In The Future, All Products Will Automatically Improve As People Use Them

Timo Elliott

This is the Tesla autopilot in action – and it’s a great analogy for the future of digital business.

As every Tesla car equipped with autopilot drives down the street, it’s using sensors and the Internet of Things to gather vast amounts of Big Data. And then it uses analytics, machine learning, and artificial intelligence to turn that data into a superior driving experience.

That’s what digital business is all about – using the latest technologies to create better products and services. But Tesla is taking things to a whole new level. The data from every car is sent to headquarters and shared with every other car on the road. So your car knows what to look out for even if you’ve never been on that street before.

This means that Tesla has essentially turned itself into a massively parallel learning machine. The Tesla customer experience now improves automatically the more you drive and the more other people use the product.

In addition, the company is gathering detailed information that can be used for many of other business opportunities in the future. And that’s perhaps why Tesla is now the most valuable U.S. car company, eclipsing General Motors, even though GM makes over 100 times as many cars.

These types of self-improving products are now starting to take over the world. For example, AlphaGo’s algorithms shocked the experts last year by beating one of the world’s strongest Go champions. And it’s been steadily improving ever since – to the point where one Chinese Go master says it now “plays like a God.”

So imagine — what if your products and services automatically improved as more people used them?

New customer chatbots are a simple example of self-improving interfaces. Modeled on consumer services such as Siri and Alexa, these chatbots are poised to make all interactions with computers easier, from purchasing items online to working with internal business applications. You can simply ask things like “what colors are available?” or “what are the details of this order?” and the chatbot will respond. And because these services leverage machine learning, the quality of the responses will automatically improve over time and as more people use them.

And that’s just the start. Machine learning can now be embedded into every customer experience and operational process. For example, chemicals giant BASF was able to use machine learning on repetitive decisions in the finance function, improving invoice matching from 70% to 94% – and that score should rise as the algorithms master the remaining variations.

But there’s one key prerequisite for this vision: good data

Artificial intelligence works best when you have large amounts of high-quality training data, applied to a specific, clearly defined business problem. So the first step to introducing self-improving products and services to your customers is a single, consistent, governed view of all necessary information, no matter where it’s stored, inside or outside the organization.

The self-fulfilling cycle of data, AI, and better product experiences

More data means better artificial intelligence algorithms, which means better customer experiences, which means better customers, which means more data… The winners in digital business will be the ones who first unleash this virtuous circle of self-improving AI and “self-driving business.”

Autopilot Full Self-Driving Hardware (Neighborhood Short) from Tesla, Inc on Vimeo.

Develop a machine learning strategy that will change the basis of competition in your industry. Learn Why Machine Learning and Why Now?


Timo Elliott

About Timo Elliott

Timo Elliott is an Innovation Evangelist for SAP and a passionate advocate of innovation, digital business, analytics, and artificial intelligence.