BI in the age of artificial intelligence

 BI in the age of artificial intelligence


>> We are in an era of abundant data and opportunity. To compete in today's landscape, organizations must have the skills and talent to leverage all that data effectively. They must acquire the right technology to empower people to do more. . >> AEG is a number of industries. We have AEG sports, we have AEG presents, we have AEG facilities. >> We have an immense amount of data flowing through our engagement strategies. Being able to analyze that data and have a workflow that services it and puts it in the hands of our relevant department heads in a way that they understand and can act on it, is the most important thing.

>> Microsoft Power BI is delivering new experiences. Powered by advances in cloud computing and artificial intelligence that help every user more easily explore and understand their data. >> Each time the car is going around the track we are processing gigabytes of data that can be analyzed and they will be a competitive advantage for any teams that will be capable of implementing some artificial intelligence into analyzing all those data. >> I think being smarter because we master our data and the way those data flows seamlessly into the team, will make a great difference in the future. >> Unlock the opportunity of abundant data through innovations in AI and the power of the cloud. Power BI from Microsoft. >> Please welcome Kamal Hathi. . >> Good afternoon and thank you so much for joining me today here at Ignite. I'm so happy to talk to you all. And those also online streaming. Really, really glad to be here. The video we saw was just really interesting from the point of view of all the things that are going on with data and AI, we saw, and Power BI, the use of the data.

In the end, it's all about one thing, and that is to get information, unlock that information, and really take advantage of it and add value to organizations, and then in the marketplace. As you think about information, I want to tell you a little story and go back a bit in time and talk about how this information got unlocked and we made available to us, in the first place.

So, you could take a trip back to 1440. Any thoughts, guesses, ideas, what 1440 happened that year? Anyone? Maybe? I can't hear very well. But anyway, I get all kinds of people say it's Columbus. But it turns out this is the year when Gutenberg invented the printing press. This is the year the printing press, we believe, was invented. And the prime reason for that we know is that there's a record of a lawsuit, IP infringement. And we know there's a lawsuit it has to be real, so this was the year we believe the printing press was invented. Now, before the Gutenberg press, getting books was only available to very, very, very few people. Books, indeed, were written by hand. They were literally copied by hand by monks sitting in monasteries, literally, in cold dark rooms in the back somewhere. They had nothing else to do. All day, all they did was write on vellum, animal skins, or on paper or parchment.

And they wrote out these books. As a result, very, very few books and mostly on religion. And you had to be somebody super rich, super special to be a monarch or some royalty or rich person to get these books. The printing press changed all that, literally, overnight almost, it changed all that, because it made mass production of books possible.

So what we saw is that from 1440, about 150-160 years, the price of books, because of mass production, just plummeted. Literally, just went down massively. And now books were affordable. Anyone could buy it. So think in what we see today with the online sales et cetera, having impacted books. Nothing, nothing compared to what happened when the printing press came out. Not only did the price of books drop and people were able to afford to buy these books, but the variety of books also changed drastically. It went from essentially, Monks writing about religion and people reading about religion to variety of books on science, on knowledge, literature, romance. All kinds of things that are coming out because both the costs went down and the variety started spreading. So as a result, everyone started to get into reading. And we can see the impact of that as the world started to move, skills started evolving.

And if you look at this chart, it's a session on BI, so we always have graphs and charts, I suppose. Look at this chart, we think from 1500 or so roughly unto today, there was a steady increase in the literacy rate. People who are reading in the population that started going up and roughly by 1800 or so, about a third of the population was now literate. That's a big change. From a few people who are getting books through monasteries and monks to a third of the population now is able to read. You have people who are specialized authors, who are writing books. And suddenly, there's an industry around this whole thing happening. But still, it's only about a third of the population. Then, in 1800 or so, something really interesting happened. The literacy rate drastically changed. It just shot up. We see it go to 99% plus, it just goes through the roof. Question is, what happened? What caused literacy to become universal? But today, I see you're illiterate, that's almost an insult, right? What happened? What happened was the economic circumstances that drove literacy changed. There's this thing called the Industrial Revolution, where machines started to come into play.

Steam Machines, other kinds of mechanical machines. And these machines, in turn, required people who could understand instructions, written instructions, could read. And literacy started to pay off economically. And there was an incentive to be literate. In fact, governments started to invest in schooling. And nations and their prosperity was defined by literacy. As was economic dominance in the marketplace was defined by literacy. Really we saw this evolution that starts happening, that everybody now starts reading.

In fact, we are at a point now where it's not just about reading and literacy, but we're all writing as well. I bet everyone in this room today has written an email of some sorts. As an author, we write Word documents, PowerPoints all the time. In fact, this creation of information is not just in those realms, if you look at social media, that's an explosion. No matter what your age is, no matter what demographic you belong to, you're posting on social media. This creation of information is just rapidly increasing more and more. And as we look at what happened with those machines, the steam machines and the mechanical machines, the water and electrical machines.

They too, are creating information. They're creating data. We now are in this stage of life in our industry, where digital devices are driving this notion of a digital transformation where data is everywhere. And so just as in the steam and the previous industrial revolutions, literacy was what drove competitive success. Now, data and the notion of working with data, data literacy is what will define the ability to succeed in this new age.

So this is the fourth industrial revolution that we are in, where data and data literacy is driving success, working with data, big data, AI. These are the differences in success and failure for large companies. In fact, this notion of having everyone in organization be able to work is this notion sometimes called data culture, and this spread of data literacy is what is required to succeed in this new age. So now if you ask, what is the enabler? What will make data literacy happen? Was there a moment actually, that we saw this, if you will, printing press moment for this new digital transformation, this new industrial revolution? And in fact, there was. And this thing here, how many of you know what this is? Any guesses? It's VisiCalc. VisiCalc was the first spreadsheet. And spreadsheets in many ways are the printing press for the digital age, for data. Before the spreadsheet, I was in IT, and I sat in a cold dark room in a IT monastery, if you will, like a monk. And my social life was that different than that of a monk really. And I controlled data going out to my constituents.

That's how information was going out. But the spreadsheet changed everything overnight. Suddenly everyone's able to access data and started making sense out of it, add big numbers up, draw graphs, draw charts, and overnight suddenly we started having the same kind of impact in some senses as a printing press had on literacy. The spreadsheet started impact this notion of data literacy. In fact if you look at the evolution over time of how this data literacy is spread, you can look at maybe this notion what we call classic BI, where IT has provided information to end users. We have seen or we actually are in the middle of the stage where analysts are working with tools like spreadsheets, unable to take the power in their own hands and pass this information, it's much like book authors, onto other people.

But we're still in some senses now, to look back, we talk about literacy, that curve is pretty much at the stage that we were with the 1800s, they're just about flattening out there. We haven't hit that knee of the curve that just takes off, and makes this motional data literacy universal. And so what we are pushing for now at Microsoft, with Power BI, with our use of artificial intelligence and machine learning, is to drive this notion of a data culture.

This data culture is where data literacy is universal. Where you are able, in your organizations to use data at every level, stockroom to boardroom. Everyone can take data, make decisions, and drive those decisions to really transform how your company works. And we're enabling this for everyone. The business user who really doesn't know how to go and configure servers or work with sophisticated and analytical tools, they want to be able to get value out of data using these kinds of tools, using the value of AI and machine learning.

And this goes across the spectrum. For the IT individual, giving them the power to do a better job in running the infrastructure. For data scientists, and helping them to be able to go in and understand and make sense out of data in a better fashion. The analyst giving them better tools, and certainly the business person and having them work with data to have assistance to make the better decisions. So in some senses, we talk about artificial intelligence, we talk about machine learning, this notion of AI is not just a way of talking about something artificial, but it's a system, it's a system intelligence AI. Where these technologies are not about replacing anyone, but helping them to make better decisions and get to a place, competitive advantage in this digital transformation that we're in, and in this age of the next industry revolution.

So what I want to do now is invite on stage in a couple of minutes, one of my colleagues to show us some demos. And talk to you about how Power BI, which is the product we're going to show you, and the capabilities in it with the AI and machine learning that's built into it, how that can help really make the power of AI be available to everybody. This notion of democratization of AI and how we can achieve that. And so Patrick Baumgartner, I will invite him on stage. Patrick is a part of my team.

And Patrick is going to talk to us about how we can take advantage of AI, with Power BI. Thank you Patrick. >> Thank you. This concept of data literacy is really interesting. Kind of makes sense right? I think we all wish that in our companies, in our lives that the people around us would use data to make better decisions. I think we've all been faced with a circumstance where maybe somebody is making a decision, kind of made up and it doesn't go as well as it could have.

We want to think about how do we actually help everybody use data better. And it's not going to be some simple solution, it's going to be a combination of access to data, training, and certainly better tools. And one thing with Power BI that we want to be doing and across all the Microsoft stack is baking in intelligence and everything that we do, to simplify how people have access to data, all across, from all different personas. Let's take a look here at a demo we can play with some of that. And there's a lot of data out there that when you see it, it hits you with a strong impact. And one data set that I wanted to pick to talk to you about is actually traffic fatalities. In 2015, there were 35,000 people that died in a traffic accident. And when you dig into this data and you play with it, and you understand it, some things that you take for granted, I know in the past I was guilty maybe on doing a short trip to the store I wouldn't buckle my seat belt right? After you play with this data, you go, that's just crazy, why would I ever do that? And when you start thinking about the impact of if everybody had access to information and could explore it and play with it, whether you're a business user or a data scientist, you can really actually make an impact to make better decisions.

The first part of that is actually just creating reports people can use. Here I'm looking at the total fatalities, I can see it by state, Florida and Texas and California are really, really big states, they have a lot of fatalities. I can just click here and do a population adjustment and see that actually it's Wyoming and Montana, some of these bigger states with lower populations where maybe people have to drive more actually, have a higher incidence of traffic fatalities over time. But I might want to play with these speeds here and I can see something like if the speed goes up, the average age goes down. Maybe younger people maybe drive faster and more recklessly and cause more accidents. If I look at the hour of the crash and I say, what about crashes just in the early morning? I can see alcohol involvement jumps up, intuitive things, so people probably coming home from the bars, there's more alcohol involvement. So this gives me just a really easy sense of playing with the data. But let's go a little bit further here.

Let's actually unleash, explore the data, and see how we can bake in some intelligence along the way to make this even easier for people to play with data. So I should mention, this is Power BI desktop, the tool that I'm using. This is a free download, it's also now available in the Microsoft store. And it's an amazing tool for mashing data together from a variety of sources, in this case I have traffic fatality information state demographics. But it's also a great way to do visual exploration of data.

And so here I can see the overall trend of fatalities. And it was actually in the early 2000's, it was a lot higher, it was actually 42,000, 40,000 people a year were dying in traffic accidents. And there was this huge improvement, it went down. Turns out this drop, a lot of it was due to gas prices were high, so people drove less, the economy went through a bit of a recession, so people drove less. But we're going to come back to this in a second because there's this really disturbing uptick. And they're about to publish the 2016 numbers just in a couple of weeks, and it's going to be really interesting to see how last year turned out. It looks like this trend is going back up, and we want to explore that and try to understand it. But let's go ahead and play with this report and see how easy it is to explore this data. So maybe I just want to look at again my metric here, I'm just going to look at number of fatalities, maybe I want to look at this by my type of vehicle.

And again, data should be easy to play with. And I'm just dragging and dropping, typing the names of fields, and immediately I can just lay this out and maybe cross highlight, I can see the trend for passenger cars. Maybe I see time here, and I want to go ahead and drill down. And again a very intuitive simple things everything is wired together, I didn't have to be an data expert to start building these reports. And I can see interesting things, like if I look at a motorcycle, the seasonal trend is way bigger because people drive their motorcycles less in the winter. So some interesting insights just right off the bat. But maybe I want to start doing more advanced analysis, maybe I want to get a forecast of what's going to happen in the future. And again, how can we make capabilities like this available to everyone? I don't want to have to be a data scientist to understand machine learning and different techniques, could I just turn on a forecast and immediately see a quick projection of what's going to happen in the future with confidence intervals. But we can go further.

Let's take a look at some of the vehicles we all drive, and maybe we can find some insights there. Let's look at the make, let's do a scatter plot here. And let's look at the speed vehicles are going. This is the average speed. And let's look at the survival rate. So if there were multiple people in the accident, how many people survived or didn't survive. Let's make this a little bit bigger. It's easier to see. And format this a little differently. So let's do the fill point and the category labels and I've got to a Yugo and a Renault and so, let's maybe, let's filter down to a little more recent because I don't think there's too many of Yugos on the road today. You know what a Yugo is? It's Yugoslavian little car. Anyway. So I can see here just really quickly kind of layout of my data and actually let's just pop this out full screen so we can really see it. Down here I've got Ducati, Kawasaki, and Suzuki. These are all my motorcycles, right? They go pretty fast in, it looks like they're pretty dangerous vehicles when they're in an accident.

Down here Gillig, does any one know what a Gillig is. It's actually a mail truck. If it's in an accident, it was probably going pretty slow and fortunately looks like most people survived. But I see some kind of natural clusters in my data here. I've got some some heavy trucks and these motorcycles. And again, I don't want to go and tag all these records individually. I mean those tens of thousands, actually hundred thousands of accidents in this dataset. What would be great if I could just automatically find clusters in my data. So this little pop out here says automatically find clusters.

And I'm going to click okay. And this is going to run a bunch of machine learning behind the scenes. I don't have to understand CaImAn's analysis and other complicated things. And it went ahead and found some natural clusters in my data that I can now use. These are just part of my data model that I can keep doing analysis with. Again, lower the bar of entry for how we think about moving, maybe a person who's just kind of a simple analyst and let them start behaving like a data scientist in everything they do.

This is really great because I can take this, this again cluster here I could drop it. I can cross highlight by it so, I can look at maybe this cluster too, which is mostly motorcycle's with a couple other type of vehicle types by those makes. And again, so it's something really easy to play with. But let's see if we can even simplify this a bit more. And let's come back to that big question we had right at the beginning. So, I'm going to turn off my forecast here and I'm going to pop back up to the top, maybe zoom back out so we can see this whole line here. We see this spike.

And what's going on here? Why did this spike? What's going on? And this is where the power of AI gets even more powerful. Before you saw me baking a little bit of machine learning, and again, I didn't have to know what that was, it was just part of the tool. But what can we start doing now with the power of the cloud and in computational power and maybe I see this increase and I wanted to just ask, explain the increase, right? We can go ahead and trigger this and start looking for reasons that this might be gone up. And I think one of the most powerful things about this type of analysis is, it went off and it scanned, we ran many, many different algorithms searching for what would be the best explanation of this kind of time change from period to period.

And we go ahead and rank these, this big list of them. And one really interesting thing happens. Is most people only do analysis, they start with a hypothesis, right? And they kind of start trying to prove that hypothesis, disapprove it or prove it. One of the great things about baking in AI is it causes you to find stuff you didn't think about, right? Because the machine is just looking for statistically interesting subsets of data. In this case, it found something that really stuck out to me, which one of my hypothesis would have been that maybe drunk driving is going up. But it turns out actually that drunk driving actually went down between 2014 and 2015. And again, this is AI just automatically found that for me and laid it out in a really nice little waterfall chart that I didn't have to know how to create, I didn't have to know how to break this down by different periods, and I can just go ahead and add that to my report, just like that I can add that and have it as something that's fully interactable and something I can keep using.

But let's keep looking here. I found this second item that popped up. It might have been really statistically significant, zero for drunk drivers, but it's actually hard for me to understand. And one of the other interesting things we can do is build in something like voting. And I can tell this wasn't an interesting insight, maybe it was a poor choice of fields, because I really don't understand what this zero is. It's hard to read. And I just gave the system feedback. And that's feedback that we can take and we can learn how to improve the feature in general, but also start building up an understanding of data within the organization and the data that I use and start learning from that over time, so we can suggest better and better insights over time.

And I can keep scrolling through, here's another interesting insight. It says that there were some big changes amongst states but not a big relative change amongst them. And I can look at other visualizations out of the box here. Here's a scatter plot of the year over year change and I can see almost all the states, follow that same trend. It is interesting Florida got a little bit worse. Texas actually went down. And again, this is actually if I wanted to build this chart, it actually would have taken me a while to set that up and the formatting and create the calculations to make it. And again, the AI just helped create it for me on the fly. But there's one other insight that popped up when we are doing, when I was prepping for this demo that was really interesting to me.

So I'm going to add this other one here and pop it out and talk about it for a second. And again, we think about a hypothesis, I came in with a hypothesis about what was happening with accidents, and I think of myself as a really safe driver in general, right? I've got young kids. But then I saw this chart kind of pop up in the list and again little things built into the product, I really love this. This is this waterfall chart again, and I can just kind of group it together, and it automatically does another category for me. So, let's just call out this one big bar.

It turns out between 2014 and 2015, the biggest increase in fatalities was actually not collisions with two vehicles, it was just one car involved. So they either hit a pedestrian or they drove off the road and rolled their car over. And again, I had this hypothesis coming in with this dataset that it's the other person. I know I'm a good driver, I'll stay out of the accident. And then you look at this and you start seeing statistically, well, there's a big chance that the accident that's going to get caused is actually me as the driver. And again, that sticks out with you in it and it's helping people find those delightful little insights with data that helps people buy into making better decisions with data. And so, you saw here, I mean this is pretty cool stuff, right? I mean hopefully you want to get your hands on this. Like I said, you can go get it in the Windows Store today, it's a free download.

It's a really cool way to start powering up what you do with your analysis. But not everybody is going to go play with data and even explore, as easy as we make it. A big part of this idea of data literacy and data culture is also, how can we format reports and make them really interesting to our consumers, right? The great thing about Power BI is, we have a really modern data visualization stack and we have a huge library of custom data visualizations. You can develop your own data visualizations if you like, and you think about this literacy again theme, how can I create reports that are compelling to read, that people want to read because that's certainly the demand side of it is people want interesting things.

And so I can build really visual reports like this where I can still interact and say, let's look at just passenger vehicles. Let's look at where the impact points were happening in these accidents. Let's look at the fatality rates by seat and let's make some things that people want to play with, right? And, again we can go even further again. Maybe I want to do a what-if analysis. One of these new features we just shipped in Power BI is I can actually package in a model into Power BI with parameters that you can pass into the model and do a what-if analysis. And so, here I'm looking at increase in our, excuse me, so I'm looking at 35,000 people die in traffic accidents in 2015, I could go change the year if I wanted.

But it turns out 15,000 of those people weren't wearing a seatbelt or a helmet. And, again, you think about this, it seems pretty obvious, right. And when I'm playing with this data it seems even more obvious. And 5,000 those folks were involved in accidents where there was a drunk driver. And so this little report here, I can come into play and I can say, what if we had a 50% increase in people that were in these fatal accidents wearing seatbelts. I can see we would have saved based of the survival rates across this whole dataset because again there's I think 500,000 accidents recorded in this, so we can get a pretty good idea what different cars would be like, with seat seatbelts or not seatbelts.

And you can see we'd save a lot of lives. But what was also really interesting is you see something like the change across all these accidents and survival rate. If you're wearing seat belts, a boost seat belt people, people wearing seatbelts, these high speed accidents get, people are safer in those. This really stuck out to me, it was actually low speed accidents. I told you at the beginning, I'm guilty of sometimes running to the store and not buckling up and it's not anymore. I actually see that if you're going to die in a low speed accident, it's probably because you weren't wearing a helmet or a seatbelt. And again, you can see motorcycles really high here.

But the same thing, I can just click on passenger vehicles, the car I drive, right? And so cross highlight and see just the impact of that vehicle that I drive. And again it stays that really high, that really high bar there. So really interesting ways to play with data and you can think about packaging this up. Now this is like a traffic fatality. Hopefully I didn't come across as too preachy about wearing your seat belt. I think my mom is really happy that I did this demo for you because it certainly made me a safer person. But you can imagine this with maybe we're doing an awareness campaign here, we're trying to boost awareness and make a decision about where we put our funds. Certainly, I can be doing a marketing campaign or I could be allocating servers. And all of these things are now stuff that you can take to your organization and help all of your users use data better. It turns out we can make this even simpler, right? Again, I'm still here in the report, so, it takes a user going to a report.

And we also want to ask ourselves, could we do, make it even simpler overall. So, I switch my iPhone here. And here I am, I'm in my Power BI mobile app. We have a mobile app for Android, Windows Phone, and I'm on my iPhone here. And in all of those contacts, this is a great way, I'm going to go to my travel analysis, I want to go to my traffic fatality analysis report. I've got the same data, the same metrics, I can take it anywhere I want just right in my pocket. This is something that really appeals to our business users because they can take everything with them wherever they want to go.

And one of the great things we have built in here is natural language. So people just want to ask questions, they want to use natural language and express. And Power BI has really been a leader in bringing natural language into the BI and data space. So let's ask about seat belt use, right? So you just, we're talking about seat belts. So I just want to ask you about seat belt use, and it turns out you know across this whole dataset from 2000 till today, people involved in traffic fatalities, only 52% of people were wearing seat belts, it's a little mind-boggling. But a really cool thing happened, not only did it fetch the data for me, it ran that same AI we were playing with and Power BI Desktop.

We also have that running up in our service and it found a thing that said, "Hey, seatbelt use is trending upwards", right? So, I can see really quickly now it found an insight for me that pulled me through and said, fortunately the average across time was 52 but actually now on average is up to 65% or so with a bunch of seasonal variation or change over time. And again this is still even conversational I might say, for 2015. In Florida, yes, giving the state abbreviation. So, I can just ask a follow-on question and say, for 2015 in Florida, and it turns out actually in 2015, at least in Florida, it is actually 70% which was good to see, but what's really cool is that same AI was then running and it tried to find some other insights that I might be interested in.

And if I scroll all the way over here, it actually said that speeding amount, which is this is the amount above or below the speeding limit, had outliers. And every time I play with this tool, I find something else I want to go follow up on. Because now it found, actually, there's a couple of days here where there was apparently a bunch of accidents where people were going below the speed limit, there were a couple of days with really abnormally high speeds. And now I want to like, this is interesting, let's go follow up on it. So, I can add a quick annotation and say, "What was going on here? Do I want to follow up?", and I could go and share this with my business colleagues.

And so, I think what you saw here, hopefully what you saw, is how we can take AI and really start baking it into all of the experiences that we build. We want to pair that with a system that makes it really easy to build engaging reports and dashboards and engaging ways to play with data. And hopefully, in your organizations, this inspires you to go make data more available, create better reports, but also help your users come and find data in more interesting ways. And hopefully then, everybody make better decisions whether it's traffic fatalities or how you run your business. So, thank you very much and I'll hand back off to Kamal.

>> All right, thank you Patrick. Thank you Patrick! That was cool, wasn't it? That's pretty amazing to see him just to ask those questions, but that's how transparently we use AI and machine learning just inside the product. But as an example how, on a personal level, you can go in and jump in these tools and get value and start to just really make sense out of data and this notion of your data literacy for everybody. But then I want to change it a little bit and talk about how organizations, how companies can take advantage of similar kinds of things and start to really transform themselves using those kinds of things that Patrick showed. And so I'm going to do for that is actually have one of our customers, Real Madrid, in a minute join us. Real Madrid, for those who don't know, is the world's most valuable sports franchise. I mean literally, any sport you're talking about, Real Madrid is the world's most valuable franchise, no question, bar none.

And so, I'm going to ask the global director, digital director for Real Madrid, Rafa delos Santos, and the principal program manager for Power BI, Mark Gregara to join me on stage. Rafa, Mark, please join me. >> Hi, guys. >> Thank you for joining us. Mark. >> Hi, thank you Kamal. Hi, everybody. As the global digital director for Real Madrid, it is very important for us to connect with our fans and the fans are in all the digital channels that you can find out there. Instagram, Facebook, you name it, the Real Madrid app, the website. You know that if you follow a sports, you will see that there are many, many rivalries. If you follow baseball, probably the rivalry between the Red Sox and the Yankees is the most important one.

If you follow rugby, you know that Australia versus New Zealand is a big thing. If you follow cricket, you know that when India plays against Pakistan, it's a big thing. But when it comes to soccer, the game between Real Madrid and Barca is the greatest rivalry game that you can find out there. If you look at everything that we do, and that game is called el Clásico, and more than 400 million people follow the game in the stadium, in television, many different places. Just to show you a little glance of it, let's roll the video. >> This is the El Clásico. >> Brilliant. Ronaldo puts it home. >> Swatted. >> Genius. >> Oh, what a goal for Real. >> Astonishing Messi. Magnificent from Los Blancos.

>> I am rivalry. I've grown up. Well, this is a little video that shows the rivalry that we have against Barcelona. And this is not only on the pitch. If we look at everything that we do, we compete against Barca, and Italy is no different. For the past year, we had the big race to see what was the first team to reach the 100 million fans in Facebook. Facebook is a very important tool for us because it's the way we can engage with our fans, the way we can send them the content and generate the passion around Real Madrid.

So, Mark, can you show us a little bit about how we were doing in Facebook compared to Barca? >> Oh, I'd love to. So, here we have a Power BI report about the growth of the Facebook fans in Barcelona and in for Madrid. So, we're going to see the line here shows the growth in Barcelona from February to March six months ago, February 1 to March 31. You can see the line starts for Barcelona in red at 96 million and ends at the end of March at million. And we're going to see dynamically with this really, I love this custom visual, the pulse chart, how well Madrid has done in the same time period and as we move forward, we're going to see the corresponding postings that you have done. So, we're going to start the time here and we're going to see what happens here on February 5. You've got a pretty good increase right there. Can you tell us a little bit about what happened on February 5? >> Well, actually that was Cristiano's birthday.

You know, Cristiano is one of our best players. He's huge, he's massive in social media, so when it was his birthday, we decided to run a campaign. We target the demographics, we target the people that was following Cristiano, and we decided to run this campaign. And you can see that the increase that we had was massive. Even if you look at the posting that we did, we were using Arabic as a language to address to the people that is following Cristiano in that side of the world. So, that was probably the main reason for that peak. >> That's right. Then you had like 400,000 fans that were a gain. And so, in the next few days, you have about legends like Beckham, you have highlights of games, and you basically have postings about the team's updates, when they win a game, about some milestone about players, and you can see the blue line little by little narrowing the gap with the red line here.

Bale is coming back. This is one of the few games you've lost, so we'll go fast about that one, you win in Villarreal, and slowly but surely, you narrow that gap up to, I think it's on March 8. You win against Naples, you qualify for the Champions League the last eighth, but Barcelona has a huge job. Do you know why? >> Yeah. Actually that day was a bad day for us. They won against their PSG. They had their first leg that was awful. They lost against PSG four-nil. And then in the second leg they made this huge comeback that never happened in Champions League before they won 6-1. For them just represented like, it was like winning a Champions League almost, in terms of increasing their the fan base following them. So that was basically what happened, it was an unexpected victory. >> That's right. And so that's really why they grew. And then you're going to continue your linear growth, highlighting legends like Xabi Alonso and victories and using a lot. We'll see later those posts that are not random. What you post, the players you post about, the topics, the demographics and you are able to go linearly pretty well close to 96 million.

And you have also a jump with Sergio Ramos. Now, we see Barcelona having a significant growth after March 21 until the last 10 days of March. Any insight why they had such an amazing growth? >> Well, the reality is that there were no reason for that at that time. So when that happened we realized that this was the moment when they decided to go for the 100 million mark. We were panicking at the beginning, we saw how rapidly they were increasing the number of fans that they were out into their Facebook account. So what we basically did was pick up the phone, we call CEO and I had a conversation with him and I said, "Jose Angel, we are in deep in deep trouble." Sorry. We are in trouble. >> He didn't say it. >> And we see Barca trying to catch up a little bit to the 100 and we have to react. And that was the moment where he called the president and Mr.

Florentino and he had a conversation with the president. And the president said, "Do whatever you need to do to reach this 100 million." So at that moment, because of us having all this information and having all this access to data and using the tools and the mechanism that we have and the Power BI resources that we have, we begin thinking about doing the campaigns. The problem here was that we had very, very limited time. What we had at the beginning, maybe two months to reach this 100 million. We knew that because of the rhythm of everything we were going to have only a few hours, maybe a couple of days at the most. >> That's right. So, you had basically two-three days to get three and a half million fans because you know Barcelona was going to get there. And actually the line shows that they got to 100 million two-three days later on April 2.

So can you tell us, you had very little time to figure out what to do. Can you tell us a little bit about your strategy to get that many fans in two-three days basically? >> The reality is that we have the developed up a lot of insight from our Microsoft data platform and we knew a lot of information that we were looking on a daily basis into our Power BI report. So basically, we had an idea on the things that we needed to do first. We leverage in four different areas. We leverage with our players, our current players. We leverage with former players of Real Madrid that were big stars in the past; David Beckham, Ronaldo; the Brazilian one, the other Ronaldo. We also leverage with many of the famous people that follows Real Madrid. Famous athletes that are following Real Madrid and that they are die hard Real Madrid. And we also drive some contests, some contest with prizes that will enable us to gain people into our Facebook account. >> Okay. So here similar visual than before but the access is March 31 after first three days.

And you see the line from Barcelona, it reaches 100 million. So let's see how they did it. So we started with Sergio Ramos' birthday and here we're going to see what happens. It tells us what he has planned for his 31st birthday. Is this significant or? >> Well, actually data told us to do so. We know that Sergio is an amazing player for the club and it was his birthday and we were looking at the data. So we knew that we needed to do something to boost the campaigns that we were doing. >> That's right, so data made you do it basically talking about Sergio Ramos. And so here we can see the different steps. You talked about Ronaldo, some legends, current players, Toni Kroos, Fernando Alonso who won the Formula 1 race. So that's one of the different campaigns that you go. And here we have again some Kaká and we have a little bit of a contest here that you did, right? >> Yeah, that's right.

That's actually one of the things that we know it made a huge difference. It was a little detail but it made a huge difference. We were thinking about running some contest as Barca was doing. The only thing is that they believe that the best thing that they could do at that time was doing a contest with tickets, giving away tickets to people to come to the stadium and watch a game. And because of the data that we were looking at, we know that less than 3% of the population that are following Real Madrid in Facebook is doing it in Spain. So we know that 97% of our fans are outside of Spain. So imagine winning a contest and being in Indonesia and telling them to come to watch a game. It's more than a contest, it's a big thing that can happen in your life.

So we decided to just do it with a jersey and we knew at that point that we were making a lot of good comeback from that contest that we put in place. >> That's right. So here we're going to see progress Kaká who plays for Orlando here helps with Brazil fans. And then the current players like Gareth Bale, again you emphasize the contest and you start gaining momentum. And a lot of the players are participating from the current team. We see Marcelo.

We're going to see Sergio Ramos and you are gaining momentum. We see obviously again the contest that you promote, Cristiano Ronaldo himself. And then one here that is surprising, Rafa Nadal, the best tennis player in the world. I thought his uncle used to play for Barcelona. >> Yeah, that is correct. He played for the wrong team. And then all the family except for Rafa supports Barca but have Rafa is a diehard Real Madrid fan. He likes football, he loves Real Madrid. And it was right before, we call him up and we said that we were in this, the El Clásico as it was called later by the press. And we called him up and literally the guy was in the locker room just changing and getting ready for his next game in the Miami Open. And he said, "Yeah, I'll do it. And he posted this picture where you can see him in his locker room just supporting the team to reach the 100 million fans.

>> So those four combination; legends, current players, contest, and all of that is starting to pay and you start catching up. David Beckham starts helping you and then you have some relevant videos and then you get to 100 million. What time exactly was that? >> Well, I remember that as if it was yesterday. I was in the kitchen in my home watching my computer, watching the phone, talking to many people.

What's happening with everyone? And this happened exactly, that was the moment. It was pm, that was the moment when we reached 100. >> So you screamed goal when you got to- >> Pretty much, almost. Almost, yeah. >> And so, it's incredible but here nothing was done by luck. There was a very, if we can switch to the PowerPoint. There was a very deliberate strategy and I'm going to go back on how you went about this. Can you talk a little bit first about the infrastructure that you use to get to those insights? >> Well, pretty much, I'm quite sure that many of you is going through this very same process. We are first gathering the information. We have 10 different sources that are the core sources where we get this information, which are basically the website, the application or our social media channels.

Then we mix these with other sources where we get information, we get information out of 70 different sources. We massage that information, we put that information into our data lake. We transform that information into insight and then is when we start personalizing the message that we want to send to our fans. And we target and we geo-target everything and this is when we begin doing our campaigns. We are expecting to send out these campaigns and get a return, a return on the investment that we're doing and in revenue. That is hitting the B2C and also the B2B.

And in all this, nothing can be done without the proper organisation which is something that we have in place now. And of course the technology that we've been using all along. >> That's right. And when you bring in new sponsor, this is extremely valuable and helps you in negotiation. So let's talk a bit about technology in itself. Can you talk a little bit about the data sources and how you process that into insight? Well, as I was saying before, as you can see there, we have different channels where we're getting this information. Then we're using out of the box reporting tools like the ones that are coming in with Facebook or Twitter or other companies like Shareablee that are giving us a lot of insight.

We get all this insight. We cross link with the information that we have in our digital platform and then we generate by using the elements that we have for us Excel and Power BI. To run their campaigns and send out all the campaigns that we want to, we want to make it a value proposition for the people that is following us.

>> Okay. And so maybe we're going to make it real. Let's look at what kind of information you get about a fan. And you pick an interesting fan here, right. >> So, this is, you know, I'm exposed now, I'm a fan as well. Tell me a little bit about the kind of info you got about your fans? >> Well, actually, this nice guy here, maybe you recognize him, his name is Mark. He's a diehard Real Madrid fan. So as you can see here, we get a lot of information and a lot of insight from different sources as I was saying before. We try to slice and dice exactly the kind of fan that Mark is. As you can see here, he is in his fan profile. He's a gold member because of the things that he does, the interaction that he does. He's also a good influencer with the community that he's got around in social media.

He shares, he participates in the things that we do. So we begin having a lot of information, we have a lot of access to the kind of things that he likes, the kind of things that he's doing with Real Madrid, the kind of interaction that he's doing with the club. So this allow us also to make decisions in our end. If you look at the lower graph there, you will see that we are also getting a lot of insight from our application.

We get all the telemetry, the things that he's consuming, the video that he's watching, where is he going. We build a map where we can see where he moves around when he's connected with Real Madrid. So all this is giving us a lot of insight and a lot of information. >> That's right. And that's Micro. Now, we're going to look at Power BI as well because you use Power BI, right? >> Yes, we do. We do it in a very extensive way. It's very easy for us to use it. More than anything, it's very easy to us to share it with the people that is working at Real Madrid internally, our senior staff. So everything that we do, we do it based on the data that we're getting from Power BI.

>> That's right. So maybe if we can switch the screen to Power BI and I can show you an example. So this is something you can get today, it's a Facebook solution template that is available on powerbi.com, solution, solution template, and you go to the Facebook. The insights and analytics that Rafa use are available to all of you for any Facebook page that you own. I'm going to give you an example of the kind of information we have. This is something we built specifically because when you get a solution template, you get reports out of the box, but you get all the data here in Power BI desktop, which you can build that info. Here, we know that the countries that are the most popular, India, Indonesia, India, Bangladesh, have a high growth and a strong base. We know the time of the day where it's better to post. We know which language. In Indonesia, there is million people who want to have the local language. We know photos and videos are the best medium and we know the demographic, males between 18-24. So, we build that page very quickly because when you get the solution template, you get all the data.

What you get out of the box in basically seven clicks, it's not just five minutes to wow, it's five minutes to wow in seven clicks. You get all the information about the audience, in this case, the jump with Cristiano Ronaldo, the top post. My favorite page is this one because there is a funnel in Power BI. You have Real Madrid, the 513 posts between January 1 and March 21. 700 people liked it, those are the storytellers. They created pages, and through their friends, they reached 21 million people that they generate the page impression and all of that generated 185 million page clicks on the Real Madrid Facebook page. So you see the entire funnel and through that, they got four and a half million fans. You can see the source of those fans. They can come through an app developed by a third party like the Real Madrid app, they can come through a page that was suggested, probably through the campaigns that Rafa did.

And here, there's a lot of depth here. You can look not just at new fans but you can go to people reached and you can have all these richness and you see all these visual that change, how many people, the frequency, the story type, is it organic or paid. We have a lot of information about demography, so you're going to get that out of the box, again, in seven clicks where they are, the post performance. And one of my favorite, too, this is what the Real Madrid use. You want to know which post is effective? Well, here, we have the history of all the post. We can pick a time line, we can expand that time, and we can see the page, the post that generate the most people reach, or we can change that. And the ones that generate the most reaction in all the post underneath are going to change, and you're going to see which post were the most effective and you have the link.

So when you click on the link, it's actually going to open and show you the post. So all of that is available to you out of the box in seven clicks, and I'm not exaggerating. Powerbi.com, solution, solution template, you're going to scroll down. I think the machine was asleep, there's a lot of other great solution template like Configuration Management, you click on the Facebook one, you can see the actual views that I'll just show you. You click on GET IT NOW, it's going to take you here, and in literally, in seven clicks, you have the same file that I have for whatever brand you have. Walmart, Nordstrom, seven clicks. I'm not exaggerating. So, Rafa, I think-. >> I told you he was a diehard. >> Nobody has those glasses. So the only way I think you were able to reach three and a half million fans in three days because you mix the best sports team with the best technology organization and that's how you were able to bridge that gap. Hope this was a good use of your time.

Thank you. Let's do a selfie. >> Can I join in? No, it's okay. Thanks, Mark. Thanks, Rafa. That's amazing. I mean, if you've ever been to the Clásico, you can really appreciate what this means and the notion of a digital Clásico really takes you to the next level. And this example of Real Madrid is really again about this one thing we've been talking about which is democratizing of AI. The ability to take artificial intelligence, machine-learning, all these things and really making it transparently available to really make a transformational impact in the business you use. Real Madrid, you saw what they did, being data driven, being incredibly precise and able to go and do the right things, the digital Clásico, and they did it in a most amazing way, all based upon leveraging Power BI, the Microsoft platform, AI, all these things together. So now, the question is how do you, and the audience, if you want to go off and do the same thing, how would you do it? And so for us here at Microsoft, in the Power BI team, we've had this one goal, this one approach that we've been driving at over and over again.

And the goal is the following. We are striving to make sure that in five minutes, you are able to go, get Power BI, start to work with it, and get to this moment where you say, "Wow, I got something out of it, I got some value, got an insight out of it." So we call this, "Five minutes to wow", and this, "Five minutes to wow" is this mantra, this thing that drives us relentlessly over and over again. Everything we do, design decisions, what the product does, the features we pick, how can we make it possible for users in any organization to very, very quickly get to this moment where you get value with the product? So five minutes to wow.

This is a design point, this is what the product is driven by. So, Power BI allows you to connect to pretty much any kind of data source. Cloud, on-premises, any kind of source. And out of the box, it provides you the ability with these content packs, the solution templates, what Mark showed what Real Madrid did, to be able to get very fast, effectively create complete solutions end-to-end. And it can do things that most other tools just cannot do, real time analytics literally streaming in the product, natural language, you saw what Patrick did with natural language.

You know, obviously working with custom visualizations, you saw the car visual and all kinds of things you can do there, working with Office and really making it possible for you to get data from anywhere, work with anything you want, and then deliver the results out to the web, to mobile devices. We work very deeply with Excel, embed it into any product you want, work with Cortana, pretty much anything you want. You can get data from anywhere and anytime you want and deliver it anywhere that makes sense for you.

Now Power BI is not just about drilling in the cloud. Power BI is a complete solution in the cloud and on-premises. It also allows you to work with data on-premises and delivering in the cloud, this complete hybrid solution. This is BI on your terms. If you want all in the cloud, it's there for you. You want to run on-premises? It's there for you. You want to have a hybrid solution? It's also there for you. So Power BI enables you to succeed in any way that works for your organization. And Power BI has succeeded. It is a massive success from a market point of view. We hear this from analysts. So Gartner is a top analyst company. Many of you, many organizations go to Gartner to hear about what is it that they recommend. What are their advice on what to use and where to go. And this is the Gartner magic quadrant, as they call it, for BI and analytics.

And you can see, Power BI is definitely a leader in this magic quadrant. And this picture really says everything. So, just take a look, let it sink in for you, you can see where Microsoft stands. Right? And its not just about analysts. Power BI is used across the globe in hundreds of thousands of companies. Hundreds of thousands of companies, organizations, not users, in pretty much any corner of the world you can imagine, all the countries in the world, Power BI is pretty much being used. And Power BI is being used not just in certain niche segments. It's used in every segment of the market you can imagine, from retail to distributors, manufacturing to education, to healthcare, banking, everywhere, Power BI is being used and these logos you see are representative of a few of those. If you really put in every company we've got, this wouldn't be, complete gibberish, to make sense what we're showing here. So really, used all over the world in all kinds of segments of the industry and being used by the who's who of the corporate world. Now what I want to do next is to take one specific example and let them talk to you via video about how Power BI has really helped their business.

And that is with, might get the slide to go further. Can someone advance the slide? Try the different clicker, maybe that will work, let's try that. Okay, it's with Heathrow Airport. And in Heathrow, I fly a lot. And I go through Heathrow all the time. That's an amazing airport. All kinds of things happening, there's so much of traffic. And at the heart of what Heathrow is doing, you'll see, lies Power BI. So please, watch Heathrow and how the big advantage of Power BI. >> Going through an airport is a stressful experience emotionally. When you track our emotional journey, we do everything we can to reduce the stress, let people enjoy the experience a bit more. We are dealing with 250,000 passengers every single day. The 400 companies, the 90 airlines, the 1500 flights. 75,000 people that support Heathrow all need orchestrating, organizing, planning in a predictive manner and that take some huge amounts of data. Every single day, we're handling disruption as part of our normal business.

And the challenge we have is giving the right data to the staff predicting disruption. So there's no point knowing about this in real time. We have to know about it several hours ahead of time. We took a look at what do we need from a strategic platform to handle the data on the requirements that are emerging over the airport. And so, when we looked at all the different platforms available, Microsoft came out as a clear winner.

We focused on Power BI to provide the data to the frontline colleagues who are making the decisions out there on the day to day basis. >> We have a range of employees both internal and third party external who need that data, whether they are on a bus or in the terminals or in the baggage haul, they need to have access to that information so that they can quickly understand what needs to be done and when and where. >> The real opportunity and part of the reason we chose Power BI is the seamless integration and the seamless experience.

The real benefit to the colleagues across the airport and part of the whole ecosystem is it's making a huge difference in their ability to make decisions ahead of an event. So they're not having to react to something that is happening now. And so it takes a lot of stress away from the job. You plan a problem to disappear before the passenger knows there is a problem, so their experience is seamless and so they can come through Heathrow with no interruption. That's our job and that's why we're here. . >> As I said, every time I fly through Heathrow and I think about Power BI running that operation, man, that's just amazing. So the question is how can you take some of these capabilities, maybe transform your organization. And for that again, I'm going to invite Patrick Baumgartner who just did a demo for us to come back on stage and Patrick will show us some more Power BI goodness. Thanks, Patrick.

>> Thanks, Kamal. Alright. So last time I was up on stage we were looking at exploring data and AI builtin and asking questions on my phone and a lot of pretty advanced things. Hopefully we made it simple to use. And then you saw Real Madrid and Heathrow Airport really transformed their organizations using data and integrating it to everybody. But again, those are running an entire airport and that seems as this seems so complicated and massive. What I want to show you today is that we know what we're really committed to in the Power BI team is really simplifying how you deploy reports and dashboards and intelligence throughout your organization. And it all starts with, "Hey, I can develop amazing reports." This is something anybody can do, you can connect to data. This is the report we're looking at. And how do we just make it incredibly simple to deploy these reports and dashboards to everybody.

And so I've got the same report that we were working at earlier. And here's that same workbook deployed in Power BI report server. So one of the great things we haven't and Power BI is an offer called Power BI Premium. When you get Power BI Premium, you can use this either to deploy in the cloud or to deploy on-premise using Power BI Report Server. So here's that report deployed in my organization locally. Here's the same report deployed in the Power BI service of course. But we make it incredibly easy. If I wanted to just embed this into SharePoint and so my alerts can pop up right in the middle of my demo.

But I can put this in a Sharepoint app if I wanted to put it into a Sharepoint app. If I wanted to put it into Microsoft teams, I can put it in Microsoft teams. Anywhere that you want to take one of these reports and deploy it we make it incredibly simple. One of the really cool things that's going to come in October is even the ability to take Power BI apps in the cloud service and share it to other companies. So you can do business to business sharing, which is a highly demanded feature.

And again simplifying taking any of these reports and deploying it the way you want. I can go even further, right. I can think about integration with exporting to Powerpoint. We can even come into Cortana and ask for, the same dashboards now it just automatically show up in Cortana. So this is the web view of the dashboard we were looking at. So I can configure a web view automatically shows up in Cortana. And again, the point here is to say you can develop these reports, you can develop these dashboards but they're only good if they're used, right? And again Power BI with Power BI Premium makes it incredibly easy for you to to deploy these reports, whether it's integrated into your own apps or pushed through the cloud or even Power BI Report Service on-premise.

Incredibly easy. One other really cool feature I do want to make sure I highlight is the administration capabilities for Power BI Premium. This is, you would expect from Microsoft all the great management capabilities, you saw this phone demo. If I wanted on this phone, I can do conditional access and in full administration of how people access the data. And I have full usage reporting and metrics. So this is looking at the dashboards, I can look at the report usage over time, I can look even how this ranks in my organization in terms of people using it, and all these amazing capabilities to really keep tabs on how data is being deployed.

And one of the really fun things, is I can edit these reports, this comes built-in for any artifact. And we love reporting so much that we even let you do reporting on your reporting. So I've got my full edit-ability here so I can look at all the other fields that we might not show out of the box. And again, just really powerful ways to take data to deploy it to my whole organization. In addition to just integrating into these other tools, using power be either integrating into SharePoint, we also focus quite a bit on integration of other tools into Power BI. Because we know that if you're going to make data and distribute your organization, you need to integrate seamlessly with all of your tools. And so for that I'm going to switch to a slightly different data set and I want to show you a demo of how Power BI is really helping integrate with a wide variety of other Microsoft tools. So here you can see I've got Excel content pinned up into my dashboard, is a really seamless way to get different Excel types of visualizations.

But I want to drill into this data here and the scenario we're looking at is where a customer or I'm a company that deals with customers asking for quotes. So they come in and say, "Hey, I want to buy something from you. How much is it going to cost?" And we prepare a quote and we have a whole process that we go back to that customer. So a little lighter I think than the traffic fatalities. So we'll we'll look at their customer quote processing. And I've got my report here and I'm going to hit at it. And let's think about how we might help our end users do a better job managing this process. And so I've got my estimated account opportunities by size. I've got some information on how long this process is taking. Maybe the first thing I want to do is just visualize this a little bit better. This is the average end-to-end response time for my different accounts. I want to do that a little bit differently. I'm going to drive that into the color saturation. And I can see Lucerne Publishing the amount of time it's taking is green, it's pretty good but this Parnell Aerospace is red, it's taking a little bit longer.

And up here I've got this table and this is all of my process steps. So this is the process we go through with the customer. And it turns out a table is not a very good way to visualize a process. Fortunately, at Microsoft we have one of the greatest tools available for visualizing a process which is Visio, right? So I can lay out each step and understand that and how great would it be if we could just integrate this directly into the report my end users is going to play with and see so they can understand their data. So I'm just going to take the link to this report and I'm going to switch this over to the Visio Power BI visualization and drop this in here and hit connect.

>> And just like that, I've gone ahead and see that visio diagram integrated in Power BI, I'll make it a little bit bigger. And you see these colors kind of changed, that's because we automatically looked at the different fields and bound it to the data because the names matched up, and wherever the names matched up, we automatically bound it to the data. This is a really cool visual because I've got the full flexibility to say change the KPI colors, maybe instead of blue for this threshold, we'll make it yellow, and I can see that color automatically change. And this sales queue process step, it didn't get bound to anything. Maybe we're going to say, "Hey, that leads and the opportunities." And I can even if I, I could do all the matching here, right in the tool so I can line up that data to integrate it together.

And again, the great thing about being integrated into Power BI is that cross highlighting just comes naturally, so maybe I click on Parnell Aerospace, and I see, really visually now my problems apparently are in the sales queue, and telesales, in contract approval, that's what that one says right there. And so I can see very visually, see where my process is having problems. And again, a lot of power when we integrate things together, and we think again back to the theme of data literacy and helping people understand data. And so now I know I've got a problem and traditionally, what do I do, I call somebody, I go search for some app. How do I actually make a change? Well, this is another opportunity where we can integrate things together and really streamline processes with other tools that you love. It turns out, we have a power app that lets me manage my customers and the different process flows that are going to go through for customers. If you're not familiar with power apps, it's a codeless way to build apps that I can distribute on phones or in the web, and now, I can coming very soon deploy these right in my Power BI apps.

So I could come here in this app on the web, and I could drill down, and I can play with these fields. But again, it would be a lot more powerful if I could just integrate this directly into my Power BI report. So, I'm going to grab the app ID. I'm going to come back here. And I'm going to add a power apps visual. And all I have to do is drop my customer ID. Drop that in, hit submit. And just with a couple of clicks, I've now integrated this app right into my report experience that I could send to my users. And again, I can deploy, and this is a feature that would work in our cloud service. So let's let the app get loaded up for the first time, it takes a couple of seconds. And now, I can see Parnell Aerospace, and again I could go to the Lucerne Publishing where things are doing pretty well, and I can see that the lead screen is prioritized for the sales team, and it's eligible for automatic approval.

So now I can come down to parnell aerospace, and maybe I want to toggle that, set this to prioritize for field sales team and hit submit. And now, it's using power apps to go to right back trigger a workflow. And if you're not familiar with power apps in Microsoft Flow, it's an amazing orchestration engine that lets you really connect with your users through integrating app experiences. And you think about what we did here. I really connected insight into action, so I found an insight about a customer really quickly and seamlessly let people take action in this report. And again, something very easy to deploy with Power BI. But it's more than just integrating with other Microsoft tools. We also are really focused on integrating with any tool that can bring interesting rich data and visualization to your customers, to our customers and to your internal users. And so, one scenario that comes up quite a bit is geospatial. We have a wonderful partnership with Esri, which is the leader in geospatial analysis.

And so, let's keep going with the whole sales scenario. Now I have leads, and I want to understand my leads, and maybe do some geospatial analysis across the different leads that I have. So I'm going to take this table, and I'm going to switch it to the Esri visualization that's built into Power BI. So this is available for everybody in Power BI Desktop, and it gives you an incredible set of rich geospatial analysis capabilities. So, here I can see my leads plotted, they're all around San Francisco, but let's look at some of the things I can do with this Esri visual here. So I'm going go ahead and edit it. And Esri has an amazing set of capabilities, one, just to visualize things differently, maybe I want to have a little bit of a darker background as a base map.

Maybe I want to change the symbol style here because these are all just single points, and sometimes circle the size, so maybe I'm gonna change that to just the single data point. I've got a full set of control for colors and outlines and types. Maybe I want to add a reference layer, which is a pretty cool capability, so I can come in and Esri provides, in this case, out of the box. The U.S. demographic data that I can go ahead and overlay, so maybe I want to put disposable income, which is an interesting metric. When I'm thinking about my leads, and maybe where I want to go focus on doing analysis. And so, I have disposable income overlaid on the map, so darker colors more disposable income. And I can keep going further, it's just amazing capabilities built in here. Maybe I want to drop a pin on the map, and do some geo fencing and other type of analysis. So let's say, I've got a cluster of leads here right around Oakland. Let's see what would happen if I focus myself right in the center of Oakland, California.

So I dropped a pin here, right. And now let's do a drive time. So let's select this pin, and say, what's everything within, say 15 minutes drive time of downtown Oakland? Right. So again, right into this visual, I've got this drive time map right, it's pretty cool huh? So if I come here and go back to my report. Now, this is again, one of the things we really emphasize is integrated, flexible, and always interactive. So I can come here and I want to be able to select my drive time area, and I can just like this, click this drive time area, and now I'm seeing all of the leads that I have within this drive time area. So I can be dropping pins and doing other geo fencing.

And again, this is all just integrated with our partner Esri, right in the Power BI Desktop. And one other reason I wanted to show you this, is that we have a really cool announcement with Esri, is that coming this year is the ability to buy a plus offering from Esri right inside of Power BI. So you get additional base maps, more geocoding information for worldwide capabilities, and really extend these set of capabilities you get in Esri. And it's amazing example of how Power BI is not only developing all of these amazing capabilities ourselves but also partnering actively with everybody in our ecosystem to bring amazing capabilities in and really offer a full offering. So what I wanted to make sure that you took away from these demos here was that, we've made it really, really easy for you to take rich dashboards, and reports, and interactive capabilities, and deploy them how you want, whether it's on premise, whether it's in the cloud, whether it's embedded into your own internal applications.




And we really streamline that with a full management set of capabilities. We're really excited to see what you guys can do with Power BI. And thank you very much for your time and I'll let Kamal finish off here. Thank you. >> All right. Thanks again, Patrick. That was good. And so, you saw again, Patrick, show us some really interesting and amazing capabilities that can be done. And we talked a lot about how to dig Power BI and deliver these kinds of capabilities at scale inside the organizations. And you saw what's possible with Power BI Premium to enable this at scale, embedding it inside things like teams, Sharepoint, et cetera.

And if you look back in time, and even back in the media announcements for Power BI back in May, we also have talked about a unified API surface area for Power BI. And today, I'm really excited to announce that API surface area is available now starting in October in Azure. And so, if you are an ISV, someone's building a solution, you'll be able to go in Azure along with all of the great things you can do there, use this API to deliver even more amazing solutions out to your customers. But that, wrapping up, as I finish off I have one ask a few, I hope you learned something interesting today in terms of AI transparently inside Power BI, and be able to go and deliver this inside your organizations.

And as you go back into your, back to your day jobs after the conference, think about how you can take the things you saw today, how can you drive that digital transformation using this data literacy that we talked about, and really make it possible to take Power BI, the AI and the ML that's fused into it and make a difference as you go back. Thank you very much for your time.

Thank you for joining me. Hope you have a great conference. .



BI in the age of artificial intelligence, IGNT17-GS08