Data and Technology Are Useless Without Clear Value
We live in the era of Big Data. Every day, we create around 2.5 quintillion bytes of data — and that pace is only growing. The past two years alone account for more than 90% of the total data we’ve accumulated.
With numbers like that, it’s no surprise that so many technology companies are prioritizing collecting massive amounts of data and analyzing it. This can be seen in the mobility industry, where many view Big Data as the key to solving long-standing problems such as congestion, accessibility, and environmental impact.
Yet amidst all the excitement, we must ask this question: Why are we collecting so much data?
It’s tempting to think that if we can just get enough data into the cloud, we can come up with big, revolutionary solutions. But that’s going about it backwards. We need to start with a problem that needs solving — not with terabytes of potentially useless data.
The truth is, many data-driven people in tech rarely stop to ask that question, if ever. And that’s a mistake. Big data and technology are only useful insofar as they help us address specific problems that people experience, just like design thinking leads us to believe. If we’re not guided by specific goals, we’re just hoarding bytes of data.
Take, for example, a city that installs “smart” streetlights on all major streets. Now every streetlight has Wi-Fi capacity and is transmitting data to the cloud. Implementing that system consumes a lot of resources. But to what end? Are people in the city safer, or able to travel where they need to go more easily? Without a clear plan for implementation, the answer is an obvious no.
A focus on big data rather than problems has real costs. After all, storing massive amounts of data in the cloud consumes resources. The challenges posed by cloud computing means that accessing that data quickly can be difficult. It makes no sense to devote so many resources to storing data without a clear plan about how to use it.
Trending AI Articles:
1. TOP 100 medium articles related with Artificial Intelligence
2. Neural networks for algorithmic trading. Multimodal and multitask deep learning
3. Back-Propagation is very simple. Who made it Complicated ?
Technology companies may say that the important part is to collect data now. We’ll figure out how to use it in the future. But if we keep collecting data without clear goals, when is that future going to arrive?
There’s a better, more workable approach. Instead of starting with the goal of collecting big data, identify a real problem that people face. Create a solution that can address the problem. Ideally, the solution shouldn’t require massive amounts of data nor massive computational power to run. Then, implement the solution from the bottom up. Test it and refine it.
As connected vehicles become the norm, automakers would do well to keep this idea in mind. Supporting a pilot problem for a single use case can be done at a very low investment. It’s really not necessary to make a $300 million acquisition to enter this space. Instead, we can focus on coming up with real solutions to real problems at a low marginal costs with already existing hardware and infrastructure inside the vehicle itself.
It’s tempting to think that if we can just get enough data into the cloud, we can come up with big, revolutionary solutions. But that’s going about it backwards. We need to start with a problem that needs solving — not with terabytes of potentially useless data.
Don’t forget to give us your 👏 !
