Jeff Immelt is the source of strategy around digital: GE’s William Ruh
Latest News »
- Galaxy Note 8 is the new weapon in Samsung’s arsenal
- Making bad ideas bigger doesn’t make them better
- Deals Buzz: Demerger of SRL Diagnostics from Fortis Healthcare deferred
- Right to Privacy: A timeline of events leading up to Supreme Court ruling today
- Train derailments triggering fear, say rail passengers
William Ruh, chief digital officer of the $123-billion multinational conglomerate General Electric Co. (GE), is credited with shaping the digital vision of CEO Jeffrey Immelt into one of the most talked-about “industrial internet” models out there. A 30-year industry veteran, Ruh also acts as CEO of the company’s $6-billion digital unit, GE Digital. In an interview, he talks about the birth of GE Digital, ingredients of a good digital strategy and the tie-up with Reliance Industries Ltd (RIL). Edited excerpts:
Tell us about GE Digital, referred to as the world’s largest start-up.
It really is a true transformation, that’s how I like to think about it. The way a company exists for 130 years because it has the ability to reinvent itself through process. And that reinvention is a difficult thing because it requires you to rethink everything that is making it successful. The good news is that we could see the threat…there are all these fantastic consumer start-ups: Amazon, Google, Alibaba and so on. They have undermined and taken over other businesses and I think it gave us a moment of reflection: What if somebody else is able to know more about our machines than us? If you start to think about the threat of somebody transforming you rather than saying, like the taxi services, “No one will ever be able to displace us.” And then you have an Uber come along! But once you think, “No, I’m afraid somebody will,” and you do it early enough, it gives you the opportunity to be the leader.
So, I was fortunate to reach out to Jeff Immelt and he could see that there were others who were beginning to use data and analytics in a new way that undermined other businesses. And he chose to ensure that no one would do it in our business. But it’s a hard thing. The fortunate thing is that he is much like a venture capitalist: he is willing to invest and, over a series of years, to turn it into a successful thing. From that standpoint, it’s got all the makings of a start-up but it comes with all those complexities of a big company—and you can’t ignore that.
What are the ingredients of a good digital strategy?
Being outcome-oriented is the most important thing. Especially in the industrial world, an outcome is very practical: say, I want to save 10% on fuel or I want to increase safety by 20%. The consumer world also has outcomes, but they are more like: I want to make my life more convenient or I want to talk to more friends or have more friends. So, when you look at the differences, the practicalities of the outcomes have to come first. You have to be truly able to deliver those outcomes. Think about ERP (enterprise resource planning): ERP actually promised no outcomes. ERP does allow for scale, but that would no longer be enough. It’s not like the ERP world where I walk up to a company and say, “Buy my technology” and you need an ERP. In a digital world, what you are going to have to say is, “I’m going to make your business run better by X amount.” I think that outcome is extraordinarily important for strategy.
The third thing I think is talent. This is the one people often ignore and the reason business people do it is because they do not know good software talent from bad software talent. That is what has made the start-ups we all know and love so successful. They attract talent and they know how to select talent. I think that’s a lesson we learnt early on. Getting the right talent is where most companies struggle.
The last thing: you have to have access to data. You know machine learning—everybody loves machine learning! But machine learning is useless without data. The data is the valuable thing. Look, you can pick up anybody’s algorithm, it doesn’t matter. It’s a question of who has the best data that allows you to get insights you couldn’t before. I think those are the recipes we see.
While you have all these designations—chief information officer (CIO), chief marketing officer (CMO), CDO, chief financial officer and, of course, the CEO—who do you think should own the digital mandate?
I know my title is CDO but Jeff Immelt is actually the chief digital officer of GE—he just has a better title, CEO! But the key point is that the business person has to own the strategy and in our company, the reason we are successful is that Jeff Immelt is the source of strategy around digital. He really has the outcome mentality and he knows that if he focuses on the outcomes, then everyone underneath him will figure out the right thing to do to achieve those outcomes. While we will continue to see a flux in CIO, CDO, CMO roles, the company that thinks its digital strategy is an IT strategy is doomed to fail. That doesn’t mean that the IT department has nothing to do with it; it’s just that you have to rethink your IT department so that it becomes the digital department. The IT has to be coupled with digital business.
You have this concept of ‘digital twins’ of products. Could you please expand on that?
Machine learning and artificial intelligence (AI) are quite important but they are typically good at reasoning once you put in a lot of data on what has happened. Unfortunately, it doesn’t give you a perfect answer because it’s not meant to. In the real world, you know that something is happening and then decide what to do about it. Now, the doing part of it is not going to be AI and machine learning—it’s going to be modelling and simulation. A digital twin is when you couple machine learning and AI to look at the past with modelling and simulation to predict all possible outcomes and pick the best. When you couple those together with a steady stream of real-time data where the digital twin is operating, you literally are able to have an instantaneous view of that machine or that process, and then you can always tell the human operators what the most optimal solution is. In the end, what you get is the crystal ball, because it predicts the best outcome that can occur for you. It sounds like magic but then the reality is, so did the cellphone 30 years ago.
How is your Predix platform evolving? And what is the status of your tie-up with RIL for the same?
We released Version 2 of the Predix platform in February 2016 and about a year after that we have 22,000 developers and over 250 applications running on Predix. We believe we have built a very unique platform for industrial assets and services, and for making them better. So, not only can you bring your data and do analytics, you also have a way to do reliability management, to create a service order—and all of that is what you want to do to act on the outcome.
With RIL, we are working with both the oil and gas, and the Jio businesses. In oil and gas, we are working with them on asset performance management, service automation and how the company’s oil and gas operations can continue to be at the leading edge. We are also looking at how we can take this to market. We are also working on the details of running the Predix applications in the networked environment of Reliance Jio. We are looking at a deeper relationship with RIL.