How Amazon leveraged data early to conquer E-Commerce
In our new podcast about the rise of Amazon, called Jeff: Ambitionz az a Trillionaire, we liken Jeff Bezos to a rapper from the 90’s. This is a re-written excerpt from the Podcast with links to relevant publications provided for context.
Most people forget that Amazon started in a garage. Jeff Bezos drove the books they sold to the post office himself in the early days.
A few years after Amazon was launched, Bezos realised the value of data. He knew that if he was going to achieve true scale he needed to understand his customers.
Bezos is relentless and will stop at nothing to get what he wants. In may ways, that’s what he did when he hired a specialist data team in the early days of Amazon.
Amazon and The Data Mindset
Two hours away from Seattle, in the mountains, is the Sleeping lady mountain resort. The name comes from the shape of the mountains that tower above its beautiful wooden cabins.
In January 1997 Jeff Bezos rallied his team, and together in a bus, they drove to the resort where they spent the night. In the morning, they spent some time creating algorithms to try and measure customer enjoyment.
James Marcus, an in-house book reviewer at Amazon recalled how Jeff always had a mania for quantification.
Bezos understood the value of data. He knew that if they were going to accomplish all the goals they set for themselves, they needed to have a good understanding of their customers.
Most online customers feel unseen and unheard. If you’ve ever been in charge of an ecommerce website, one of your main tasks is figuring out how to get someone from your homepage or a landing page to the checkout page as quickly as possible, without any distractions.
eCommerce business owners are infatuated with increasing their conversion rate and lowering their bounce rate.
After all, that’s how they make money.
Bezos wanted to go a little bit deeper though. And he did. The eCommerce landscape was about to change forever, and only Bezos and his team knew it.
The first hire into the data team was Andreas Weigend.
He was an artificial intelligence expert who had published over 100 scientific articles, he started one of the first ever music recommendation systems, and created an application to analyse online trades in real-time.
This guy was the real deal. As great as he is, he wasn’t any different from the other early Amazon employees.
Bezos always hired the best of the best. He knew how to get people who wanted to be challenged. In August 1994, Amazon’s first known job listing was posted online.
“Well-capitalized start-up seeks extremely talented C, C++, and Unix developers to help pioneer ecommerce on the internet. You must have experience designing and building large and complex yet maintainable systems, and you should be able to do so in about one-third the time that most competent people think possible. You should have a BS, MS or PhD in computer science or the equivalent. Top notch communication skills are essential. Familiarity with web servers and HTML would be helpful but is not necessary.”
The job ad then explains bits of information about co-workers, compensation, equity and provides some contact information. At the end is a quote by Alan Kay, “It’s easier to invent the future than to predict it.”
Jeff Bezos certainly invented the future when he created the one click purchase system.
Amazon Changed The Game With 1-Click
The idea of entering your billing, shipping and payment information just once and then simply clicking a button to purchase an item in the future was unheard of before Amazon secured the patent in 1999.
Today we’ve got Apple Pay and Google Pay, but this was a breakthrough for online hassle-free shopping when Amazon did it.
When Amazon first patented one-click, it was still primarily an online bookseller. But according to R. polk Wagner, a professor at the University of Pennsylvania Law School and an expert on patent law,
“the 1-Click patent allowed Amazon to create a very strong position in the market.”
Even Apple began licensing the technology from Amazon for use on its website and iTunes.
But the real value of one-click became evident to the world one year after they secured the patent.
Amazon expanded its business to include the Amazon marketplace, allowing third-party buyers and sellers to sell new and used products on their platform.
Having a large database of consumer information they acquired using the one-click system gave them leverage in enticing users on both ends.
Amazon could simply say, we can give you access to a large number of customers literally with one click. At that point, Amazon’s growth went through the roof! No one could keep up.
Bezos was ahead of the game by miles!
Today, shopping cart abandonment is a huge issue for online retailers.
Several studies suggest that average shopping cart abandonment is around 70%. At that time, no other retailer had that kind of customer payment or preference information.
It became a huge asset for Amazon. Together with their recommendation system, quick shipping, easy returns policies and low prices, no one was even coming close to Amazon.
Going Further With Customer Data
Back at the Sleeping lady mountain resort, Bezos and his team were creating algorithms.
“Jeff’s algorithm wasn’t that much better than anybody’s algorithm that day,”
James Marcus recalled.
But it was later, after Andreas became chief scientist at Amazon that their data game went from zero to one hundred.
Andreas would have weekly meetings with some team members where they just looked at clickstream histories, to figure out why people actually behave the way they did on the Amazon website, why on Earth would they click here?
Clickstreams are a trail which Amazon follows to see which sites users come from, how they travel through its own pages and where they go to next.
Andreas, Amazon’s chief scientist hired an ad-tech specialist named David Selinger to work on customer behaviour analysis. They wanted to know each customer, who they were and what they wanted.
Their work led to personalised homepages for each customer, based on their past usage history, and tailored emails in their inboxes.
Bezos was still not satisfied with the data he was getting from Amazon’s customers. So step two of his plan? figure out what everyone else’s customers wanted. And he did just that.
Jeff Bezos is notoriously long-term in his thinking, and once Amazon figured out inventory, fulfilment, and basically how to ship stuff anywhere, Amazon’s team came up with a plan to reach out to huge companies and convince them to outsource their e-retail operations to Amazon.
And guess what? They agreed!
According to reports, the strategy was internally known as Launch and Learn.
Massive companies like ToysRUs, Borders, Waterstones, Marks & Spencer and Target were among household names to sign up to the program which saw them hand over the keys to their online kingdom.
Amazon would handle all their e-retail operations and in return, Amazon promised a boost in earnings.
What they didn’t realise at the time was, it probably wasn’t a good idea to let Amazon get that level of access to their online customers. Their payment information, their purchasing habits and so much more.
See, this is why Bezos is the OG. He knew at some point these companies would eventually realise how important ecommerce was becoming, so he came up with a plan to take them out without them even knowing.
According to the BBC, John Rossman, who was head of the programme at the time explained how traditional retail bosses didn’t understand the potential for e-commerce and digital business, and they essentially just viewed it as, ‘Hey here’s additional revenue,’”
“They really gave away the keys to a kingdom.”
When Amazon partnered with these companies, they would study their sector’s value chain and eventually expand onto their patch.
This went on for years. Yes, really!
Years later, these companies realised the value of what they were giving away.
Carl Casey, a former Chief Strategy Officer for Target later complained, saying “They learned a tonne on our dime, and we didn’t learn much”.
Lessons To Learn From Amazon’s Use of Data
The Amazon marketplace started slow, but it became a massive success. It allowed third-party sellers and even large retailers to sell products alongside products sold by Amazon. It presented a huge opportunity for sellers. There was just one problem.
Amazon gets to keep all the customers.
The marketplace is successful because Amazon is willing to share a great deal of data with their sellers.
In depth analytics allows customers to grow their stores and find new growth opportunities.
But only Amazon gets complete access. All the customer data stays with them.
Some merchants are completely fine with this. Amazon gives them an opportunity to get their products in front of millions of people.
Roland Brana, a motorcycle clothing business owner decided to join Amazon, and within months, his business boomed. His story and relationship with Amazon is well documented in a house of commons debate.
After years of solid sales, Amazon reached out to him and persuaded him to form a retail-manufacturer relationship with them. They wanted him to focus solely on supplying his goods and Amazon would take care of selling.
It was great, but after a short while, his orders dried up. This was odd. Roland logged into his merchant account, and saw that Amazon had received new stock, but he didn’t supply it. What he finds out next is in the third episode of our episode. Listen now to find out.
It is a strong lesson for entrepreneurs about the value of ownership and using data to make your own data driven decisions to drive growth for your own business.