GE’s Jeff Immelt on digitizing in the industrial space

In this first installment of our two-part interview, General Electric chairman and CEO Jeff Immelt explains why industrial companies are now in the information business—whether they like it or not.


Five years ago, US corporate giant General Electric had to decide how to respond to the digitization of the industrial sector: build or buy? As chairman and chief executive officer Jeff Immelt reveals in this interview with McKinsey’s Rik Kirkland, GE decided to build, and has since added thousands of staff and embraced data and analytics as the driving force transforming its operations globally. It’s a bold push driven by Immelt’s belief that the storied company, founded 123 years ago in Schenectady, New York, must shape its own future in a rapidly evolving industrial landscape. An edited transcript of his comments follows.

Interview transcript

Digitizing in the industrial space

We think about this as digitization of the industrial world. We as a company didn’t go to bed one night and say, “We can’t be an industrial company anymore. We need to be more like Oracle. We need to be more like Microsoft.” It happened more on an evolutionary basis, really based on the industries we’re in and the technology we serve.

You think about a jet engine today or a locomotive or an MRI scanner. A new jet engine might have a hundred sensors on it. These sensors have the capability to take continuous data about the heat of an engine, fuel consumption, the wear of the blades, the environment it’s taking off in—a series of things. And one flight between New York City and Chicago produces a terabyte of data.

So industrial companies are in the information business whether they want to be or not. This is going to happen in the industrial space. Now, add to that a series of decisions every company needs to make: “Do I outsource all of that? Do I do it myself? Do I change my business model accordingly?” The decision we’ve made is that we just want to be all in.

We want to treat analytics like it’s as core to the company over the next 20 years as material science has been over the past 50 years. We can hire the talent. We can evolve our business model accordingly. We need to treat our service agreements to share outcomes with our customers the same way an IT company might approach that in the future. So, in order to do that, we have to add technology, we have to add people, we have to change our business models. We have to be willing to do all those things.

But the point I’d make to people is, if you think about today, 15 percent or 20 percent of the S&P 500 valuation is consumer Internet stocks that didn’t exist 15 or 20 years ago. The consumer companies got none of that. When you look at retailers, banks, consumer-product companies, they got none of that. If you look out 10 or 15 years and say that same value is going to be created in the industrial Internet, do you as an industrial company want to sit there and say, “I don’t want any of that. I’m going to let a Newco or some other company get all that”? Is that really what you’ve relegated yourself to?

So I think all these things led us to say, “Let’s build it. Let’s see if we can be good at it. We may be wrong. We don’t think so, but we may be wrong. But let’s not sit back and just say, ‘Look, that’s somebody else’s job,’ or ‘We’re not good enough to do it,’ or ‘We can’t change.’” We’re unwilling to take that as a fait accompli.

Building versus buying

We went through a process of “make versus buy,” “in versus out.” We basically said, “Look, do we want to make a big acquisition in analytics or IT?” And we analyzed a bunch of different cases and said, “We don’t have the foundation inside the company to do a big acquisition. Do we want to partner, or do we want to do it ourselves?” We have lots of good software partners, but, basically, we said, “We need to do this ourselves. Let’s err on the side of seeing if we could approach it in that way.” That was 2010. So we brought people in from the outside. We built a center in California. We started populating our businesses. Roll forward, we started doing applications with customers. We started building it into our service business, things like that.

I could give you a bunch of different analogies, but in the case of our locomotive customers, they have a phrase called “velocity.” Every CEO of a railroad could tell you their velocity. The velocity tends to be, let’s say, between 20 and 25 miles per hour. This tends to be the average miles per hour that a locomotive travels in a day—22. Doesn’t seem very good. And the difference between 23 and 22 for, let’s say, Norfolk Southern, is worth $250 million in annual profit. That’s huge for a company like that. That’s one mile [per hour]. So that’s all about scheduling better. It’s all about less downtime. It’s all about not having broken wheels, being able to get through Chicago faster. That’s all analytics.

Being a platform company and an app company

Inside in the company, we’re about $5 billion in revenue—this is from software, analytical applications, things like that. We’ve built up a population of applications; we’re approaching $500 million of productivity a year. Now what we’re trying to do is push that back inside the company. We’re selling it, but we want to get our own internal company on the same basis, on the same platform, using the same skills—what we call the “digital thread.” We want the digital thread to go from engineering all the way through our installed base.

We’ve made the decision that we’re going to try to be both a platform company and an application company. So we have a platform called Predix, and we’re building applications on top of that. We’re probably the only industrial company that’s actually trying to do its own. And we’re opening up our platform to our customers. We’re saying to our customers, “Look, if you want to write apps, applications on Predix, you’re free to do it.”

I always think risk first, like most CEOs. I basically say to investors, “Look, if all we did is we got more productivity, higher service sales, applications, you guys are going to love this. If we end up having the platform that works, it’s a whole new company, you know? So you get that for free.”

But that’s why I circle back and would say to any CEO, industrial or nonindustrial, that where we are right now is going to be the most important thing that you’re going to work on, at least in this era. And you give up your latitude at your own peril.

Hiring the right people

We have probably hired, since we started this, a couple thousand data scientists and people like that. That’s going to continue to grow and multiply. What we’ve found is we’ve got to hire new product managers, different kinds of commercial people. It’s going to be in the thousands.

So what you’re going to do is you’re going to blend them with the GE team, and then we’re going to recruit differently. When we go to college campuses, we’re going to look for different skills. We’re going to put them in different training programs. It’s a combination of we’ve got to bring our culture along, but that’s not enough. We’ve got to bring thousands of people in from the outside. That’s the only way we’re going to get there fast enough.

This is something I got wrong. I thought it was all about technology. I thought if we hired a couple thousand technology people, if we upgraded our software, things like that, that was it. I was wrong. Product managers have to be different; salespeople have to be different; on-site support has to be different. We’ve had to drill and change a lot about the company. And I just think it’s infecting everything we do. It’s infecting our own IT. It’s infecting our own manufacturing plants. It’s infected everything we’re doing, I think in a positive way.

Changing company culture

When you think about internal change—culture, people, leadership development—again, here’s a time where multiple things happened at the same time. We started our digital initiative maybe five or six years ago. We’ve also, as a company—and I don’t think GE’s unique—lived through the financial crisis. And we’re an old company. We live in highly regulated industries. What we found was our culture was too complicated to get the work done the way we needed to get our work done, both in terms of how we were trying to digitize and how we were trying to survive in terms of a more highly regulated world. And just think about our footprint and the complexity of running a global operation: even since I was CEO, GE has gone from 70 percent inside the United States to 70 percent outside the United States.

So what we’ve tried to do inside the company is really just drive what we call a “culture of simplification”: fewer layers, fewer processes, fewer decision points. We’ve adapted the lean tools in what I would call a Silicon Valley approach, what we call “Fast Works.” We’ve embraced some of the Silicon Valley tools in terms of putting everything on the clock, bringing commercial intensity into the company. The way I describe that is, like most big companies, we’re willing to take all kinds of market risk so that we don’t have to take internal risk, right? We try to say, “Look, let’s actually be aggressive in the markets, and let’s count on our own execution to risk reduce inside the company.” And broadly, getting to digitization, we’re democratizing information inside the company; getting IT tools that were contemporary in a mobile setting, and we call these things the culture of simplification.

My notion is we’re in a permanently complex world. And this historical organization chart with lots of processes is a thing of the past. We’ve basically unplugged anything that was annual. The notion is that, in the digital age, sitting down once a year to do anything is weird, it’s just bizarre. So whether it’s doing business reviews or strategic planning, it’s in a much more continuous way. We still give a lot of feedback. We still do a lot of analysis of how you’re performing. But we make it much more contemporary and much more 360-degree. So somebody can get interactions with their boss on a monthly basis or a quarterly basis. And the data you get is being collected by your peers, the people who work for you, in a much more accurate and fluid way.

About the author(s)

Jeffrey R. Immelt has been the chairman and CEO of General Electric since 2001. Rik Kirkland is the senior managing editor of McKinsey Publishing and is based in McKinsey’s New York office.

More on Organization
Interview - McKinsey Quarterly

Leadership and behavior: Mastering the mechanics of reason and emotion

Executive Briefing - McKinsey Quarterly

The CEO guide to customer experience