Google Cloud Next '19: Day 2 Product Innovation Keynote

>> Welcome back to an amazing day two of Google Cloud Next. Wasn't yesterday awesome? [Applause] Today we have a really packed agenda, and we're going to try and top yesterday with great, great new announcements in lots of product areas. Yesterday we talked about our vision as Google for cloud computing, to give you global scale distributed infrastructure, a digital transformation platform, and industry specific solutions that facilitate digital transformation. Today we have a packed show. Going to start off talking about some new things we're doing in infrastructure and security, talk about data management, smart analytics, advances in AI and Machine Learning, what we're doing in collaboration, and industry specific solutions. To kick off today, let me welcome Urs Hölze, senior Vice President technical infrastructure. Urs? [Applause] >> Good morning! Every day, Google Cloud connects to over a billion IP addresses all over the world, not for Google, but for GCP customers. And those bits travel on our network, and 99% of the time, they get handed off directly to the last mile ISP, so we can control security and performance all the way. Now, our scale has taught us that we can always do better to do more to help our customers succeed. And that thought guides us at Google when we think about capital investments, when we think about product priorities. As Thomas and Sundar underlined yesterday, we're spending an enormous amount of capital and human resources to build out that vision: $13 billion just this year, just on data centers, just in the U.S. And of course, there's more. So there might be better forms of hardware for Machine Learning. There might be advanced, efficient data centers, better data storage, smarter analytics. So we're committed to bring a lot more cloud to you, not just in our data centers, but also in your data centers. We believe in a cloud with a lot less hassle. The cloud is too hard today, and it's our job to make it simpler, through easier policy management, through consistent APIs, through open source software that works where you need it. And that's the guiding principle between the cloud services platform, which we announced last July, to manage containers and services both on premise and on GCP. And just eight months later, yesterday, we announced a final product, Anthos, and simply put, I believe that Anthos is the future of Cloud. That's a big statement. Why? Because Anthos is a single platform that lets you develop secure, release, monitor, operate, debug your services. It's mature. It's robust. It has advanced functionality, even today. So, the stack, the Anthos stack, includes container orchestration, service management, config and policy management, dev ops tooling, and a robust marketplace, as you saw yesterday, so that customers can have a unified experience when they consume the products and services of independent software vendors. And it works equally well for your existing as well as new software so you don't have to rewrite to get the benefits. And of course, it works wherever your teams are working. And last but not least, since it's fully managed, even on premise, you don't have to worry about operating and updating the platform, so you get a Cloud experience in your data center as well. If you want to know more, I invite you to join the ask me anything session tomorrow morning at 9:00. There were also lots of sessions yesterday that you can recap on video. Now, customers started using the CSB beta several months ago and now are starting to use Anthos in production, and the feedback has really been excellent. They're working faster with fewer impediments and more freedom. And because it's based on open source, it's a safe choice for years to come. It's not a single network choice. So that's why I believe that Anthos really is a great choice for the cloud strategy of any enterprise. And with Anthos, we're making hybrid and multi cloud the new normal. You see that in our new traffic director, which seamlessly balances your traffic across both on premise and GCP instances and soon, across instances in other clouds as well. It works for both VMs and containers. It routes traffic over our backbone so your application gets unparalleled resiliency of that network. Or take our approach to serverless computing so functions, you know, the ability to run code without managing VMs or containers. Cloud Run, a new product that you will hear a lot more in this afternoon's developer keynote, combines the speed and ease of use of serverless computing of functions with the flexibility and portability of containers, so you can use any language and any runtime in Cloud Functions, not just what your vendor happens to provide. And of course, it's based on open source, Knative, and it will be part of Anthos, so it will work everywhere you want it to work. Now, the cloud isn't just about deploying and managing applications. It's also about data. And one of GCP's biggest strengths is to help you do more with your data. You can ingest, integrate, analyze, predict both with traditional analytics and with Machine Learning. And of course, we offer a cloud platform with natural language understanding that's built in both into your apps and into our apps. So that's a purposeful AI. Like an assistant for your call centers or for your enterprise for a better customer or employee experience. So we make AI accessible not just as technology and tools but also as solutions. And that's also true for our collaborative applications. You will hear more later but using Gsuite, teams communicate and deliver faster, so you get better workforce engagement and better productivity/ There's always up to date documents, for example, with live commenting. There's video conferencing that just works. And there's AI to boost your productivity in spreadsheets, documents, and e mail. But now let me turn to security because we know that no company can serve its customers without the greatest trust in the reliability and security of the platform that it's running on. And at Google, security is really something that we have built in from the start comprehensively. It's not something that's bolted on later. And as Thomas mentioned yesterday a little bit, we start with the principle that your data is your data and only your data. It's the lifeblood of your company, and no one should access it without your knowledge and consent not Google, not vendors that you engage, not employees that you don't want to have access. And we're the only cloud that offers access transparency logs products access transparency logs for virtually all of our products that document such administrative access in near realtime. We also provide powerful data security tools like the data loss prevention to discover and classify sensitive data wherever it is. And we have VPC service controls, it's the new feature that allows you to segment multi tenant resources of cloud services to mitigate data exfiltration risks. So it allows you to combine the benefits of and strengths of such managed storage without exfiltration risks, and so it combines those services with the security of a private, truly private cloud. And you can only get them on Google Cloud. In fact, you'll see 30 more announcements at just Next that are related to security. So I can't mention them all, but I will call out just a few more. Cloud Security Command Center provides a single view of all of your assets, buckets, containers, and their risk profile. And Event Threat Detection let's you is a new service we're introducing that lets you spot signs of compromise or malicious activity in your logs. And last but not least, policy intelligence is something I'm very excited about. It's a Machine Learning based service that gives you the confidence that you actually specified the right set of access controls and that you're not accidentally giving overly broad access to people. Now, while it's critical to secure servers and services, we also make it easy to secure your users and end devices as part of the end to end strategy. First, Chromebooks are the most secure and easy to manage enterprise laptops on the market, period. Because you have no OS patching and no back ups to worry about. Like we do all of that. And with hardware verified goods in every single laptop, they're virtually immune to virus attacks. We also developed security keys, now an industry stand, to help protect accounts from phishing and to make strong authentication as easy as touching a security key. Because e want as many users as possible to take advantage of that. Today we're announcing a software security key that will work in most Android phones. So, soon, a billion users will be able to take advantage of it and secure their accounts like never before. So let me now briefly show you how all of these security features work together for end to end protection. So let's assume you're taking an existing application and you're moving it into Anthos, on premise or in the cloud, doesn't matter, it's the same. So, you immediately pick up all of the built in security features of Anthos like strong service identities and encrypted communication between services, centralized policy and config management. You get that out of the box. And then by putting our identity or our proxy in front of it, you now get single sign on with security key second factor for this application. And you can configure further context of where access policies for that application. And if you have business customers, you can use Apigee API management to expose services to other businesses in a B2B setup. All of that without touching your application, your existing application. And since Anthos has centralized audit logging, all of the logs can be fed into BigQuery or event threat detection to give you critical insights about this set up. So it's a simple, consistent way to modernize not just the application but modernize your security posture and move to a zero trust model. Now all of this is built in close coordination with our customers and partners, as you have seen a growing list of some of the world's largest Enterprises. And we're really honored to have their trust. Now, while there are many new features that we're announcing today, they're all part of a larger, single consistent vision: A cloud platform that's uniquely built around open standards with production grade code. So you pick your environment, on prem or in the cloud, and you don't have to change a thing. And you see that commitment to up time and to reliability in the Google Cloud as well. In fact, recently, two leading third party research firms published their findings about 2018 cloud reliability. And I want to say this research was neither sponsored, nor influenced by Google. But GCP demonstrated. [Applause] Thank you. GCP demonstrated by a substantial margin the highest reliability of the three major clouds, right? It's not even close. So, to wrap up, at Google Cloud, we're investing a lot in security and in your future. You can see that really throughout all of today's announcements. And of course you get great reliability and great security on GCP, and you reduce the complexity of your IT so that you can move faster. So, thank you for choosing us as a partner for your journey to a simpler, more secure, naturally hybrid cloud. Thank you, and now back to Thomas. [Applause] >> Thank you, Urs. In addition to security, we at Google take you as customers and your trust in Google extremely seriously. One of the issues we want to address at the conference is the issue of privacy and our stance on it. It's very simply put: Your data is your data and no one else's. Your AI models are your AI models and no one else's. No one at Google will access your data without your permission. We do not have a back door to allow any agency to access your data without your permission. You can encrypt the data at rest or in transit into our cloud. You can use your own encryption keys to encrypt that data. If you want Google to help you access that data for a support purpose, we will stream, in near realtime, the logs of every operation our people are doing to help you with the support request. No other cloud provider provides you these facilities. And for us, that's a measure of providing you the comfort and trust that Google will never, ever use your data for any other purpose outside of delivering you the service in GCP that you're using. [Applause] Now, one of the key areas that people wanted us to provide great technology for is the area of data management. We offer a number of solutions for data management: To rehost enterprise databases in our cloud, to help you migrate databases from on premise data centers to our cloud, and great new solutions to build new scalable applications using the broad portfolio of assets that we as Google provide for data management. Today we're going to show you how to move Microsoft SQL server, Windows, and active directory to our cloud. There are many other capabilities we're introducing in data management. To show you that with a demonstration, please welcome Deepti Srivastava from product management. Deepti? >> Thank you, Thomas. I'm excited to be here today to talk to you about databases, which we know are a critical part of your business. Google Cloud offers a wide portfolio of database services for transactional, operational, and analytical workloads, all protected from threats by advanced security. We know it can be challenging to build new applications and to bring existing workloads to the cloud. That's why we are committed to providing a first class experience to you as you move to Google Cloud, including for your Windows workloads. So let's take a look. Here we have a typical Windows app, Adventureworks. Just like many Windows applications, this one runs on a Virtual Machine and uses SQL server as its database back end. You might have thousands of Windows applications, from small simple ones like this. To large, critical applications running in your data centers. To help lift and shift windows workloads to the cloud, Google Cloud provides tuned Windows and SQL server virtual machine images. You can bring your existing licenses and run them on sole tenant to stay in compliance or buy licenses through us. Now let's turn our attention to SQL server. Our cloud SQL managed data service offers support for MySQL and Postgres. Today I'm excited to announce that Cloud SQL will offer fully managed Microsoft SQL server later this year. [Applause] Thank you. It works with all of your familiar tools but better because it's fully managed. This gives you and your teams the opportunity to focus on building what's next. Now, let's modernize this Adventureworks app, starting with the database. Using SQL server management studio, we'll take a full backup of the SQL 2008R2 database used by this app, we'll simply run the command in SMMS to generate the file. Once completed, we can upload this to our Google Cloud storage bucket. We've already uploaded the generated file to the appropriate bucket, as you can see in the console here. Next, we'll restore the back up to Cloud SQL for SQL server. We'll select the managed instance we have created, select import, and select the doc BAK type. From here, we will then find and select the GSC bucket where we have uploaded the file, type in the database name, and select import to initiate the restore. Once the restore is complete, all we have to do is copy the IP address and just change the connection string in the configuration file for the application. That's it! You can refresh, and now our Adventureworks app is up and running on managed SQL server for the latest version of SQL server. [Applause] Thanks. So, to recap, Cloud SQL for SQL server is a fully managed database service that handles the mundane tasks of operating a database for you so you can spend more time innovating. But we know that SQL server isn't the only critical part of the Windows ecosystem. Active directory is the standard for authentication for authorization in Windows environments. Today you can run active directory yourself on GCP and federate AD users to GCP with cloud identity, our identity management service. But we wanted to enhance this experience for you. So, I'm happy to announce that managed service for Microsoft active directory will be coming later this year to GCP. [Applause] This new managed service runs actual AD domain controllers and makes it easier to migrate your directory dependent applications and servers to the cloud. Just like with our managed SQL server service, your IT and security teams can focus on higher value tasks by spending less time on server maintenance and security configurations. You can keep your user and password data on premises using methods like one way force level trust, while allowing users to authenticate against the managed service in the cloud for lower latency and geo availability. Here you see us logged into SharePoint running in a GCEVM that is domain joined with managed Microsoft AD. As you can see, the user identity still originates from on prem active directory. So we now have SQL server and active directory running as managed services on GCP. These services work with your familiar tools. This will help you focus more on delivering higher value experiences for your customers and employees instead of managing and securing infrastructure. Managed SQL server and managed active directory are coming to GCP later this year. Now, let's take another look at what's always on everyone's mind: Security. To help secure your Windows workloads, GCP provides defense in depth. We offer shielded VMs for Windows analytics which are hardened by a set of security controls that protect Enterprise workloads from treats like remote attacks, privilege escalation and malicious insiders. This is a simple checkbox item when creating your VMs. This is just one example of the tools you have at your disposal to help secure your workloads. We covered a lot of topics this morning, so let's recap. You want to spend more time delivering innovative products to your customers and less time managing infrastructure and security. You can do this by choosing managed services and advanced security for your Windows workload on GCP. Thank you, and back to you, Thomas. [Applause] >> Thank you, Deepti. To talk about their experience using Google Cloud, please welcome Justin Arbuckle from Scotiabank. Justin? >> Thank you. >> Tell us a little about Scotiabank. You're ten times older than Google, a 200 year old bank. Tell us a bit about your business and how you have been a market leader for so long. >> Well, we're Canada's most international bank. We're in over 50 countries, and we have a strong retail presence in both Canada and the Latin American markets, which allows us to balance the stability of the financial services environment in Canada as well as this incredible innovation that we have in the Latin American markets. You know, over the last few years, we have opened a number of innovation centers, our digital factories. And I'm pleased to say that at the moment, we have over 25% of all of our global retail transactions running off of a digital platform that runs in the cloud. >> Thank you. How are you approaching your transformation at Scotiabank? >> Carefully. You know, the thing we have to realize about transformation is that it isn't bimodal. You can't transform half of your organization and not the other. And so we don't have a bimodal transformation strategy. We want to have a unified digital transformation for our entire organization. And so for this we have to re imagine how we build applications for the cloud, to be a cloud first bank. And to do this, we've built a platform called PLATO. And PLATO, of course, runs on Google. And what PLATO gives us is the ability to be far more responsive, to be high velocity. And I'm, again, happy to say, that In the next few years, we will be able to be migrating 40% of our applications globally to the cloud on PLATO. >> You know, data plays a key role in banking. You're doing some amazing stuff. Can you tell the audience a little bit more about what you're doing? >> Yes. I mean, you spoke about privacy before. Customers give us their data and they trust us with it. And trust is a critical part. It's the core of our business. And so looking after that data is important, and we take that extremely seriously. But also, we need to be able to use that data, the PII data and transactional data, etc., to be able to create complex queries that allow us to give customers the kind of service and experience that they need and want, as well as manage bank risk. And so we are moving our data to a global data platform that runs on PLATO. And what this allows us to do is be far more responsive again to our customer and to manage the quality of the data that we actually are using to make these important decisions. But you know, also, something that we have to remember is that that data, fundamentally, is the most important asset that we have. And so the same platform PLATO is also providing real time data stream of all of our transactions globally, and what that allows us to do is do some really clever things with anti money laundering, fraud, etc. Just as a final couple of points, I would just like to say, if you don't mind, Scotiabankers around the world are wearing pink to celebrate the International Day of Pink which is to celebrate the fight against homophobia, transphobia, and all other forms of bullying. I hope you join me. [APPLAUSE] And finally, I would just like to say, I'm sure the speakers would agree, we're up here, we get to be up here because of the incredible teams that are behind us. Some of mine are in the audience and some of them are back home. I would just like to say thank you. >> Thank you very much. [Applause] Thank you so much. Another part of Google's platform that many, many customers around the world use is our analytics foundation. People historically have used us for three important reasons. First, they want to do realtime streaming analytics, looking at realtime coming in realtime streams to the cloud. Second, they use us to augment and enhance their Enterprise data warehouse using BigQuery. And third, they use us to run Hadoop and Spark clusters without having to run and manage them themselves. Today at Google Cloud Next, we have some amazing new capabilities in our analytics foundation. It's predicated on three important things. First, to allow you to use AI models and AutoML to automatically categorize data and do predictions so that a business analyst can use AI as part of their work. Second, we have introduced a number of new capabilities to make it much easier to move data into our cloud. Third, we've enhanced BigQuery with even more features and a new capability called BI Engine to retain its performance, scale, and computational advantages. We're also grateful for a number of partners who have worked with us to integrate their tools and technologies with our analytics foundation so you can use your favorite ETL tool, your favorite visualization tool, your favorite metadata management layer. And to show you some of these advances, please give a warm welcome to Julie Price, a big data specialist in Google Cloud. Julie? >> Thank you, Thomas. Hello, everybody! Now, I would like for you to think for a second about the last time you sat at home eagerly awaiting the delivery of a package. Now I want you to think about the incredible amount of data that goes into ensuring that millions of packages are delivered as expected across the globe each and every day. Data needs to be pulled together from so many different places, often in siloed systems, both internal and external. That's what Cloud Data Fusion is all about. It's a fully managed and cloud native integration service that's being announced today. It's got a broad library of open source connectors sorry open source transformations and over 100 out of the box connectors for all sorts of systems and data formats. So let's take a look at how data fusion shifts an organization's focus away from code and integrations to insights and action. So all of our shipping data is in an on prem system, and we want to move it into BigQuery so we can further analyze it and combine it with other date. This is incredibly easy to do with data fusion. Here we have defined our on premises connection, and we can see all of the tables that are available. We can simply type in to select shipment details as that's the table that we want to see. And so if we click on it, we can see the data in the table and decide what type of transformations we want to do. So let's go ahead and do a simple transformation and delete one column. We could do far more complex transformations all through this UI, but we're going to keep it very simple for this demo. Now we operationalize the pipeline by clicking on create and we're going to set it up as a batch pipeline. Now that we've created the pipeline, we need to define where we want to land the data. So if we want to load the data into BigQuery, we just drop the BigQuery connector as the endpoint and connect it to our transformation step. That's it. That's all we have to do. And ow we can deploy and run this pipeline. So we've manually run this pipeline, but of course you can schedule your pipelines. You can also have them triggered. And now if we head over to BigQuery, we can see that we've got all of our data loaded and ready to be analyzed as we have deployed and run this pipeline in advance. Now, you've just seen how with a few clicks, we can create a pipeline to move data from an on premises system into the cloud without writing a single line of code. And as a quick reminder, organizations across the globe love using BigQuery to perform analytics at scale. Because it's completely serverless, folks can focus on insights rather than infrastructure and operation. BigQuery also allows enterprises to continue to use their existing BI and dashboarding tools like Tableau, Looker, Click, and Google Data Studio. And with our shipping data in BigQuery and ready to go, we'd like to be able to monitor our global operations and make very quick decisions. That's why today, we're announcing the BigQuery BI Engine. BI Engine is the native BigQuery accelerator that enables you to visually analyze and interact with your data at the speed of thought, all with incredible scalability and security. It's now available through Data Studio, and we've got other BI partners coming soon. So let's have a look at our shipping data in data studio dashboard that's powered by BI Engine. Now, right away we can see that there's something going on with Brazil. It looks like Brazil has significant delivery delays, but packages that are shipped from Brazil are not necessarily responsible for as much revenue. Now, with each and every click on this dashboard, we are visualizing and processing millions of rows of data with subsecond response time, all thanks to BI Engine. Now, if you're in charge of logistics at this company, you've got to figure out what's going on in Brazil. So how would you do that? Many folks would use spreadsheets to perform this type of analysis, but we're talking about millions of rows of data so that would be impossible in any spreadsheet environment. But not in connected sheets. Connected Sheets is a new feature of G Suite that is being announced today. It brings together the power [Applause] Thank you. Connected sheets brings together the power of BigQuery with the familiarity of the existing Sheets products. So let's head over to Sheets so we can take a look. Here we've got Sheets connected to our shipment dataset in BigQuery, and with the BigQuery data in Connected Sheets, we can take advantage of traditional functionality, like charts, pivot tables and functions but across the entire BigQuery dataset. So we've got over 100 million records in our data set but connected Sheets can process billions of records. Let that sink in for a second. Now, what's really awesome as well is that you will be able to accelerate connected Sheets with the BI Engine that we just talked about. So let's go ahead and look at our summary dashboard tab. Now, what you can see here is a number of standard functions that are referencing a combination of BigQuery data and data that we have entered into the sheets. So you can see we're using input of min and max distance on the left that we've entered into sheets to calculate averages across data that actually exist in our BigQuery dataset. But we really wanted to look into Brazil's shipping delays, so let's move over to the country breakout tab. Now, here we're using standard pivot functionality but, again, across our entire BigQuery dataset. We're slicing and dicing the on time delivery percent by country and the number of zones that each package crossed. And it looks like packages are getting delayed even when they don't leave Brazil, so there might be something going on in the country that we need to look into. Now we can share this analysis with folks in Brazil using standard collaboration functionality common in all G Suite products. Our colleagues in Brazil determined that the delays were due to highway construction, and a simple update to the routing software will take care of those issues So now let's take this one step further. Since we have the data in BigQuery, how about we use this data to predict which routes might be problematic in the future so we can make adjustments and ensure that the packages get delivered on time? Now, as you heard earlier, thousands of enterprises are already using BigQuery to store and analyze up to petabytes of data. Every company wants to understand and use machine learning but not every company has access to ML experts. And that's why today we're also announcing AutoML tables [Applause] AutoML tables will enable anybody that knows data in your organization to automatically build and deploy state of the art ML models at massively increased speed and scale. So let's jump over to AutoML tables and take a look. Here you can see we've connected to our BigQuery dataset. And on the right hand side, you can see the full table schema that AutoML has pulled in. And on the left, you simply decide which column contains the field on which you want to train your model. And this becomes what gets predicted when you feed future data in. That's all you have to do. You don't have to be a ML expert. We didn't write a single line of SQL or TensorFlow code. AutoML tables will find the best model for you. So let's move over to the predict tab. Now, to save time for the demo, we've already trained the model. And once the model is trained, then you can easily run it against new data by configuring the input data fields. So we're pooling the data from BigQuery, the new data we want to feed in, and then you can feed the resulting predictions back into BigQuery for further analysis and visualization. So let's have a look at that in Data Studio. Now we can augment our BI with AI. We can begin to enhance our historical dashboards with predictive insights. So this dashboard might look pretty familiar, but our map is now showing predicted delivery delays instead of historical. And since we solved some of the issues in Brazil with rerouting around the construction, now we can focus on where we expect the next delays to come in which, as you can see, are predicted to be in India. Now all of this was done with just a few clicks and no code, allowing you to move from raw data to predictions in no time. So you've just seen some of the amazing new additions to Google Cloud's fully integrated and serverless platform. Cloud Data Fusion, BigQuery BI Engine, Connected Sheets, and AutoML tables will make it easier for folks across your organization to quickly and seamlessly turn data into mission critical insights. With that, thank you, and I would like to welcome Thomas back to the stage. [Applause] >> Julie showed you a logistics example using our analytics foundation. A real customer in this business that all of you know and love is United Parcel Service. To talk about how they're using Google Cloud to transform their business, please give a warm welcome to Juan Perez, chief information officer of UPS. [Applause] >> Thank you, Thomas. Thank you. All right. Good morning, everyone. At UPS, every day we solve a challenge of epic proportions. How do we efficiently deliver 21 million packages every single day? And by the way, this challenge grows even bigger during our peak time around the holidays when we deliver more than 32 million packages a day. The question is simple: What is the best, most efficient, and most profitable way to do it? Well, we have just a few ways to deliver just one route. By the way, if you're really curious about it, the number of possible solutions to deliver one route at UPS is 199 digits long. But in every case, at the end, only one route is the best route. We address this challenge head on, taking our data analytics practice to the next level. We designed routing software that tells the delivery driver exactly where to go every step of the way during 120 pick up and delivery stops in any given day. And by the way, it even accounts for the most important thing of all, and that is taking that very important lunch break. This software saves UPS up to $400 million a year and it reduces our fuel consumption by 10 million gallons a year. That, quite frankly, is the power of analytics at UPS. And using this as a foundation, we determined that the power of our data opened significant opportunities to continue to improve our operations, and most importantly, our customer service. Today, with support from partners like Google Cloud, we're creating the UPS Smart Logistics network, a combination of experience, expertise and technology, powered by realtime data that transforms our physical assets into the world's most integrated network. This is not just simply when drivers are delivering packages on the road. It's the loading docks. It's where we sort packages at UPS. Everything behind the scenes that gets that package delivered, that one package to your doorstep on time every day. Data plays a key role in the Smart Logistics Network at UPS. Forecasting our package volume levels is critical to running an efficient network. Let me tell you how we leverage BigQuery for the most precise volume forecasting in UPS's history. We gather and analyze more than a billion data points every day, including package weight, shape, the size of our packages, as well as the location, the facility capacity we have across the network, and customer data. Google technology allows us to perform analytics on this massive dataset while providing the platform and the capability to run Machine Learning, different models across all this data. The outcome for us is really critical. The outcome of the forecast tells us exactly how to load our delivery vehicles, how to maximize the use of our assets, and how to ramp up our network around the holidays. Most important of all, it helps us make more targeted operational adjustments, minimizing the forecast uncertainty that comes with the unforgiving demands of e commerce. The more realtime data we can get about the state of a package, the better visibility we can get into potential challenges, and ultimately, the better we can serve you, our customers. It truly is a brand new day for a smarter, more agile, and ultimately, a more forward looking UPS. Our future follows a path paved by prescriptive analytics. That's where we're going. Which allows us to achieve true supply chain visibility and efficiency in moving goods between 220 countries and territories faster than ever imagined. We are grateful for the opportunity to collaborate with great partners like Google in a way that lets us use our joint expertise to bolster visibility across supply chains around the world. Thank you again for the opportunity to be here. [Applause] >> Google has invested many years of work in AI and Machine Learning. We have focused on building an AI platform that provides a very deep understanding of a number of fundamental kinds of data: Voice, language, video, images, text, translation. On top of this platform, We have built a number of solutions to make it easy for our customers and analysts around the world to build models, one example being AutoML, where the software itself helps you build a model. And finally, to speed how companies and organizations can use AI and ML to transform a business, we are delivering a number of business solutions that can be very quickly adopted and can change how your organization uses AI and ML to run its business. To explore a number of new advances that we're announcing today, please welcome Rajen Sheth from our product management organization for AI and ML. Rajen? >> Thank you, Thomas. So we believe that AI will transform every business and every organization over the course of the next few years. And our goal is to empower all of you to be able to transform along with us. And we're seeing it happen over and over again with big businesses and small businesses alike. This is a testament to the power of AI but also reflects our commitment to deliver that power to everyone in your organization, whether they're builders or business users or decision makers. So let's start with the builders. Builders are everyone from machine learning experts all the way through to developers who are just getting acquainted with AI for the first time. And if you're a builder, you're on the forefront of AI. You're facing the most complex challenges and solving problems in a brand new way. And we have spent years working to understand your needs, and we have used those insights to build the most sophisticated platform for AI development anywhere. So today I'm delighted to announce our Google Cloud AI platform. The AI platform is what happens when every step of the AI development process meets the best of Google. Everything from robust end to end Machine Learning pipelines from beta all the way through to production, to things like samples and tools for everyone, hosted Jupyter notebooks, and a hosted environment to make deployment and management of machine learning easy. And importantly, we believe that empowerment means giving you the most flexible hybrid cloud so you can deploy machine learning in our cloud, on premise, on other clouds, and at the edge. And this is made possible by what we call Kuplo, an open source framework for running AI anywhere that Kubernetes runs and can be managed by Anthos. We have also added the AI Hub which is a one stop shop for all of your AI resources. It allows you to use prebuilt AI resources from a wide range of sources and also share your own solutions within your organization. In addition to that, we're committed to making it available for more and more developers to be able to access AI, and as Thomas mentioned, AutoML is really doing this. AutoML makes it easy for you to able to create build Machine Learning models on your particular dataset. Now, when we introduced AutoML last year, we made it easy to do things like image classification, natural language processing, and translation, and make it accessible to everyone, regardless of technical expertise. AutoML is one of the highest quality and simplest solutions anywhere for building custom Machine Learning models and training them on your data. And this year we're delighted to introduce two new members of the AutoML family: AutoML video and AutoML tables, which extend the power of AutoML to video content and structured tabular data. So now to demonstrate what builders are capable of, we've invited one of our biggest and most ambitious cloud AI customers to join us, so please welcome up Binu Mathew, head of digital products for Baker Hughes GE. >> Hello. I'm Binu Mathew, and I'm the head of digital products for Baker Hughes, a GE company. I run a team of developers and data scientists that work to create advanced analytic products to solve complex industrial problems at a massive scale for customers, predominantly in the oil and gas industry. We work with Google Cloud for this very reason. Our AI based solutions developed on the Google Cloud give us the scale and flexibility to build our analytics models. And of course, on the deployment side, it gives our customers the choice to deploy on the Google Cloud platform. So combined with the other Google technology we use, such as cube flow pipelines for managing end to end ML work flows, Kubernetes and TensorFlow for deep learning capabilities, we view the Google relationship as very strategic in the oil and gas industry. So what are we trying to solve? In the oil and gas industry, one of the most persistent problems we face is something called non productive time. And essentially this means the machine is shut down. And that sounds simple but it's not often for the reasons you would think. There's a system of machines connected to more machines and those systems are connected to more systems, and the problem is usually an interaction between these thousands of machines. And non productive time can mean millions of dollars lost, if not more. And the way that that has been addressed in the past is not scalable. You've often had an engineer modeling data in a spreadsheet or an operator relying on past experience instead of realtime data. And so Baker Hughes's cloud based AI capabilities can now analyze hundreds of terabytes of data and at a level of entire systems and continuously learn unsupervised from normal behavior. That significantly increases the accuracy, scale, and predictive power of our models and gives us a much better chance that system issues that result in non productive time can be found in advance and circumvented entirely. Now that also takes significant computation, which Google Cloud easily accommodates by fluidly scaling across GPUs, and GPUs are critical to AI. And for our customers, this kind of rapidly scaling AI means fewer machine problems, safer operations, less down time, and money saved. Thank you. [Applause] >> Thank you, Binu. It's amazing to see what a difference AI can make when builders think big. But hat if you're a decisionmaker and what you care about is how this can make a difference for your key business goals? You can benefit from AI, too. Which is why we're offering a growing range of AI solutions to help solve some of your most pressing needs. Business decision makers can benefit from AI with a variety of different types of solutions, and I want to talk to you about three that we're introducing today. So first we're launching what we call document understanding AI which helps organizations automatically extract insights from documents but also make your business processes easier. This could be from everything like invoice processing to identifying content within contracts, so many, many, many other use cases. And so for instance, Commerce Bank is working with our launch partner, Iron Mountain, to explore ways that this can help simplify GDPR compliance for HR. We're also looking at challenges faced by specific industries, and we're starting with retail. And for retail customers, today we're launching what we call recommendations AI. And it draws upon Google's deep experience in personalization to deliver product recommendations at scale. And we have seen some pretty amazing results so far, with revenue on recommended items increasing by as much as 50% for some of our first customers. And finally, I want to talk to you about contact center AI which makes it possible to scale phone support without compromising the customer experience. Its virtual agents can automatically help callers with the most common tasks, but in addition, its agent assist tool can automatically feed information, relevant information and work flows, to human agents for those more complex calls. And it's based on dialog flow which makes it easy to build intelligent chatbots that can answer question and perform tasks through natural language. Now, one thing that's important is we want to make sure this integrates into what you already have. And so what we've done is we've integrated this with the top telephony providers and contact center providers market, so this should just insert directly into what you have today without rip and replace. So to talk a little bit about what we have, I'm excited to welcome up Sarah Patterson, Senior Vice President of product, marketing and strategy from Salesforce, and Karen Van Kirk, VP of viewer experience at Hulu. [Applause] Thank you, Sarah and thank you, Karen, for coming to Google Next. I want to talk a little bit about what you're doing. And so, Sarah, can we start with you, and can you tell us a little bit about Salesforce Service Cloud? >> Absolutely. Thanks, Rajin. So I've been at Salesforce for a little over ten years, and its developers and IT leaders, like everybody who is here today, and everyone who is watching online, you have really been our lifeblood. You've actually made Service Cloud the market leader because you have taken our platform and you've used it to help your companies differentiate their offerings with service. Service is becoming your brands. And now our customers are asking us to make it easier to deliver great service using intelligence, which brings us to the partnership between Salesforce and Google Cloud. >> Absolutely. And we're really excited about the work that we're doing together and the fact that today we're announcing that partnership with Salesforce Service Cloud to build the contact center of the future. And so can you tell us a little bit about how you're using Google Cloud AI technologies? >> Google Cloud allows us to accomplish two powerful goals. First, chatbots that we all know and love. They provide a really great way to help companies to engage with customers and scale your support without scaling your costs. And we call our AI powered chatbots Einstein bots. We wanted to give them a fun little name. Service cloud users can now bring Google's dialogue flow into Einstein bots to augment or natural language processing capabilities and add in Google's extensive multi lingual support. We're also excited to announce that we're integrating Google's contact center AI with Service cloud. Phone, believe it or not, is still one of the primary channels for support. And by combining contact center AI with our ecosystem for telephony partners, Service Cloud users can now elevate their phone support all through the interface that they're already familiar with. >> That sounds great. And so, Karen, can you give us more about how this is making a difference at Hulu? >> Sure. Hulu is one of the largest streaming services with over 25 million subscribers in the U.S., and we're growing really quickly. That growth is great, of course, but it brings customer support challenges as well. A big part of our solution is our partnership with Salesforce and 59 with technologies like chatbots, but even the best bot is not a replacement for humans, so that's why we offer phone support as well. And we really believe that AI is the next logical step to bring it all together. And that's where Google's contact center AI comes in. It improves the customer experience because each call will start with a conversation with a virtual agent rather than fumbling with a phone tree. And it will make our human agents more productive because knowledge articles and workflows will surface realtime based on what is going on during the call. And finally, we'll have better data quality and analytics. We can monitor language and sentiment realtime to flag incidents. And then we can also dive in to better understand what is driving customer satisfaction. >> These are amazing capabilities. And so can you give us an example of how this helps the customer experience in ways that you haven't been able to do before? >> Yeah. Customers will often mention a show they're interested in watching, and that can be a great signal for the agent to know how to work with them; for example, suggesting a change in their subscription plan to better meet their needs. So let's say a customer is thinking about adding live TV to their account and they happen to mention an interest in basketball. Agent assist can automatically recognize the topic and surface information to the agent to help them have a better conversation with that customer, perhaps a reminder that the NBA playoffs are about to start. From there, agent assist will surface personalized recommendations and the workflows needed to make the change to the plan. >> That's incredible. It seems like a new level of customer engagement. >> Yes. With Service Cloud and Contact Center AI, we'll be able to improve that customer experience and drive efficiency at the same time. >> And we're here to empower everyone here today to build the next generation of service experiences. >> Great. Thank you both for joining us. We really appreciate it. [Applause] >> Thank you. >> So you know, for a long time, building AI products has seemed like bringing things from the future which often seem like science fiction and bringing it to the present day. But these stories really show you how AI is making a difference for customers today. The future is now. And so let's go make it happen together. Thank you. [Applause] >> Many companies are using AI today to transform their businesses. When you go back home, think about how AI and our technology can help you change your business. 15 years ago, 15 years and nine days to be precise Google launched a product called Gmail. It was launched with a very simple premise: Every human should be able to do mail from wherever they are. All they'd need is a browser. Today, that product is used by over 1.5 billion people every day. We believe, however, we're poised for the next wave of change in collaboration. Our vision is very simple: Enable collaboration for everyone in the world, not just white color workers but the 2.2 billion people who are called front line workers: Nurses, pilots, aircraft technicians, repair workers, all of whom now have a digital device called a smartphone, allowing them to collaborate from everywhere using the devices they have, using all of their senses. They should be able to speak to people instead of having to type, see things using the camera instead of having to always enter things on just text. They should be able to collaborate in context, and using AI, we should be able to augment how they collaborate. We have an amazing new set of capabilities in G Suite. To show you that and explain our vision for it, please welcome Amy Lokey, Vice President user experience for G Suite. Amy? >> Hi, everyone. It is fantastic to be here with you all today. I lead user experience for G Suite, and I'm excited to talk to you today about how we are elevated human accomplishments and helping businesses transform the way they work. I love working on these products because it's incredibly rewarding for me and my team to hear from our users that we help people accomplish their most important goals with our cloud native collaborative tools. G Suite is comprised of apps that over 1.5 billion people use, like Gmail, Docs, Sheets, and Drive, and it includes solutions specific to the enterprise, like Hangouts, Meet, and Chat. We also provide solutions for students and educators. And I'm excited to show today that more than 90 million students and faculty are using G Suite for education. [Applause] Today, more than 5 million businesses use G Suite. And this includes digital innovators like Netflix, Spotify, and Lyft who built their businesses with G Suite and continue to scale with us today. More traditional enterprises like Verizon, Neilsen, and Car4 who need to adapt to a changing landscape also turn to G Suite. Companies with large front line mobile workforces are increasingly turning to G Suite. So, for example, Air Asia uses G Suite to connect air crews, flight operations, and ground staff, helping to improve their on time performance. These customers are turning to us because they know that this transformation provides tangible business impacts. According to a recent Forester total economic impact study, customers moving to G Suite saw, on average, 1.5% topline revenue growth as well as efficiency improvements, equivalent to saving each employee one month of time. These are just a few examples of businesses that are re imagining how they work. So let's dig in on how G Suite empowers teams to make it fast, make it smart, and make it together. Here are some examples. First, make it fast. Teams need to be able to move quickly, whether it's responding to a customer request or getting a product to market. We've built AI into G Suite to remove repetitive mundane work that gets in the way. The Google assistant can now help people get more done at work the same way it can at home or on the go. I'm incredibly excited to announce today that the assistant will be able to access your G Suite Calendars so you can see what your day looks like and stay on top of any schedule changes. Let's try it out. Hold on, hold on. Okay, Google, what's next on my calendar today? >> Next up, you have G Suite spotlight session today at 11:00 a.m. >> Awesome. But I might be running late coming from this keynote. Okay, Google, tell the meeting attendees to save me a seat in the front row. >> Here's the message to the attendees. Do you want to send it or change it? >> Looks good. Let's send it. >> Okay. Your message was sent to the attendee. >> Okay. How cool was that? Time for the applause. Great! [Applause] Next, let's talk about how G Suite helps teams make it smart. With G Suite, everybody has better access to the knowledge and the information and people that help them make well informed decisions. Two years ago, we introduced Cloud Search which brings the power of Google search into your business. We also operate as a stand alone solution so that even businesses who have not yet adopted G Suite can harness the power of Google search. Today I'm excited to announce that we are making third party connectivity in Cloud search available to eligible G Suite customers. Cloud Search will not index data sources like SAP, Salesforce, and SharePoint, in addition to G Suite, enabling employees to search and find every digital asset and person in the company. Finally, G Suite helps teams make it together. Individual contribution is almost always one piece of a larger puzzle that requires multiple people to solve. Last year, we introduced Hangouts Chat, our messaging platform built for teams. I'm incredibly excited to share with you today that we are bringing chat into Gmail so that all of your team communication can be in one place. [Applause] Now from Gmail, you'll have the people, the rooms, and the threaded, rich content right at your fingertips. Now, I've gotten to use this for a couple months internally, and I have to say that it has changed how I communicate. Everything is easier, it's more efficient, and I feel much more connected to my team. Now, as much as we rely on chat and e mail, sometimes the easiest way to connect with people is just to pick up the phone. I'm excited to share today that Voice, the cloud telephony service built for G Suite, is now generally available. [Applause] Thank you. Voice gives each employee a phone number that works from anywhere on any device, and it's easy for admins to manage and provision. And with our text to speech technology, it's easy to set up a welcoming, interactive menu to greet the people that call your business. Now, Thomas and I have both talked a bit about how G Suite connects frontline workers to the back office to accelerate access to data and information, but what does this look like in practice? Let's take a look at the retail industry. This is inspired by how we see our customers using G Suite today. So this is Frank. Frank is a store manager at a national apparel chain. He's had three customers ask for the same sneakers in one day, so he realizes these shoes must be hot, but Frank is worried his store is losing sales because these shoes are not in stock. So Frank logs into a shared Chromebook tablet from the store floor, he jumps into chat, and he pings a room with other store managers to see if they're noticing the same thing. Frank quickly confirms the other stores are experiencing the same spike in demand. Well, it turns out a YouTube celebrity recently sported these shoes, setting off an instant trend, but unfortunately they don't stock these shoes on the west coast. With chat, Frank and his community of store managers feel engaged and empowered to influence the store's strategy. Frank and his colleagues quickly send a heads up to corporate using Google forms. Their input is immediately captured and visualized in sheets, enabling headquarters to detect the issue and adjust their distribution strategy. While retailers have a lot of data on the products that they stock and sell, it is hard to gauge customer demand for a product that's not on your shelf which creates a blind spot. With G Suite, Frank and his colleagues on the store floor can share this valuable data that might have otherwise been missed. Back at headquarters, the team uses Hangouts Meet to quickly reach a decision on how to supply these sneakers. And as we watch their meeting in a moment, you will see that features like live captioning, a new capability that is important for accessibility, makes it easy for a diverse group to participate. Let's check it out. >> Good morning, everyone. It looks like customers on the west coast are looking for a new sneaker that we don't stock there. Katie, are you seeing this in your stores as well? >> Yes. I'm definitely seeing this across my ten stores. These shoes are on fire. >> I have pulled the latest inventory data from SAP into the sheet. Let's see if any warehouses are overstocked where we can pull product from. Fortunately, looks like we have a ton of stock in the Midwest. I'll run the numbers and work with our partners at UPS to get that shipped ASAP. >> Great. Thanks for the quick work, everyone. >> As you saw, artificial intelligence in sheets means that employees can analyze data through a conversational query and data can be easily pulled in from databases like BigQuery and SAP. Teams across the retailer, from the store floor to headquarters, were connected and data moved instantly, enabling the company to quickly get the shoes on to the feet of west coast customers like myself. So in closing, I'll leave you with this quote from one of our customers ATB Financial. That's pretty incredible. After moving to G Suite, nobody at ATB Financial would ever go back because it's changed who they are. G Suite is not only a catalyst for elevating the way that your teams work together and accelerating their access to information, it's an investment in your people and your culture. I'm excited to see you make the jump to G Suite, and I am confident you will never want to go back. Thank you. [Applause] >> A company that made the jump to G Suite and has decided never to go back is Whirlpool. Would you all give Michael Heim, chief information officer Whirlpool, a warm welcome to Google Cloud Next? Mike? >> I'm fairly certain I just sold a dryer backstage. So it's clear that Google Cloud is helping me drive that, so I appreciate that. >> Thank you. Mike, you're an early enterprise customer of G Suite. What made you decide to join Google Cloud? >> Well, we started our implementation in 2014, but the story goes back before that. We were emerging from a general recession in our business. We saw tremendous innovation on our product pipeline, and we really had the ambition to get that product to market faster. That required us to change the way we work, the way we engineered across our global engineering centers, and we embarked on a mission, partnering with our friends in HR and our colleagues and facilities, to renovate our buildings, change our culture, and then put in place a new productivity suite to underline all of that. We called that strategy the winning workplace, and it was really about we put our consumer in the center and it was really all around how do we live differently digitally and then how do we work differently digitally? So that whole strategy really began to drive our thinking around what is the right platform? And we got a little bit crazy. Not only do we want to change consumer experiences but we wanted to change our business partners experiences with us. And we decided to conduct a series of tech fairs. We brought in three competing products, and we let our business partners come in, our colleagues, and try them, essentially kick the tires on these products and see which they might like. And overwhelmingly, they chose G Suite. >> What's been the effect? >> Tremendous, I would say. The way in which we have driven collaboration, productivity inside our environment, everyone is just thrilled is the quickest way to say that. And maybe a couple of stories of what people can do on these products. We have a plant in Mexico to build a whole productivity visual dashboard around plant operations on their own in G Suite. We got tremendous support. And if you haven't been in these products and you don't understand the challenges of versions versus no versions, you have to really try it. So we had one of our best supporters is the CEO's executive assistant. She was tired of presentations coming to the executive committee or going to the board, changes, revisions how do you keep all of those things current. And she really drove, no, we're going to do it one way, we're going to do it in Google, we're going to set that up and when you're done, you're done. And the beauty of these products is there are no revisions. When you're operating on the same when you see an error, fix it. When you are operating on those platforms, when you're done, you're done. We also had another example where we acquired a company about the same size as our European business. We needed to integrate them quickly. And the first step in that integration was first to give the team managing the integration the tools because that was very early in the deployment. And then we very quickly brought that whole business, if you will, into our business and made them part of our family. And when you can collaborate that way and share in that fashion, you really increase the speed at which you can get things to market. >> That's amazing. What else have you got plans for, and how else have things changed? >> You know, we chose one of the big reasons we chose G Suite and Google, we believed we would get a constant stream of innovation. And clearly you've seen that again this morning, and we've benefitted from that. So we've not plateaued. We started our journey in '14. We're five years in. We're still growing. We've got tremendous additional productivity through your next mile program. We thank you for that. And we see just continuing opportunity to drive the way in which we work and the change that is associated with that. So and maybe second one that I will just mention, this was our first big push towards the cloud and both for the enterprise, for our business partners, and for the IT function. What we have done as a result of that and the success that we have experienced that really emboldened our strategy to go cloud first and from that moment until now, we've got over 80% of our workloads in the cloud, and it really all started with the success that we drove from G Suite. >> Thank you so much, Michael. It's wonderful to have you. [Applause] . >> So the last piece of our portfolio, specific solutions to enable digital transformation in specific industries. One of our important assets at Google is Geo. And to show you how companies and Enterprises are using Geo to transform their business, please welcome Jen Fitzpatrick, senior Vice President of Geo at Google. [Applause] >> Good morning, everybody. I'm happy to be here to share with you today what we've been up to in our Geo work at Google. For more than a decade now, we've been mapping the world and translating that knowledge and understanding into meaningful experiences for our users. We started many years ago by mapping roads so we could help people reliably get from point A to point B. But over time, what we've seen is our users showing an appetite for more and more types of knowledge and understanding about the real world. They asked us more and more complex questions, and they fully expected maps to be able to provide helpful answers or experiences in new types of contexts. So we sought off to deepen our understanding of the world, to bring new types of knowledge and understanding to the map as well as more dynamic and realtime content. The real world changes fast, and keeping up with the pace of that change can be a challenge. But in recent years, advances in technology have helped us get smarter about the way that we build maps. Using the power of Machine Learning and what we know about the world through things like imagery and user contributed content, we're able to keep Google Maps fresh, accurate and realtime for more than a billion users all around the world. Let's take a quick look at how we do that. We use Machine Learning algorithms to look for changes in satellite imagery which allows us to automatically detect new roads. Based on this approach, over the last nine months, we have added half a million kilometers of new roads in more than 20 countries. With a similar technique, we can automatically detect and draw building outlines, which has allowed us to more than double the total number of buildings on the map around the world in less than a year. We continue to push the bounds of what it means to map the world and to have an incredibly rich and up to date data model. Now, what does this have to do with all of you? Google maps platform is the means by which we give our enterprise customers access to the same rich and complex real world information that we use to power our consumer products. This has long been an area of focus for us. In fact, our original maps APIs are well over ten years old. What we see happening today is a real acceleration in businesses finding new ways to use location and real world insights to transform their customer experiences as well as how they operate. Let me share a few examples. Transportation has always been an obvious area of focus for us given our core emphasis on helping people navigate the world. And as the transportation industry continues to transform, we're partnering with many different players, from automakers to ride sharing to public transit. In the automotive space, we've partnered with automakers for years to bring information from Google maps into cars, but in recent years, we've begun going deeper, partnering with auto companies like Volvo and Mistusibhi to bring a fully embedded voice driven, auto tailored version of Google maps and the Google assistant to the car, and we're excited to see the first cars with this new experience hit the streets next year. But really, this is just the start. Cars are quickly becoming more connected and gathering larger and larger volumes of sensor data. So whether it's making sense of telemetry data from cars or spotting part performance trends or helping manage in car machine learning and compute, we see many ways that we can partner together. Ride sharing is another industry that we partner closely with. Today we power more than 800 million routes each day for customers like Ola, DD, Go Jack, Lyft, MyTaxi, and more. Beyond just navigation, though, we're helping to provide insights that can drive realtime decisions and optimizations for their vehicle fleets so operations get more efficient and riders and drivers can have a great experience. We're excited to keep innovating with our partners to unlock even more benefits to the transportation industry, as well as for cities and public transit agencies all around the world. The information that we know about places, businesses on the map, and points of interest all around the world also bring great opportunity for enterprises of many types. Let me share a few more examples. Last year we introduced a tailored solution for the gaming industry, letting game studios use the power of Google maps to build real world games, creating virtual game worlds based on real life locations or creating game play in the real world based on the Maps' knowledge of where the best spots to play are. The game studio Mixi was an early customer and partner for us here. And when they added a real world experience to their already successful game Monster Strike, they saw a 30% increase in daily sessions per user. And this experience isn't unique to gaming. As the world has gone mobile, location based capabilities are repeatedly transforming customer experiences across a wide range of industries. The travel industry is another area where knowledge of real world places is absolutely critical to success. And so we're working with travel booking sites to add our deep understanding of places directly into their customer experiences. Servicing the right information about what places are nearby when someone is trying to make a decision can help drive conversions and turn casual explorers into real customers. These are just a few of the innovative ways we see our customers taking advantage of the Google maps platform. Now, many of my examples today have been about enterprises using Google Maps platform to provide better experiences for their customers, but we're also working with companies around the globe to provide real world insights that help drive operational efficiency. From retail to insurance to finance and more, we're helping with things like where to best target your promotional spend, where to best locate your next storefront, and many, many more things beyond. So no matter what industry you're in, location based data and insights can have a powerful impact on your business, and today more so than ever. We encourage you to think about how you can use the power of location to transform your business. And with that, I'll turn it back to Thomas. [Applause] >> So we at Google are using our cloud to bring the best of all of Google's technologies to enterprises. And one of the industries that we've had a long history of working with is health care. Our mission is to help organize different kinds of information that health care providers need, like electronic medical record information, genomic data information, clinical trials information to help organize it so that providers, life sciences companies, and others can use that information in new ways to take care of patients and prevent disease. One of the customers that we have been very fortunate to work with is McKesson. And please welcome Andy Zitney, chief technology officer of McKesson, to Google Next. Andy? >> Thomas. >> Andy, thank you for coming. We're very excited to have you with us. Tell us a little bit about how your IT organization is solving your key business problems. >> Yeah, first of all, thanks for allowing McKesson to be a part of Google Next. We do appreciate it. For McKesson, Google Cloud is much more than the technology, right. For us, it's more about the culture and what it brings to the table of not only the IT but truly the culture of Google to the table. We're a 180 year old company. We've been dealing with suppliers, partners, customers for 180 years. It's critical for us to take a look at emerging technologies and how we change the way we do business and what we deliver to the workforce. It's my job to take a look at the emerging technologies that we can bring in for the business and enable them to deliver at speed. It's more important now than ever. And if you take a look at the technology that Google Cloud brings, it fits that space and allows us to deliver at speed. Several years ago, McKesson started a journey looking at what we should do moving to the cloud. We partnered with a people, a few other companies and had looked at selling some of our data centers, getting rid of some of the legacy equipment, and really forcing the move to cloud services to eliminate a lot of the complexity that an on prem data center gives you. Our existing cloud strategy was a lift and shift. That really wasn't getting the job done for us. After four years, we looked back. We didn't make the progress that we really thought we should have at that point in time, so we had to look for a new partner, and once again, that's where Google came in and Google Cloud. And in the meantime, we were running analytics projects and looking at POCs, those two conversations sort of came together, looking at re inventing the infrastructure but then the Google Analytics platform, too. Google, it became apparent, really understands healthcare. They understood our business. They understood the regulatory and compliance issues and complications that it brings. It brings customer focus to the table. Plus, we needed to move faster. And Google really brought that culture of moving with speed and engineering mindset to the table. >> That sounds like a strong signal. >> Yeah. It was really interesting for us, right? Google came to the table and made us forget everything that we knew, basically tore up and blew up all of our processes, challenged everything that we did, and brought a software engineering mindset to the table and an automation first mindset that really made us rethink. Google has been around for a long period of time now. It's not a start up any more, but they still have that move at speed and move with agility mindset. And that really made us start to think we can do different things. The work approach that Google brought was really a shift in our culture. Google Cloud became an actual contender for our major shift in transformation projects. They competed well by showing up much differently than the others. And it really comes back to that culture and mindset of moving with speed and enabling the business through the technology. To be honest with you, at the beginning, I think my team was taken aback by Google's approach, because it wasn't just the lift and shift. The team seemed a little bit frustrated from a partner coming to the table and telling us we needed to significantly change the way we were delivering and think differently. The leadership viewed that as an opportunity to change things significantly, much faster than we had in the past. We looked at it as maybe we should be stepping back, thinking more like a software company, more like a software engineering company as Google is. It's not what we were doing at the current point in time. Our incumbents were telling us lift and shift was the way to go. Make it a cost saving initiative. That's not the case. Google said reinvent yourself. Make yourself better. We sent the engineer teams back out. We met with our peers. We met with Google engineers. And it became very clear Google was winning a lot in this space, not just small accounts but large enterprise accounts. Pretty quickly, the resisters that were in the teams sort of gave in, saw what we could do for the company and really how much IT could shift from being a back office function to a differentiator for the business. My director of strategy, Trevor, always says, you know, we want to shift IT from being the people that just maintain technology to being people and a team that is driving differentiation to the business and enabling them to deliver product at speed. That means the move to the cloud for us and a cloud architecture needs to be put in place so that we can deploy features faster, continuous deployment, continuous testing, and deliver and fix security across the board in a seamless manner, right. Enabling that speed once again. This moves us from below the line, low value added services in IT, to above that line and delivering value to our business and not just business as usual. >> That's a great outcome. Thank you so much for your partnership, and thank you for coming to Google Next. >> Appreciate it. Thanks for having us. [Applause] If you're a media company or telecommunications company, Google can help you with audience acquisition, media asset management, media archiving, streaming infrastructure, and the ability to reach consumers directly. Not only have we built technology for the industry, we're also very proud to announce today a partnership with Accenture, announcing an intelligent customer engagement solution for a number of industries using Google's platform. Let's hear from the Paul Doherty, chief technology officer of Accenture. [Applause] >> I think the ultimate power of the Google Cloud platform is the acceleration of innovation that you get. Digital transformation is the big thing that's happening in companies with right now. Energy, utilities, travel, financial services, insurance, communications, media, technology, all have some of these common needs that we think we can solve together. Every exchange you have with a customer needs to improve trust, and all it takes is one bad interaction to destroy trust. That's where the Accenture/Google Cloud partnership comes into play, bringing together industry expertise and solution capability of a company like Accenture with a powerful technology platform with Google and accelerating customer solutions and bringing a new platform innovation to customers who are looking to draw up their business faster. >> Please welcome David Klein, chief technology officer of Viacom, a media company working so closely with us. David? [Applause] David, welcome. Thank you for coming. >> Thank you. >> Viacom is delivering an incredible number of entertainment experiences, billions of people every day. How are you using Google Cloud to make that possible? >> So Viacom creates and distributes iconic content across the globe, Paramount Pictures, Nickelodeon, MTV, BET, Comedy Central, AwesomenessTV through Viacom Digital Studios, Pluto TV, which is a new acquisition, experiences like the BET Experience, live events, KCAs, Kids Choice Awards, and then what we do on the daily events, The Trevor Noah Show, The Daily Show, what we do for things like Paw Patrol and animation. So all of that leads up to a great user experience and iconic memories. With that said, Google Cloud very much enables us to look at a whole bunch of fine points, security, reliability, scalability, enabling us to take that content and enrich it, ML/AI, and how do we at the end of the day, enable the consume experience to be awesome by being on your cloud. >> Why were we the right partner for you? >> So that starts with partnership, just in the words you said. I will tell you, to begin with, your sales group, your engineers, your dev ops teams, your ML and AI teams were very much broad in knowledge bust also came in with the understanding of really wanted to understand what Viacom needs. What's our strategy? What do we looking to accomplish? How are we looking to accomplish that? And then how can the Google tools sit on top of us and make sure that we're enabling and growing with it. So what I would say to begin with, understanding our culture. Viacom's got a very rich and traditional culture that is based on creativity. And that creativity was enabled by Google coming in and really taking some time and thinking about what can we do with you? What can we do with things in the music industry? What can we do around girls who code, which I'm a big advocate of, with women in engineering and what goes on in the space of an inequality of dev ops and how we help balance that. And then really then came in and helped us understand some of the things you've done around search, around metadata, around enhancing that content, and really coming together on a joint opportunity on how we can take our content and really enrich it, as I mentioned earlier. That's our intellectual property. Taking that content, putting it up in the Cloud, making it accessible, knowing that it's safe. Knowing that it's got the proper if you want to call it ability to enhance and entice our consumers is a big piece. And we really have just gotten started on the journey, but we are so very excited and very much looking forward to a future. >> Thank you very much for coming and thank you for your partnership. >> Thank you. It's a pleasure! [Applause] >> Lastly, retail. An industry where Google helps so many organizations. We've introduced a number of new products targeted for retail, developed these search products using images, recommendations, e commerce hosting, ordering things online with a digital shopping assistant, demand forecasting, and others. We've also built retail health care and financial services solutions specifically with Deloitte. And let's hear from chief executive officer of Deloitte. Janet Foutty. >> What I hear from customers about the Deloitte and Google cloud alliance is that they love the richness of the Google technology. The very, very deep industry and domain expertise that Deloitte brings to bear, our focus on the end customer, and, frankly, our ability to execute and execute aggressively. Financial services, health care, government, and retail are the places where we believe we can collectively have the most short term impact, centered in and around where SAP is going as a critical dimension, where customer and marketing analytics are going is a critical dimension, and then all things that surround those domains. The Deloitte and Google Cloud alliance allows us to address both the great optimism and enthusiasm about the market, as well as the anxiety and worry about can we keep pace with what's to come. And it's those fundamental that's really what it's all about. >> A customer in the retail industry that's working with us to deliver analytics and insight in new ways is Proctor & Gamble. Please welcome Guy Perry, chief data and analytics officer of Proctor & Gamble. Guy? Welcome. What business problems is Google Cloud helping Proctor & Gamble with? >> At Proctor & Gamble, our products touch the lives of billions of consumers around the world every day. Brands that many of you know of, Tide, Swiffer, Pantene, Gillette, all those brands, at their core, require data analytics as part of our digital transformation journey. We were looking at digital transformation as a way to get data as a way to disrupt our business and better improve the lives of consumers. Now, to power an enterprise like Proctor & Gamble, there are a lot of data signals that we're looking to integrate and harmonize, including purchase information, supply chain, consumer data on the media, social media channels, as well as demographics and weather. There's more data out there from more sources than you and I ever would have imagined when we first started our careers. It's very exciting but also a great challenge to scale for an enterprise of our size. Last year, we were looking for an agile leading edge platform on which to collect, organize, and analyze data. What we really were looking for is a deep technical capability to scale to our needs, and so we leveraged our experts to do an analysis of all the cloud providers including Google Cloud. P & G has a very thoughtful cloud strategy, and based on the successful technical pilot, we're excited to have GCP as one of our strategic cloud partners. >> Why did you chose GCP? What were some of the reasons you chose us? >> So at P&G we have a very deliberate data strategy, and we want reliable granular data to be integrated and harmonized and powered by algorithms to inform and democratize the data for decision makers around the world and in some cases in realtime. So what this means is we need to have data loading and prep at scale and data warehouses built and priced to handle the scale and the future data loads that we need to process by including very low hassle management and overhead. We also want to apply the best ML and AI on top of the data to deliver what we call our superiority strategy, which is irresistible superior product and package, brand communication, in store execution, everything aimed at delivering value to our consumers and customers. So while we're in the early days of our GCP relationships, we found that in the Google Cloud. And one last thing, the core of what P & G stands for is passion for consumer understanding and focus. And even in IT, we're looking for providers who understand that about us and are wanting to work with us on those terms, and we found that in the Google Cloud, too. So, we're looking forward to sharing many successes together, Thomas. >> Thank you very much, Guy. Thanks so much. >> Thanks for having me. [Applause] >> Thanks. So, what an amazing two days. So many product announcements. To bring you this technology, we at Google are committed to expanding our go to market teams, our work with partners, and deepening our industry and product focus. You will hear a lot more from us at events throughout the year. Today, I wanted to thank you all, our customers and partners, for the amazing time you have given us at Google and supporting you in our vision for transformation.

Channel: Google Cloud Tech

Views: 61,491

Keywords: Keynote, google cloud, next 19, NEXT 2019, Google Cloud Platform, GCP, live stream, livestream, Product Innovation, Justin Arbuckle, Michael Heim, Urs Hölzle, Thomas Kurian, Amy Lokey, Binu Mathew, Sarah Patterson, Rajen Sheth, Karen Van Kirk, ML & AI, data analytics, Maps, security, type: Stream;

Id: PZ1Lqxfs1yw

Length: 104min 12sec (6252 seconds)

Published: Wed Apr 10 2019