 What our approach is at Oval Edge to end-to-end data governance? So we're going to take a look at the tool at one of our very first features, which isn't one of the main features today, but a cool, nice to have. This is called a data story, and that's where I'm going to be showing you my presentation from. I'm going to show you just four slides, and then the rest will be a live presentation. So when we think about Oval Edge, who we are, what we do, the approach we take, who can benefit from it and interact with it, those are all topics I'm going to cover today. So you're in this room probably because you are interested in data governance, you have an idea of what it is, you're trying to execute it, or maybe you're helping a different company execute it. Most of the people that we talk to at Oval Edge find themselves somewhere along the lines of this diagram. I'll blow it up a little bit more for you. When you're thinking about governance, it's important to have a correct definition of what it is. You ask 20 different people what data governance is, you'll probably get 20 different answers. What we like to say is that any person that touches data is going to need to be involved in data governance and will be affected by the maturity of your data governance strategy or program. When we talk to people, this is an initial look of usually where they're at on some regard. Their environments are confusing mess, they don't know where their data's at, and even if they do know where it's at, who owns it, who's responsible for the quality of it. If you have an ETL tool that moves it somewhere else, what happens to it at that point? And then when it ends up on reports, can I trust this information? We have analytics, but is it governed information? Do I feel confident in making a decision? So we see people's governance strategies kind of convoluted in, I don't know who to go to, I don't know how to access it. So unlike, you know, we're not unlike many of our other competitors in the fact that the grounds of a good data governance program is the ability to catalog your information, right? We want to have a central repository for all of our metadata and be able to profile that data and give it to you in a platform that's easily accessible regardless of technical capabilities. That's the first thing I really want to touch on is that this tool, hopefully, you'll see that the UI is pretty intuitive. At Overledge, we believe that data governance needs to be owned and steward by the business. We want the business to drive it forward while the technology team supports. So we begin with cataloging all of your information. But beyond that, for an end-to-end approach to data governance, we need to be able to do more than just catalog it, more than a central repository for all your information. We want to be able to have Overledge as a layer on top of all your systems, applications, information, but then be able to do things like support people to be able to search and understand data, to analyze it, to give access to it, to start to standardize the terminology, metrics, and KPIs that you have all in a central platform. So you don't have to have this piecemeal solution of all these different pieces of technology because when we see that happen, adoption typically lags. So we want to give you an end-to-end platform where you can govern all of this in one space. We also have built this tool to be very collaborative so that no matter your technical expertise, we've tried to lower that barrier to entry so you can get in the tool and really understand how can I interact and understand my information and how do I start to benefit my organization through my new understanding. So how we start to do this is we really focus on three main areas. We focus on data access, data literacy, and data quality. What you're going to see me do today is try and walk you through a very fast overview of our platform. I always joke that I could demo this product for four hours. There's so much in here, but what we're going to do is we're going to take a look at the most important concepts about how do you use this and start to gain value from it, and hopefully you'll see how it will make the things you're already doing a little more efficient and a little easier to execute on the things you're responsible for in your day-to-day job. So where we're going to start, just so you have some perspective, is I'm going to talk a little bit about how do we build that catalog, how do we secure that catalog, and then how do we gain value from the information being in there. We'll take a look at data literacy, data quality. I'll run you through a couple privacy and compliance examples, and then we'll wrap up looking at the feature in our tool called the governance catalog and a few examples in there. So when we get started, the first thing in Overledge that I love to show is our connector screen, our crawler screen. So Overledge, you can use APIs, you can use JDBC drivers to go ahead and connect to all of your source systems, applications, databases, reporting tools, ETL tools. You then get to choose specifically how you want to crawl and profile those sources. You could forgo the profiling if you're not ready for that yet, and you'd just like to do the metadata collection, but we do support both. You can for every source choose exactly what do I want to crawl and what do I want to profile. That crawling is the metadata collection, and the profiling will be prebuilt queries and algorithms that we send out to your source systems to go ahead and scan for what are those minimax values, what is the density of this source, what is my null percentage, and many other metrics. We then allow you to reschedule these so that you can capture changes. So if you have a system upgrade, there's new metadata added into your environment, Overledge will be able to collect that. We have this job workflow engine that you can schedule these jobs, capture changes, and then you can review what those changes are like in many different ways. You can compare QA and production environments, or if you'd just like to simply see how is my metadata changed on a specific source since the last time we looked at it, I can click in there and see two columns were added. I would also get notes that this was modified or deleted if that were the appropriate data here. Once we start to do this, this is going to build out the data catalog and keep it current. So what you see on my screen right now is the data catalog home screen. This again can seem overwhelming, but I'm showing you from an administrator perspective, so we can hit on all the important tools. But in here you would be able to come in and see only the sources you had access to, and you can start to do in column filtering and searching to find information. Before I take a deep dive into the catalog, we'll get there in about two minutes. I'm going to quickly show you security just so you can think, okay, we've brought in all my sources. I want to make sure the right eyes are on the right things and there's not a vulnerability of, man, I might show sensitive information to someone. In Overledge we have user roles to manage that. So we give you permissions on both the metadata and the data. So you're able to interact with the tool in accordance with your user role. These user roles are assigned to people and then they're assigned to the sources. What you're seeing on my screen right now are the roles applied to the database schema level, but you could change that all the way down to that report level, the report column level, file column level, or table column level. So it's very granular sources of control here. And on that column level, if you would like to not only choose what roles could see it, but if you would like to mask or restrict values, you can do that as well and you'll see some examples of that today. But when we think about, okay, we've built this, we have the jobs running to keep it current, how do we start to discover the information in the platform? And that's what I'm going to start to show you right now. So something that you can do right away is that Overledge has an enterprise-wide global search. So this will search against everything that you've brought into your data catalog. So I can do keyword searches and find information that I've either brought into the catalog or created maybe through adding in some documentation in the data stories, business glossary terms, KPIs or metrics tracked in the business glossary. This will search against absolutely everything. I'll be able to come over to the right side and if I'd like to filter down my result set, say I know I'm looking for customer information that lives in a table, I can go ahead and filter that. And I could even go one level lower by looking at who's the identified owner or steward for that source. In Overledge, an owner and a steward must be added to every source to be in the catalog. You then see some crowdsourced ratings here. There's a five star icon here. This is one of our crowdsource and collaborative features where everyone can evaluate the information that they see and then you'll see an aggregated score and representation of how are people evaluating my data sets. You'll then see icons over here which are a byproduct of one of our governance catalog features called certification policies. You can encapsulate your business knowledge into policies, associate it to data elements, and if they pass, you can have them be certified in the correct nature. Typically this green icon means it's gone through a governance cycle, it's passed the policies, and this is going to be good trusted data that I can rely on and make decisions from. Go ahead and click on that and it drills me into the table level screen in the catalog. This is going to be that area where we did that crawling and that profiling. That's where this information is going to come. So I can come in here and see it's picked up column names, types and titles. There's profiled information down here for top values, null counts, distinct counts, min and max values. Outside of that though, there's some business aspects here that are important to note. We've been able to make a business description telling people, okay if you're not familiar with this source, what can you expect to find? So you can write descriptions on your tables and columns and schemas and sources to say, this is the information here and this is what you should see. There's also that governance component of we'll highlight everyone who's interacting with that source. We'll show you how many times they've queried it, commented on it, or added it to a project. You can see who your owner is, who your steward is. So right here now, if I'm looking for information and I have a question, immediately I know I could go ahead and talk to Mike to get my questions answered instead of sending an email to IT saying who owns this and then it get lost. And then I'm ineffective in trying to solve my questions because I can't find an answer. And we'll let you do a lot of that collaboration right out of the tool. So every screen in Overledge has this collaboration window where I could open it up and send a message and they could get a notification from that and it would take them directly to this area. In the catalog, we're also showing you at that time of crawling and profiling, we're trying to pick up what entity relationships exist for this table. And then we give you two scores along the right side of my screen, a similarity score and a join score. These are trying to help you evaluate this potential relationship. So how likely is it that there's duplicate data? How likely is it that I could join or union these sources? And that's something you could run an impact analysis on and dig into to find out your answers. We then have the capabilities to do end to end lineage in our platform. So Overledge can go ahead and connect to your pipeline tools, ETL, ELT, BI and analytic platforms. And we can parse through the underlying source code of those to see what are the handshakes between the movement of your data all the way from your source down to the production layer, whether that's a BI report through Tableau or Power BI, we can go ahead and give you that analysis here. And then this is interactive. So if I want to understand the responsible code for this ETL job into my customer dimension table, I can go ahead and click on that and see what the responsible code is, who that code's owner and steward is, any associations or references that that code makes as well. That's something I can dig into and is a part of our end to end platform. I'm going to show you an example a little later on about the BI integration here with Tableau, but that'll come up in about two minutes. Before we get there, I want to show you one final screen in the catalog. And that is our column level screen. So here it's the same information and a little bit more that you saw in the summary screen of the catalog. But here you get to see all of your columns listed as well as the metadata we've been able to collect on the column level and not just your table or schema level. We can come in and also find types of technical descriptions, like at the source, is this named as a primary key or foreign key? We can automatically populate that here. Something I want you to notice is that in here I have fields like social security number. I have fields in here that have privacy requirements, right? This needs to be protected information. And I want to talk to you about how we did that. You can see that the values are masked over here. There's a description pulled over. I can see the compliance standards it's compliant under like GDPR or CCPA. And I see that there's a term applied here. I want to walk you through how our business glossary supplements the information and helps you define, standardize, classify, and protect the information that you've been able to bring into your enterprise-wide catalog. So two examples we're going to walk through here. One will be literacy focused and one will be classification and privacy focused. So the first one I'm going to dig into is a literacy example. And this is where you'll start to see some of our integration with BI tools. I've looked for this term called average length of stay. This is an example that we got from working with a large healthcare delivery network about a year ago. We were doing a review of all of their reports in production. They had this term, average length of stay reported out on a hundred different reports. And when they did some analysis, they figured out all the values were mismatched. The KPI values, the metrics didn't match. The values were different. And why? When we dug into it, we found that they were calculating this metric 17 different ways across all of these reports. So they were having a really difficult time making decisions. So the business people who had to be responsible for knowing this and then making a decision upon it were having a really difficult time and we're making poor decisions. So something we want to be able to have you do in Oval Edge is be able to clearly define how do we define and calculate this per the business, trying to encapsulate that tribal knowledge into a central source of truth, and then be able to take those terms and apply it to the cataloged information. So how we do that? One, all terms go through a governance cycle for a workflow, right? A term needs to be created, submitted, and then it has to go through a review through the reviewer, steward, or creator of the term for it to be added into the program. Now we talk about you should have governance councils where you're reviewing these terms before it even gets to the technology piece in Oval Edge. We want to give you a framework where you can establish what your councils are and then meet with those councils and then get those terms into the tool. But when you get these terms into the tool, something cool that we can do is we can take these terms and then associate it to reports and to your cataloged information. What I just did is I clicked on a column, a report column where this term was associated. If I come to the summary view of this report, what you're going to see is you're going to see the fact that this is a Tableau report we've been able to pick up and pull into the tool. Let's let my screen load here for a second and I'll show you what that looks like. So something that we're able to do, I can just jump here first. In the tool, that literacy work that we've set up and the classification work we've set up, this is a report that's gone through the Oval Edge platform. I've switched my tab up there. It's hard to see, but I'm in Tableau online right now. And then along the right top corner of my screen, you're able to see a pop-up that says certified, meaning this has gone through Oval Edge. It's a certified report. It's passed a policy. And then in this drop-down menu, you're able to see all of the metrics on this report and how they've been defined per your governance council. So here I've pulled up that length of stay example. So as a business user, if my behavior is to make decisions off reports, we're also now giving them the confidence of this is good, certified, trusted information. So we're trying to extend it back out to the business where they're already making decisions. At this point, they also can report data quality issues if they see them. They can click on this, submit what their issue is, and that kicks off an automatic workflow in the Oval Edge environment. So my screen loaded. Sorry, that took a second, but we're here back in the tool and we can see that same report. It's coming over live from Tableau into the Oval Edge environment where there's ownership around the report. There's a description and literacy work done on the report. And then up here, you can see that it's certified. So that's a nice little tool of ours where we can extend it back out to the business. Now, if I take another example for the business glossary, if I don't do literacy focus, but I switch my hats and I'm thinking compliance privacy, there's going to be situations where you're responsible for knowing where sensitive information's at, and maybe you do know where some of it's at, right? You have a CRM or you have an HR system, you have an ERP. There's some sort of customer or personal information stored somewhere. Likely you'll know where some of it is, but maybe not all of it. And so what we want to be able to help you do at Oval Edge is be able to query all of your source systems at one time to find where this information might also live based on some standards. So say I take social security number. Social security number is something we've added into the catalog. We've given it a description, but also down here we've given it some data association rules and classification rules. Here I'm saying copy the title and description of my term to the catalog when applied. Also mask the values found and show our classification at the catalog level. What Oval Edge can do is you take terms and associate them to objects. As long as you've associated one term to one object, it's going to put you in the position to run our AI recommendation engine. What that engine does is you can go ahead and submit a job to run against all of your unclassified columns. What this is going to do is it's going to search the metadata, data, and pattern match recognition of the columns in the tool and pull back things that might also be a social security number. You then get results where you have to either thumbs up or thumbs down your finding. When you thumbs up a finding, it's going to go ahead and add it to your associated data dictionary. When I go ahead and click on that column, it'll take me into the catalog and I'll be able to see the descriptions come over, the classifications have come over, and that masking rule has been applied. So we want to help you quickly find and identify where this information is at so you may protect it. Social security is a pretty easy example because it's three, four, or three, two, four-digit patterns, right? But this is a really sophisticated tool. If I take something like county, the only similarity in the data here is going to be that it's a character field, right? So as long as I go ahead and find one piece of data, same type of thing will ensue. I can come in here, run the recommendation, and here I have 177 responses and we can then get a smart score, which is our aggregated score between the metadata match, data match, and pattern match. Now, something that's really great about this is the recommendation is very intelligent on its own, but it'll learn from your behavior. So as long as I come in here and I approve or reject these findings, the tool's going to learn from that, and so when I rerun it, it'll be even better results that next time. When I do associate that column to the term, at that time, it automatically assumes everything that we've set up on the term level to the cataloged information. Now, if I think about one step further, say we take that social security example one level deeper. I'm not looking for social security numbers in general. I'm looking for social security for a specific person. So say someone under GDPR puts in a request. I want to know every instance that you have of my name, social security, or email showing up, right? I have a right to be forgotten. This tool in our toolbox says governed data query. What that is is it's a query that you don't need to be able to write any sort of sequel or scripting. It's going to be click and point. I identify the columns I want, I identify the values within those columns that I want, and then I can execute this query against all of my sources at one time. So it's very powerful. I come in here, I execute it, and I see the results, and you can see a history of every time that you've ran it. This history is important because I can see how I'm moving towards compliance over time. The first time I ran this rule to look for Claire's information, I found 15 records across five different systems, three row records in each one. I go ahead and click those, can see what those row-level entries are. As that responsible person, my job was just very easy. I ran one query against every source, found all of the information, and now I can use Oval Edge to go ahead and go to those sources to fall in line with compliance. I can go there and take care of those records in my source system. Something we're really excited about that we just launched is a partnership with Snowflake, where we actually can go ahead and push back a mask rule to a Snowflake environment saying I found this data and I'd like to mask it. You also can mask and restrict and encrypt the information within the Oval Edge platform. So we did that for three records, and we can see the last time I ran it, there was 12 entries found, meaning we're three row records closer to falling in line with complete compliance of protecting that information. Next thing I want to show you is just some of the ways that your users are going to be able to find information really quickly and then start to ask some questions and answer their own questions through some self-service features. So in the tool, say I identify poor quality. I'm in here and I notice I'm looking through this data and I feel like this null count here is really high. I'm really familiar with this data and that's not normal. Something that we're really excited about is we have a full data quality metric and program within the tool. If I come to this nine dot icon in the service desk, I can say report a data quality issue. Before it allows me to create a ticket, it's going to make me check myself to see has this ticket already been created on this source. So every source is going to have a data quality index where it will show you the issues that have been identified and any data quality rules that have been written against that column and against those issues. So here I see when I go ahead and check the issues log, this one has already been reported that there's a high null count there. So someone else noticed the poor quality I also noticed. I can then go to the rules that were written and rules are written by those data stewards or owners or someone that has that metadata right capability within the tool. They can create these rules and associate them to the objects. These rules will then not only run against that profile information, but it will run against your source system live and survey the actual data without needing to move the data into your system. So I can come in here and see the results of did that rule pass or did it not pass. I can see everywhere that this rules been associated and then view those execution results. So we want people to be able to identify issues that are there and then we have full workflows built out to give a responsible person the responsibility to follow up on that, write rules, monitor those results and then they can final you know finish that circle by going into the source system and correcting the poor identity the poor quality that has been identified. Outside of that say there are more savvy user, they're more technical and they want to be able to run some sort of impact analysis on their own. They will see something that could be a problem. We're not really sure we need a data quality rule yet but I want to add it to an impact analysis to see what might this be affecting if I'm right and there is poor quality. You can always add sources in here from the nine dot option to an impact analysis. I can choose the impact analysis I want to add something to and then I can run that analysis. When I come to look at that analysis, let me pull that up really quick, this is going to be something where I can go ahead and look at all of my sources I've identified where I want to see what would this column table or object do to anything downstream. What are those downstream associated objects and what is the impact on them? So you can add those and then you can come in and survey what is that impact. You can come in and review the responsible code for that change. You can make it go through a review process and you can see who reviewed that and allowed it. So that's something that people can do as well. One final thing I want to show you out of the catalog is the ability that we can use our self-service query sheet. So what I just did was I was in that customer dimension table and I launched the in-app query sheet. Now this is something that's more locked down for users because this will go ahead and query your source system. It's only select statements. You're not writing, pushing, changing, creating anything like that. It just allows people to be able to ask some ad hoc questions without needing to be very technical. So when I say that, I mean they don't have to come in here and script out the sequel to ask questions. The tool is going to do that for you. So if I come in here, I can go ahead and build out what those conditions would be on those specific columns and I can build out and ask questions for myself. Now say I am a really technical user and I would like to interact with Oval Edge that way. We do give you a query window where you can come in here, write your select statements, and then be on your way and ask your own questions. What the great thing about Oval Edge is is that you can then save these queries and those are going to be viewable in the catalog as well. So if you've written a really productive query that helps you, that you want to share with other people, you can go ahead and catalog those queries and those are going to be searchable within the data catalog as well. That's really taken me through what I want to show you. So what we looked at today, I want to wrap up, is the idea of the fact that you don't need multiple different tools to achieve end-to-end data governance. We're not a modularly sold tool and we do that on purpose because we want to be able to give you a tool that will grow with you. So if your first initiative is to find all of my sensitive data and I want to protect that, do that first. And then when you're ready to maybe discover that end-to-end lineage, to manage your data quality and write those rules, maybe to define all of your KPIs and get them in the business glossary, all those things are ready for you when you're ready for them and otherwise you can turn them off until you are ready. So this is trying to give you one platform, one stop shop for the end-to-end discovery of your data and then the usage of that data. How do we use it? How do we make sense of it and how do we make better business decisions because of it? So thank you and that's all I had for you today. So.