Digital supply chains: why you need to turn your data from junk into an asset
With the world’s volume of digital information set to increase ten times by 2020, EY Global Technology and West Region Leader, Advisory, Dave Padmos says it’s time to completely rethink your approach to data
If you are a supply chain professional, here’s a challenge for you: I dare you to deliver a world-class, industry-leading, digital supply chain.
This is not meant to question your ability, but to get you thinking differently about one thing: digital data.
The volume of data will continue to grow
The world’s total digital data volume has been doubling every two years. There are 1.7 megabytes of new information created every minute for every person on Earth. In 2013, world data volume was estimated to be 4.4 zettabytes. I did not make up that word; a zettabyte is one trillion gigabytes. The volume of digital data is projected to increase ten times – to 44 zettabytes – by 2020.
You know what? I think these projections are too low.
Every minute, internet users share more than 2.5 million pieces of content on Facebook, tweet more than 300,000 times and send more than 204 million text messages.
Robots, sensors and automated processes produce plenty of data, too, and that’s not included in these texts and tweets. We should be particularly aware of the explosive growth projected for the Internet of Things (IoT) sector over the next half decade – it’s estimated there will be 28 billion IoT devices in the world by 2021.
Better data infrastructure isn’t just needed, it’s essential
These new IoT devices will not just produce floods of new data – they will also need robust data infrastructures if they are to function anything like we would like them to.
The era of big data is here now, it’s growing fast and frankly it scares me. We’re getting to the point where we need quantum computing just to sort through this stuff. If that doesn’t scare you too, it should, because big data is capable of driving us to extinction.
That’s a big statement, but here’s why I’m making it.
A recent survey of 1,500 companies found that on average, respondents could identify only 14 percent of their data as business critical. Fourteen percent!
Another 32 percent of that data was redundant or obsolete.
And 54 percent of their data was identified as “dark data.” Dark data is as sinister as it sounds: we are storing zettabytes of data, and we don’t even know what it is.
Why you need a better data strategy
Essentially, cloud technology is allowing us to be sloppy with our data. Instead of sorting through it and keeping what we need, different kinds of storage are letting us pile up a bigger draw of junk data. It’s enabling us to store everything.
That’s not a strategy.
Machine learning and artificial intelligence only offer parts of the answer. Yes, they’re important, central elements of any plan to dealing with data, but robots are not a panacea.
Technology shouldn’t put us in business, it should enable our businesses. But the way it’s being used now is making us think we're addressing these problems, when really it’s just letting us avoid confronting them. Machine learning needs hard work and enterprise to provide context and direction for its operations.
How to think about big data
Earlier this year, at the USC 2016 Global Supply Chain Summit, I shared my one simple goal: to get all of us thinking about three aspects of the big data challenge that can save us from extinction.
Specifically, I recommend focusing on three aspects of the big data challenge – controlling cost and complexity, reducing volume and risk, and improving insight. Here’s how:
First, develop a data strategy. I am not talking about something from IT, I am talking about a strategy with the same level of support and funding as any other product. That’s right: Make data strategy the new product.
Next, focus on data integration as a large part of that digital strategy. Whether your data comes from wearable devices or automobiles, the resulting data needs to integrate into your ecosystem.
Third, simplify and standardize your data to reduce the size of your data lake and prevent it from becoming a swamp.
The questions you should be asking
Finally, ask the right questions about and of your data. Gathering and storing the right data is only the first step. Ensuring you can access it is a vital second.
But none of this will add any value to your business if you aren’t asking the right questions to extract insights from that data. So ask yourself:
- How well am I managing today's data explosion?
- What is my data strategy, and how does it support my business goals?
- How much data am I willing to own and protect?
- What insights are a must and which are nice to have?
- What steps do I need to take as we move toward a more digitally connected supply chain?
And if you aren’t asking those questions, you can be sure that someone else will be, which is why today’s unprecedented volume of data is also leading to unprecedented business disruption. This trend is only going to continue.