 Welcome to the new abnormal. Yes, you know, the pandemic, it did accelerate the shift to digital, but it's also created disorder in our world. I mean, every day it seems that companies are resetting their office reopening playbooks, they're rethinking policies on large gatherings and vaccination mandates. There's an acute labor shortage in many industries and we're seeing an inventory glut in certain goods like bleach and hand sanitizer. Airline schedules and pricing algorithms, they're all unsettled. Is inflation transitory? Is it a real threat to the economy? GDP forecast or seesawing? In short, the world is out of whack and the need for fast access to quality, trusted and governed data has never been greater. Can coherent data strategies help solve these problems or will we have to wait for the world to reach some type of natural equilibrium? And how are companies like Google helping customers solve these problems in critical industries like financial services, retail, manufacturing and other sectors? And with me to share his perspectives on data is a longtime CUBE alum, Bruno Aziza, he's the head of data analytics at Google. Bruno, my friend, great to see you again, welcome. Great to see you, thanks for having me, Dave. So you heard my little narrative up front, how do you see this crazy world of data today? I think you're right. I think there's a lot going on in the world of data analytics today. I mean, certainly over the last 30 years we've all tried to just make the life of people better and give them access more readily to the information that they need, but certainly over the last year and a half to years we've seen an amazing acceleration in digital transformation. And what I think we're seeing is that even after three decades of investment in the data analytics world, the opportunity is still really out wide and is still available for organizations to get value out of their data as looking at some of the latest research in the market. And only 32% of companies are actually able to say that they get tangible, valuable insights out of their data. So after all these years, we still have a lot of opportunity ahead of us, of course, with democratization of access to data, but also the event in machine learning and AI so that people can make better decisions faster than their competitors. So do you think that the pandemic has heightened that sort of awareness as they were sort of forced to pivot to digital that they're maybe not getting enough out of their data strategies, that maybe they're whatever, their organization, their technology, their way they were thinking about data was not adequate and didn't allow them to be agile enough. Why do you think that only 32% are getting that type of value? I think it's true. I think one digital transformation has been accelerated over the last two years. I think, if you look at research the last two years I've seen almost a decade of digital acceleration happening, but I also think that we're hitting a particular time where employees are expecting more from their employers in terms of the type of insights they can get. Consumers are now evolving, so they want more information and I think now technology has evolved to a point where it's a lot easier to provision data cloud environment so you can get more data out to your constituents. So I think the connection of these three things, expectation of employees, expectation of customers to better customer experiences and of course the global environment has accelerated quite a bit where the space can go and for people like me, 20 years ago nobody really cared about databases and so forth and now I feel like everybody understands the value that we can get out of it and we're kind of getting in the sexy territory finally data now is sexy for everyone and there's a lot of interest in the space. You and I met of course in the early days of Hadoop and there were many things about Hadoop that were profound and of course many things that just were overly complex, et cetera but one of the things we saw was sort of the centralization we thought that Hadoop was going to send five megabytes of code to petabytes of data and what happened is everything came into the centralized repository and that centralized thinking, the data pipeline organization was very centralized. Are you seeing companies rethink that? I mean, has the cloud changed their thinking? Especially as the cloud expands to the edge, on-prem, everywhere, how are you seeing organizations rethink the way they, their regimes for data? Yeah, I think we've seen over the last three decades kind of the pendulum, right? From really centralizing everything and making the IT organization kind of the center of excellence for data analytics all the way to now providing data as a self-service application for end users and I think what we're seeing now is there's a few forces happening. The first one is of course multi-cloud, right? So the world today is clearly multi-cloud and it's going to be multi-cloud for many, many years. So I think not only are now people considering their on-prem information, but they're also looking at data across multiple clouds and so I think that is a huge force for chief data officers to consider is that you're not gonna have data centralized in one place, nicely organized because sometimes it's gonna be a factor where you want to be as an organization. Maybe you're going to be partnering with other organizations that have data in other clouds and so you want to have an architecture that is modern and that accommodates this idea of an open cloud. The second problem that we see is this idea around data governance, intelligent data governance, right? So the world of managing data is becoming more complex because of course you're now dealing with many different speeds, you're dealing with many different types of data and so you want to be able to empower people to get access to the information without necessarily having to move this data so they can make quick decisions on the data. So this idea of data fabric is becoming really important and then the third trend that we see of course is this idea around data sharing, right? People are now looking to use their own data to create a data economy around their business and so the ability to augment their existing data with external data and create data products around it is becoming more and more important to the Chief Data Officer. So it's really interesting, we're seeing a switch from the Chief Data Officer really only worried about governance to this world now worried about innovation while making sure that security and governance is taken care of. We call this freedom within the framework which is a great challenge but a great opportunity for many of these data leaders. You mentioned several things there, self-service, multi-cloud, the governance key especially if we can federate that governance in the decentralized world. Data fabric is interesting. I was talking to Jamak Degani this weekend on email. She coined the term data mesh and there seems to be some confusion. Data mesh, data fabric, I think Gartner's using the term fabric. I know like NetApp I think coined that term which to me is like an infrastructure layer. But what do you mean by data fabric? Well the first thing that I would say is that it's not up to the vendors to define what it is. It really is up to the customer. The problem that we're seeing these customers trying to fix is you have a diversity of data, right? So you have data stored in the data mart in the data lake in the data warehouse and they all have their specific reasons for being there. And so this idea of a data fabric is that without moving the data can you one, govern it intelligently? And two, can you provide landing zones for people to actually do their work without having to go through the pain of setting up new infrastructure, moving information left and right and creating new applications. So it's this idea of basically taking advantage of your existing environment but also governing it centrally and also now providing self-service capability so people can do their job easily. So you might call it data mesh, you might call it a data fabric. The terminology to me doesn't seem to be the barrier. The issue today is how do we enable this freedom for customers? Because I think what I've seen with vendors out there is they try to just take the customer down to their paradigm. So if they believe in all the answers need to be in the data warehouse they're going to guide the customer there. If they believe that everything needs to be in a data lake, they're going to guide the customer there. What we believe in is this idea of choice. You should be able to do every single use case and we should be able to enable you to manage it intelligently, both from an access standpoint as well as a governance standpoint. So when you think about those different and I like that you're making it somewhat technology agnostic. So if it's whether it's a data warehouse or a data lake or a data hub, whatever you want to go, data mart, those are nodes within the mesh or the fabric, right? That are discoverable, accessible, I guess governed, there's got to be some kind of centralized governance edict, but in a federated governance model so you don't have to move the data around. Is that how you're thinking about it? Absolutely, you know, in our recent event in the data cloud summit, we had Equifax, so the gentleman there was the VP of data governance and data fabric. So you can start seeing now these roles, you know, created around this problem. And really when you listen to what they're trying to do, they're trying to provide as much value as they can without changing the habits of their users. I think that's what's key here is that the minute you start changing habits, force people into paradigms that maybe, you know, are useful for years of vendor, but not so useful as to the customer, you get into the danger zone. So the idea here is how can you provide a broad enough platform, a platform that is deep enough so the data can be intelligently managed and also distributed and activated at the point of interaction for the end users so they can do their job a lot easier. And that's really what we're about, is how do you make data simpler? How do you make, you know, the process of getting to insight a lot more fluid without changing habits necessarily, both on the IT side and the business side? I want to get to specifics on what Google is doing, but the last coat of Uber trends I want to ask you about, because again, we've known each other for a long time, we've seen this data world grow up, and you're right, 20, 30 years ago, nobody cared about database, well, maybe 30 years ago, but 20 years ago it was a boring market, right now it's like the hottest thing going. But we saw, you know, bromide, like data is the new oil, well we found out what actually data is more valuable than oil, because you can use, you know, data in a lot of different places, you can use oil once. And then the term data is an asset and you said data sharing. And it brings up the notion that, you know, you don't want to share your assets, but you do want to share your data as long as it can be governed. So we're starting to change the language that we use to describe data and our thinking is changing. And so it says to me that the next 10 years aren't going to be like the last 10 years. What are your thoughts on that? I think you're absolutely right. I think if you look at how companies are maturing their use of data, obviously the first bear is how do I as a company make sure that I take advantage of my data as an asset? How do I turn, you know, all this information into a sustainable competitive advantage? You realize stop of mind for organizations. The second piece around it is, how do I create now this innovation flywheel so that I can create value for my customers, my employees and my partners? And then finally, how do I use data as a center of the products that I can then further monetize and create further value into my ecosystem? I think that the piece that's been happening that people have not talked a lot about, I think, with the cloud, what's come is it's given us the opportunity to think about data as an ecosystem. Now you and I are partnering on insights. You and I are creating assets that might be the combination of your data, my data. Maybe it's an intelligent application on top of that data that now has become an intelligent, rich experience, if you will, that we can either both monetize or that we can drive value from. So I think, you know, it's just the scratching the surface on that, but I think that's where the next 10 years to your point are going to be is that the companies that win with data are going to create products, intelligent products out of that data and they're just going to take us to places that we are not even thinking about right now. Yeah, and I think you're right on. That is going to be one of the big differences in the coming years is data as product. And that brings up sort of the line of business, right? I mean, the lines of business heads, historically have been kind of removed from the data group. That's why I was asking you about the organization before, but let's get it to Google. How do you describe Google strategy, its approach and why it's unique? You know, I think one of the reasons, so I just, you know, started about a year ago and one of the reasons for why I found, you know, the Google mission. Interesting is that it's really rooted at who we are and what we do. If you think about it, we make data simple. That's really what we're about. And we live that value. If you go to Google.com today, what's happening, right? As an end user, you don't need any training. You're going to type in whatever it is that you're looking for. And then we're going to return to you highly personalized, highly actionable insights to you as a consumer, as a consumer of insights, if you will. And that thing, that's where the market is going to. Now, you know, making data simple doesn't mean that you have to have simple infrastructure. In fact, you need to be able to handle sophistication at scale. And so simply our differentiation here is how do we go from highly sophisticated world of the internet, disconnected data, changing all the time, vast volume and a lot of different types of data to a simple answer that's actionable to the end user, it's intelligence. And so our differentiations around that, our mission is to make data simple and we use intelligence to take the sophistication and provide to you an answer that's highly actionable, highly relevant, highly personalized for you so you can go on and do your job because ultimately majority of people are not in the data business. And so they need to get the information just like you said as a business user that's relevant, actionable, timely so they can go off and create value for their organization. So I don't think anybody would argue, I mean, Google obviously are data experts, you know, arguably the best in the world. But it's interesting some of the uniqueness here that I'm hearing in your language. You used the word multi-cloud, Amazon doesn't use that term. So that's a differentiation. You sell, but you sell a cloud, right? You sell cloud services but you're talking about multi-cloud. You sell databases but of course you host other databases like Snowflake. So where do you fit in all this? Is your, do you see your role as the head of data analytics as to sort of be the chef that helps combine all these different capabilities or are you sort of trying to help people adopt Google products and services? How should we think about that? Yeah, the best way to think about, you know, I spend 60 to 70% of my time with customers and the best way I could think about our role is to be your innovation partner as an organization. And, you know, you know, whichever is the scenario that you're going to be using, I think you talked about open cloud. I think another uniqueness of Google is that we have a very partner friendly, you know, approach to the business because we realize that when you walk into an enterprise or a digital native and so forth, they already have a lot of assets that they have accumulated over the years and it might be technology assets but also might be knowledge and know-how, right? So we want to be able to be the innovation vendor that enables you to take these assets, put them together and create simplicity towards the data. You know, ultimately, you can have all types of complexity in the backend, but what we can do the best for you is make that really simple, really integrated, really unified. So you as a business user, you don't have to worry about where is my data? Do I need to think about moving data from here to there? Are there things that I can do only if the data is formatted that way and this way? We want to remove all that complexity just like we do it on Google.com. So you can do your job. And so that's our job. That's the reason for why people come to us is because they see that we can be their best innovation partner regardless where the data is and regardless, you know, what part of the stack they're using. Well, I want to take an example because my example, I mean, I don't know Google's portfolio like you do obviously, but one of the things I hear from customers is we're trying to inject as much machine intelligence into our data as possible. We see opportunities to automate. So I look at something like BigQuery, which has a strong affinity and embedded machine learning and machine intelligence. As an example, maybe of that simplification, but maybe you could pick up on that and give us some other concrete examples. Yeah, specifically on products, I mean, there are a lot of products we can talk about and certainly BigQuery has tremendous market momentum. And it's really anchored on this idea that, the idea behind BigQuery is that just add data and we'll do the rest. So that's kind of the idea where you could start small and you can scale at incredible volumes without really having to think about tuning it, about creating indexes and so forth. Also, we think about BigQuery as the place that people start in order to build their ecosystem. That's why we've invested a lot in machine learning, just a few years ago, we introduced this functionality called BigQuery Machine Learning or BQML, if you're familiar with it. And you notice out of the top 100 customers we have, 80% of these customers are using machine learning right out of BigQuery. So now, why is that? Why is it that it's so easy to use machine learning using BigQuery is because it's built in. It was built from the ground up. Instead of thinking about machine learning as an afterthought or maybe something that only did a scientist have access that you're going to license just for narrow scenarios, we think about you have your data in a warehouse that can scale, that is equally awesome at small volume, as very large volume, and we build on top of that. Similarly, we just announced our analytics exchange, which is basically the place where you can now build these data analytics assets that we discussed. So you can now build an ecosystem that creates value for end users. And so BigQuery is really at the center of a lot of that strategy, but it's not unlike any of the other products that we have. We want to make it simple for people to onboard, simple to scale to really accomplish whatever success is ahead of them. Well, I think ecosystems is another one of those big differences in the coming decade because you're able to build ecosystems around data, especially if you can share that data and do so in a governed and secure way. But it leads to my question on industries and I'm wondering if you see any patterns emerging in industries, and each industry seems to have its own unique disruption scenario. Retail obviously has been disrupted with online commerce and health care with, of course, the pandemic. Financial services, you wonder, OK, are traditional banks going to lose control of payment systems? Manufacturing, you see our reliance on China supply chain in, of course, North America. Are you seeing any patterns in industry as it pertains to data? And what can you share with us in terms of insights there? Yeah, we are. And I mean, there's obviously the industries that are very savvy or data hungry. You think about the telecommunication industry. You think about manufacturing. You think about financial services and retailers. I mean, financial services and retailers are particularly interesting because they're kind of both in the retail business and having to deal with this level of complexity of they have physical locations and they also have a relationship with people online. So they really want to be able to bring these two worlds together. I think about the scenarios of Car4, for instance, it's a large retailer in Europe that has been able to not only to onboard on our platform and they're using everything from BigQuery all the way to Looker, but also now create the data assets that enable them to differentiate within their own industry. And so we see a lot of that happening across pretty much all industries. It's difficult to think about an industry that is not really taking a hard look at their data strategy recently, especially over the last two years and really thought about how they're creating innovation. We have actually created what we call design patterns, which are basically blueprints for organization to take on. It's free. It's free guidance. It's free data sets and code that can accelerate their building of these innovative solutions. So think about the ability to determine propensity to purchase or build a big trend is recommendation systems. Another one is anomaly detection. And this was great because anomaly detection is a scenario that works in telco, but also in financial services. So we certainly are seeing now companies moving up in their level of maturity because we're making it easier and simpler for them to assemble these technologies and create what we call data rich experiences. I want to ask you the last question. How you see the emerging edge, IoT, analytics in that space, a lot of the machine learning or AI today is modeling in the cloud, as you well know. But when you think about a lot of the consumer applications, whether it's voice recognition or fingerprinting, et cetera, you're seeing some really interesting use cases that could bleed into the enterprise. And we think about AI inferencing at the edge as really driving a lot of value. How do you see that playing out and what's Google's role there? So there's a lot going on in that space. I'll give you just a simple example. Maybe something that's easy for the community to understand is there's still ways that we define certain metrics that are not taking into account what actually is happening in reality. I was just talking to a company whose job is to deliver meals to people. And what they have realized is that in order for them to predict exactly the time it's going to take them from the kitchen to your desk, they have to take into account the fact that distance sometimes it's not just horizontal, it's also vertical. So if you're distributing and you're delivering meals in Singapore, for instance, at high density, you have to understand maybe the data coming from the elevators so you can determine, oh, if you're on the 20th floor, now my distance to you and my ability to forecast exactly when you're going to get that meal is going to be different than if you are on the fifth floor and particularly if you're ordering at 11.32 versus if you're ordering at 11.58. And so what's happening here is that as people are developing these intelligent systems, they're now starting to input a lot of information that historically we might not have thought about, but that actually is very relevant to the end user. And so how do you do that? Again, you have to have a platform that enables you to have a large diversity of use cases and that thinks ahead, if you will, of the problems you might run into. Lots and lots of innovation in this space. I mean, we work with companies like Ford to reinvent the connected cars. We work with companies like Vodafone 700 use cases to think about how they're going to deal with what they call their data ocean. I thought you would like this term because we've gone from data lakes to data oceans. And so there is certainly a ton of innovation and certainly the chief data officers that I have the opportunity to work with are really not short of ideas. I think what's been happening up until now, they haven't had this kind of single, unified, simple experience that they can use in order to onboard quickly and then enable their people to build great rich data applications. You know, we certainly had fun with that over the years, data lake or data ocean. Thank you for remembering that Bruno. Always a pleasure seeing you. Thanks so much for your time and sharing your perspectives and informing us about what Google's up to. Can't wait to have you back. Thanks for having me, Dave. All right, and thank you for watching everybody. This is Dave Vellante. Appreciate you watching this CUBE Conversation. We'll see you next time.