Ninety percent of the world’s data was created in the last two years (2016-2018). That’s 2.5 quadrillion bytes of data. For reference, the whole of Wikipedia takes up approximately 45GB of storage — less storage than the cheapest iPhone 8 ships with. As time goes on the amount of data and the rate of generation is only going to increase as more ‘things’ connect to the internet.

It’s cliched to say ‘data is the new oil’ but, in many ways, it’s true: data will fuel the oncoming AI-powered future just as oil fuelled the industries of the 20th Century. Just take a look at the most valuable companies in 2006 compared 2016.

Three of the six most valuable companies in 2006 were oil companies. In 2016 only one remained — and at a lower valuation. By 2017 ExxonMobil had slipped out of the top six completely. Today the most valuable companies in the world are tech companies.

In reality these ‘tech’ companies are actually data companies. Here’s a list of the biggest and most valuable tech companies in the world (not in order) and the kind of products and services it looks like they offer and what it is they really do.

It’s easy to think of these companies as the services they provide and split them up as different beasts. Yet they all have data at their core that informs their products, services, and decisions. While they offer different things and serve different segments (with some overlap), they are all data companies. They collect, process, store, and use data to get ahead, and stay ahead, of everyone else.

Companies like General Electric (another missing company from 2006 vs 2016) and Siemens, who existed before the tech boom but have managed to adapt and thrive somewhat, now sell themselves as data companies. That’s telling.

What is it about data that makes it so valuable?

Data changes the nature of competition. It lets you build a moat around your products and services due to the data network effect.

The more data a company has the more it can improve its products (and build new ones) and the more profitable it can become. Data can predict the moment someone is ready to buy from an eCommerce site, what demand for energy will be in a certain area at a specific time, when a jet engine needs servicing, or when people are at risk of diseases. Each time a company uses data to make a decision more data is created. Decisions with positive outcomes are repeated, decisions with negative ones are not. In this way data companies outstrip their competitors because they have evidence of what works.

Tesla, which sold roughly 25,000 cars in the last quarter, is more valuable than General Motors despite quarterly sales of 2.3M cars. How can this be so? As Tesla drivers zip around the world they generate huge amounts of data. This data belongs to Tesla, which uses it to improve its cars and their self-driving functions. The more data they have the more they can improve and the more likely they are to succeed in selling more Teslas and becoming one of the biggest providers of autonomous transport. GM doesn’t have that kind of data focus.

This Tesla-specific example can be applied to any one of the companies listed above. Any company wanting to thrive in the future should take note.

Data is your competitive advantage

There are few who can compete on the scale of the data unicorns. But that doesn’t mean you can’t become a data, or a data-first, company. In fact, it’s why you most definitely should become a data-first company.

There’s sometimes the idea that more data is better and this takes over the ‘doing stuff with data’ bit of being a data company. While having big sets of data is beneficial there’s no point having data centres full of data that isn’t being used. Google’s Chief Economist, Hal Varian, told the Economist that data exhibits ‘decreasing returns to scale’. That is: data in and of itself isn’t valuable, it’s the quality of the analysis and the algorithims that crunch it that draw the value out of the data.

It’s what you do with the data once you’ve got it that matters. This is the reason data scientists are being snapped up straight out of PhD programmes at ridiculous salaries: even the data unicorns are struggling to use their data to full effect. Don’t worry about not having petabytes of data, just worry about doing something with the data you’ve got.

Data can be used to improve operational efficiency, financial performance, and customer relationships. It can also be used to enhance existing services or help create new classes of smart products specific to you and your customers.

Viewed through this lens it’s easier than it seems (but not easy) to become a data-first company. All you need are some questions and desired outcomes coupled with the data strategy, infrastructure, and techical expertise to unlock the value hidden in your data.

This was originally posted on Initial Commit.