Data Driven Leadership
Saving Time, Money, and Headaches with Mature Data Governance
Guest: Jon Sakanai, Principal Business Intelligence Program Manager, Workday
Inaccurate data isn’t just a headache, it’s a profit killer. Bad data costs companies 12% of their revenue, according to Experian. If you’re ready to start (or continue) your data transformation, Paola Saibene and Jon Sakanai are here to help.
Inaccurate data isn’t just a headache, it’s a profit killer. Bad data costs companies 12% of their revenue, according to Experian. If you’re ready to start (or continue) your data transformation, Paola Saibene and Jon Sakanai are here to help.
Paola is the Principal Consultant of Data Governance Practice at Resultant. She helps with our Solution on the Spot and shares key principles for diving into data governance.
Then Jon, former Business Intelligence Manager at Colorado Housing and Finance Authority, shares his story to put the theory into practice. You’ll hear an excerpt from a webinar where he explains how his team went from manual data warehouse releases a couple times a year to automatic releases multiple times a month. He discusses lessons learned as well as results he experienced from this digital transformation.
In this episode, you will learn:
In this podcast:
Jon Sakanai has spent the past 18 years leading data and BI initiatives in the financial, nonprofit and enterprise SaaS industries. He is currently Principal Business Intelligence Program Manager at Workday, where he drives insights across go-to-market teams in the Revenue Operations organization. Jon has presented at numerous regional and national industry conferences on how a modern analytics ecosystem can enable organizational strategy. He is a graduate of Colgate University and fell in love with data analytics after learning how to create a pivot table in Excel during his first post-college job. Outside of work, you can find him enjoying the mountains of his home state of Colorado or spending time in his ever-expanding native plant garden.
Jess Carter: The power of data is undeniable, and unharnessed, it's nothing but chaos.
Speaker 2: The amount of data was crazy.
Speaker 3: Can I trust it?
Speaker 10: You will waste money.
Speaker 9: Held together with duct tape.
Speaker 3: Doomed to failure.
Jess Carter: This season, we're solving problems in real time to reveal the art of the possible making data your ally, using it to lead with confidence and clarity, helping communities and people thrive. This is Data Driven Leadership, a show by Resultant. I'm your host, Jess Carter, and on this episode of Data Driven Leadership, we're diving into data governance programming. Specifically, we want to look at the advantages of data governance frameworks and what they can and should bring to an organization and when to use them.
We're going to get some quick themes in the first few minutes of the show and afterwards, we've got two experts who'll take a lived experience and they're going to break it down for us in a whole bunch of detail. To help me 'Solution on the Spot', is Paola Saibene, Principal Consultant of Data Governance Practice at Resultant. Hey, Paola, how are you?
Paola Saibene: I'm doing great. Thank you so much for having me here today.
Jess Carter: Yeah, absolutely. Glad to have you. Well, so before we get started on our 'Solution on the Spot', I have a couple quick questions for you. Why data governance? How'd you fall in love with it?
Paola Saibene: It's the glue. It's the glue that makes you run faster and a lot of folks just don't know how to leverage. They don't know how to implement it well. They don't know how to bring velocity from it. And as a former practitioner of IT and strategy and delivery and operations and all of that package together for many years, somebody told me, "Hey, take a look at governance," Many, many moons ago. And I remember Gardner promoting it heavily over 20 years ago saying, "Hey, business value and governance are your basic equations components." And when I started using it, I realized, "Wow, this is an accelerator. If you use it, well.
Jess Carter: What's your elevator speech when someone's like, "What even is that? What's data governance?"?
Paola Saibene: Think of a Formula One race, and think of the pit crew, the pit stop crew. So, how good are they? You're running at a really great pace. You're ahead and then you have the crew that just takes four or five minutes changing your tires. You can't afford that. So, what we bring to the table is a methodology that is fresh, dynamic, holistic, really dynamic, that actually makes us be so in sync with a client that we do not slow them down and we give them exactly what they need so they can keep on running the race.
Jess Carter: You do not know this, but I'm going to share it because I just think it's cool. I had the pleasure of being on a test pit crew in the last six months and one, it is bizarre how simple some of those things to do are but one, how hard it is to do under pressure. I dropped the gun so many times because it was like pressure was on and I was nervous. And two, the amount of improvement you can make when you do have these small isolated things that are coordinated, synchronized, critical paths out. So, I hear you on, that's a good analogy of both. There is a bunch of data around Formula One and there's a whole bunch of improvement you see from the business value and the way the data's used. So, it makes sense to me that you use that as an example. And then, how long have you been in love with it? How long have you been doing data governance stuff?
Paola Saibene: I have been in executive roles for most of my career, so specifically data governance and that alone, only for about a couple of years, but I have had to run governance programs all of my career. So, that's why I know that they need to be calibrated well. So, they need to be in sync with innovation and risk management. They need to have the controls fused. So, think of, with your analogy and your experience and what you just went through, how many steps were there from actually grabbing the equipment, to the car? And were you running in parallel with folks? Were you all trying to get to the same part of the car the same time? Did you know exactly... That orchestration is actually controls in frameworks and that could be privacy, that could be risk, that could be Cloud, that could be AI, it could be ML. So, you orchestrate all of those controls so with one movement, you can comply and accelerate at the same time with all of those simultaneous.
Jess Carter: They're going to choose to do that with someone who isn't Jess after they saw my performance cause it was so bad. But I, at least, have a cool picture that I will show you at some point of me next to one of the cars because it was just a really cool experience and it gets your blood pumping and you get excited about the opportunity. And I think that's where you probably have those moments in your whole career. Once people get latched onto the concept and the quick value you can find, it gets exciting and fun. Right?
Paola Saibene: Exactly. And it stays and they pass it on.
Jess Carter: Well, okay, you good to 'Solution on the Spot'?
Paola Saibene: Sure.
Jess Carter: So, in this scenario, Paola and I just grabbed a cup of coffee and some lunch. We're on our way to a new client site. We sit down in the board meeting. They've got tons of data, they are using it all the time. They've got a whole bunch of organizational metrics that they're looking at. It's near real time, it's crazy, it's exciting, it's a fun place to be. Paola has mentioned at a conference, "Hey, what's your governance strategy?" Or, "What is it your data governance programming look like?" And they're like, "We don't even know what that means." So, we're walking in together to have this opening conversation about where do we begin if we care about data governance and should we.
Paola Saibene: Well, one of the key questions to ask from the board and from the C-suite is, "What kind of information is extremely valuable to you and why and for how long?" So, there is information that is tied to a shareholder value or stakeholder value that maybe is good for a quarter or for six months. There's information that is vital for... And I'm staying at the information level now, haven't even gone to data yet. Information that is vital for operations, for business continuity. And then, there is information that is tied to key insights, competitive advantage or citizen driven initiatives that are going to be pivotal. So, let's do some math around those. What are the numbers? And then, finally, another question for them is, "Imagine, God forbid, that you had ransomware hit you, what are you willing to lose and for how long and why?" And once you get to those answers then you've got your structure for where to apply data maturity and data governance, and how to make sure that you are anchoring the value of those data sets associated with that information and mitigating risk, as well as, helping innovate.
Jess Carter: So, when I think about some of the dynamics at play here, too is, in that kind of a dynamic, there's also the question of "Hey, where are you getting some tension in the organization, because," and maybe I'm leading you to your next level of depth or maybe I'm missing a step, so tell me. But there's this level of, I can imagine an executive that says, "Hey, I'm also getting... I'm asking for reports and they're contradictory. They don't all align, people aren't using the same data, I guess, because I've got one person saying that this line of services is amazing and the other one is like that, it's $2 million different, it's 20 million, 200 million different. Where's my source of truth? How do I know what's going... I got data everywhere. How do I know if it's good or right or trustworthy?" Where does that enter into governance programming?
Paola Saibene: That has a simple business fix. So, a lot of folks will try to just slap on a tag that says this is a source of truth, the source of real information system application, et cetera. What you really need to do first, is to step back and get the person that knows the data the most or the people that know the data the most, that deal with it day to day.
Tell us the story. Let's start from the very beginning. How does it get captured? Who, where, for what reason? How does it get collected all the way to the final disposition? How does it get purged or not? And who touches that along the way? How many vendors or outsource parties come in and manipulate that data? How does it get transformed? In which environments? The business rules? This is all a non-technical story. You would be blown away by how few organizations have that. In fact, I can tell you that data scientists normally will say, "I'm a little hesitant because I'm building things and I wish I know the beginning of the story and I wish I knew how it's going to end. I'm passing this on with caveats and disclaimers cause I'm not sure if this is the right data for the information I'm being asked."
Jess Carter: That's amazing and so relevant. And it's interesting to your point, how much of that is not even technical yet. That's like learning the story of the data, learning the story of how the business is running based, what the interactions between the business and the data actually are. What role do all their systems play in a governance program?
Paola Saibene: The roles of the systems, the administrators and systems engineers, they are really the ones that are enabling the conduit and enabling the network, the passing through all that data. But it is exclusively the task of the business, whether it's operators, creators of data, collectors, transformers, to be able to declare the importance of it so the conduit is appropriate, the residents limit so the storage is appropriate and being able to say it is going to end up in figures that are going to change the company radically or not. So now, it needs to have a different level of curation.
Now, the metadata needs to be a lot stronger and a lot richer. And by the way, we're thinking of monetizing that. So now, the legal aspects and the partnerships kick in and the story continues. Imagine of data, we hear this all the time, data as an asset.
I don't believe it is an asset anymore. I think it is data as a product and data as a service. So, if it is a product, you are constantly shaping it and improving it and enriching it and you're making this a very healthy product because the marketplace will demand it to be better and better. That marketplace is inwardly or it is outwardly. And then if you're going to sell it, then it's data as a service to some extent. And by sell, I mean, even sharing with a partner, with a business partner. So it has to have [inaudible 00:10:24] curation.
Jess Carter: I'm thinking about all the different stories in my handful of years of consulting, and every state agency, municipality or private client I've worked with, seems to have such a different story in their data governance journey. What that looks like, what the realities or challenges are. And so, it's really interesting to hear you explain some of this. How do you keep, for this client too, how do you keep data governance feeling like something that's just going to keep exploding and taking up more and more time, energy, budget. How do you make sure that it's a well-run efficient program and you can get, kind of, organizationally, some arms around it to make it stable?
Paola Saibene: So, the business value equation will give you the key data sets and my recommendation is always to stay very narrow on those key data sets and go very deep. So, by going very deep, is get that metadata so rich that it tells a fabulous story, that it course corrects all kinds of things, that it solves mysteries that have been [inaudible 00:11:24] around in the enterprise for a long time and it solves chronic issues. But if you expand that scope of how many data sets, you have to have a compelling reason because now you're stealing time from those data stewards, that could be spending time in the other more valuable data sets. So, the business has to declare what's important now, what's important later and for how long.
Jess Carter: You brought this up in a different conversation. I mean, if we're meeting with this client and they're trying to really seek to understand what we would define data governance programming, in your opinion, what are some of the conditions in which you would say, "Hey, you really need a data governance program," or "You really need to mature it. It's time for some maturity and around some of these things." Like how are you tinkering with that as a consultant?
Paola Saibene: Nowadays, the answer, on my part, would be different from maybe five years ago. We have such a strong reliance on instance and forecasting and predictions, that one of the questions I would ask is, "How well are you doing on your predictions," Right? "Is that over 50%? Is that 60? Is that 80?" And it could that be that you don't really know the whole story of what's happening with the data? So, data governance is about decisions on the data and about rights on the data.
Decisions are very driven by goals and objectives and then, access rights are driven by transformation needs and usability, et cetera.
What you have when you have people ask, "Well, how much or how little sure should I implement into governance?" That's not necessarily the right question. Is your pain point something that you're willing to live with? And if the culture accepts it, then you're willing to live with it. And which ones are the areas where you know have to pivot, you have to change, and just start there. We right size it for them so that they want to come back for more because they see the value right away. There's depth [inaudible 00:13:22].
Jess Carter: So, say they sign at the dotted line, they're like, "Oh my gosh, in this situation we know we need it," or "We know we need some maturity around our data governance."
In your experience looking at organizations that are trying to either, mature it or, stand up a program, what's it feel like? What's the experience for a client? Like, give this client a sense of, "Here's how it's going to feel, here are the times it's going to be really exciting, times it's going to be really tricky."
Paola Saibene: I think that they have a predisposition to think that it's going to be very boring and disruptive, and it's going to take a lot of time so, we have an obligation to make it exciting. It's not a choice. And what we do is, we are very clear on the time commitment. We actually start with logistics, from so and so, just an hour and a half a week and we will honor that. From these other parties, just 30 minutes and those are just emails and decisions. From these other parties is two times a month that you're going to meet and that's a couple of hours. So, that perimeter gets carved and it puts them at ease so that their thinking is not, "Okay, I like it, but how long, how bad is it going to be with everything else I'm holding?" Once we have that perimeter carved out, then we go with that deep assessment of what is important.
From there, we walk away with an exact quantification and qualification of how every single aspect of our data governance activity is going to map to a KPI. If we cannot map it to a business KPI on that activity level then, it's not worth doing. And this is why somebody will be asked a year later, "We funded that, what did you guys do?" Then they would also say, "I know what you did, but what is it worth? How does it help the execs?" And then they'll say, "I don't see the results."
So, you work your equation backwards again, map it to the KPI, be obsessed with showing value exactly where it matters, and then make sure that you frame and you scope the amount of time and effort placed into it and quantify the use of good tools. You can do this manually, it's a nightmare, or you can use good tools and not every tool is good for that organization. Look at the culture. Do they like to read? No. Do they like to watch videos? Maybe not. Just fine tune it to get the right tool for them and they will adopt it.
Jess Carter: That is amazing. Wrap up the 'Solution on the Spot', and wrap up with this client. Is there anything I haven't asked or we haven't talked through that you think would be important to mention?
Paola Saibene: Normally, the regulatory obligations are such that, they may rely on friendly auditors and that they're hoping that they don't get asked certain questions. And those obligations may actually be coming from the security side of the house and be coming from plain compliance. In the case of manufacturing or transportation, they have very specific regulations and many other industries and sectors. But what the data governance function can do very easily is, help them comply with those quickly, by virtue of having active metadata. So, imagine a lawyer that is in charge of running a records' management program. They don't have staff, they know that they have to declare records, and by records, I'm using the legal term here. That normally gets postponed and badly tackled.
Why not have a steward right off the bat? Just put an R on the metadata for that tag and say, "This happens to be a record. And by the way, throw in the InfoSec classification, throw in the privacy regulations and while you're at it, start ranking. Does this make it to a dashboard that the top executive looks at? Now it has a score of X. Okay, so the more that the stewards come in, why am I bothering to come and curate this data because it has a score of seven? So, take a look and spend time." It begins to form into a nice fabric of meaning behind the data.
Jess Carter: You brought up metadata, which we hadn't talked about, yet. And some of the experiences that I've had around data governance are large entities, usually in the public sector, that have both some of their own data but also some federal data, and how those are protected and the rules around how you can use them are highly complex. And they found that, if they had that metadata governance, they could really easily manage and understand what data was highly desired and what the scenarios or use cases were in which they could actually allow access to that data or, how could we design meaningful solutions where people could see it, but it was protected and it was autonamized and you couldn't re-identify somebody.
And I think that metadata piece is really important because if you can start to understand, to your point, and maybe you need to do a quick definition of metadata. To me, just as a pragmatist, it's been data around the data, it's the data around the data that we're using. And so, you have a source system with that data, but all this stuff about, "When was it entered and what is it used for and what kind of reporting..." Having the data in a system, say you are in the public sector, where you can identify which fields directly impact a federal report. That makes someone's life so much easier.
Paola Saibene: Tremendously. Having worked in government for almost 12 years, I can tell you that even down to the level of getting better grants or, fulfilling those grants or, being able to show to the citizens and residents that you have indeed created good outcomes and impact from those efforts.
Metadata is that fuel behind and it describes the data, but it also is able to get quantified and qualified along the way. So, it runs its own story that also allows people that are trying to protect it, rightly so, but up until this point, because I understand that what you're going to be using is a derivative or a calculation, or I understand that you're going to be anonymizing to such a degree that is beyond what HIPAA would require, that is very anonymization and that it can even be potentially converted into synthetic data or add some noise to it. So, the data scientists come in and are able to help with the solutions of that.And so, it takes something actually simple, which is having the right parties talk to each other and tell the story. Think of it as a family dinner time. When you have everybody at the table, everybody tells a story and now we all heard it and we have a picture of what happened but if they're all showing up at the table at different times, then that story doesn't get told. A text doesn't do it.
Jess Carter: Amazing analogy. That's awesome. Absolutely. Well, Paola, thank you so, so much for 'Solutioning on this Spot' with me. This has been a pleasure.
Paola Saibene: Likewise. Thank you so much for having me here.
Jess Carter: Now for the deep dive on implementing a data governance program. The two experts you'll hear from are Dave Haas, executive at Teknion and Jon Sakanai, business intelligence manager at the Colorado Housing and Finance Authority.
In this conversation, Jon shares CHFA's inspiring data transformation and how they now leverage a unique data governance program and framework. Super, super helpful to hear both. I think Paola's stories about themes she's seen across experiences over time, and then hear this really specific story in the depth of the detail here.
Dave Haas: Today I'm with Jon Sakanai. Jon is with the Colorado Housing and Finance Authority or CHAFA, and super excited to have Jon here today to share CHAFA's digital transformation story. Jon, would you mind introducing CHAFA and helping us get started.
Jon Sakanai: CHAFA is a quasi-governmental nonprofit housing finance authority for the state of Colorado. We are involved with housing for low to moderate income Coloradans and the single family, multi-families and business finance space. And we've been around since about 1974, helping with 22 billion invested in Colorado. We've done about 121,000 single family loans in that time. We've helped out 6,500 small businesses and we've, through tax credits or loans, provided at least 70,000 units of affordable multi-family housing in Colorado.
Dave Haas: Jon, I'm going to start by just asking you, really, where did it all begin? Where did you guys start in your journey around data transformation?
Jon Sakanai: Well, when I started at CHAFA, we had a good data operation going. We had a couple report developers, ETL developers, and our data was very useful. It was organized with one legacy data warehouse for one line of business. But it was all hand coded ETL. We had two developers that had to, literally, write out in SQL, all of our ETL transformations for that data warehouse by hand. And then, all of our data was spread out around 300, 350 classic operational SSRS reports. So, the data was very useful to people, but it was slow to deploy. Anything new that came in was very difficult. We only were able to do, perhaps, two or so data warehouse releases per year. And the data was scattered in reports there, all the logic was buried within each report. It wasn't so much contained within the data warehouse.
So, we realized at that time with increasing data and importance and our business was increasing quite rapidly at the time, we really needed a solution that helped us be more agile and unlocked the potential of the skill of our workers by letting them all work on higher level activities a lot more than the maintenance and going in and troubleshooting line by line, SQL code by hand. So, that's when we went out to RFP for a new solution, and that involved implementing WhereScape RED, for data warehouse automation and Tableau for our data analytics. And it was just a complete paradigm shift, really, for us, on transforming to a more modern and agile data warehouse and dash-boarding reporting solution.
Dave Haas: So, that really was the big goal that you guys had from the start was, I guess, I heard to automate, Jon, as well as just to make your operations more efficient. Is that fair?
Jon Sakanai: That is fair, yeah. We're smaller organization. So, any little bit of help we can get to really just enhance and leverage what we already have, is what we were hoping to do.
Dave Haas: And did you guys encounter any roadblocks along the way that were, kind of, stumbling points? And if you did, what were they?
Jon Sakanai: I think, as anyone could tell you, change management is always a bit of a concern and a challenge for any organization that's trying to move along on a digital transformation. So, really, kind of approaching that with finding champions in the business is how we mostly approached that, at first. And we did run into some silo issues when we did that because we had worked with certain champions that were more on board with a more agile approach, but using them as then champions for other parts of the business, we really helped to work through that. And, of course, there's just a new learning curve for skills around our new tools.
Dave Haas: So, what were some of the secrets as far as how you overcame those obstacles, Jon?
Jon Sakanai: A big part of it was kind of doing a lot of road shows for high impact analyses. That's how we really brought in a lot of that buy-in from our leadership on down, was attacking a problem we knew that they had and showing them how much more quickly we were able to do things with our new models.
Dave Haas: What would you say... What's the vision? What's the big picture before we dive into some of the other specifics of where you guys are headed, Jon?
Jon Sakanai: For our BI and analytics teams, we always talk about getting the right data to the right people at the right time and, really, just focusing on continuing to improve our agility and our mastery of these skills and our data, to deliver even more quickly and to get a little bit ahead of the curve to even moving more to advanced analytics, perhaps, and machine learning so we can do a lot more forecasting instead of purely analytics of what happened in the past.
Dave Haas: So, let's shift gears a little bit. Let's talk about data governance or why did you decide to build on a data governance practice and can you elaborate a little bit around that about what you did with data governance?
Jon Sakanai: I think as we were moving along in our data, just our general data and analytics charity levels, and we had so much new data being used by so many people, we started realizing just at all grassroots level, that we needed some better processes to handle it. And also, as you know, a lot of new regulations have come in, in the data space and we wanted to make sure that we were ahead of the curve for privacy and regulatory reasons.
And what really happened is, a group of data stewards came together to talk about, "We need to start governing this data better." And we started having these regular meetings to talk about how we were going to start approaching it, what we might start doing and we all agreed we were doing data governance to a certain level but very informally and, again, and maybe in a siloed and not cross-functional manner. So, the drive came almost from a grassroots effort, but we did perhaps, not quite know how to approach getting started on the entire framework.
Dave Haas: And how did this group of data stewards kind of come about? Were you guys informally, like you said, maybe informally starting to do data governance or you just realized there was a group of people that just realized you needed a formal process around your governance?
Jon Sakanai: That's really what it was, is that the people who were informally already doing these processes, getting together and talking about how we might start to formalize it. We have a privacy officer that was also really involved in helping us understand the new regulations that were coming through. And we realized we couldn't do it just all informally like we had been before.
Dave Haas: Where would you guys say you were along the data governance maturity curve? Was it fairly early?
Jon Sakanai: Yes, I'd say definitely fairly early still. Like I said, we started understanding that we were doing a lot of these activities informally and we probably didn't understand all of the terminology even that went along with it. And we certainly didn't know exactly how to get a long-term framework in place to get a formal data governance program running.
Dave Haas: And you entered into... You and the data stewards basically kind of got together and started to think about, "Okay, we need a data governance program in place." What were some of the problems and the challenges you guys were facing?
Jon Sakanai: Well, as you know, kind of that data trust issue that always comes up with people getting different numbers for the same thing, would rear its head here and there. And we never ran into an issue where it, ultimately, caused a problem with, say, a board or an external report.
It certainly came up in a lot of our processing to where somebody had to come along and dive into it and fix it. And again, once the mandate started coming through regulate and the new regulatory and privacy space, that really drove us to realizing we could be at risk in the future if we didn't start getting a formal process around that.
Dave Haas: So, what was your vision for data governance?
Jon Sakanai: What we were hoping to do was just get a roadmap to understand, "How do we get a formal data governance program up and running?" That's where it became overwhelming from just a simple grassroots level. We didn't know where to get started. I think our main analogy that came up was, we were sitting down at the data governance restaurant, but we didn't know what was on the menu.
So, we needed somebody with some outside expertise who had been through this before to help us focus that and just understand what was available and what we should probably be doing that fits our business.
Dave Haas: I love the analogy at the restaurant that's... I've never heard that before. That's really cool. So, let's shift gears here a little bit. I'd like to, next, talk a little bit about data trust, is a really big topic right now. I want to talk a little bit about a solution called Validator.
I'll start off, just if you're not familiar with, Validator is, essentially, a data quality and testing automation tools. So, it's one of those things that you're going to find is, more than likely, going to be essential for any organization that is managing medium-size to large amounts of data volumes and needs consistency, for testing and for business services to make sure that your data, I think Jon, as you had mentioned earlier, is essentially what you expected to be, right? From source to its destination. So, with that in mind, talk to me a little bit about why you guys chose to use data testing and automation tool.
Jon Sakanai: Remember how I was saying, back in the old days, we hand coded everything and we could only do two data warehouses releases per year.
So, once we were on board with WhereScape RED and a lot of the new data modeling techniques, we realized we could start amping that up to one, two, sometimes three or four releases per month, instead of one or two per year. Except, we realized, [inaudible 00:31:17] need to find some way to trust the quality of that data and to test that data. Validator really is what the tool that came along and helped us get into that, almost a data ops type of agile cycle. Because now, we could use Validator and WhereScape together to automate that data warehouse and the testing of those releases. So, all of that source to target, and that's what really got us [inaudible 00:31:41]. We didn't have to have a person go and manually testing every single one of these releases. We could automate it, we could save those use cases, we could run them in the future.
It really helped us get on that agile cycle. We run all these tests nightly, we run some on demand for releases and we've found quite a few where there might be some type of funky data change in a source system that we weren't aware of, or it's even [inaudible 00:32:04]. Maybe an analyst made a mistake and put in some completely unrealistic number or even, there might be some type of little bug, not that we ever make those, in our ETL process that it catches right away that, again, we were [inaudible 00:32:19] had to avoid or had somebody manually go and test for and we're catching them up [inaudible 00:32:24]. So, that ability to just kind understand what's coming through our data, catch it ahead of time before it's a problem, fix it, and then have now an automated test case that we can just run again every single day until we catch it again in the future [inaudible 00:32:39]
Dave Haas: Thanks for sharing. So, this is the part that I was super excited about. Let's talk about the results that you guys have seen. How has your digital transformation changed CHAFA, in general, at a high level?
Jon Sakanai: Agility and the ability to process new data, make new analyses, and make them quickly at the time the business needs them. That has been our biggest win. Now, before, people just had to wait for data and the expectation is they were going to wait a fairly long time until, perhaps, we can integrate some data, run a report, or run an analysis, and now the expectation is, if there's data available or there's something that we need for the business, they know they can come to my team and we're going to get it to them quickly and in time for them to make that critical decision.
Dave Haas: Wow, amazing. And you spoke to it, about how it's affected internally in your team. How about externally in your customers?
Jon Sakanai: With our customers, and again, it's mostly an internal customer type team, but just that expectation and trust that they have in the data. They know they're getting much more consistency, much better quality in the data and the self-service that it's enabling.
We're able to get a lot of our biggest data users on board and trained on how to interact with our data warehouse, particularly through Tableau. And they no longer have the bottleneck and we need to go to IT, just to find some data and to know that they can trust that data that's coming out. That's been a great win with our customer relationship experiences.
Jess Carter: Thank you for listening. I'm your host, Jess Carter, and don't forget to follow the Data Driven Leadership wherever you get your podcasts and rate and review, letting us know how these data topics and stories are transforming your business. We cannot wait for you to join us on the next episode.
Insights delivered to your inbox