Learning from Amazon: data, AI and automation
Cast your mind back 20 years (if you are old enough – like me) and imagine a day in 1996. You are bored, so what do you do? Boot up your desktop computer, make the tea while you wait for a minute (or more) for your dial up to connect and, bingo, you’re online! You’ve got the world at your fingertips – kind of. There’s no YouTube, Huffington Post, or Candy Crush. There’s no Google, Twitter, Facebook, or Wikipedia. Slim picking by today’s standards. You can however, order a book from a company called Amazon (it’ll never take off – after all why wouldn’t you just go to the book shop?). Fast forward two and a half decades and customers can do a whole lot more than order a book from Amazon.
Jeff Bezos has systematically grown the company to become the most valuable public company in the world. But that does not mean that its founder believes it to be infallible. In fact Bezos even recently predicted the company’s demise telling staff that one day Amazon would go bankrupt. For many retailers that would be a dream come true. Unfortunately for them, however, Bezos’ one goal is to delay that day for as long as possible by focusing on its customers.
So Amazon isn’t going anywhere soon. This means retailers need to find ways to thrive in the $1 trillion + Amazon world or face their own extinction. Research from the Yale School of Management shows that the average lifespan of an S&P 500 company in the US has fallen from 67 years in the 1920s to just 15 years today. And it is predicted that 75 per cent of firms in the S&P 500 now, will be gone by the year 2027. Not a rosy outlook. It’s the same story in the UK although corporate lifespans are a little longer. The main reason that companies die (beyond M&A) according to the study is that they fail to anticipate or react to:
- New customer demands.
- New technology.
- Competitors with new ways of doing things.
To counteract these, the answers lie in data, AI and automation. These are the three ways that retailers can stay in the game and compete with Amazon today.
Transforming data to gain a more granular understanding of your customer
Understanding what your customer needs, what motivates them and predicting what they want in the future is critical. Research from Deloitte shows that customer-centric companies are 60 per cent more profitable. In order to be customer-obsessed you need good quality customer data. With new legislative restrictions over what data you can collect and use, this means managing the data ecosystem has never been more complex.
Customers trust Amazon with everything from their personal information, buying habits, viewing habits to the literal conversations they have in their house as a result of Alexa.
But for retailers not lucky enough to have this wealth of information, you have to make the most of what you do have by being incredibly fast and efficient in bringing together disparate data sources, such as EPOS, social, CRM, customer, ecommerce, web browsing and app data and continually building persistent profiles of what next action is right for every customer. Once the data has been ingested and organised it can be used to build an incredibly powerful and predictive segmentation which goes beyond the scope of traditional segmentation models. To compete in today’s market, it is vital that segmentation becomes a tool that helps create the future.
But Amazon’s data is so rich for so many people, and with such breadth due to historic data, that maximising the value of your owned data to compete is no longer enough. You need to be as smart about your future customers as you are your current ones and that means building a rich data asset that allows you build predictive models to drive both your strategy and your tactical actions in growing customer engagement. Our focus on building a GDPR compliant highly predictive dataset, to combine with retailer data is part of the process in equipping retailers to compete on an equal data playing field.
Maximise the value of data with Artificial Intelligence
AI and machine learning are dominating the headlines as some of the most powerful innovations to drive business growth. And it’s not surprising. Amazon has been doing it for years. Its recommendation engine; love it or hate it, is responsible for 35 per cent of Amazon sales. Of course it’s powered by knowledge of the customer’s past behaviour and also an algorithm that predicts what the customer might want or need next.
However, AI is no longer the domain of global giants like Amazon, it is becoming increasingly mainstream with increasing number of businesses, particularly retailers, investigating how it might lead to better business outcomes. One way it does this is by moving segmentation from a descriptive tool into a predictive one. This means that in a retail context AI can help retailers understand:
- The different types of customers within and across brands
- What motivates different customers across the brand, product and channel offers
- Where the biggest opportunities are to grow customers, and how to do it
- Where the right future customers can be found
- How to drive personalisation and content engines
This brings to light actual, real-time customer trends which can then be used to build and action more relevant business plans tailored to how specific groups of customers are behaving, at scale and speed. Not only that, but with all this in place it is also possible to consider data monetisation by providing insight to suppliers; something else that Amazon is also very good at.
The time element is crucial. Retail is one of the fastest moving sectors and being able to keep up with the customer is critical. The complexity of decision making to meet the needs of the myriad of customers and business needs can only be streamlined by using AI to find patterns and trends that humans could not and enable people to make better decisions. Which is where our third point comes in – automation.
Automate to enhance business outcomes
Amazon is also renowned for its convenience and speed. Amazon Pay removes the need to fill out lengthy payment detail forms. Amazon Go removes the need to checkout at the supermarket. Amazon Prime is the greatest loyalty mechanism the world has seen. Automation of every part of the customer experience through data and AI drives their success, it runs very large parts of its business with very few people and machines shaping the customer experience. Too many retailers aiming to compete with Amazon have large teams working on legacy technology. The answer used to be in unpicking legacy technology, where now it is now possible to create new AI and cloud-based technologies incredibly quickly to automate the process of creating and delivering every aspect of the data process. From ingestion, transformation, data science and modelling to then deliver data and insight to all the right people, places, systems and processes. People must be used to create the future and to wrestle with the complex choices of competing in a new world order – the day to day job of delighting customers has to be delivered by ruthlessly efficient, smart, agile technology if you are to compete on an equal cost basis.
The old adage is true. If you can’t beat them – join them!
What is clear is that Jeff Bezos clearly knows a thing or two about what generates good business and I suspect Amazon will still be thriving in 2029. So, when he extols data and AI as an ‘enabling layer’ that will ‘improve every business’ it is madness not to not sit up, take notice and start competing on an equal footing.
CEO at Outra helping brands connect to their data and their customers. Previously 25+ years at dunnhumby helping retailers and brands put the customer at the centre of every decision including 6 years as CEO. As data and insight grows every customer interaction must be personal and relevant to build…View full profile ›