What is our primary use case?
We're a medical company and we have our own source systems that process claims from multiple organizations or health plans. In our world, there are about 17 different health plans. Within each of those health plans, the membership, or the patients, have multiple lines of businesses, and the way our company is organized, we're in three different markets with up to 17 different IPAs (Independent Physician Associations).
While that is a mouthful, because of data governance, and our having own data governance tool, we understand those are key concepts and that is our use case: so that everybody in our organization knows what we are talking about. Whether it is an institutional claim, a professional claim, Blue Cross or Blue Shield, health plan payer, group titles, names, etc., our case represents 18 different titles. For us, there was a massive number of concepts and we didn't have any centralized data dictionary of our data. Our company had grown over the course of 20 years. We went from one IPA and one health plan to where we are today: in five markets, doing three major lines of businesses, etc.
The medical industry in general is about 20 years behind, technology-wise, in most cases; there are a lot of manual processes. Our test use case was to start from fresh after 20 years of experience and evolution and just start over. I was given the opportunity to build a data strategy, a three-year plan where we build a repository of all sources of truth data used in governance. We have our mapping, our design, our data linkage, principles, business rules, and data stewardship program. Three years later, here we are.
How has it helped my organization?
erwin DI needs the Data Modeler, obviously, to be able to harvest the data directly from an existing database, or even a brand new one as you're designing it. That is a huge step in the right direction, although erwin has been known for that for 30 years. But the ability to take that model and interface it directly to the data governance makes it an easy update. It makes it simple for me to move from a development/design stage, for each environment, and into production, and to update the documentation using the data harvester and the Metadata Management tool and data cataloging module. That really brings it all together.
If I were to note any downside, it's that there are multiple modules and you can't have one without the other if you want to be world-class. But when you have them all, it makes life really easy for something like data profiling of an existing database to know if you want to keep it or not, given that there are so many legacy changes all the way through. The way we do it, when we make a change to a database or we add a database, the model is mapped, we import it, and then we have the data stewards populate any of our descriptions in their glossaries. The tool allows us to see all that instantly, unlike before.
I mentioned we have a data steward program, which is not part of the tool. While the solution has ways of using issues and for requesting data access within it, we're still stumbling with that. Sometimes it's just easier to talk to people. But we find that getting requests, getting data, and updating it, is actually a much easier process now.
In addition, the fact that I can always refer back to a centralized location with executive approval has helped me.
For our business analysts and data analysts, especially for some of the wannabes and the data steward program, we have been able to centralize a tremendous amount of data into a common standard. One of our mandates was to have a Tableau-type business-intelligence component. We went live with our entire enterprise data warehouse, all the tools, in January of 2019, even though we started in 2016. We spent most of the year in massive amounts of discovery just around our organization's members. We didn't even get to claims or provider-contracting because they are so complex. The tool itself has expedited our getting to brand-new levels we've never seen with our members, because now things are becoming standardized.
People can refer to an inventory of reports and they can see that we don't have the same report in 20 different places, having 20 people support them. Now, there is one report in Tableau with one dataset. That dataset has become a centralized dictionary/glossary/ terminology inside the tool. Anybody who needs to get access to our data can access it.
It's enables efficiency. Just in our marketing department alone, the number of new ways they have to think about our membership and growth has completely changed. They have access to data to make decisions.
Executives can now look at what we call a scorecard of our PCP because we now have standardized sales. Everybody knows what they mean, how they are calculated.
Very high-end statistics and calculations are now easily designed. Anybody can go look at them, they know where to go. And if they want something because it helps make their business grow, it's almost a 24-hour turnaround, as opposed to a four-week SDLC process. It has expedited our process. The goal was to build a foundation and then, for the next couple of years, to really expand it. We hit that and I don't think we could've done it without these tools.
Recently we had to bring on a brand-new entity, a brand-new medical group. One of the minimum requirements was that we had to take 10 years of historical data from whatever system they had and to convert it, transform it, map it, and log it into our existing source of truth. We did this about four years ago for an entity, and it took us almost nine months just to get a dataset that somebody could use. This last time, it took us three weeks from start to finish because, outside of the governance tool, we have erwin's Mapping Manager and harvester. It also allows us to do source-to-target, so we have all our target mapping to our own repository, and then we have all our targets to EDW already mapped. Our goal was to bring a 100 percent source of truth. We had a complete audit, from when it came in from outside the building, to a location in the building. Then we would transform it into our EDW to whatever attributes, facts, or dimensions we wanted to. The tool allowed us to do that almost in hours, compared to what used to take months.
Another thing with their DI, not necessarily governance, but some of their other tools — which, of course, all feed back there — is that as soon as we do it, it's available to anybody. Not that a lot of people look at it, because a lot of times they just come and ask us, but the difference is that we're giving them the right answers within minutes. We don't have to tell them, "Well, let me go back and search it for six days."
We have downstream departments, like our risk department which manages our Medicare patients, and makes sure that we are taking care of them, which involves a very data-intensive process. Our ability to bring in historical data from an old system, a different type of a computer system, and convert it to make it look just like ours, no matter what it looked like before, is all because we have a data governance program. People can look at the changes from before and after and determine if they need certain data.
A year ago, if somebody in our company's "left hand" brought in new data, no one but that left hand would know about it. Today, if somebody brings in data, all my data stewards know about it and they can choose to subscribe to it or not, today or later. And that is a matter of a flip of a switch for them, once we have brought it in and published it to anybody in the company. That's really important, for example, from the point of view of a human being. If someone has been around for 20 years it would be nice if we had all their records. Because of our data governance and what we built, all those records are maintained and associated with that person, and that's huge from a medical point of view. Data governance is helping us become even a better company because we know our data and how to use it.
The fact that erwin DI for Data Governance has affected our speed of analysis is a given. The DBAs are starting to use it more and even some of our executives are wanting to get to it for the data dictionary. It can happen that somebody from one of our departments sends them something and it doesn't make sense to them. Our goal was that if that happened we would try to find out and try to centralize it. We ended up creating our own dashboard reports on our Tableau server and published them to the same parties, so we could get rid of old habits and focus on new ones that have now been validated and verified, with the rules checked.
The data governance allows us a real-time inventory. Every time there's a new request or a new ask, we put it in there and we track it and we make sure that our attributes are the same. If they're not, we have an explanation with a description for the different contexts in which the data is being used.
In addition, part of our ingest of an ask is that we take a first look at it and we provide as-is documentation so that the functional design can be tracked. That's a huge advantage. That has saved huge amounts of time in our development cycle, either for data exchange or interfacing, or even application development. The ability to just pull up the database, to be able to look at the fields and know what's important and what isn't important, note the definitions — we're able to support that kind of functionality. I'm one of the data architects here, and we work with everybody to make sure that our features and our epics are managed properly. For me to be able to quickly assess something, within a few minutes, to be able to say, "Here's the impact, here's what we have to do," and then hand it off to the full-blown design teams; that saves a month, easily. And that's especially true when there are 10 or 15 requests a week.
As for how the solution’s data cataloging, data literacy, and automation have affected the data used by decision-makers in our organization, on a scale of one to 10, I would give it a seven. It depends on which stakeholder or executive we're talking about. But has it had an impact? Every one of them has brand new reports, reports that didn't exist a year ago. Every one of them now sees data in a standardized format. The data governance tool might not have a direct impact on that, but it has an indirect impact due to the fact that we now govern our data. We treat data as an asset because of the tool. It's not cheap, it's an expensive tool. But my project has a monthly executive steering committee and, for 36 months, they never had a question and never second-guessed anything we did, and they loved any and all tools. So being able to sit with them and say, "Hey, we had an issue," and immediately give them a visual diagram — show what happened with the databases and what somebody may have misinterpreted — is huge; just huge. For everyone from our chief operating officer to our network operations, physicians' contracting, our medical management group, and our quality improvement group, it definitely has impacted the company.
We've only taken it out to about 50 percent of what it can do. There's so much it can do that we still don't do, because we ourselves are maturing into the program. It really has helped when it comes to harvesting or data profiling. For those processes, it's beautiful — hands-down the best so far. I love the data profiling.
What is most valuable?
For me, the biggest benefit with erwin DI is that I have a single source of truth that I can send anybody to. If anybody doesn't know the answer we can go back to it.
Just having a central location of business rules is good. That has come up a lot.
The solution gives us data lineage which means we can see an impact if we make a change. The ability for us to have that in this company is brilliant because we used to have 49 data stewards from some 23 different groups within six major departments. Each one of those was a silo unto itself. The ability to have different glossaries — but all pointed to the same key terms, key concepts, or key attributes — has made life really simple.
It has given us better adaptability in our EDW and a more standardized way of looking at data, as opposed to all of the different formats.
Our experience with the solution's Smart Data Connectors is still limited, but so far we're impressed. For us, it's mostly been reverse-engineering. But because we got to start this whole project from scratch it was really about forward-engineering. One of the advantages is that we wanted to go back through the entire enterprise, and start mapping everything legacy or future. Ultimately, the future is to move to the cloud and rearrange all our data and reprioritize the most important attributes and not have multiple replications of data to our many silos. So the Smart Data Connectors to engineer code have been spot-on. Harvesting reverse-engineering allows us to test and verify some of the things that are odd.
It's got a proprietary format that works really well within all the systems and I can export to any format, including my data profiling, my mapping integrator, and any and all of my governance stuff, whether it's within the business rules, the policies, the dictionaries, or the glossaries. I can export that into a CSV or Visio depending on what it is.
What needs improvement?
There is room for improvement in automation, no question.
Also, the fact that I sometimes have to go in and out of different applications, even though it's all part of the whole erwin suite, perhaps means it could be architected a little bit better. I think they do have some ideas for improvements there.
But regarding the data governance tool itself, for me there was a huge learning curve, and I'd been in software development for most of my career. The application itself, and how it runs menus and screens when you can modify and code, is complex. I have found that kind of cumbersome. I had one guy make an error and it costs us a few days because it had an impact on a whole slew of options and objects because he didn't know what he was doing. That was not their fault; it was purely my fault, allowing that to happen. For me, that was a struggle.
For how long have I used the solution?
We started about three years ago when we started our data warehouse project.
What do I think about the stability of the solution?
We haven't had any problems with anything. I haven't had to worry about updates. If anything is done in terms of updates, it has no impact, to my knowledge. Every day, one of three of us on it and we haven't had any problems.
What do I think about the scalability of the solution?
In terms of its scalability, I'm a really simple person. Two plus two equals four. If I know that I can always expand, it's basically a configuration engine or I don't feel like I'm painted into the corner, I feel I'm covered for scalability.
I feel that way about everything with the Data Governance tool. If I need to grow it, it's just a few more licenses. I have to pay a little extra, but I can get another hundred if I need that many. So I have no worries about being able to add users. I'm also not worried about size and space on the system.
So far, I have no limitations in terms of scalability. I'm good with it, but I can't say that I have pushed it to the limit yet. We have 500 people here and I have a hundred licenses. Of those, about 85 are currently active, between our users in the building and our IT department.
It's used extensively by the people who need to use it. I don't think the end-users are using it as much as are the data analysts, the business analysts, or the data stewards. But that was the goal. The data stewards are the ones who should use it. The data stewards are the ones who are managing it. They're the ones who get the requests from the users within their department or their group.
As far as using it more extensively, it's just a matter of if we grow as a company and we need more data stewards, but I think we're in a good place. Everything feeds accurately. As far as visibility and reading dictionaries, that's something that we would want to do more of. As users adopt it, as departments go from a mom-and-pop mentality to "corporate America," I would say our goal over this next year is to double our growth.
Self-service is one of our goals: get people to know how to use it themselves. People here ask IT for a particular report and that report goes to a bunch of different people. The same report is used for different reasons in different groups but no one knows it. With erwin, the goal would be that if they want a report they would be able to go in there, see the BI reports which are inside the Data Governance, determine if it's what they want, and make a request; and that day, they would have it, filtered to their needs. If we had to create a brand-new one, they would know the elements and would put the request into the hopper and we could turn that around. Even today, in some cases, because we still have BusinessObjects and we have SSRS, with some of that stuff it can still take up to six weeks to turn a report around. If they use the new system and they use the data, I can put a Tableau report on their desktop within 48 hours.
How are customer service and technical support?
Their helpdesk ticket itself is pretty great. The team of people, whether it's Tammy or Susan or a few others that my associate has worked with, is very good, no matter what the tool is. It's very comparable to our own helpdesk. You can get stuff done and, in some cases, there are almost too many emails. That's just how good they are.
Which solution did I use previously and why did I switch?
We had Excel and Word documents, but nothing else.
When we started our data warehouse project we were looking for a data governance tool. That led us to buy their data governance 1.0. They've upgraded to the 2.0 and we're working with that now. We then ended up buying the modeling software, so we got server licenses for some copies for our modeling, which is the foundation. Then we purchased their Mapping Manager harvester, their data integrator, their metadata package, and we've just recently even purchased their architectural software, so we're working with their entire stack.
How was the initial setup?
The initial setup of the software was a combination of straightforward and complex. For me, it was complex, as I'd never seen it and there were a lot of components to their instructions. But we ended up doing it pretty much ourselves, with a few interactions with some of their technical people on the installs. The clouds were, obviously, easy, but the making sure the Data Modeler was there and dealing with the architectural software within our organization wasn't as easy.
But erwin DI for Data Governance, itself, was pretty straightforward. Adding users required a little learning curve but no one taught me. It was trial and error more than anything. I'm sure we could have had great training from them, but they have good videos and good tools on the web. For us, there was probably less than a 5 percent interface with erwin for our installs, configuration setup, and even configuring the databases to house all the data.
Being that it's a cloud solution, in the beginning, the firewalls became an issue but we got those resolved right away. There really wasn't anything bad. We self-taught; we didn't have a whole lot of Professional Services on this. It's intuitive enough that we run our entire data strategy on it. With a group of three people, we support another 50 developers, systems analysts, and data analysts outside of my group.
We didn't really have an implementation strategy. We didn't know what we were doing three years ago. We just knew that we were tasked to create a data warehouse, and I'd never done one. We tried to do it the right way. If it was something that I'd been doing for 30 years, there would have been more of a strategy for how I would do it.
We had a consultant come in in 2016 and create what they called an IT map of applications and data strategy plan. We called it the Imap. They laid out the groundwork and the framework in which to build this whole thing. One of their areas was data governance. It just so happened that I was doing research and erwin's Data Governance tool came up in my research. I knew erwin's responsibility, so I put in a request. Around Christmas of 2016 it got approved. We didn't expect it because the year was out. That became the first thing we started with. From there, it just kept building with all the other modules. It just kept growing with us.
The advantage erwin had with us is that we were new at it. So even if there was something wrong, we wouldn't have known the difference. But I knew what was right in terms of what the data means to the company and how we run our business. For me, it just fit perfectly. It just kept falling into place. Each time we need to get more money for another license or a different module, I could integrate it pretty simply because it fit the narrative. Maybe that's the best way: the technology just simply fit, purely by accident. I'd love to tell you that I'm such a genius and that I planned this all out, but I didn't. It did help that erwin purchased some companies and ran right along with what we were trying to do.
Because of the erwin Data Governance software, and trying to figure out and follow their MO as far as key concepts, key terms, and attributes were concerned, we took 27 of the most important people in our organization from the president on down, and sat them in a room for three hours so they could define the term "member." Who was a "patient," a "member," or a "consultant," meant a lot. It changed the direction of the company. They didn't even know what data governance was three years ago. Now everybody talks about it.
What about the implementation team?
We worked directly with erwin. If we had any questions, we'd go through their helpdesk. Sometimes we'd have conference calls. A lot of times, they seemed to go above and beyond to help us, especially when it came to the database configurations. We ran into a few things with the Mapping Manager and the harvester, but their team was great. They worked with our DBAs with no questions asked and no hesitation.
What was our ROI?
Our labor costs are half what they would have been. And then there was the lack of quality or the lack of productivity. So it's had a huge impact on all those things. That's why the money is more than recouped.
My three-year plan was to recoup the $3,000,000 we spent in the previous three years, and this year we have already recouped $2,400,000 of it. We have two more years to recoup to break even. I don't know if that's directly related to just the Data Governance. It might be that because of governance it has allowed us to do all the other things. The ultimate goal was to get data into the hands of the decision-makers and have it as accurate as it could be, so they could make better decisions.
We have improved our time for report servicing and our capabilities in turning things around quickly.
One thing we missed in our estimates of cost savings was the reduction in the number of requests or the time it would take to do a request. The issue was that we created a standardized report, and it worked so well that we stopped getting requests altogether. We never thought they would never send a new request, so we saved even more. But that's because we knew the fields, we got it right the first time, and we created standards around governance that allowed us to really simplify our business. I have 29 standardized reports around memberships, providers, and claims, and they are used throughout the organization now. Those are reports we didn't have before.
In terms of time-to-value, I wouldn't say standing it up was quick, because a day would be quick. But it was under a month. It could be set up pretty easily, especially once you understand all the components you need and all their modules. The erwin governance solution is on the cloud while the modeler is on-premise and the suites are also on the cloud. And the fact that it's cloud-based made it simple and straightforward. It was just "boom," we got our logins and we were fine, for the Data Governance software.
What's my experience with pricing, setup cost, and licensing?
The whole suite, not just the DI but the modeling software, the harvester, Mapping Manager — everything we have — is about $100,000 a year for our renewals. That works out to each module being something like $8,000 to $10,000.
Which other solutions did I evaluate?
Everybody was evaluating other options. I got in trouble because I picked erwin after I had looked at 20 other things out there. Based on price and the size of our company, and the fact that we were brand-new to this whole endeavor, I didn't want to spend a fortune on something like Informatica, and have a master data management system.
Another big difference between Informatica and erwin is interfacing. Informatica is the top-of-the-line, upper-quadrant for MDM solutions, whereas erwin was more about just managing data and not necessarily manipulating it, moving it, interfacing with it, etc. That was the big difference. But erwin allowed us to get our footprint into it and really learn it. It was just the right solution at the right time.
What other advice do I have?
Our first goals were data literacy and data as an asset. Those were our two big, ultimate goals three years ago. Data literacy turned out to be 10 times more important than a data warehouse. We could look at existing data sets and, just by educating people, it gave them an advantage almost immediately. The fact that the data governance was able to put a framework around data literacy helped us focus on the right answer, even if it wasn't the first one given. In other words, sometimes we'd have the same answer three or four times, and it would shift until we nailed it. But without governance, we would never have done that and we would have stayed the same.
The secret to the success of this project was that we had a vision and we stuck to it. Governance was important to us, no matter how other people might have thought about it. In my very first data steward meeting I was introducing everybody to these brand-new terms they'd never seen, and someone in our analytical group totally derailed the meeting. So be aware that it's not going to be easy, but have a vision.
And make sure that governance is important or don't bother. It's not something that a lot of people add value to at all. When you say, "Oh, I want governance so we can have a data dictionary and you can go look at it," they'll say, "I don't want to look at it, just give me a report." But the ability for those who need to do that is huge. Have a vision and stick to it and be willing to take a step back, sometimes, to go two forward.
The neat thing is that we've pretty much done all of this with two to three people, for our entire organization. We do have three data teams that are using the Modeler for development — ETL SSIS stuff — but we have a pretty serious "wash, rinse, repeat" standard. If anything is in doubt, we just go back to the business rules and see what our rules are. What are our principles, and are we meeting them?
As far as automating the changes through the environments, it has helped, but not a lot. It's not like it was a silver bullet. We need help there, because there's so much. There's the model, but once you promote that in different environments, sometimes you miss it because you only get three or four days to get out of QA to get it into stage.
Obviously, you mitigate risks with automation. It does have an impact. As a company, we just haven't been able to take full advantage of it now, but that's our hope. We're only into it for about a year-and-a-half, even though we have run with the suite for almost three years. We're still immature. I wish I had everything at the push of one button, the "Easy" button. Some of it's over our heads. We could use some new training and we could use some additional support. erwin has been great with us, but it's also a matter of the appetite and the resources. The biggest issue is that I don't have a team of people doing what a team of people need to do to accomplish what we would like to. It's done by a small number of people on a consistent basis, and not full-time.
The solution's generation of production code through automated code engineering would reduce the time it takes to go from initial concept to implementation, but we're a Microsoft shop and most of all that is done inside TFS or Visual Studio. That's how we manage all our codebase, including release management. That's all done separately and is automated. We're trying to create some interfaces between the two. We just haven't gotten there.
In my three years using erwin, besides actually getting approval for the money to purchase the software, I don't think I've had a struggle with it. They've been great. When we first got on and we had some questions, they got me to the development team in England and set it all up with us without question, no extras. They just tried to make sure it worked.
I would rate erwin DI for Data Governance at eight out of 10. I never give a 10 because I have yet to see perfection. It has some gaps, but I definitely think it's in the top third. As far as rates go, I don't have a lot to compare it with. It's easy now but it took going through a learning curve, but that's the case with any software. Does it need to mature a little? Possibly. But that would be it. With their roadmap, they're buying companies, and changing things, and doing things. I've been pleased.