 Live from London, England, it's theCUBE, covering AWS Summit London 2019, brought to you by Amazon Web Services. Welcome back to London, everybody. This is Dave Vellante and you're watching theCUBE, the leader in live tech coverage. We're here at the London AWS Summit 12,000 people here for a one day summit, which is typically the size of a large tech event that we cover in Las Vegas. John Lovelock is here. He's the vice president and analyst at Gartner. He's essentially Gartner's chief forecaster. John, thanks for coming to theCUBE. Pleasure to have you. Thanks for having me. It's a great show today. Great event. I'm happy to be here. Yeah, you're in from Toronto, and yeah, I'm very impressed with the crowd here. Obviously a developer crowd. You and I are in ties. They see us coming. They think we're trying to sell them something. We seem to have a monopoly on all the ties in the room. We have a very diverse group here, but they're all very enthusiastic to be here. It's been a great conference. So everywhere we go, we hear numbers. Obviously people want to talk about the size of the market. It's growth. That's your job to figure that out. I mean, I've heard numbers that it's a multi-trillion dollar market now growing faster than GDP. I'd love to get your thoughts on that. Where do we start? Top level macro. What's the picture? Top level macro cloud in all of its forms is the fastest growing tech that Gartner is tracking. There is definitely IT spending in there. We are in the 20, 25% growth globally. Nothing else comes close. The overall growth rate for total IT spend this year is 1.1%. Cloud at 25%, it is moving the market. The only way it's doing that of course is by taking money away from legacy lines of business. You know, it's about the switch in IT spending preference from legacy IT and moving that into cloud IT in all of its forms. So it's a share shift you see going on. So you've got the total market growing below global GDP. Is that a fair statement? It's just below global GDP. It usually tracks pretty close to you would take, right? I mean, that's logical that it would. Actually, there's almost no correlation between GDP and IT spend. Really? It is one of the biggest things that we have to fight against. So that's a myth. Absolute myth. Here to tell you, it is dead. There is a slight correlation, but there's no causational move between GDP and IT spending, just not there. So that makes your job even harder. It does. We have to watch what the vendors are selling and what they hope to spend. But most importantly, it's about what the demand side is doing. What are people doing? Why are they buying what they're buying? How much are they spending on the stuff that they have? What's getting retired and what gets replaced with something new? And that's the whole big shift that we're seeing. There's a lot of things that are being retired out of the CIO's bag of tricks and a lot of new things coming in. So the spending shift that we're seeing, it's all down to where is the CIO in their journey? How quickly are they able to move from legacy IT to the new IT? How quickly is their business moving into being a digital business? So, okay, so it's one plus percent growth on what? We're talking two trillion, three trillion? I mean, what's the four trillion? Four trillion dollars by 2020. Okay. And you said cloud computing is growing at 20 to 25%. AWS, a $30 billion run rate business now growing at 42% in constant currency. So growing at nearly, or maybe even slightly more than twice the market. That's astounding that, I mean, they're basically adding nine to $10 billion a year. They are right in the sweet spot for cloud growth. Do you think they hit the law of large numbers? I mean, people have been predicting that for years. Can accompany that size in your experience continue to grow at that pace? Absolutely. There is nothing stopping AWS from taking advantage of this market. We are nowhere near saturated for cloud changes. Most of software spend is still on legacy and maintenance of software on-prem. There's still a great deal of money being spent on servers and infrastructure and networking equipment and all of that gets bled out into the cloud eventually. Where they have opportunity to shift? Is almost limitless. You know, the amount of money that is being spent by enterprises on cloud is different around the world. In the US, where cloud basically started, where the infection started and is spreading around the world. Back in 2016, they were about 16% of the overall enterprise spend was on cloud. The rest of the world is tracking towards that. We have countries that are close. The UK, Canada, one, two years behind. France, Germany, three, four. Most of Europe in the three to five years behind. We have some countries that are lagging a little bit further and several that are just resisting that are not on track to get to cloud. We don't see them getting to cloud even in the 10 year time span. But the fact that cloud spend in the US still makes up over 50% of global spending on cloud, but only 25% of global spending on IT, a lot of money's still left to move over to cloud. That's interesting. Those are the facts that suggest that there is a delta in cloud adoption between the United States and the rest of the world. The vendor narrative would not have you believe that. Am I getting that right? Is it not only slower adoption, but are they as sophisticated in their adoption, or is there a delta there as well? There is a delta also in the sophistication. We know that there's a skill gap when it comes to cloud. Everywhere in the world faces the skill gap of the number of people they need with the new skills in cloud and the people they have with the skills that they have. Many companies are missing the fact that some of their COBOL programmers are the ones that should be developing their new cloud applications because it's about changing the business and nobody knows their business better than the guys that have been writing the legacy apps that have been running the business for the last 20 years. So the training opportunity is actually with their COBOL programmers, with their long-term programmers, we're not seeing that getting into the market as much as we'd like. So your job, a very difficult job, especially the consolidation makes your job harder in a way because it's harder to squint through. Companies want to tell you what they want to tell you, but you got to figure out what the truth is. When you think about cloud, AWS relatively straightforward, it's a pure play. They now report their numbers. That must have helped you a lot, but a lot of vendors will throw everything, the kitchen sink numbers for cloud. So you have to parse through that. You have to come up with common definitions across. I mean, a good example, certainly IBM, Oracle broke it out earlier, but now they sort of consolidate everything. One wonders, okay, where are they trying to hide? Not to pick on people, but they're large established legacy companies but they want to show their investors, oh, we're growing at this high rate. So how do you parse through that and squint through that and then come out the other end with the real numbers? Well, we have a lot of advantages at Gartner. We spend millions of dollars every year on surveying out globally. We get responses back from CIOs from around the world. We do the largest CIO survey every single year. So we're getting feedback on where the money is being spent. We also have interviews that we do with our clients every single day. We do over 250,000 inquiries with clients every year. So we're getting a great deal of feedback from where the money is being spent. And we have to reconcile both sides of it, what the vendors are expecting to make, what they're telling us that they're making, and reconciling that with what we're being told is being spent. So we have multiple sides to get to this angle. And again, when you start with a vendor, you start with their global revenue. It has to parse out from there. It's got to match the income statement somehow, but so you've got the empirical data from your surveys, you've got the vendor data, you bottom up, you can do that, and you've got the anecdotal data from your inquiry, you know, your corporate memory. And your job is to put all that together. Yeah, and we're tracking what we call our peer insight data where we're asking our clients, you know, when they're making a choice, which vendors are they choosing, which friends are they considering, why did they make the choices they are? We have our talent neuron database where we are scraping job postings from around the world. So we have somewhere over four billion job postings covering the last five years. So when a company is telling us that they have a large new division, we can go back and say, I don't see you ever hiring those people. So we do have multiple points of light that all really have to come together. It is a tremendously interesting job and a bit of a challenge, but it's one that keeps me up every day. I often joke, will Dr. Oz, oh, sorry, Dr. Watson replace Dr. Welby. And the answer comes back, well, you won't replace Dr. Oz because you still have to have that nurturing and that interaction. Do you feel as though machine intelligence, based on what you know, your Gartner analysts, you got experts and many I'm sure that follow artificial intelligence, machine intelligence, do you feel like you guys can start applying AI, deep learning, et cetera to identify patterns to make your job easier, more effective, more science than art? What are your thoughts on that? Well, we have taken a different route. Artificial intelligence requires a lot of good, bad data going into it in order to make the right decision. IT is changing so quickly, it's difficult to get enough data points together to train an artificial intelligence. We do do some augmentation, we do have tools that automate certain processes for us and feed us results from multiple millions of data points. But at the end of the day, it's not about coming up with $4 trillion that's interesting to anybody. It's the why is it $4 trillion? Why is it a different $4 trillion than last year's $3.9 trillion? And what's the changing environment that is going on? And the story behind it, the segments, the share shifts and all those other trends that you're seeing. Because everybody on this floor, all of these ISV startups, they desperately want to make my numbers wrong. They want to change the market in such a dramatic way that they disrupt all of the IT spending. I can't train an AI to watch for that. Is your background in econometrics? Are you an economist? Are you a math whiz? Are you a computer scientist? All of that. I have degrees in economics and statistics. I have 40 years almost in computer programming been through this cycle for many, many times. Been a great job for me as all of my skill sets coming together. And you're obviously not a one-man band. You mentioned you spend millions of dollars on surveys, 250,000 inquiries, but still hurting all that data and actually making sense of it is a challenge. How do you manage that? How are you evolving your systems, your models? I mean, what you use today, the tooling is different than it was 10 years ago and you've got to stay current. Yeah, our forecasting model, generically we call them market dynamic models. And what they do is build out user behavior. Where is demand coming from? How are we fulfilling on that demand? What do we do with the investments that we've already made? These models run from 1980 through 2030. It takes somewhere in the neighborhood of 800,000 calculations to come up with one segment forecast for 43 countries. We have over 250 segments that we forecast so you can see the complexity that we're getting into. There are over 250 analysts, the gardeners who are working on from what we call our technology and service provider research group to help our vendor clients know where their market is, know where it's going and the partners that they should be looking for. Do you factor in or how do you factor in? If at all, geopolitical trends, tariffs, things of that nature, what do you say? You know what? We're going to do a clean forecast and let the market figure that out. How do you handle that? At the end of the day, there's two very important pieces within a model. They break into signal and noise. The signal is the shifting buying patterns. When the demand level changes, there's a signal there. When a choice pattern changes, instead of buying licensed software, I'm starting to buy cloud. That's a signal change. Those are the things that we focus on. The stuff that you were talking about, the economic situations, Brexit, tariffs, China, those are all noise. They're important. They have to be taken account of in the model, but they're not the most important thing. All right, Brexit right now is depressing the European spending on IT. It is below that 1.1% growth rate because of the uncertainty. People are keeping their hands in their pockets when it comes to big changes in IT. But the big shift is still happening. We are still seeing movement towards cloud. We are still seeing movement towards digital business. All of those big signals are there. They're dampened a little bit by the noise of the economy. So the rip currents, obviously cloud. You mentioned that digital business, which I interpret as data orientation toward a business. A little bit more, yeah. But please add some color to that. And what are some of the other rip currents that you're seeing? Artificial intelligence is another rip tide that is moving through it. It's a big trend that is changing what's expected of technology at every level. Digital business is changing what's expected of customer interactions at every level. Digital business ecosystems, where companies are able to interact in a way that moves data from one organization to the other without necessarily having trust, commitment, or a contract is a major change that we are seeing. It reduces the friction of handoffs between one business and the other, speeds the process, drops the cost. A lot of your clients are large established businesses. Gartner's well known for advising those businesses. Many of those businesses, their data lives in silos. They have legacy infrastructure, technical debt, call it whatever you want it. And they're getting disrupted by these, the guys who are doing cloud native, all the guys out here that want to make your forecast wrong. How does Gartner see, just sort of anecdotally, those guys closing the gap, the traditional, the incumbents closing that gap. And to a certain extent, they don't have to, right? Certainly their size is going to give them longevity, whether they make the change or not. They will see their influence on the market chip away if they don't start to. They don't have the same urgency as the small vendors that are moving quickly. Where we see them doing things is very patiently and incrementally. They're taking different processes and moving them to the cloud. It is very common to see them take something that they're already doing or comfortably doing and moving that to a new platform and improving that small piece. Incremental change. The world gets better with incremental change. Where we love to see them do something is where they actually change the business model first using the technology that's going to enable that. We have the company in China who has managed to get home food delivery, cheaper than buying it in a restaurant. Because they change the business model first. They work with the places that are selling the food, they're doing Groupon, they're doing direct cash ordering, they're doing guaranteed sales so that they can get food less expensively. They're using artificial intelligence to work out delivery routes and pickups so that multiple deliveries are made at the same time. In most of the world, that's not been the model. They've changed one part of delivery. We're going to make it easier for you to order food on your phone and then we're going to charge you for the delivery and we're going to charge you more for the food that's coming in. That's incremental. It's nice, it's helping, but when we change the model first, the outcome is so much better. So last question, U.S. largest market, right, in terms of- Largest market for cloud, 58% of cloud spend globally. And IT spending generally, correct? Correct. China, when do you think, do you think China will overtake the U.S. as the largest market for IT spend? China right now has got almost double the growth in cloud spending of the U.S. It is, as a percentage of spend's still well below. But they are the only country that is breaking the trend of following the U.S. They are on a much steeper incline. They could be above the U.S. spend by 2025, even with the growth rate that the U.S. is on. Wow. John, awesome having you on. Thanks so much for coming to theCUBE. Thank you for having me. Really a pleasure having you. Great insights from Gartner Analyst, John, love luck. And you're watching theCUBE. We're bringing it all to you live from London. This is Dave Vellante. We're right back, right after this short break.