Data Governance Best Practices with Malcolm Hawker

Unknown Speaker 0:00
Hey

Unknown Speaker 0:15
what’s up this is Georgie and the time is right welcome today’s guest strong and powerful. Malcolm Hawker. Malcolm, are you ready to do this? I am ready to do this, sir. All right, let’s go. Malcolm is the head of data strategy at Prophecy. He’s a thought leader in the field of master data management and data governance. Malcolm, tell us a little about your personal lives more about your work and why you do what you do.

Unknown Speaker 0:39
Personal life Well, I’m originally from Western Canada. So I recently became an American citizen about two years ago, which is, which is a big deal. I was here under a green card for well under a well over a decade, almost 15 years on a green card. And I never took the plunge. I never became a citizen. But I wanted to because I thought it was important to be able to vote. And lo and behold, I’m looking down, you can’t see me, I got my first jury notice today. So yes, it would be my first jury appearance ever at age 54, I get to go serve on a jury. So as as annoying as that is, for many people, I’m actually kind of excited about it. That means I’m just more legit, and I’m part of society. So I’m originally from Western Canada, born and raised in Western Canada went to school there. But I did come to graduate school here in the US. And we’re a proud graduate of the University of Florida Go Gators. Live in a bunch of places lived in the DC area. For a long time, I worked for this little internet startup called AOL for over a decade, laying the plastic tubes with with Al Gore and others laying the foundations for the internet. I’ve had an amazing career as an IT leader, as a consultant as an analyst, most recently with a company called Gartner which was a lot of fun. And now I’m the head of data strategy with prophecy. So what I do is I help companies navigate the increasingly complex waters of the data world. We sell data management software, so we help companies kind of wrestle with rather difficult data related problems. So why I get up every morning. And why I do what I do, is really a function of helping companies solve complex problems that on the surface seem really simple and seem really easy. And I’ll talk a little bit more about that. But

Unknown Speaker 2:34
maybe that makes me a little bit of a status, I don’t know, but but the harder the problem, the more I want to help other companies try to figure it out. And the more I want to help other people try to figure it out. So if you’ve ever, you know, had an experience where you called up a support line, and you had called three or four or five other times, but when you called, it was like the first time they heard from you, or maybe you went to a hotel and a hotel that you’ve been many times before, and they treated you like you were a new customer, or conversely, when you went back to a hotel, and they knew exactly who you were, they knew exactly what you prefer to eat, you knew they knew that you preferred a feather pillow. So those are extreme ends of a spectrum, where what end was that hotel didn’t know who you were, and didn’t know that you had a previous relationship there, and the other one who did, and the type of experience that you can have and the two ends of that spectrum. So where my company comes into play, and what I do for a living, I really help companies develop what we call best practices around data management, is we solve for a problem that’s called a lack of Master Data Management. That is master data is the is the nouns that matter most for your company. What are the complex things like customers assets, locations, employees? George, you’d be amazed how hard it is for extremely large companies to understand whether Jeff Smith and Jeffrey Smith are one person or two people, right, you’d be amazed at how hard it is to come up with a single version of the truth when interacting with customers or suppliers. It’s the same thing like maybe you’ve ever wandered down the aisle at Home Depot. And in the nuts and bolts aisle at Home Depot. Maybe this is more of a male example here. But where we’re, you know, the Fed all of these labeling standards, and you’re able to understand because of the picture and the label, this is the exact boat that I want. And there’s one experience within the other one is that there’s no label, there’s no description, there’s no understanding what this thing is, like that’s a bit of a less powerful metaphor here. But getting your hands around data as a company is core to understanding your relationships with customers, with employees with suppliers, extremely relevant in today’s world around supply chain and where your goods are coming from and how long it takes to receive them and all the risks that may be present along the way. So you

Unknown Speaker 5:00
That’s what I do day in and day out. I like helping companies solve those very, very difficult related data challenges. Nice. That’s really well explained. Malcolm, thank you.

Unknown Speaker 5:11
Thank you. That’s not necessarily an easy thing to, to boil that down. So that is well done. Interesting that you’ve the the examples were super helpful. And some experiences I really want. I love going to a really nice hotel. They’re like, Oh, welcome back. Mr. Mr. George, we’ve got all your favorite things. But also, I really don’t want to be tracked in certain circumstances. Right. Right. I think that’s gross. Yep. So I wrote down in advance of our conversation, is there a good versus evil conversation that kept companies need to have? Do you know what I’m getting at? Yeah, there most certainly is, there most certainly is. Because there’s a there’s a fine line between exceptional customer service and getting into something that’s a little creepy. Right. And it’s, I mean, it’s, it does come down to security and privacy and ultimately having control over your own data. Right. And that’s a role that my company software would play and other companies software would play in terms of managing privacy settings and managing what you opt in for and what you opt out of. But at a very, very high level. I mean, there is an interesting academic discussion to be had here about convenience and service and being tracked when you don’t want to be tracked. Typically, it’s a situation where people, it’s good until it’s no longer good.

Unknown Speaker 6:36
Right? And and you know, it’s no longer good when it goes over that line, right? Where you didn’t, where you didn’t allow that use of the data, or you didn’t opt in for something or you didn’t receive a positive experience from from something. So, you know, I think, you know, the rule of thumb here, yes, absolutely, positively, there are privacy regulations in place, there are laws that are in place, and that’s kind of the minimum bar, right? Companies have to adhere to the law, they have to adhere to things like the California consumer protection Protection Act and have to align to other data protection acts. And then if you kind of come up the chain, there, there is more of a gray area, where perhaps you opted into something and you didn’t even know you opted in. It was one of those user agreements that was 45 pages long. And they asked you, did you read the user agreement? And click click? Yeah, I did. Yeah, just let me download the app. So I can go do what I want to do. And maybe somewhere along the lines, you did opt in for something, but you didn’t know that you did, because it was the convenience overwhelmed any desire for conservatism when it came to managing your own data. So there is certainly a bit of a gray area there. I think we’re we’re marching towards a future where there is more self determination over your data, separate issue, probably separate conversation, I do think that actually, believe it or not, blockchain and some advanced technologies have have a lot of promise here around the management of your own profile, the management of your own information, the management of how your information is disseminated. But today, it’s an interesting mix of again, of of convenience that people are opting in for very often, but when a company goes too far, they need to also make sure that they have the capability to allow you to say, hey, wait a minute, that what that wasn’t cool, because I I know, I may have agreed to that in the past, but it’s no longer cool. And I want to control my own data and my own experience.

Unknown Speaker 8:32
That makes a lot of sense. So

Unknown Speaker 8:35
your opportunity, your challenge is to figure out the inner workings of the organizations you’re serving, and to then say, Okay, here’s sort of what we want. And we want to make sure we have all the products on the shelves, and we have happy customers. So how do we how do we bridge that gap? Yeah, well, it’s a great question. So you know, data is a good way to look at data is like its exhaust. And a lot of the data practitioners who may be listening to this, the data science people or other kind of data centric people like me would be like, clutching their pearls right now. Oh, wait a minute. No, because data can be incredibly, unbelievably valuable. Right date data can be you know, there was lots of articles recently talking about data is the new oil. But ultimately, data is exhaust from a business process. So what you said George was accurately, you know, working with companies to figure out the best way to manage relationship to manage experiences and manage data, but that comes through business process management, right? How are you? What is your desired outcome for managing that customer experience when they check into your hotel? What is the experience you’d like somebody to have when they’re buying something on your website? How much privacy how many privacy options do you want them to have? How much control do they do want them to have over this data? And on and on the list is very, very long here. But ultimately, we’re talking about optimizing business processes. So data in its best form here can be used to feed back into the

Unknown Speaker 10:00
As processes to optimize them, because if you’ve got data that suggests, okay, something is wrong here, data is incorrect, the name has been incorrectly incorrectly added, chances are, there’s a reason why, right? There’s a reason why you may have three versions of Malcolm haukur In your customer database, right? Maybe I did that and created them under three different email addresses, because I really didn’t want to receive any more spam from you. Right? So a lot of different ways to optimize his business processes. But But yes, you’re absolutely right. It’s about really understanding what customers want, what customers don’t want. And then building your business around that data is key to understanding that. But I would argue that understanding who your customers are, right, Jeff Smith versus George Smith versus G. Smith, is that one person? Or is that three people? That is the kind of the core building block, that the prophecy really helps focus on and understanding, you know, who are your customers? And what are the depths of your interactions with them? And managing that those core nouns, customers, products, suppliers, locations, vendors, on and on, that’s kind of like the the core building blocks of your data estate within your organization?

Unknown Speaker 11:09
And is that is that the fundamental work that needs to be done? And the solutions are easy?

Unknown Speaker 11:16
I wish well, maybe I didn’t, because I wouldn’t have a job. But the solutions are. In theory, they are they are easy, but in practice, they are difficult.

Unknown Speaker 11:30
The solutions aren’t easy here. Because one can one could argue that corporations need different views of data need different instantiations what we say in the data world, and maybe there’s a reason why you need two different customer records for whatever reason, and maybe there’s a reason why business process a it looks different than business process B classic use case here merger and acquisition when one company buys another company, and where those customers may actually overlap. And we’ve got two different data sets. Now all of a sudden that you’re trying to combine these data sets, maybe one hotel buys another hotel chain, maybe, maybe you know, Marriott buys Starwood and they need to find a way to treat those, those two separate customer sets as one customer set. So there’s a lot of different reasons why data can be kind of become broken as, as it were, or at least disparate, or at least siloed. That’s the most common use cases where there are silos of data across a large organization, they start to they come in are created rather naturally, as companies grow and evolve and acquire other companies become highly specialized within their operations, it’s quite normal for the silence to pop up. But solving for that can be very, very complex, because you have to have some fairly complex business rules to solve for. Right, you need to understand when when you know what we call entity resolution when one thing or when three things and three things are. But one thing is one thing, you need to understand some basic rules around data quality, but you expect how do you know when something is accurate? How do I know if you’re if your address or business address is up to date, and when it’s not up to date? So there are policies and procedures that sit under Data Management that can actually get fairly complex? They sound simple, like, Okay, how do I get correct information around Acme Incorporated? I do a Google search and give me information about Acme Incorporated. But there’s a lot of nuances sitting underneath that. Are you talking about Acme Incorporated? Are you talking about every division or department or operating unit of Acme Incorporated? Example would be Dairy Queen? Give me all the information on Dairy Queen? Well, what do you mean, Dairy Queen as a corporate entity or individual Dairy Queen stores, maybe individual Dairy Queen stores within an individual state, right or an individual city? Or maybe you’re talking about Berkshire Hathaway, which is the corporate parent of Dairy Queen, or maybe you’re talking about all of it, right? So that’s an example of In theory, it seems difficult, but in practice understanding and defining all the policies and procedures and understanding you know, how to get the most value out of that that complex set of data is is really what I do for a living. Why get out of bed every morning. In theory, it seems complicated. And in practice, it’s also very complicated. Welcome, and I just want a vanilla ice cream cone from Dairy Queen. Now that sounds delicious. So do you know why Dairy Queen didn’t know that about me? Welcome. Okay. All right. I’ll do my best.

Unknown Speaker 14:20
Nice. So

Unknown Speaker 14:25
how, how does one get started? Just begin. Oh, what a great what a good question. So I didn’t necessarily set out

Unknown Speaker 14:34
to become a what we call MDM Master Data Management or a, you know, an expert in the field of data management. I didn’t set out to do that. But I’ve been doing this for about 20 years now. And I kind of fell down this rabbit hole when I was I was a consultant. I had just kind of done a tour of duty with a small startup based

Unknown Speaker 15:00
out of Austin, Texas, I was the chief product officer. And we were building some really cool software. And it’s pretty neat. And I transitioned into this more of a consulting role when that when that software company was bought off by another software company, it’s it’s so often happens with startups. And I went to this consulting role. And I went back to my rolodex of people that I knew. And I got hired by this big publicly traded company based out in Northern Virginia, where I previously lived.

Unknown Speaker 15:24
To solve for a question of how many customers do we have? That would that was that was my consulting gig, right? And I thought, oh, man, this is gonna be a layup. Right, like, I’ll just I’ll just pull some reports. So I’ll build some dashboards and this this, this should be this should be a layup, right. And it’s like, okay, oh, wait a minute. Hold on a second. No, it’s actually not. And that was that was like, that was kind of like my, maybe my kryptonite, I guess it was like, again, something that sounds so simple, but it really, really wasn’t because I would go and start having conversations with people in the company. And I’d say, Well, how many customers? Do you have chief revenue officer? And the first thing they would ask is, well, what do you mean by that?

Unknown Speaker 16:06
Is that trial accounts? Is that previous accounts? Is that current accounts? Is that anybody who’s ever given us money ever, or is only people giving us money now, right? And then I was like, Okay, well, all right. Let me get back to you on that. And then I would go ask to the Chief Financial Officer, the CFO, well, how many customers are there?

Unknown Speaker 16:26
Well, we have 100, that wasn’t the answer. But we have 100. I’m like, Okay, well, how do you define a customer and the CFO would say a customer is the people where we have recognized revenue that hits the book this year. Right that like that, like that’s it. But make a long story short, here’s everybody out talk to run, the organization had slightly different definitions about who’s a customer, right. And at multiple levels, it was it was common, everybody had their own definition of customer. Ultimately, it was the CFO who had the final say, because there were reports that were going to Wall Street and a publicly traded company. And those those those were bank on, of course, because they were audited every year. But there were these kind of natural differences that existed across the organization. And, and that’s what I was drawn to was just like, wow,

Unknown Speaker 17:13
this is this could be a real real problem here. Because it could most certainly create the hotel example that I had before, right. And I don’t want that as as a customer, I would want I would want a seamless experience. And I would want the customer service line to know who I was, and how often I’ve talked before and what my previous problems were and all those other things. So that’s really kind of how I ended up getting into the into the data space. Was Was that one engagement where it was like, Okay, this, this, this is right up my alley, people are running away from this stuff, but I’m going to run towards it. Because that’s just in my nature, the heart of the problem, the more I’m attracted to trying to solve for it. I love it. I appreciate the work. So I think we’ve all had experiences where been very, very frustrated, where you’re calling in for the fifth time, and you really wish that they would have a record right on their screen, potentially about the last time we spoke. So Right. It’s interesting, you should use that metaphor because my first job Believe it or not out of grad school was as a customer service rep. Working for America Online. I was actually a technical service tech rep. In 1995, doing tech support for AOL. I helping people get connected to the internet internet in the early days of the internet. So it’s interesting that that’s that’s always an example that I give because I come at that knowledge from from firsthand knowledge when it was you know, hates modems and internet connectivity at 14 Point point 4k baud, which is probably before the time of most anybody listening to this, but it was a it was a lot of fun being a frontline representative working for a large company. I bet a limited number. unlimited number of funny, funny encounters.

Unknown Speaker 18:55
Yeah, no, I heard the cup. The cup holder is broken. Right. I heard that one right. And the couple there was the disk tray that popped out and you plop your AOL disk that you got in the mail because everybody was getting them back then. But my cup holder was broken. Oh, I heard that when I heard all sorts. My funnest experience was

Unknown Speaker 19:15
talking to somebody who had a very talkative pair a parent in the background who pretty much dominated the conversation, who would mimic a lot of the things that this person was saying can’t get online.

Unknown Speaker 19:28
But then we would add live like all of this stuff, like in you, this parent would mimic a dog his parent would mimic all of these sounds in the background and I was on the I was on the phone I was just cracking up this parent just had me completely raptured was fantastic. That’s awesome.

Unknown Speaker 19:45
That’s an interesting study right there. The kind of human being that would want something like that in their house. That’s not me, but different strokes for different folks welcome especially when especially a pet has a very good chance of outliving

Unknown Speaker 19:58
their wrinkles

Unknown Speaker 20:00
Oh, yeah, apparently I mean that’s a lifelong commitment.

Unknown Speaker 20:04
I love it. Welcome. Thank you so much for coming on where can people learn more about you? How can they engage with you and prophecy? Two ways the best way is Find me on LinkedIn. So ha Wk er Malcolm is there’s I think there’s three of me on the planet. There’s three Malcom Hawkins talking about kind of entity resolution and disambiguation. I’ve been thinking there’s only three. So please find me on LinkedIn. I’m publishing content, almost daily, best practices, insights, videos, you name it on how to become more data driven within your organization. So find me on LinkedIn number one and connect with me. If got questions, I will happily engage it’s my job. Second way is prophecy.com. So p r o f i s e prophecy.com. Same thing. We’ve got a ton of resources there about data management best practices, data quality best practices, establishing data strategies, again, becoming more data driven as both a person and a company and we’re happy to help however we can love it. Well, if you do as much as I did, Chanel, come your appreciation and share today’s show with a friend who also appreciates good ideas, find them on LinkedIn, Malcolm haukur, ma L. Colm Hawke er, and then go to prophecy.com or ofise.com. And and check it out. One last question, Malcolm, you’ve been seeing data. How do you feel about data? Either way, I can go either way. I’m not a purist. I think I think technically data might be correct. I think data may actually be correct, but I’m a data guy, but if that makes me I don’t know. I go either way doesn’t bother me.

Unknown Speaker 21:47
Thank you again, Malcolm. Thanks so much, George. And until next time, remember, do your part by doing your best