belated investment in tech has also shone a light on the potential for AI, and especially machine learning, to drive the efficiencies needed to survive leaner, tougher times.
Finding ways to grow revenue, expand market share and reduce costs while minimizing risks are challenges for any company. Leading-edge digital-focused companies are increasingly turning to AI to tackle these challenges head-on, without affecting customer experience.
Consumers might have become familiarised with the idea of AI in their day-to-day lives for quite a few years now, in areas like self-driving vehicles, web search, facial recognition and social media platforms. But with digital and business transformation happening at such pace and scale, there’s a growing understanding of AI’s expanding role in less obvious areas. One sector that’s using AI particularly effectively to transform the customer experience – from the back-end – is retail and e-commerce, with profound knock-on effects on the entire ecosystem.
We’ve come a long way from Amazon’s use of AI and machine learning to show only relevant products to shoppers, based on previous searches, purchases and views. Where once its recommendations were notoriously odd and inaccurate for individual users, it’s now an integral part of Amazon generating additional sales and showing it understands shoppers and their needs.
Retail and e-commerce boom
The Artificial Intelligence in Retail market was valued at 1.80 billion USD in 2020, and is predicted to reach 10.90 billion USD by 2026. One fascinating trait of AI technology is that as it’s embedded in more areas of life, and machine learning can develop in tandem with it, the future use cases and benefits to retailers are almost literally exponential. Even now, AI is enabling e-commerce businesses to increase their engagement rate, conversion and decrease time per transaction.
There’s a plethora of data available to retailers to essentially plug into machine learning and reap those benefits. Businesses of all types have more data than ever at their disposal, from sales, supply chain, procurement, CRM systems and marketing campaigns. For many, the issue has been having the talent, time and resource to glean actionable insights from all the data. Many have therefore outsourced a lot of the more complicated competencies to third parties and agencies, which can be costly and also sacrifices control over valuable, owned data to others.
At risk of being reductive, data’s fundamental power for ecommerce brands is principally twofold: highlighting popular products and identifying high-value customers – and the personalisation that stems from combining both can boost sales significantly. Boston Consulting Group (BCG) reports that retailers who have implemented personalisation strategies see sales gains of 6-10%, a rate two to three times faster than others. Other top benefits of AI in e-commerce include: enhancing products, making better decisions, informing the creation of new products, optimizing processes, identifying new markets, automating workflows.
Behind the scenes magic
We’re going to continue to see a sort of dichotomy with AI for some time. There’s the new shiny ‘front of house’ applications of it that garner the headlines – and then the ‘behind the curtain’ use of AI that is actually driving big changes and efficiencies for both businesses and consumers alike. A few are straddling the divide capably, like Amazon, but it’s tricky when for many mid-sized companies, the data and competencies to really drive change are outsourced.
Amazon Go recently opened its first checkout-free store in London, using tech to allow customers to “just walk out”. Instead of paying at the checkout, sensors track the item customers collect from the shelves and put into the cart/basket and charges are automatically applied to the card at the exit via the Amazon Go app. Although computer vision and AI-equipped Dash Carts that are prevalent in the US aren’t in the London store just yet, this AI technology is working backstage to allow the simplest shopping routine.
Another example of behind the curtain AI and ML magic is Volvo’s use of data to uphold its safety reputation. Sensors are fitted into the cars to detect driving conditions and the data is uploaded to Volvo Cloud and also shared with the Swedish highway authorities. Also, part of their analytic strategy is improving driver and passenger convenience, using AI and ML to monitor the use of application features to see what’s working and what’s not.
Where does AI go next?
For Amazon Go, AI is the product, and for Volvo, it’s a core component in improving the product. But what about AI use for the average SME, where the toughest part is getting your brand in front of an audience? How can ‘behind the curtain’ AI really make a difference?
Until recently, the answer was, it probably couldn’t. With agencies handling their data and top-of-funnel marketing activity, it hasn’t typically been an area where AI has had a chance to prove itself. Yet for all the difficulties Covid has presented, it’s also created opportunities to refocus and evolve. Many businesses have lowered or are lowering their marketing budgets, meaning the products and services that consumers are primed to consume digitally (after a year at their laptops) won’t have as much spend behind them to be seen by prospects and established customers.
Simply put, that means there’s a lower cost to gaining market share with sensible investment in advertising – amidst this greater consumption of digitally promoted products and services. It also means many growing businesses have brought costly agency support in-house, where it’s vital that the saved costs don’t impact results.
Software-as-a-service (Saas) is an area that is likely to explode as we exit lockdown and return to normal, but only such services actually save time and resource, and this requires the right mindset to implement. Saas platforms for marketing campaigns, underpinned by ‘behind the curtain’ AI, can offer the ease-of-use and control that SMEs really need right now, but coupled with machine learning which tests and learns what campaigns work, and for who, in real-time.
Maintaining control over data, saving money, optimising on-the-go, and all without needing to hire an agency or full in-house specialist data-marketing team is the sort of background implementation of AI that will drive the next stage of growth for ecommerce businesses and beyond.
In today’s ultra-competitive, fast-paced digital landscape, efficiency and accuracy are imperative, not a nice-to-have. AI on its own isn’t going to be the saviour of the high street. Yet an ecosystem of people, data, technologies and processes, all working in harmony to raise business profile, at a lower cost than has been historically possible – that is where the next wave of behind the curtain and front-of-house AI use cases will combine, and really take to the stage.
Mikael Kreuger is the founder and CEO of the AI-powered advertising platform, Match2One. Mikael has more than 10 years of leadership and marketing experience, and has held senior roles within blue-chip multinationals, growth and scale-up stage start-ups – with a proven track-record in driving high-performance products and teams across multiple disciplines and verticals.