>>welcome back. And thanks for joining us for our second session. External data, your new leading indicators. We'll be hearing from industry leaders as they share best practices and challenges in leveraging external data. This panel will be a true conversation on the part of the possible. All right, let's get to >>it >>today. We're excited to be joined by thought spots. Chief Data Strategy Officer Cindy Housing Deloitte's chief data officer Manteo, the founder and CEO of Eagle Alfa. And it Kilduff and Snowflakes, VP of data marketplace and customer product strategy. Matt Glickman. Cindy. Without further ado, the floor is yours. >>Thank you, Mallory. And I am thrilled to have this brilliant team joining us from around the world. And they really bring each a very unique perspective. So I'm going to start from further away. Emmett, Welcome. Where you joining us from? >>Thanks for having us, Cindy. I'm joining from Dublin, Ireland, >>great. And and tell us a little bit about Eagle Alfa. What do you dio >>from a company's perspective? Think of Eagle Alfa as an aggregator off all the external data sets on a word I'll use a few times. Today is a big advantage we could bring companies is we have a data concierge service. There's so much data we can help identify the right data sets depending on the specific needs of the company. >>Yeah. And so, Emma, you know, people think I was a little I kind of shocked the industry. Going from gardener to a tech startup. Um, you have had a brave journey as well, Going from financial services to starting this company, really pioneering it with I think the most data sets of any of thes is that right? >>Yes, it was. It was a big jump to go from Morgan Stanley. Uh, leave the comforts of that environment Thio, PowerPoint deck and myself raising funding eight years ago s So it was a big jump on. We were very early in our market. It's in the last few years where there's been real momentum and adoption by various types of verticals. The hedge funds were first, maybe then private equity, but corporate sar are following quite quickly from behind. That will be the biggest users, in our view, by by a significant distance. >>Yeah, great. Thank um, it So we're going to go a little farther a field now, but back to the U. S. So, Juan, where you joining us from? >>Hey, Cindy. Thanks for having me. I'm joining you from Houston, Texas. >>Great. Used to be my home. Yeah, probably see Rice University back there. And you have a distinct perspective serving both Deloitte customers externally, but also internally. Can you tell us about that? >>Yeah, absolutely. So I serve as the Lord consultants, chief data officer, and as a professional service firm, I have the responsibility for overseeing our overall data agenda, which includes both the way we use data and insights to run and operate our own business, but also in how we develop data and insights services that we then take to market and how we serve our dealers and clients. >>Great. Thank you, Juan. And last but not least, Matt Glickman. Kind of in my own backyard in New York. Right, Matt? >>Correct. Joining I haven't been into the city and many months, but yes, um, based in New York. >>Okay. Great. And so, Matt, you and Emmett also, you know, brave pioneers in this space, and I'm remembering a conversation you and I shared when you were still a J. P. Morgan, I believe. And you're Goldman Sachs. Sorry. Sorry. Goldman. Can you Can you share that with us? >>Sure. I made the move back in 2015. Um, when everyone thought, you know, my wife, my wife included that I was crazy. I don't know if I would call it Comfortable was emitted, but particularly had been there for a long time on git suffered in some ways. A lot of the pains we're talking about today, given the number of data, says that the amount of of new data sets that are always demand for having run analytics teams at Goldman, seeing the pain and realizing that this pain was not unique to Goldman Sachs, it was being replicated everywhere across the industry, um, in a mind boggling way and and the fortuitous, um, luck to have one of snowflakes. Founders come to pitch snowflake to Goldman a little bit early. Um, they became a customer later, but a little bit early in 2014. And, you know, I realized that this was clearly, you know, the answer from first principles on bond. If I ever was going to leave, this was a problem. I was acutely aware of. And I also was aware of how much the man that was in financial services for a better solution and how the cloud could really solve this problem in particular the ability to not have to move data in and out of these organizations. And this was something that I saw the future of. Thank you, Andi, that this was, you know, sort of the pain that people just expected to pay. Um, this price if you need a data, there was method you had thio. You had to use you either ftp data in and out. You had data that was being, you know, dropped off and, you know, maybe in in in a new ways and cloud buckets or a P i s You have to suck all this data down and reconstruct it. And God forbid the formats change. It was, you know, a nightmare. And then having issues with data, you had a what you were seeing internally. You look nothing like what the data vendors were seeing because they want a completely different system, maybe model completely differently. Um, but this was just the way things were. Everyone had firewalls. Everyone had their own data centers. There was no other way on git was super costly. And you know this. I won't even share the the details of you know, the errors that would occur in the pain that would come from that, Um what I realized it was confirmed. What I saw it snowflake at the time was once everyone moves to run their actual workloads in this in the cloud right where you're now beyond your firewall, you'll have all this scale. But on top of that, you'll be able to point at data from these vendors were not there the traditional data vendors. Or, you know, this new wave of alternative data vendors, for example, like the ones that eagle out for brings together And bring these all these data sets together with your own internal data without moving it. Yeah, this was a fundamental shift of what you know, it's in some ways, it was a side effect of everyone moving to the cloud for costs and scale and elasticity. But as a side effect of that is what we talked about, You know it snowflake summit, you know, yesterday was this notion of a data cloud that would connect data between regions between cloud vendors between customers in a way where you could now reference data. Just like your reference websites today, I don't download CNN dot com. I point at it, and it points me to something else. I'm always seeing the latest version, obviously, and we can, you know, all collaborate on what I'm seeing on that website. That's the same thing that now can happen with data. So And I saw this as what was possible, and I distinctly asked the question, you know, the CEO of the time Is this possible? And not only was it possible it was a fundamental construct that was built into the way that snowflake was delivered. And then, lastly, this is what we learned. And I think this is what you know. M It also has been touting is that it's all great if data is out there and even if you lower that bar of access where data doesn't have to move, how do I know? Right? If I'm back to sitting at Goldman Sachs, how do I know what data is available to me now in this this you know, connected data network eso we released our data marketplace, which was a very different kind of marketplace than these of the past. Where for us, it was really like a global catalog that would elect a consumer data consumer. Noah data was available, but also level the playing field. Now we're now, you know, Eagle, Alfa, or even, you know, a new alternative data vendor build something in their in their basement can now publish that data set so that the world could see and consume and be aligned to, you know, snowflakes, core business, and not where we wouldn't have to be competing or having to take, um, any kind of custody of that data. So adding that catalog to this now ubiquitous access, um really changed the game and, you know, and then now I seem like a genius for making this move. But back then, like I said, we've seen I seem like instant. I was insane. >>Well, given, given that snowflake was the hottest aipo like ever, you were a genius. Uh, doing this, you know, six years in advance. E think we all agree on that, But, you know, a lot of this is still visionary. Um, you know, some of the most leading companies are already doing this. But one What? What is your take our Are you best in class customers still moving the data? Or is this like they're at least thinking about data monetization? What are you seeing from your perspective? >>Yeah, I mean, I did you know, the overall appreciation and understanding of you know, one. I got to get my house in order around my data, um, has something that has been, you know, understood and acted upon. Andi, I do agree that there is a shift now that says, you know, data silos alone aren't necessarily gonna bring me, you know, new and unique insights on dso enriching that with external third party data is absolutely, you know, sort of the the ship that we're seeing our customers undergo. Um, what I find extremely interesting in this space and what some of the most mature clients are doing is, you know, really taking advantage of these data marketplaces. But building data partnerships right there from what mutually exclusive, where there is a win win scenario for for you know, that organization and that could be, you know, retail customers or life science customers like with pandemic, right the way we saw companies that weren't naturally sharing information are now building these data partnership right that are going are going into mutually benefit, you know, all organizations that are sort of part of that value to Andi. I think that's the sort of really important criteria. And how we're seeing our clients that are extremely successful at this is that partnership has benefits on both sides of that equation, right? Both the data provider and then the consumer of that. And there has to be, you know, some way to ensure that both parties are are are learning right, gaining you insights to support, you know, whatever their business organization going on. >>Yeah, great one. So those data partnerships getting across the full value chain of sharing data and analytics Emmett, you work on both sides of the equation here, helping companies. Let's say let's say data providers maybe, like, you know, cast with human mobility monetize that. But then also people that are new to it. Where you seeing the top use cases? Well, >>interestingly, I agree with one of the supply side. One of the interesting trends is we're seeing a lot more data coming from large Corporates. Whether they're listed are private equity backed, as opposed to maybe data startups that are earning money just through data monetization. I think that's a great trend. I think that means a lot of the best. Data said it data is yet to come, um, in terms off the tough economy and how that's changed. I think the category that's had the most momentum and your references is Geo location data. It's that was the category at our conference in December 2000 and 12 that was pipped as the category to watch in 2019. On it didn't become that at all. Um, there were some regulatory concerns for certain types of geo data, but with with covert 19, it's Bean absolutely critical for governments, ministries of finance, central banks, municipalities, Thio crunch that data to understand what's happening in a real time basis. But from a company perspective, it's obviously critical as well. In terms of planning when customers might be back in the High Street on DSO, fourth traditionally consumer transaction data of all the 26 categories in our taxonomy has been the most popular. But Geo is definitely catching up your slide. Talked about being a tough economy. Just one point to contradict that for certain pockets of our clients, e commerce companies are having a field day, obviously, on they are very data driven and tech literate on day are they are really good client base for us because they're incredibly hungry, firm or data to help drive various, uh, decision making. >>Yeah, So fair enough. Some sectors of the economy e commerce, electron, ICS, healthcare are doing great. Others travel, hospitality, Um, super challenging. So I like your quote. The best is yet to come, >>but >>that's data sets is yet to come. And I do think the cloud is enabling that because we could get rid of some of the messy manual data flows that Matt you talked about, but nonetheless, Still, one of the hardest things is the data map. Things combining internal and external >>when >>you might not even have good master data. Common keys on your internal data. So any advice for this? Anyone who wants to take that? >>Sure I can. I can I can start. That's okay. I do think you know, one of the first problems is just a cataloging of the information that's out there. Um, you know, at least within our organization. When I took on this role, we were, you know, a large buyer of third party data. But our organization as a whole didn't necessarily have full visibility into what was being bought and for what purpose. And so having a catalog that helps us internally navigate what data we have and how we're gonna use it was sort of step number one. Um, so I think that's absolutely important. Um, I would say if we could go from having that catalog, you know, created manually to more automated to me, that's sort of the next step in our evolution, because everyone is saying right, the ongoing, uh, you know, creation of new external data sets. It's only going to get richer on DSO. We wanna be able to take advantage of that, you know, at the at the pacing speed, that data is being created. So going from Emanuel catalog to anonymous >>data >>catalog, I think, is a key capability for us. But then you know, to your second point, Cindy is how doe I then connect that to our own internal data to drive greater greater insights and how we run our business or how we serve our customers. Andi, that one you know really is a It's a tricky is a tricky, uh, question because I think it just depends on what data we're looking toe leverage. You know, we have this concept just around. Not not all data is created equal. And when you think about governance and you think about the management of your master data, your internal nomenclature on how you define and run your business, you know that that entire ecosystem begins to get extremely massive and it gets very broad and very deep on DSO for us. You know, government and master data management is absolutely important. But we took a very sort of prioritized approach on which domains do we really need to get right that drive the greatest results for our organization on dso mapping those domains like client data or employee data to these external third party data sources across this catalog was really the the unlocked for us versus trying to create this, you know, massive connection between all the external data that we're, uh, leveraging as well as all of our own internal data eso for us. I think it was very. It was a very tailored, prioritized approach to connecting internal data to external data based on the domains that matter most to our business. >>So if the domains so customer important domain and maybe that's looking at things, um, you know, whether it's social media data or customer transactions, you prioritized first by that, Is that right? >>That's correct. That's correct. >>And so, then, Matt, I'm going to throw it back to you because snowflake is in a unique position. You actually get to see what are the most popular data sets is is that playing out what one described are you seeing that play out? >>I I'd say Watch this space. Like like you said. I mean this. We've you know, I think we start with the data club. We solve that that movement problem, which I think was really the barrier that you tended to not even have a chance to focus on this mapping problem. Um, this notion of concordance, I think this is where I see the big next momentum in this space is going to be a flurry of traditional and new startups who deliver this concordance or knowledge graph as a service where this is no longer a problem that I have to solve internal to my organization. The notion of mastering which is again when everyone has to do in every organization like they used to have to do with moving data into the organization goes away. And this becomes like, I find the best of breed for the different scopes of data that I have. And it's delivered to me as a, you know, as a cloud service that just takes my data. My internal data maps it to these 2nd and 3rd party data sets. Um, all delivered to me, you know, a service. >>Yeah, well, that would be brilliant concordance as a service or or clean clean master data as a service. Um, using augmented data prep would be brilliant. So let's hope we get there. Um, you know, so 2020 has been a wild ride for everyone. If I could ask each of you imagine what is the art of the possible or looking ahead to the next to your and that you are you already mentioned the best is yet to come. Can you want to drill down on that. What what part of the best is yet to come or what is your already two possible? >>Just just a brief comment on mapping. Just this week we published a white paper on mapping, which is available for for anyone on eagle alfa dot com. It's It's a massive challenge. It's very difficult to solve. Just with technology Onda people have tried to solve it and get a certain level of accuracy, but can't get to 100% which which, which, which makes it difficult to solve it. If if if there is a new service coming out against 100% I'm all ears and that there will be a massive step forward for the entire data industry, even if it comes in a few years time, let alone next year, I think going back to the comment on data Cindy. Yes, I think boards of companies are Mawr and Mawr. Viewing data as an asset as opposed to an expense are a cost center on bond. They are looking therefore to get their internal house in order, as one was saying, but also monetize the data they are sitting on lots of companies. They're sitting on potentially valuable data. It's not all valuable on a lot of cases. They think it's worth a lot more than it is being frank. But in some cases there is valuable data on bond. If monetized, it can drop to the bottom line on. So I think that bodes well right across the world. A lot of the best date is yet to come on. I think a lot of firms like Deloitte are very well positioned to help drive that adoption because they are the trusted advisor to a lot of these Corporates. Um, so that's one thing. I think, from a company perspective. It's still we're still at the first base. It's quite frustrating how slow a lot of companies are to move and adopt, and some of them are haven't hired CDO. Some of them don't have their internal house in order. I think that has to change next year. I think if we have this conference at this time next year, I would expect that would hopefully be close to the tipping point for Corporates to use external data. And the Malcolm Gladwell tipping point on the final point I make is I think, that will hopefully start to see multi department use as opposed to silos again. Parliaments and silos, hopefully will be more coordinated on the company's side. Data could be used by marketing by sales by r and D by strategy by finance holds external data. So it really, hopefully will be coordinated by this time next year. >>Yeah, Thank you. So, to your point, there recently was an article to about one of the airlines that their data actually has more value than the company itself now. So I know, I know. We're counting on, you know, integrators trusted advisers like Deloitte to help us get there. Uh, one what? What do you think? And if I can also drill down, you know, financial services was early toe all of this because they needed the early signals. And and we talk about, you know, is is external data now more valuable than internal? Because we need those early signals in just such a different economy. >>Yeah, I think you know, for me, it's it's the seamless integration of all these external data sources and and the signals that organizations need and how to bring those into, you know, the day to day operations of your organization, right? So how do you bring those into, You know, you're planning process. How do you bring that into your sales process on DSO? I think for me success or or where I see the that the use and adoption of this is it's got to get down to that level off of operations for organizations. For this to continue to move at the pace and deliver the value that you know, we're all describing. I think we're going to get there. But I think until organizations truly get down to that level of operations and how they're using this data, it'll sort of seem like a Bolton, right? So for me, I think it's all about Mawr, the seamless integration. And I think to what Matt mentioned just around services that could help connect external data with internal data. I'll take that one step beyond and say, How can we have the data connect itself? Eso I had references Thio, you know, automation and machine learning. Um, there's significant advances in terms of how we're seeing, you know, mapping to occur in a auto generated fashion. I think this specific space and again the connection between external and internal data is a prime example of where we need to disrupt that, you know, sort of traditional data pipeline on. Try to automate that as much as possible. And let's have the data, you know, connect itself because it then sort of supports. You know, the first concept which waas How do we make it more seamless and integrated into, you know, the business processes of the organization's >>Yeah, great ones. So you two are thinking those automated, more intelligent data pipelines will get us there faster. Matt, you already gave us one. Great, Uh, look ahead, Any more to add to >>it, I'll give you I'll give you two more. One is a bit controversial, but I'll throw that you anyway, um, going back to the point that one made about data partnerships What you were saying Cindy about, you know, the value. These companies, you know, tends to be somehow sometimes more about the data they have than the actual service they provide. I predict you're going to see a wave of mergers and acquisitions. Um, that it's solely about locking down access to data as opposed to having data open up. Um to the broader, you know, economy, if I can, whether that be a retailer or, you know, insurance company was thes prime data assets. Um, you know, they could try to monetize that themselves, But if someone could acquire them and get exclusive access that data, I think that's going to be a wave of, um, in a that is gonna be like, Well, we bought this for this amount of money because of their data assets s. So I think that's gonna be a big wave. And it'll be maybe under the guise of data partnerships. But it really be about, you know, get locking down exclusive access to valuable data as opposed to trying toe monetize it itself number one. And then lastly, you know. Now, did you have this kind of ubiquity of data in this interconnected data network? Well, we're starting to see, and I think going to see a big wave of is hyper personalization of applications where instead of having the application have the data itself Have me Matt at Snowflake. Bring my data graph to applications. Right? This decoupling of we always talk about how you get data out of these applications. It's sort of the reverse was saying Now I want to bring all of my data access that I have 1st, 2nd and 3rd party into my application. Instead of having to think about getting all the data out of these applications, I think about it how when you you know, using a workout app in the consumer space, right? I can connect my Spotify or connect my apple music into that app to personalize the experience and bring my music list to that. Imagine if I could do that, you know, in a in a CRM. Imagine I could do that in a risk management. Imagine I could do that in a marketing app where I can bring my entire data graph with me and personalize that experience for, you know, for given what I have. And I think again, you know, partners like thoughts. But I think in a unique position to help enable that capability, you know, for this next wave of of applications that really take advantage of this decoupling of data. But having data flow into the app tied to me as opposed to having the APP have to know about my data ahead of time, >>Yeah, yeah, So that is very forward thinking. So I'll end with a prediction and a best practice. I am predicting that the organizations that really leverage external data, new data sources, not just whether or what have you and modernize those data flows will outperform the organizations that don't. And as a best practice to getting there, I the CDOs that own this have at least visibility into everything they're purchasing can save millions of dollars in duplicate spend. So, Thio, get their three key takeaways. Identify the leading indicators and market signals The data you need Thio. Better identify that. Consolidate those purchases and please explore the data sets the range of data sets data providers that we have on the thought spot. Atlas Marketplace Mallory over to you. >>Wow. Thank you. That was incredible. Thank you. To all of our Panelists for being here and sharing that wisdom. We really appreciate it. For those of you at home, stay close by. Our third session is coming right up and we'll be joined by our partner AWS and get to see how you can leverage the full power of your data cloud complete with the demo. Make sure to tune in to see you >>then

>>Yeah, >>yeah. Hello and welcome to Day two at Beyond. I am so excited that you've chosen to join the building a vibrant data ecosystem track. I might be just a little bit biased, but I think it's going to be the best track of the day. My name is Mallory Lassen and I run partner Marketing here, a thought spot, and that might give you a little bit of a clue as to why I'm so excited about the four sessions we're about to hear from. We'll start off hearing from two thought spotters on how the power of embrace can allow you to directly query on the cloud data warehouse of your choice Next up. And I shouldn't choose favorites, but I'm very excited to watch Cindy housing moderate a panel off true industry experts. We'll hear from Deloitte Snowflake and Eagle Alfa as they describe how you can enrich your organization's data and better understand and benchmark by using third party data. They may even close off with a prediction or two about the future that could prove to be pretty thought provoking. So I'd stick around for that. Next we'll hear from the cloud juggernaut themselves AWS. We'll even get to see a live demo using TV show data, which I'm pretty sure is near and dear to our hearts. At this point in time and then last, I'm very excited to welcome our customer from T Mobile. They're going to describe how they partnered with whip pro and developed a full solution, really modernizing their analytics and giving self service to so many employees. We'll see what that's done for them. But first, let's go over to James Bell Z and Ana Son on the zero to search session. James, take us away. >>Thanks, Mallory. I'm James Bell C and I look after the solutions engineering and customer success teams have thought spot here in Asia Pacific and Japan today I'm joined by my colleague Anderson to give you a look at just how simple and quick it is to connect thought spot to your cloud data warehouse and extract value from the data within in the demonstration, and I will show you just how we can connect to data, make it simple for the business to search and then search the data itself or within this short session. And I want to point out that everything you're going to see in the demo is Run Live against the Cloud Data Warehouse. In this case, we're using snowflake, and there's no cashing of data or summary tables in terms of what you're going to see. But >>before we >>jump into the demo itself, I just like to provide a very brief overview of the value proposition for thought spot. If you're already familiar with thought spot, this will come as no surprise. But for those new to the platform, it's all about empowering the business to answer their own questions about data in the most simple way possible Through search, the personalized user experience provides a familiar search based way for anyone to get answers to their questions about data, not just the analysts. The search, indexing and ranking makes it easy to find the data you're looking for using business terms that you understand. While the smart ranking constantly adjust the index to ensure the most relevant information is provided to you. The query engine removes the complexity of SQL and complex joint paths while ensuring that users will always get thio the correct answers their questions. This is all backed up by an architecture that's designed to be consumed entirely through a browser with flexibility on deployment methods. You can run thought spot through our thoughts about cloud offering in your own cloud or on premise. The choice is yours, so I'm sure you're thinking that all sounds great. But how difficult is it to get this working? Well, I'm happy to tell you it's super easy. There's just forced steps to unlock the value of your data stored in snowflake, Red Shift, Google, Big Query or any of the other cloud data warehouses that we support. It's a simple is connecting to the Cloud Data Warehouse, choosing what data you want to make available in thought spot, making it user friendly. That column that's called cussed underscore name in the database is great for data management, but when users they're searching for it, they'll probably want to use customer or customer name or account or even client. Also, the business shouldn't need to know that they need to get data from multiple tables or the joint parts needed to get the correct results in thought spot. The worksheet allows you to make all of this simple for the users so they can simply concentrate on getting answers to their questions on Once the worksheet is ready, you can start asking those questions by now. I'm sure you're itching to see this in action. So without further ado, I'm gonna hand over to Anna to show you exactly how this works over to you. Anna, >>In this demo, I'm going to go to cover three areas. First, we'll start with how simple it is to get answers to your questions in class spot. Then we'll have a look at how to create a new connection to Cloud Data Warehouse. And lastly, how to create a use of friendly data layer. Let's get started to get started. I'm going to show you the ease off search with thoughts Spot. As you can see thought spot is or were based. I'm simply lobbying. Divide a browser. This means you don't need to install an application. Additionally, possible does not require you to move any data. So all your data stays in your cloud data warehouse and doesn't need to be moved around. Those sports called differentiator is used experience, and that is primarily search. As soon as we come into the search bar here, that's what suggestion is guiding uses through to the answers? Let's let's say that I would wanna have a look at spending across the different product categories, and we want Thio. Look at that for the last 12 months, and we also want to focus on a trending on monthly. And just like that, we get our answer straightaway without alive from Snowflake. Now let's say we want to focus on 11 product category here. We want to have a look at the performance for finished goods. As I started partially typing my search them here, Thoughts was already suggesting the data value that's available for me to use as a filter. The indexing behind the scene actually index everything about the data which allowed me to get to my data easily and quickly as an end user. Now I've got my next to my data answer here. I can also go to the next level of detail in here. In third spot to navigate on the next level of detail is simply one click away. There's no concept off drill path, pre defined drill path in here. That means we've ordered data that's available to me from Snowflake. I'm able to navigate to the level of detail. Allow me to answer those questions. As you can see as a business user, I don't need to do any coding. There's no dragon drop to get to the answer that I need right here. And she can see other calculations are done on the fly. There is no summary tables, no cubes building are simply able to ask the questions. Follow my train or thoughts, and this provides a better use experience for users as anybody can search in here, the more we interact with the spot, the more it learns about my search patterns and make those suggestions based on the ranking in here and that a returns on the fly from Snowflake. Now you've seen example of a search. Let's go ahead and have a look at How do we create a connection? Brand new one toe a cloud at a warehouse. Here we are here, let me add a new connection to the data were healthy by just clicking at new connection. Today we're going to connect Thio retail apparel data step. So let's start with the name. As you can see, we can easily connect to all the popular data warehouse easily. By just one single click here today, we're going to click to Snowflake. I'm gonna ask some detail he'd let me connect to my account here. Then we quickly enter those details here, and this would determine what data is available to me. I can go ahead and specify database to connect to as well, but I want to connect to all the tables and view. So let's go ahead and create a connection. Now the two systems are talking to each other. I can see all the data that's available available for me to connect to. Let's go ahead and connect to the starter apparel data source here and expanding that I can see all the data tables as available to me. I could go ahead and click on any table here, so there's affect herbal containing all the cells information. I also have the store and product information here I can make. I can choose any Data column that I want to include in my search. Available in soft spot, what can go ahead and select entire table, including all the data columns. I will. I would like to point out that this is important because if any given table that you have contains hundreds of columns it it may not be necessary for you to bring across all of those data columns, so thoughts would allow you to select what's relevant for your analysis. Now that's selected all the tables. Let's go ahead and create a connection. Now force what confirms the data columns that we have selected and start to read the medic metadata from Snowflake and automatically building that search index behind the scene. Now, if your daughter does contain information such as personal, identifiable information, then you can choose to turn those investing off. So none of that would be, um, on a hot spots platform. Now that my tables are ready here, I can actually go ahead and search straight away. Let's go ahead and have a look at the table here. I'm going to click on the fact table heat on the left hand side. It shows all the data column that we've brought across from Snowflake as well as the metadata that also brought over here as well. A preview off the data shows me off the data that's available on my snowflake platform. Let's take a look at the joints tap here. The joint step shows may relationship that has already been defined the foreign and primary care redefining snowflake, and we simply inherited he in fourth spot. However, you don't have toe define all of this relationship in snowflake to add a joint. He is also simple and easy. If I click on at a joint here, I simply select the table that I wanted to create a connection for. So select the fact table on the left, then select the product table onto the right here and then simply selected Data column would wish to join those two tables on Let's select Product ID and clicking next, and that's always required to create a joint between those two tables. But since we already have those strong relationship brought over from Snow Flag, I won't go ahead and do that Now. Now you have seen how the tables have brought over Let's go and have a look at how easy is to search coming to search here. Let's start with selecting the data table would brought over expanding the tables. You can see all the data column that we have previously seen from snowflake that. Let's say I wanna have a look at sales in last year. Let's start to type. And even before I start to type anything in the search bar passport already showing me all those suggestions, guiding me to the answers that's relevant to my need. Let's start with having a look at sales for 2019. And I want to see this across monthly for my trend and out off all of these product line he. I also want to focus on a product line called Jackets as I started partially typing the product line jacket for sport, already proactively recommending me all the matches that it has. So all the data values available for me to search as a filter here, let's go ahead and select jacket. And just like that, I get my answer straight away from Snowflake. Now that's relatively simple. Let's try something a little bit more complex. Let's say I wanna have a look at sales comparing across different regions, um, in us. So I want compare West compared to Southwest, and then I want to combat it against Midwest as well as against based on still and also want to see these trending monthly as well. Let's have look at monthly. If you can see that I can use terms such as monthly Key would like that to look at different times. Buckets. Now all of these is out of the box. As she can see, I didn't have to do any indexing. I didn't have to do any formulas in here. As long as there is a date column in the data set, crossbows able to dynamically calculate those time bucket so she can see. Just by doing that search, I was able to create dynamic groupings segment of different sales across the United States on the sales data here. Now that we've done doing search, you can see that across different tables here might not be the most user friendly layer we don't want uses having to individually select tables. And then, um, you know, selecting different columns with cryptic names in here. We want to make this easy for users, and that's when a work ship comes in. But those were were sheet encapsulate all of the data you want to make available for search as well as formulas, as well as business terminologies that the users are familiar with for a specific business area. Let's start with adding the daughter columns we need for this work shape. Want to slack all of the tables that we just brought across from Snowflake? Expanding each of those tables from the facts type of want sales from the fax table. We want sales as well as the date. Then on the store's table. We want store name as well as the stay eating, then expanding to the product we want name and finally product type. Now that we've got our work shit ready, let's go ahead and save it Now, in order to provide best experience for users to search, would want to optimize the work sheet here. So coming to the worksheet here, you can see the data column that we have selected. Let's start with changing this name to be more user friendly, so let's call it fails record. They will want to call it just simply date, store name, call it store, and then we also want state to be in lower case product name. Simply call it product and finally, product type can also further optimize this worksheet by adding, uh, other areas such as synonyms, so allow users to use terms of familiar with to do that search. So in sales, let's call this revenue and we all cannot also further configure the geo configuration. So want to identify state in here as state for us. And finally, we want Thio. Also add more friendly on a display on a currency. So let's change the currency type. I want to show it in U. S. Dollars. That's all we need. So let's try to change and let's get started on our search now coming back to the search here, Let's go ahead. Now select out worksheet that we have just created. If I don't select any specific tables or worksheets, force what Simply a search across everything that's available to you. Expanding the worksheet. We can see all of the data columns in heat that's we've made available and clicking on search bar for spot already. Reckon, making those recommendations in here to start off? Let's have a look at I wanna have a look at the revenue across different states for here today, so let's use the synonym that we have defined across the different states and we want to see this for here today. Um yesterday as well. I know that I also want to focus on the product line jacket that we have seen before, so let's go ahead and select jacket. Yeah, and just like that, I was able to get the answer straight away in third spot. Let's also share some data label here so we can see exactly the Mount as well to state that police performance across us in here. Now I've got information about the sales of jackets on the state. I want to ask next level question. I want to draw down to the store that has been selling these jackets right Click e. I want to drill down. As you can see out of the box. I didn't have to pre define any drill paths on a target. Reports simply allow me to navigate to the next level of detail to answer my own questions. One Click away. Now I see the same those for the jackets by store from year to date, and this is directly from snowflake data life Not gonna start relatively simple question. Let's go ahead and ask a question that's a little bit more complex. Imagine one. Have a look at Silas this year, and I want to see that by month, month over month or so. I want to see a month. Yeah, and I also want to see that our focus on a sale on the last week off the month. So that's where we see most. Sales comes in the last week off the month, so I want to focus on that as well. Let's focus on last week off each month. And on top of that, I also want to only focus on the top performing stores from last year. So I want to focus on the top five stores from last year, so only store in top five in sales store and for last year. And with that, we also want to focus just on the populist product types as well. So product type. Now, this could be very reasonable question that a business user would like to ask. But behind the scenes, this could be quite complex. But First part takes cares, or the complexity off the data allow the user to focus on the answer they want to get to. If we quickly have a look at the query here, this shows how forceful translate the search that were put in there into queries into that, we can pass on the snowflake. As you can see, the search uses all three tables as well shooting, utilizing the joints and the metadata layer that we have created. Switching over to the sequel here, this sequel actually generate on the fly pass on the snowflake in order for the snowflake to bring back to result and presented in the first spot. I also want to mention that in the latest release Off Hot Spot, we also bringing Embraced um, in the latest version, Off tosspot 6.3 story Q is also coming to embrace. That means one click or two analysis. Those who are in power users to monitor key metrics on kind of anomalies, identify leading indicators and isolate trends, as you can see in a matter of minutes. Using thought spot, we were able to connect to most popular on premise or on cloud data warehouses. We were able to get blazing fast answers to our searches, allow us to transform raw data to incite in the speed off thoughts. Ah, pass it back to you, James. >>Thanks, Anna. Wow, that was awesome. It's incredible to see how much committee achieved in such a short amount of time. I want to close this session by referring to a customer example of who, For those of you in the US, I'm sure you're familiar with who, Lou. But for our international audience, who Lou our immediate streaming service similar to a Netflix or Disney Plus, As you can imagine, the amount of data created by a service like this is massive, with over 32 million subscribers and who were asking questions of over 16 terabytes of data in snow folk. Using regular B I tools on top of this size of data would usually mean using summary or aggregate level data, but with thoughts. What? Who are able to get granular insights into the data, allowing them to understand what they're subscribes of, watching how their campaigns of performing and how their programming is being received, and take advantage of that data to reduce churn and increase revenue. So thank you for your time today. Through the session, you've seen just how simple it is to get thought spot up and running on your cloud data warehouse toe. Unlock the value of your data and minutes. If you're interested in trying this on your own data, you can sign up for a free 14 day trial of thoughts. What cloud? Right now? Thanks again, toe Anna for such awards and demo. And if you have any questions, please feel free to let us know. >>Awesome. Thank you, James and Anna. That was incredible. To see it in action and how it all came together on James. We do actually have a couple of questions in our last few minutes here, Anna. >>The first one will be >>for you. Please. This will be a two part question. One. What Cloud Data Warehouses does embrace support today. And to can we use embrace to connect to multiple data warehouses. Thank you, Mallory. Today embrace supports. Snowflake Google, Big query. Um, Red shift as you assign that Teradata advantage and essay Bahana with more sources to come in the future. And, yes, you can connect on live query from notable data warehouses. Most of our enterprise customers have gotta spread across several data warehouses like just transactional data and red Shift and South will start. It's not like, excellent on James will have the final question go to you, You please. Are there any size restrictions for how much data thought spot can handle? And does one need to optimize their database for performance, for example? Aggregations. >>Yeah, that's a great question. So, you know, as we've just heard from our customer, who there's, there's really no limits in terms of the amount of data that you can bring into thoughts Ponant connect to. We have many customers that have, in excess of 10 terabytes of data that they're connecting to in those cloud data warehouses. And, yeah, there's there's no need to pre aggregate or anything. Thought Spot works best with that transactional level data being able to get right down into the details behind it and surface those answers to the business uses. >>Excellent. Well, thank you both so much. And for everyone at home watching thank you for joining us for that session. You have a few minutes toe. Get up, get some water, get a bite of food. What? You won't want to miss this next panel in it. We have our chief data strategy off Officer Cindy, Housing speaking toe experts in the field from Deloitte Snowflake and Eagle Alfa. All on best practices for leveraging external data sources. See you there