 And here we go. Hello and welcome. My name is Shannon Kemp and I'm the Chief Digital Manager of DataVercity. We'd like to thank you for joining the latest installment of the Monthly DataVercity Webinar series, Advanced Analytics with William McKnight, sponsored today by Looker. Today, William will be discussing trends in enterprise advanced analytics. Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. For questions, we will be collecting them by the Q&A in the bottom right-hand corner of your screen. Or if you'd like to tweet, we encourage you to share highlights or questions via Twitter using hashtag ADVAnalytics. And if you'd like to chat with us or with each other, we certainly encourage you to do so. Just click the chat icon in the bottom middle of your screen for that feature. And if you'd like to continue the conversation after the webinar, you can follow William and each other at community.datavercity.net. As always, we will send a follow-up email within two business days containing links to the slides, the recording of the session, and additional information requested throughout the webinar. Now, let me turn it over to Elena for a brief word from our sponsor, Looker. Elena, hello and welcome. And it looks like you're muted there. Wonderful. Thank you so much. Good morning, everybody. Today, I guess we're good afternoon, depending on where you may be. It's not news to anyone that the market is shifting at breakneck speed. Technologies that promise to revolutionize the way we use data have come and they've gone. And still, you know, we're sitting here thinking about how we can put the data we collect to work for our businesses. The importance of being able to take advantage of new and changing technologies really cannot be overstated, which means that the tools that we use have to be flexible enough to adapt to changes in the market. And our stacks need to be modular enough to let us swap out technologies when they no longer stack up. Our mission has always been to empower people through the smarter use of data. Today, over 2,000 companies partner with Looker to harness the power of their data and empower their people to drive successful business outcomes and deliver better experiences for their customers. This need for better experiences for customers has been the catalyst for digital transformation and really the technologies that support it. And the Looker platform is part of the solution that people are choosing for this. It's because of the ability to deliver timely, actionable insights and a few data into business workflows that's really made this a go-to choice for these companies. Digital transformation is a profound shift in the way companies use technology to deliver value to their customers. It is, without a doubt, the number one priority for technology leaders today. And as you can see in this graph, out of all the areas that drive digital transformation, the number one area of investment is data and analytics, proving just how pertinent this topic is for us today. Forster calls the companies that are capturing this and doing this really well inside driven companies. These are the companies that are using data to drive growth and create differentiating experiences, products and services. The idea is that data is not just something that we analyze, but actually something that can be infused into everyday processes to make employees smarter, give machines better data, and drive new streams of revenue. We're taking the concept of BI effectively and bringing it into the modern era, the era of the data-driven experience. Now, what is a data-driven experience? A data-driven experience can represent any number of ways in which companies drive value from data. That could be internal value by increasing revenue streams or optimizing costs or external economic value by monetizing data and generating new value streams. Looker's platform offers a unified surface to access the truest, most up-to-date version of your company's data. With this unified view into the business, you can choose or design the experience that makes the most sense for what you need. And ultimately, this creates the freedom to send governed data wherever you need it, whenever you need it. So, Looker powers a number of data-driven experiences beginning with the tried-and-true modern BI. Looker's world-class BI experience puts trusted, actionable insights into the hands of decision-makers and helps cultivate a data culture throughout the organization. But, as I said earlier, BI is just one data experience. And in all honesty, it's a small slice of a company's data strategy. We see companies working with data in incredibly innovative ways. They're weaving it into every part of their business in ways that go far beyond reports and dashboards. One of our customers' future play is actually a great example of another type of data-driven experience. They've built a fully automated, AI-powered bid bot that optimizes their ad spend by automatically adjusting bids based on real-time performance. And all of this is being powered by data from Looker. Another one of our customers is feeding both legacy systems and data science and more modern data science workflows with governed data at a massive scale. This company basically modernized the IT services layer for their old systems plus added new AI workflows all from a single platform. We also see leading organizations deliver data-driven experiences by just building a data product to truly capture the immense value of digital transformation now and get ahead of incompetent markets for even starting to see companies take that next step and begin to monetize their data. Ultimately, they're creating new revenue streams from the wealth of data that they're already sitting on. They rapidly build custom data solutions that fits the needs of their customers adjusting as they go to account for changing trends in the market. For example, shifts in 2020. These case studies that I've told you about and that you can see at the top of your screen are perfect examples of how leading companies are delivering data experiences and shifting their focus from traditional reporting tools to platforms that integrate data into business operations and into data products. A modern approach to data and analytics requires a modern technology foundation. Looker was designed to take advantage of evolving trends in data infrastructure. The technology powering looker is based on three main pillars and that's these three concentric circles you see. Our in-database architecture leverages the power of modern cloud MPP databases that are significantly more powerful, far faster and much cheaper than the predecessors. This live connection to the database provides a complete view of your data without having to move any of it into the application layer. Instead, we query the database in real time so data is always accurate and it's always fresh. And what this ultimately does is it changes the ways that people are actually moving data into that database. It changes the way they're doing ETL to focus more on transforming the data at the time of query and less on in-flight transformation into the warehouse. Now, this paradigm is supported by the next concentric circle, our version-controlled semantic modeling layer that acts as a centralized definition for all of your business rules and business logic. Effectively, we're separating business logic from physical data which allows you to reliably apply consistent definitions across KPIs. So both the technical and non-technical can work with trusted data and also allows more flexibility as the data in your warehouse changes and grows. Wrapped around all of this are our APIs for data. These APIs are really the secret sauce that brings it all together because it means that you can give data and send data where it's needed most. These APIs give builders the tools that they need to deliver any experience anywhere, custom apps, proactive rule-based learning, scheduled reports, automated workflows, whatever it is. And all of these capabilities are built on a flexible multi-cloud architecture. We strongly believe that organizations need to modernize at their own pace and be able to control their stack. We have a strategic commitment to multi-cloud approach because we believe it's critical for companies to choose the environment that works the best for them. And that's both today and in the future. So I wanted to talk a little bit more about our architecture and I wanted to call out a few of the key pieces that really drive this ability to keep up with modern data trends. The first thing I want to point out is this bottom layer. This is the idea that Looker is in database and does transformation at the time of query. The key piece of this is that it allows you to maximize the value of the investments that you are already making in very advanced data stack by adding to it and augmenting it as opposed to pulling data out of those systems to try and put it into another space. Additionally, as you are able to upgrade underlying technologies as changes need or as things come and go, that's also possible because we're separating the business logic from the data engineering. The other piece of the stack I wanted to quickly call out is our API for data. The ability to supply governed data at scale across IT services teams means that your data is unlocked to the potential that you want it to have while still being secure in your system. The world of data is changing incredibly quickly and so having a architecture that can support that is incredibly important. And I would like to hand it over to William to talk to us more about the changes in coming up in 2020. Elena, thank you so much for this presentation and thanks to Looker for sponsoring and help make these webinars happen. If you have questions for Elena, feel free to submit them in the Q&A section in the bottom right hand corner of your screen as she will be joining us in the Q&A section at the Q&A portion at the end of the webinar. Now let me introduce to you our series speaker, William McKnight. William is the president of McKnight Consulting Group. McKnight Consulting Group focuses on delivering business value and solving business problems, utilizing proven streamlined approaches and information management. His teams have won several best practice competitions for their implementations. He has been helping companies adopt big data solutions and with that I will give the floor to William to get today's webinar started. Hello and welcome. Hello and thank you, Shannon. Thank you, Elena. That was really good. And I think a lot of the trends are inherent in your presentation. And it's good to see that Looker is doing some of this. You mentioned Elena, you mentioned digital transformation as a big area of investment. I see that as well. I see that as the label for where a lot of organizations have to level up today to be competitive. And really what that comes down to, I like to say, you said it a little differently. I like to say that that's really all about improving your data maturity. And as you mentioned, the number one area of investment in digital transformation is in data. So we're all in the right spot here in our careers. And I thank you all for joining Advanced Analytics here in the new year. Now, a lot of people come up with their trends, their predictions and so on in December, but here at Advanced Analytics, we like to wait until January when you're back to work, when you put the holidays behind you, when you're really starting to think about this stuff. So here we go. Yeah, okay, a little bit about me. I think Shannon introduced me. We do strategy, we do training and we do implementation work and all of the things I'm gonna be talking about. Okay, so what are the trends? And I'm gonna start with why are the trends important so that I get your attention for the rest of the presentation number one, but so that you know why you need to start putting some of these things on the radar. I'm also gonna get to what were my trends from last year and see, well, how did I do? And just give you a window into how far we've come. You know, sometimes it's hard to see how far we've come, how fast we've come when you're focused day to day and you don't ever step back. Well, this is a little bit of a step back kind of presentation, although I hope I'm putting things in play for each and every one of you, especially in terms of planning and doing some of the important things like platforming your future applications and so on. So why are trends important? Now, maybe you've heard me say this before, but I like to say it wherever I can, beyond the mountain is another mountain. That's a Haitian proverb. And what I like to say about this is that as you see the mountains in the graphic here, it's similar to what we're going through right now. There's always another mountain. Now, unlike these mountains, there are not new mountains obviously out there being grown before our eyes, but in technology, there are. And you're never to the end and that's gotta be okay because I talk to a lot of data professionals and they are waiting for this or that implementation to occur and then things will be settled. I don't think so. I think this is an area where it's gonna be unsettled for quite a while. It's gonna get more complicated, even more complicated before it gets less complicated. And there are some signs on the horizon of some things that might uncomplicate our world and I'll get to some of them, but still I think we just gotta be prepared. We gotta be prepared. Now, a trend to me, one of these 10 trends I'm gonna be sharing with you, a trend to me is not something that's here today gone tomorrow. I've discarded those. The trends that I'm gonna be getting into are the ones that you really should think about putting in your roadmap and having a plan for each and every one of them in your company because there's a lot of investment out there from the vendor community, VC community and so on that's gonna go into these trends and the trends that are going to stick around are ones that you wanna be a part of. You wanna at least acknowledge what you're doing in every one of these trends. We're in the business of data. This shows that our information is exploding, it's real time, it's becoming real time all the time and it's not just selective data that's becoming real time. We're bringing analytics to bear at that point where the customer is standing right there at the cash register or on your website, et cetera and making instant decisions and it has to be filled with analytics and to some degree we really gotta work on the backend where the data resides and make that data cultivated to the point where it can be effective in this kind of environment. But it's our information that differentiates us from our competitors, our information quality that impacts everything and our information is used and reused and finally information is the key business asset which is gonna lead to my first trend for you here in about five minutes. Data maturity is highly correlated to business success. Now this isn't the maturity presentation but in my maturity model and other maturity models there's like one, two, three, four, five and what I like to say about this is well, first of all, where are companies? Don't fret, there are five in my sense of work in this area. Companies tend to fall into one of five distinct camps and that's why it's not an even distribution and there's many more ones than there are fives, very few fives out there and I can explain one, two, three, four, five maybe in another form but I found about half or more firms are at this very basic level of data maturity. Data is certainly, certainly an area where you can make investments or make investment and if it's a wise investment it's probably the best place that you can put investment in your company today and by the way, every one of you should be striving to get to level three and I'm not saying four or five some of you need to get four, frankly but everybody needs to be at a level three I think just for competitive purposes because this is digital transformation this is the area of competition today and you can define for yourself what that number three means but I think we kind of all as data professionals inherently know where we stand on this in our shop and if we each will work on our respective areas bringing them up collectively we might have a three and we might be able to sustain that. That's my hope for you. Maturity modeling like this and I'll be quick about this but it should give you a sense of priority what comes next, what's important how do other companies do it what has been their path because every one of you and even if you're a best practice in some area I look at my clients some of them are best practice in some areas nobody's best practice across the board there's always a leader out there that you can be looking to as something and that should give you a sense of priority what's the path? What's the path to that? By the way, you can't skip levels and say okay we're one let's be a five this year you know just it takes a while at least six months I'd say to move a level and that's if you're doing things right maturity levels tend to move in harmony yeah yeah whatever you categorize things I use technology, strategy, architecture and processes and your maturity will tend to move in harmony across those categories you can't be like a one in in governance and then a five in technology just doesn't work that way mid-size and smaller companies I give you a little bit of grace here you might have you might strive to be a two but for most people interested in enterprise data it's a three all need to be at a three momentum is paramount companies who need to be at four are those in highly competitive industries and some of you are in those so what maturity modeling have to do with trends well has a lot to do with trends because the things I'm going to place in your future today will level you up in terms of maturity our goal with this nice little slide here is to raise foundation of our company we must however we're doing things today we cannot still be doing that at least from a technology perspective the same way next year we all need to be leveling up our area of data management so this is the reason we exist as technology professionals so here's some good news for every one of you okay every one of you is the best practice so congratulations but what what year are you a best practice for so that's the caveat to that right okay so if you're a best practice for 1995 that's not so good you're not a best practice today okay hopefully that was just just humorous and taken in the right way but I want you to be a best practice for sometime in the last few years at least if not this year so that's what we're striving to do raise the foundation of our company so that we can get the year closer for what we're best practice for the money tree does not exist speaking for myself 20 years in consulting now 20 plus years I've never been given a budget for leveling up a company's maturity in data it's always for something that the business wants and the trick is to do it in the right way a scalable way an efficient way something that gets them a timely result but also does it right so putting these trends into the picture is something that you are challenged to do and oftentimes the business will come to a tech team and and I'm making a little differentiation here I know some tech teams are quote unquote in the business but what I mean by that is a technology professional wherever they may be okay so the business will come to this technology professional or tech team with a solution and sometimes the method to get there and sometimes that's but sometimes it's not so I want to put you on your toes with these trends so that you can look for the opportunity to introduce these trends into your shop many executives believe that data is an asset and want to be a data driven company but you can't get there without tending to some of these trends I'm going to be talking about so why do we want to do this we're just being asked to do a job right well here's what I found here's what I believe you know from for my data technology professionals and the ones at my clients I believe that it's not simply good enough for a data professional to create user satisfaction that's a part of it that's a big part of our job but we also must level up our area with data maturity and doing data maturity things within our area so that we can get longer term things like a longer term user satisfaction things that drive ultimately business ROI and I believe that for many organizations people like us data professionals that really need to be putting the projects on the table into the mix for the budget and not sitting back and waiting for the business to do that they don't know what we know we are on that precipice of innovation and we are here today on this webinar etc we're getting aware of what the trends are and we're looking for opportunities to introduce them so I would say go beyond looking but actually be proactive and figure out how these things can play in your organization so with all that preamble let's look at last year's trends I was on this webinar about a year ago and these were my trends from last year I don't have as many this year not that I couldn't but I just I guess I got carried away last year but how did I do and more importantly than how did I do how did we all do in terms of the industry and what are some of the things that did happen in our industry that we need to be aware of and where have we come from and I'm going to be quick about this my hope is not to spend a lot of time on last year's trends sensible division of analyte platforms meaning we know that there's a difference between a data warehouse and a data mart and some other analytical database and operational database etc MDM and we're putting things much more sensibly into the right platforms I'll give myself a check mark on that one cloud storage overtakes HDFS I think for sure that happened at least in terms of provisioning for new things multi-cloud becomes the norm okay so far so good the year of master data management 2019 I declared it the year of master data management I think I was a little premature on that I think MDM continued to grow at its pace maybe even excel at an accelerated pace from previous years but not to the point where I may should have made that declaration it continues to grow forward it might pop up in a trend or two here but maybe 2020 will be that year data virtualization being very important providing the enterprise data fabric I think we're all doing it now I think we just can't have all data everywhere and so virtualization over the top has become pretty important whether we have the special tool forward or not is another matter 2019 the year of the graph yes I think I did it on that one it was the year of the graph and it continues to go forward as the well the decade the graph I guess now yeah so much to say about graph stream processing begins to supplant ETL I don't think it's supplanted in ETL I think I misworded this recommendation here because I think for new applications that involve lots of big data yes we are doing that by stream processing and cloud storage etc but I don't I don't see a lot of replacement so far so the words supplant was problematic in that one the newest highest use will be training AI algorithms no it did not it did not become the highest use it's slowly becoming the highest use as a matter of fact I'm going to roll that one forward into 2020 so that's going to be one of my 2020 trends and we're going to roll that one forward data visualization yes it's footprint escalated so did footprints of everything but I will have something more to say about straight up reporting and our love hate with that self-service takes off I didn't I wouldn't say it it took off in reality I would say that it's taken off in terms of planning and in terms of what people want and what people now know they want and know they need in their organization because there's so much for data teams to do that doing the query and the reporting is not part of that charter anymore but it's going to take some time for it to take off in the side of organizations but we're on the right path for that chief data officer going mainstream yes we're doing much more that organizations acknowledge information architect chief analytics officer maybe a B on that one a lot of organizations do acknowledge this now and see the need for technology at a high level whether they're embracing the CIA or the CAO or not is another matter data science pioneers lock in now this is something we're not going to know for a while whether the ones who are embracing data science today are the ones that are going to ultimately succeed it takes some time I believe that's true I believe that 2019 and 20 here right now we're in that sweet spot of claiming claiming some ground claiming some territory around data science for our organization in our industry and now it's the time to do that acknowledgement of the need for data deployments to be near the business unit this has to do with people yes much more much more distributed in organizations reductions challenges poised by internal brist this has to do with as a consultant coming into shops trying to make change getting a lot of internal brist and resistance to change and seeing that kind of fall by the wayside because organizations must have this kind of change in order to survive and I think that is true and LA skills go into the operational environment yes I think that's true I think that's happened and I think that the the line between operational and analytical it's getting really blurry out there getting really blurry and there's a lot of applications that really straddle that sense operational big data platform selection to see more sequel with the rise in what sequel can do less no sequel I think this is true but I'm not I'm certainly not signaling any death toll here for no sequel no sequel has grown tremendously as well it's a very solid way to deploy data in an organization so it's still seeing its heyday but we're going to see more sequel sequel is growing databases are growing and we're seeing even some pull back from cloud storage and things like that into good old databases all right and a new asset bio data didn't quite happen the way that I had hoped for there is a lot of evolution in healthcare we're getting predictions now at the genome level we have some leaders out there companies that are using genomic analysis to learn diseases and provide preventive methods and so on and I think the evolution of AI will continue to improve the quality of treatment and so on and a lot of that will have to do with bio data but we're going to have to keep an eye out for bio data in our data sets of the future okay now let's get to them top trends I got 10 for you first of all data take steps towards the balance sheet let's just start out with a doozy here right this has been talked about sort of I don't know at a low level for quite some time I am certainly not an expert on the balance sheet and et cetera things but I did sit down with my accountants for about an hour and talked about this and ferreted out a few things I want to share with you because I do think that this is coming down the pike and I think we're going to see a major move in this direction this year and here's what he told me there are rules about assets okay and one of them has to do with putting a cost on the asset and this is the big problem this is the big problem with making data an asset for the balance sheet because it's really hard to do or do you put in costs for the hours that you spent collecting the data that doesn't seem to make a lot of sense there's really no accounting model for today and that's the big challenge moving on another one is depreciation the asset must be able to be depreciated it must have a useful life well data retains its usefulness for quite some time and depending upon what you're doing with it it can retain that usefulness for decades and but it does it does tail off after some time so this one is a hurdle but it probably could be could be solved with a with a great model context is another one this has to do with the assets being utilized in a similar manner across different circumstances across different companies and somewhat this is true in that we're all doing reporting analytics etc with our data we're solving compliance issues and so on but there's a lot of things that we do with data it's certainly no one size fits all except at a very high level will that win the day I don't know but the the key to this becoming true is the ability to assign a value to data that's going to be the challenge and we do have some things that have happened like in some acquisitions some M&A acquisition that that we've seen out there value has been placed on data such as the $26 billion acquisition of LinkedIn by Microsoft which was widely accepted that the value of data had a significant role in that business obviously so the prediction here is at least somebody's going to try this this year to boost their value and they may or may not succeed but it will be a major step in the direction of having data sitting on our brown machines all right so the next prediction is about sensor based time series data time series data is taking off and it has a lot to do with internet of things and collecting data at edge points and by the way I'll have another recommendation or another prediction about edge points are coming up but time series data think of it as a bunch of data points that measure pretty much the same thing over and over and over again in time order with the time stamp on it it's that and a lot more but that is the basic way to think about time series data and we are starting to need a lot more time series data to do the things that we want to do with data the data that arrives is recorded as a new entry it's not overwriting data and the data typically arrives in time order in sequential order according to time so time is a very primary access for this kind of data so this is a different kind of data because we are tracking changes to the overall system as insert and update I should say insert because it's not updating it's not deleting it's just continually accumulating and so what I like to say about it is that it really gives you a sense of where the system is where the system is going and how the behavior has changed over time because you're getting data at a very fine grain level like for example if you have a web application every user is going to log in you're going to track all that login activity that's time series data so you're getting a lot of information there about your website and this is true for a host of different very interesting applications out there today like self-driving cars trading algorithms smart homes retail tracking assets and supply chain and so on and tracking all sorts of vehicles so time series databases or I should say databases that have great time series capabilities have steadily remained one of the fastest growing categories of databases so keep an eye out for that business intelligence interfaces are undergoing an upheaval we are getting away from reporting asking a user today in 2020 to go find the data and build reports that just seems really antiquated and I think that things like NLP natural language processing natural language inputs are going to change how we interface with our data tremendously now Bill Gates said earlier this year that natural language inputs and AI voice assistance will improve to the point that they might be able to fill the role of a human secretary so we're all kind of secretaries when it comes to our data trying to find data trying to make sense of it so AI voice assistance yes being more Google-like chat boxes yes that's going to be another way that we're going to interface with data it's just going to change I mean if you think reporting hasn't changed it has for example we're not doing a lot of printing of reports and having somebody with a kind of a grocery cart go around the organization dropping off a print out I hope we're not doing that but you know slowly that has changed and I think that this is sort of the next evolution in that where users will have to be able to ask a question and get an answer and not be concerned about well let's do I need to go to the warehouse of the lake on this one or is that over here or over there we just need to create a better infrastructure for them because this is going to be the demand so we'll also see if if tracking things like your eyeballs against the monitor is going to be something that is interesting but that is something that Elon Musk of all people is working on he has a startup called NuraLink which is one outfit working on brain computer interfaces that use our thoughts as input mechanisms rather than taking the time to type speak or gesture our commands how about that so I'm not saying that going to happen in 2020 but out there is potentially getting beyond even speaking but tapping into our thoughts to give us answers very interesting ETL will be nearly automated this is happening already the data discovery part of it is happening already and auto generating pipelines based on global experiences so you can bring to bear the experience of 10, 20, 100, 200, a thousand different experiences out there that needed that source to target map and did it already and it isn't cloud did it with a provider that has access to all of this and brings that to bear for you the job of the ETL architecture I should say the data integration architecture is drastically changed it's getting much more automated and even the joins across data is becoming more automated all of this is being done with the help of bots so once the data is found it's pretty straightforward from that point we're getting away from manual interfaces manual design the machine can learn from the things that we do and automate that for the future any ETL products that you are considering to be a foundational piece of your future should be doing this leveraging cloud object storage for data lakes yeah okay I talked about cloud object storage being important and overtaking HDFS last year this trend is also bringing the data lake aspect into that and trying to say that data lakes themselves are becoming very important in organizations so we're going to see a lot more of this I still think there's a problem I think the there's a problem still between data lakes and data warehouses and I'm working on it I'm thinking about it I don't have the answer if you want to share some thoughts with me feel free but right now we are collecting a lot of data in the data lake pushing a subset of that onto the data warehouse I see it as very necessary today for most organizations but it's it may be overkill in the future but for 2020 let's build those data lakes our data scientists needed frankly our data warehouses need that kind of a great staging area and the cloud offers many capabilities for our data lakes so more achievable separation of compute and storage compute resources can be taken down scaled up around or interchange without data movement storage can be centralized but compute can be distributed et cetera what you see there yeah some vendors also have remote data replication to ensure redundancy and recovery so all of the great things that we want are finding their way into cloud object storage definitely be utilizing cloud object storage in 2020 more edge ai so here again I'm kind of combining things here more ai for sure but more that's going to happen out on the edge so we're getting away from just having some some file system on our edge devices and we're having real databases out there so now truth be told if we're looking at databases across the world and what databases are employed for embedded databases would be far far beyond what we're doing with non-invented databases there's just so many of them and now that we're growing to a projected 75 billion devices out there I'm not not saying all of them will have an embedded database on them but many many will and the utility of having a database there is the beginning of having artificial intelligence being able to happen right there at the edge and then communicate back with the the mothership of course but a lot of AI is just going to happen right at the edge pushing it pushing it all the way down there so examples of this are like mobile airline applications with online check-in boarding pass retrieval etc yeah these databases will check in with again the mothership maybe a data warehouse something like that but the other point the big point I want to make here is about AI in in these devices alongside the database and that's where you have the power when you have AI and data side by side able to do things in real time with a ton of smarts you have chip startups out there like samba nova grass core cerebrus and cintiaz these are companies that are developing architectures to handle artificial intelligence they're building high-performance AI chips known as neuromorphic or brain chips in case you see that these mimic the structure of the brain and process the top AI algorithms so you get the data together and the AI is coming and this leads to my next prediction for you here data's new highest use will be training AI algorithms I said this last year I mentioned before I'm rolling this forward some I tossed although most of the predictions were good for last year this one's probably good for this year they just knew as highest new highest use will be training AI algorithms now running reports now building dashboards but training AI algorithms I hope I'm not too early with this one once again but this is where AI steps in and what we want to do with AI is know what the behavior is going to be changes if we want to and know that we're changing it to the way that to the behavior that we want from the counterparty so data collection is foundational to this and put that up there data is the foundation of AI data I mean AI is on the data maturity spectrum as far as I'm concerned so you have to have your data maturity to a certain point before AI becomes viable but it's never going to become viable until your data is at that level of maturity and data for AI where's it going to be it's going to be in a great architecture a great data architecture it's going to be in the data lake the aforementioned data lake it's going to be in your data warehouse it's going to be manageable it might be cataloged which is not one of my top 10 trends but it could very well be the 11th the emergence of data catalogs we're cataloging data at a rapid pace we're understanding what data is in there we're doing this for compliance reasons we're doing it for regulatory reasons we're doing it because it's great practice it serves the user community it helps create that self-serve environment that we're also striving for so there's a lot there the hardest part of AI is the data it's cleaning up your data so organizations are going to become algorithm libraries where they're trading algorithms they're trading what works for cataloging they're getting into ML ops and AI is becoming very foundational to their future speaking of AI another one on AI is explainable AI to reduce bias in the algorithms it's going to it's getting to the point where we just can't say to regulatory committees with regulations like GDPR and CCPA while the computer told us that we should take the fashion we have to get in there and know and I don't think this is actually going to slow AI down at all because I think what from what I can tell this is being worked on and this is being done so this is going to help reduce bias in the algorithms and reduction in so-called lazy AI where we don't even really know how the algorithm works we just throw it at the data and trust the results we may not even be trying multiple algorithms and testing the results but anyway transparency is becoming part of compliance regulation transparency remains a hot topic and will continue into the new year as companies aim to ensure transparency, disability and the trust in AI and all AI assisted decisions we're going to see a lot of development and expansion in the so-called explainable AI movement and that will be coming to a shop near you a couple more Kubernetes and containers well this one is already moving in a trend direction but I thought I'd put it out there because I think it's got a ways to go and we'll take off here and really be part of the wallpaper in 2020 where data analytics back will go to Kubernetes for both open source and commercial winners will go from thought to POC quickly and this is really one of the big advantages of Kubernetes and containers is giving rise to the ability to spin up clusters for maybe big data data warehousing and machine learning what have you on a task related basis it's ephemeral it's enabling a mentality of serverlessness and the architecture is the substance of a lot of new platforms like what Cloudera is doing like what Google is doing on their k8s and so on ultimately what this type of rapid prototyping is doing is it's enabling new workloads and new capabilities within organizations so it's going to go forward every single client I have is doing this to some degree or another and I think it's just a matter of time before everything is thought of that way okay my last prediction and by the way as I give you this one if you have any questions for us Elena or myself about looker or about these predictions or anything on your mind in relation to data we're going to take that Q and A here in just a bit but my last prediction slash trend for you is about hybrid databases hybrid databases kind of a kind of a throwback if you think about it back in the early days of databases I was around then the database was the database and it was used wherever you needed a database there wasn't a there wasn't a big distinction between operational and analytical there became one and it became very very defined and truly databases most of them out there today you just don't want to get it wrong I'll put it that way you don't want to be using an analytical database for an operational purpose or vice versa but what about these hybrid databases that are coming on the scene now and saying we're pretty good at both could hybrid databases take us back to that one size fits all do the complexity of the current situation we have to carry multiple skill sets we have to carry multiple vendors and so on operational analytics is really the and here we got some other names for it hybrid transactional analytical processing htap translitical or hybrid operational analytical processing H-O-A-P all right these are some of the terms for this new breed of application that fits that sense straddling position that I talked about earlier when I said that the lines getting kind of blurred between operational analytics today you can't really do operations without analytics for example in a lot of things so operational analytics cannot be done on a data work house it's loaded in batch from the previous day it requires real-time analysis without data movement and so today a real focus and interest of operational analytics includes streaming data in chess and analysis in real-time so this is about real-time that's that's part of it for sure and you've got companies like Splice Machine MemSQL Actian these are databases for those types of applications you may or may not need asset compliance in that check on that but it's also for those shops that are interested in deploying a single database across the enterprise and some do this by deploying two types of data stores so how does this how do these databases get away with it? Well it might have an in-memory row store for their OLTP and a disk-based column store for their OLAP and we all know by now column stores are best for analytics so they differ in both storage format and the storage medium so format being column or versus row medium being RAM versus disk or flash or SSD so you can do joins across the various stores that that may may come out of a hybrid database type of strategy but it's all about real-time so now that you know some more trends and you're going forward as a data professional in your organization you're not off the hook this is the point to bring this type of thinking into some of the new plans maybe have to discard some of them for now but hopefully not all because as I mentioned before trends are important now before I leave you and we move to Q&A there's more maturity in moving in perfectly than in merely perfect perfectly defining the shortcomings a lot of people can perfectly define the shortcomings but what do we do about it? And will it work? Will the plan work to move us forward beyond where we are today? We must build credibility within our organization not every organization is quote unquote data driven yet and if you're if you're unlucky to not be in one of those shops well make your organization data driven show them the power build the credibility you might have to go slow at first because this may not be expected out of each and every data professional in an organization to do this but I think it's it may not be verbally expected or written down but I think it actually is expected so don't get caught up in that don't talk yourself out of beginning be open-minded no plateaus are comfortable for long back to the mountains thing and that resistance that you get about making progress it's it's not about having the progress it's about the journey so you got to spell out the journey so that you can do it bit by bit step by step and get into these trends and not be left behind which brings me to the end of my part of the presentation I'll turn it back to Shannon see if we have any questions William thank you so much for this great presentation and a great kickoff to the new year just a reminder that if you have questions for William or Elena to submit them in the bottom right hand corner of your screen and to answer the most commonly asked questions just a reminder I will send the follow-up email for this webinar by end of the day Monday with links to the slides and links to the recordings of this presentation Elena that question came in when you were chatting about Looker how much work is needed to build the semantic layer if I haven't if I have a semantic layer built another place such as Calibra or another package such as Calibra can I bring it into Looker that's a great question and it kind of goes back to that ability to have multiple tools working at once Calibra is actually a partner of Looker and you know their data catalog data cataloging functionality is incredibly strong loading it into Looker I'm not sure on the specific details on that one we could we could get that for you if you wanted as far as how much needed how much work is needed to build the semantic layer we have a generator for Look ML so as soon as we connect to the database it generates a model based on the tables that it identifies which will get you really off the ground and running if you now like the magic of Looker and of Look ML is being able to go in and customize that and that can be as easy or as complicated and long as really as you want to the great news is that Look ML is based is an abstraction of SQL and so it's a highly efficient language and also you know very very familiar to people who know SQL um so from that point of view it's certainly a you get off the ground easy and then you invest in your future love it so uh diving in here you know William when you were talking about the data maturity are you referencing the CMI data maturity model or if not is there a different discipline or a specific discipline you're mentioning well I mean I could be I was just really referring to data maturity as a concept that one certainly is is one that you could place right in there and say yes that's that's one that we will you know target but there are others I have one for example and so I was thinking about that in a little bit as well but the point is to get to get to that point where you are you know at least on the lower end of really mature and different models can help you in different ways I'm not saying don't think but I'm use it as a catalyst to help you think about where you need to be and and steps to get there and how to sequence all the stuff that you need to do anything about data maturity you want to add there I think I really liked what William said about not being able to you know skip a step I think that it's really really tempting to go straight to the really like kind of sexy stuff like right straight to like AI and feeding that into things and which is super valuable and by no means has to be you know left to the end but I think investing in a strong foundation as you continue to go off that maturity curve will you know serve you in the long run to be able to really create that vision of digital transformation and do either of you see any trends you know in data maturity standards as it relates to this William maybe you want to kick us I mean I just think that there's a renewed interest in data maturity because it's people are realizing that it's synonymous with digital transformation which is what the what where companies need to level up their organization and so there's a renewed interest in it I see more organizations maybe informally but coming together around this notion of maturity and even information organizations are being given more leeway to bring that maturity up in absence of spending months to show the direct ROI of various things or various approaches that you might take with business efforts as the executives of those organizations start to acknowledge that they need to be a data driven company I'd second that I think the other thing that we've really seen is in the past I would say it's you know in the past years there's been a explosion of staff tools on the market and I think people had this really strong desire rightfully so to be able to analyze their data and get insights from it which caused a you know everybody had seven different staff tools that we're trying to do on seeing and people are realizing that a centralized you know unified approach to this that's more strategic and more long run focus is ultimately going to be the winner in the game as opposed to trying to do the quick fixes of the staff tools Perfect and we I think we have time for a couple additional questions here we've got a couple minutes left there was a question here William you mentioned early on and this was one of your big predictions from last year is Master Data Management it's been around for years of course so why are companies still not doing it and what's the foundation of the resurgence great question and not for great reasons necessarily but because it's hard because to do true Master Data Management there's always another way you can do it that's sort of a halfway half measure you know you can do it the same old way you can I mean or everybody's I like to say everybody's doing UMDM they just may not be doing very well because who doesn't need a master customer list or a master product list but the unfortunate part is many organizations have 15, 20, 30, 50 such lists that are different and that means they're all out of whack and that just right there it's going to set the whole organization back as we can imagine from an efficiency perspective from an effectiveness perspective and so on so the reason it's it's a slow go or has been a slow go is because you have to get different business departments to work together on something the budget may not come from one department and if it does it's not true master data management it's just serving one application which is not really you know enterprise MDM which is I think what we're all talking about do you want to add to that I think that William wrap it up very nicely perfect well that does bring us to the top of the hour so just a thank you to both of you William and Elena for these great presentations and this information and thanks to all of our attendees for being so engaged and everything we do just a reminder I will send a follow-up email by end of day Monday with links to the slides and links to the recording and I hope you all have a great day and happy new year thanks everybody thanks William thanks Elena thanks the looker thank you