Compliance Leader and Ex-Marine Patrick McGowan Digs Into the Predictive Power of Data Analytics

On this episode of There Has to Be a Better Way?, co-hosts Zach Coseglia and Hui Chen talk to Patrick McGowan, senior director for global compliance auditing and monitoring at industrial tech innovator Fortive. With a JD, MBA and background in forensic accounting, Patrick discusses how using quantitative methods to analyze seemingly small compliance issues can lead to big human insights.

Transcript:

Zach Coseglia: Welcome back to the Better Way? podcast as we continue our journey to find innovative ideas, disruptors of the status quo, and those who, like us, just feel like at times there has to be a Better Way. I’m Zach Coseglia. I am joined, as always, by the one and only Hui Chen. Say hi, Hui.

Hui Chen: Hello, everyone—welcome back.

Zach Coseglia: Today, we are joined by a very special guest, Patrick McGowan. Patrick, say hello to everyone.

Zach Coseglia: Thank you for joining us. Patrick, why don’t we start by just getting to know you a little bit better. Tell us: Who is Patrick?

Patrick McGowan: I can rewind a little bit on my professional background. I’m currently with Fortive Incorporated, which is an industrial portfolio company based in Everett, Washington, and I’m the senior director for global compliance auditing and monitoring—I’ve been there not quite two years. Before that, I was with Raytheon Technologies, but we were only Raytheon Technologies for about a year before I left. I was there for nine years with United Technologies, and the 2021 merger of United Technologies and Raytheon, became Raytheon Technologies. So, it was kind of the same place for about eight years while I was there, United Technologies, and that last year it had transformed significantly. I’ll say I had some really great opportunities there. I joined that company in a role, I think, at the time was called “fraud manager”—we changed that to “forensic accounting manager.” Within about two years, I had the opportunity to stand up a team—I made up the idea on the back of a placemat and really wanted to do a lot more of this work but felt like we needed more capabilities and bandwidth. Before that, I was with Pfizer in the internal audit department in the corporate audit compliance group for about five years, mostly doing investigation-type work around compliance risks, obviously heavy around bribery, corruption, and FCPA, but some other very interesting things, as well. Before that, I was in the client services space with Ernst & Young in their forensic practice group out of Chicago. And before that, that’s when I was in graduate school, my post-Marine Corps career transformation, where I did an MBA and law school at Indiana University. So, that’s the high-level quick overview of how I wound up here.

Hui Chen: Let me just highlight a few things that actually jumped out at me there—not that I didn’t know this before because, full disclosure, Patrick and I have been friends. Patrick is a former marine (thank you for your service, sir), a JD, an MBA, an auditor, and a compliance professional.

Zach Coseglia: Literally all the things.

Patrick McGowan: Yes, double threat—I can sing and dance.

Zach Coseglia: Patrick and I didn’t actually meet while at Pfizer—although we share that on our resume—we met because Hui used to organize these salons of like-minded folks who shared an interest in a particular topic, and we met years ago when Hui organized one of these at AB InBev with Matt Galvin (who’s now at the DOJ) with the two of us and with just a number of other folks who were living in the compliance space and interested in analytics. We’re on this journey for Better Ways, and analytics and data have very much been a part of my search and journey for Better Ways, and it’s been part of yours, too. Tell us about that, Patrick.

Patrick McGowan: There’s a lot to unpack there because, as I started learning in the profession and in this field, it did strike me at some point that the two real lesser-explored frontiers were this area of analytics (call it what you like: data science, data analytics, compliance analytics), and I’d say behavioral science. And there’s a really important coupling between those two areas, but I could see that other disciplines had been doing the solutions. Even if we just talk about the analytics side of things, there are other disciplines and other industries that were doing so much more than I felt like we were, so it seemed like we were leaving a lot of money on the table. I think people have maybe a bit of a stigma about auditors and accountants and think that it’s their strength in math and their love for mathematics or something that gets them into that—it really isn’t. My goodness, have marketing professionals developed some bat speed in the analytical space and used that in incredibly impressive ways that I feel like run laps around some of the things that I’ve seen in the compliance domain in some ways.

Zach Coseglia: The most important word I feel like you just said was “mathematics,” because I feel, and I’d love to know if this misconception has popped up in your life, which is this idea that analytics or even data science is reduced to dashboarding. This concept that the thing that we’re doing is all about creating a Tableau dashboard or it’s about a [Microsoft] Power BI, and it’s not. It’s all of the amazing, wonderful, exciting, sometimes complex things that are happening behind the scenes that often involve math. So, how do you define this discipline?

Patrick McGowan: Great question—great frame for it. I think that’s true—how you define it is a tough one, and I don’t know if I can do it, but I’ll say that the vast domain of data solutions/data analytics (I’ve started to refer to it as “compliance analytics” in our domain) is just the use of quantitative methods to solve problems. Some of the most interesting stuff that I’ve seen, and interesting solutions, are pretty simplified stuff. In my MBA program, one faculty member, Wayne Winston, well known around the world for his quantitative methods and things that he’s done, including a book he wrote in 2009 called Mathletics. There was a trend of Moneyball and a new awareness about the use of analytics in sports, which I personally love, and I think is a lot of fun. Wayne, in a couple pages in one chapter, did a great simplified analysis of just trying to answer the question: “Was this NBA referee fixing games in some way?” Wayne Winston did a two-variable analysis by creating an expected value of the number of fouls for all NBA games. He explained the computations—he used sound statistical techniques to create an expected value, and also compared that to which official was officiating a game and whether the Vegas line moved from the opening line (implying that there was a lot of betting on those particular games). It’s, in two or three pages, such a succinct, clean analysis that is very stark in its results. I think the solution domain today, in terms of statistical techniques and data solutions, is so much bigger than what is being applied today—there’s a lot of dashboarding. So, I also don’t want to say you shouldn’t do that, because I think people are getting a lot of functionality out of that, but it does leave a lot of money on the table. The way I see it is that there’s limitations to that—there’s a lot of things that are interesting and worth exploring, but maybe don’t get the attention that they should because there’s so much reliance on these new data visualization tools and stuff like that.

Hui Chen: Patrick, there’s a couple of things that you talked about that ring back to a lot of the conversations that we’ve had. Back in 2010, 2011, I was working with you at Pfizer, and we had a market that didn’t have any issues in the past, or at least for the past few years before I joined, had been quiet on investigations. Suddenly, several investigation matters popped up in that same market. As I was making trips to that market, I said, “I wonder what’s going on here? What is up with this sudden spike of cases?” And someone turned to me and said, “Patrick predicted that two years ago.” And so, that was the first time I had ever heard of somebody doing predictive analytics—again, we’re talking about a decade and a half ago. Is there a way you can help people like me understand how did you do that? What kind of data did you look at to make that kind of predictive work possible?

Patrick McGowan: It was an interesting little evolution, so a couple of things. One is just the reflex that trying to use data to solve problems, focusing on some type of clean quantification and comparison that would be meaningful. I’ll say that, in the client services space, there was, I think, a healthy discussion that we were having about our audit clients and our riskiest audit clients around, “How do we know how effective their compliance program is? Do they have a functioning whistleblower mechanism?” The Association of Certified Fraud Examiners, they do their annual survey—many people may be familiar with it. If not, they do an annual global fraud survey that every year one of the prominent answers is that the leading way that frauds are surfaced is through whistleblower tips, and it’s just been that way the entire time I’ve been in the profession, which always meant two things to me. It meant, one, we want that whistleblower mechanism to function really well. There’s more than just having it—there’s other elements of policy, non-retaliation, having it in different languages, accessibility, and all that stuff that matters. But the other thing that was important in my mind was that everything else that we’re doing is a little bit in the medieval stages—that if the best way that we’re finding stuff is someone ambles through the door and blurts it out, makes you think, “Really? All the auditing, all the other things that we’re doing aren’t having a better effect than that?” So, I just started thinking about, “How can we get ahead of it a little bit?” Because it just felt like we were always playing defense, that things get reported and we’re the last to know whatever was going on—maybe it’s been a long-running issue, and it seemed painful, expensive, and disruptive. I thought, “How would we know where to target better?” Then, maybe we could decide to do something there, if we know where there’s radio silence and stuff like this.

I started with trying to get a clean comparison of where compliance issues surfaced, how they’re reported, and stuff like that. I tried many different modeling techniques that went absolutely nowhere, so a lot of blind alleys before it. Ultimately, a fairly simple, straightforward, two-variable analysis gave me a lot of insight. So, just taking a per capita measurement, it wasn’t enough to just say, “How many issues have been reported?” but having a comparability of that to measure it by the population and having enough people in a given jurisdiction was necessary—I think they had to have a minimum of 100 people or employees or something in a given country and then do that measurement over six years. It’s just, again, a simple quantification, and then, comparing that at first to just the Transparency International Corruption Perceptions Index in an ordinal ranking. So, not the absolute scores that they get from all the different indexes that they mash together to get that, but how they rank the corrupt jurisdictions from one to 180 or whatever that was. When I did that (plot those along the X and Y axis and scatter plot it), it was a shotgun blast. Just looking at that shotgun blast, the couple of things that really stood out to me were, “There are some places where we have nothing going on in six years, and quantitatively, that’s like you’re Lance Armstrong with your compliance program—you’re just winning every time in a really difficult business environment.” To me, that did seem a little Lance Armstrong-like, that it’s not possible—something should be happening here, and it’s not. So, now, how do we shake the bushes a little bit and see, “What is not surfacing? What are our other indicators just insensitive to?” In that regard, it struck me as there’s got to be something there that’s undiscovered, so that was one insight from that.

From there, there were many different permutations of that particular model taking not just that Corruption Perceptions Index (CPI), but there were so many other things that could be layered in there. I started taking that same model and using it that way and started layering in other things that I think most people in the compliance domain recognize that its marketing expenses tend to be the more susceptible, vulnerable dollars in the bribery and corruption space. Looking at quantifications of marketing expense to OpEx or revenue and taking all of those into that model—starting with a fairly simplified two-variable model, but then adding in other elements or substituting other elements—was getting some new and interesting insights. So, that’s how that started. Later on, I named that analysis the “Cassandra Analysis,” after Cassandra in mythology who had the gift of prophecy, but no one would believe her.

Zach Coseglia: I just want to break it down for folks, the original version of the Cassandra model. Along one axis, you had a ratio of reports to the hotline and the size of the jurisdiction or the subsidiary with a minimum number of employees required for it to be in scope. So, that was one axis—is that right?

Patrick McGowan: Yes, with the caveat that it did it that way and with all issues, whether it came through the hotline or not. So, anything that surfaced an issue, whether it was a hotline report, an internal audit found it, or something like that.

Zach Coseglia: Then, on the other axis, at least initially, you used the Corruption Perceptions Index?

Patrick McGowan: Yes, just the ordinal ranking.

Zach Coseglia: You described it as a shotgun or shotgun blast?

Patrick McGowan: Yes, it just covered the page. I was like, “I don’t know if that helps me at all,” because if you’re looking for correlation, you’re looking for this concentration along either a linear axis or some type of defined curve, and you can see the mathematical relationship between a couple of variables that way.

Zach Coseglia: Once you started analyzing the data, what were the insights that you derived from it? What was the human insight that you were then able to communicate to a leader to say, “I think there’s something here”?

Patrick McGowan: It was a couple things. It started, as I said, as a shotgun blast. Can you imagine just the X and Y axis with a blast of data points appearing at random? When I took the mean value, the average value, on both the X and Y axis and put that onto that chart, I said, “We can look at this as maybe four quadrants. We’ve got one quadrant here where it’s a less corrupt business environment, and there’s not a lot of issues being reported. Maybe that’s what we expect, and maybe that seems to be about what we want, but we’ve also got a less corrupt environment and higher-than-average surfacing of these issues—what explains that?” There may be a reason why we’ve got persistent issues that haven’t been fully remediated or whatever. I was looking from west to east on the chart and it would be the more corrupt business environment where we have higher-than-reporting or higher surfacing of issues—that’s probably what we expect, too. Then, I think it maybe prompts the natural questions, “Are we doing enough there to get in front of those things? Are they the kind of things that we would expect in that space?” But at least we could be confident that we’re hearing things. It’s that last quadrant where it’s a more corrupt business environment and low reporting over time—it kind of defies the odds, so it’s the most interesting quadrant in my mind. It’s a little bit like if I flip a coin ten times and I tell you that I got seven heads and three tails, no one should be too surprised. But if I flipped a coin 1,000 times, and I told you that I got 700 heads, there’s no way—there’s a regression to the mean quantitatively that’s just impossible. It just seemed unlikely that we have the same kind of, let’s say, comparable business operations in places that have real challenges in their business environment, that we’re that clean and that we’re just doing it better. It strikes me as that’s defying the odds in a way that’s a little bit surprising, and so, much more likely it seemed that we ought to be looking there. I think for experienced business leaders in compliance or not, everyone recognizes these things just don’t get better with age, that if they’re left to metastasize for long periods of time, it just gets worse and worse. As a business leader, wouldn’t you rather focus on driving the business than dealing with these compliance-issue headaches as they get worse? And so, I think there was at least something about that that didn’t require complicated mathematics or computation for that notion to be accessible, appealing, and to see the utility of it.

Zach Coseglia: What I love about this example (and this also goes back to this point about analytics, compliance analytics, being more than just a dashboard), is that it’s not always the case that that visual has all the answers—sometimes that visual is there so that we know which questions to be asking.

Patrick McGowan: Yes, I totally agree, Zach. This entire domain is as much about the interpretation of the results of whatever computation or model you’re doing as it is the model or the results themselves. I think that’s where the real meaning is and maybe the starting point to, “How do we use that information to make our organization better?” It shouldn’t just be a science fair project that we show to our friends who like data science—it should be something that says something meaningful to functional leadership or general management.

Hui Chen: I also love this for a number of other reasons. One, it’s about storytelling—you have data, you plot it out in something that didn’t look like there’s any sense at all, but then you really try to dig in and figure out the story that this data is telling you. The other part is that I love this very scientific thinking about, “I’m going to try these different things.” Frankly, I love the fact that you even gave it a name. I have to say, Patrick, I don’t know if I told you, but when people started telling me about the data analytics work that you were doing, they would add this description that, “Patrick is something like a compliance ninja. He just does this in his spare time.” There’s another project that you did that I know of that I actually have urged you to write up as an academic paper, and my understanding was that that actually helped you show your management, “Talk about leaving money on the table.” So, tell us about that other project that you did.

Patrick McGowan: Yes, I know the one you’re talking about, and it’s amazing to me sometimes the mileage that I got out of that particular analysis. I remember in that case, I was new at the company, and to be perfectly honest, they didn’t know what I should be doing, and so, I started out just with a vague assignment of some expense review. What I was learning about this expense data in this case was a couple of things. One was that in this transactional data, people could use their company credit card, or they could use cash or a personal credit card for any expenses, so I was either getting system-generated transactions or manually entered expense transactions—and we’re talking about hundreds of thousands of lines of data in this case. Another interesting element was that they had a receipt limit, like many companies do, and it was $25.00, so I just was curious which expenses were most common under that. But then, when I sliced it by the manually entered ones, definitely the most were the self-dinner meals, so I did a histogram, which very simply is a type of frequency chart. I was doing a bar chart that would show the number of meals between $0.00 and $4.99, the next bar in the chart from $5.00 to $9.99, and so on and so forth so that I had everything, and importantly, the count of the number of self-dinner meals that were $20.00 to $24.99. It sounds pretty boring so far, but when I separated the system-generated transactions, something that’s coming from someone’s company credit card, and did that bar chart along these $5.00 bins, I got a nice smooth curve kind of like the sledding hill that I used as a kid. Data nerds like me and Zach know it’s a log-normal distribution (this long sloping to the right), so you see that meal expenses have a predicable distribution of the frequency in these different amounts.

Then, when I did that with the manually entered transactions by comparison, it looked very different—it was a steep slope in those lower dollar amounts in those bins, but really peaked at $20.00 to $24.99. Maybe some people think, “Isn’t that obvious?” But again, the quantification and comparison are important here because it’s a little bit like flipping the coin and it doesn’t work out 50/50. We’re seeing a stark difference in the frequency that people have a self-dinner meal that costs between $20.00 and $24.99, but only if they manually entered the amount—if it’s system-generated, that’s not true. So, to me, this was a really surprising result—there’s a very different behavior that’s implied by this. In my mind, food costs the same whether you manually enter the amount, you use a credit card, or anything else, so this just didn’t seem right. I started looking at the other expense types that also had a lot just below the receipt limit, and most of those didn’t look the same—some did, and some didn’t. For example, just to contrast office supplies, that was one that also had a lot of manually entered transactions just below that receipt threshold, but there was no similar stark difference between the frequency that had a cutoff at the receipt limit. My explanation, looking at the evidence was, “If I have to buy office supplies, I need whatever I need—if I need toner cartridges, I’ve got to buy them. But if I’m buying food, I can always go get a $5.00 foot-long and say it was $24.99 or something.” There are low-cost substitutes all over the place—but for some things, there are not low-cost substitutes. I did so many permutations of this and it continued to reinforce the proposition that there’s probably a fair amount of dishonesty in the incremental difference where people are manually entering a transaction or just keeping it under the receipt limit versus a system-generated one.

Ultimately, I did this analysis with, I think, something like 1.3 million meal expenses, and if I took everything that was not denominated in U.S. dollars, it was something like 63 foreign currencies. It had the exact same model characteristics, and it had that threshold avoidance behavior, if you will—people that are using a manually entered expense just below the receipt threshold but in any one of 63 foreign currencies, and it looked exactly the same. You’d see the smooth curve with system-generated transactions, and you’d see that disrupted curve, if you will, for manually entered transactions—and that just shouldn’t be true. In just a mash of dozens of different foreign currencies. $25.00 shouldn’t mean anything, but it did. Once we had that insight, it was like, “It’s not all of the employees doing it.” And when I broke it out by business unit and I charted it out, every chart looked the same in every business unit—you could see that something is going on there. And as you excluded other possibilities, like people just losing receipts, it became pretty convincing that there’s a lot of people who are maybe cheating a little bit. So, this wasn’t necessarily a huge amount of money for any one person, but it was something—it showed a cheating behavior that was pretty clear.

Zach Coseglia: What I think’s really interesting about this, though, is that even over the years as we’ve pushed back on it, there’s this tendency to think about rules, to think about compliance as a series of rules, and to then build even our analytics programs around a series of rules. I think one of the takeaways, for me, from this example is that there are probably a lot of programs out there that think about this issue as, “How many people are actually exceeding the threshold and not submitting a receipt?” Or, in other scenarios, “How many people are exceeding the amount that they’re allowed to spend, full stop?” Or, “How many people are paying a certain number of consultants more than they’re allowed to pay them?” What you’ve identified here—and this is the phrase that I love, “threshold avoidance” (it’s not “threshold violation,” it’s “threshold avoidance”) is not just modeling data, you’re actually telling a story about human behavior, which is far more interesting.

Patrick McGowan: That’s really the point, that there’s this coupling with human behavior (there’s

Hui Chen: What I got out of that was that you use data to find a story about how humans are behaving in the company, and then you use that behavior to try to identify other areas where this type of behavior could create even more risk than just a few dollars from dinner. It’s the data leading to behavior, and behavior identified greater risk, and then an evolving cycle of allowing you to apply this in different ways to find different risky behaviors. I find that fascinating.

Zach Coseglia: I do too. But it’s interesting because I don’t know that the Better Way here is analytics—maybe that’s the bigger Better Way, but the other Better Ways that I’m hearing from this are you said, “I was curious.” You were curious, and you did something about it—that’s a Better Way. Let’s actually satiate our curiosity and explore data in ways that help us answer important questions.

Patrick McGowan: Yes, I’m 100% with you, Zach. I think what you said, Hui, also is really important and interesting. Certainly, don’t compare some of the things that I’ve done to this, but I remember reading a great book called Micromotives and Macrobehavior by Thomas Schelling. Maybe not a lot of people know him—he’s a Nobel economist, so he certainly did some important work. If you read that book, there’s lots of really interesting stuff in there, but the intro to the book is enough, I think, to highlight this point that you’re making, Hui. He describes, as I recall, that he was just standing in an auditorium and people were filing in to sit and listen to his lecture, and he started just making little observations about the way that individual people chose their seat in the auditorium, whether they wanted to be next to someone or didn’t want to be next to someone, wanted to be in an aisle, wanted to be close, or wanted to be further away. He was looking at this and it was occurring to him that there was just the net effect of all these internal rules that everybody was going by, but then there’s an aggregated effect, and that’s the micro motives and macro behaviors that Thomas Schelling then ultimately made some incredible observations about population movement and other things, and that was kind of the origin of agent-based modeling and some powerful analytical techniques and models. So, I think that being curious is an incredibly valuable element in this. I think for someone that isn’t really interested, it’s not going to work—I think you’re going to have a hard time with all the challenges that you have in this business and in this field. You really have to have something of a relentless curiosity gear that helps take you to where you’re going to find some value add in this space.

Zach Coseglia: Let’s talk a little bit more about this, because I love the phrase “relentless curiosity.” I think that’s definitely a quality that you want in the people who do this. I also believe that part of the reason why we find ourselves still at the earlier stages in the application of analytics in the compliance space is because we don’t have the right people. Relentless curiosity is part of that equation, but there are other skills that you need, as well. I’d love to hear a little bit more from you about the skills that you think we need to be able to do this work well and right.

Patrick McGowan: It’s a very interesting point, because it does feel like there’s a lot of lawyers and auditors in this space and, like I mentioned earlier, I really was starting to feel that it was the data analytics and the behavioral sciences, those were the two lesser-explored frontiers in our field. I think people are doing amazing stuff with this, and there’s probably people in our field doing some remarkable things.

Zach Coseglia: Gosh, if only someone would create a, I don’t know, lab, like, in a law firm.

Patrick McGowan: How about that? And some people have.

Zach Coseglia: How about that? We may have just saved this, folks.

Patrick McGowan: There are some really great things, no doubt, that people are doing here, but if you look at the broader practice, it’s certainly not everywhere, and it’s certainly maybe not predominant in this space. Still, I think, the solution domain in both the behavioral science and the data science space is so much bigger than what we’ve applied to our problem domain as compliance professionals in the aggregate. So, what do we do about it? It’s a great question. Hui and I have talked about this quite a bit in the past, and it still strikes me that after all this time in this field and going to many of the conferences, there’s many that are good, but there’s many where sometimes I’m just feeling like we’re just mulching over the same things that I’ve heard—there’s a little bit of calcification there. And just by contrast, about ten years ago or so, I started going to a different conference. I got into a data analytics working group in my company that led me to a group of statisticians working in one business unit—they lured me into attending the Joint Statistical Meetings (JSM), which is an annual conference, I think, primarily hosted by the American Statistical Association, but many other counterpart organizations from other countries, like the UK, India, and every other country that has a professional association like that. There are approximately 6,000 attendees that attend this six-day conference—all statisticians, mathematicians, data scientists. And there was nothing in the compliance domain in this—out of 4,700 sessions, there was nothing in the financial or accounting fraud space or anything that you could say “compliance.” Even the business and economics section wasn’t predominant in this. The thing that was amazing to me was I really felt like we were a little bit backwards in our profession, that we’re getting all of our compliance professionals together and we’re having panels of mostly lawyers and a couple of maybe auditors or accountants. I feel like there are so many Ph.D. candidates and professionals that are excited about data solutions and novel applications of them and are ready and willing to go deep into those, exploring those data techniques, so in a way, I’m a little bit surprised that we haven’t found our way to merge those.

Hui Chen: We actually had contemplated at one point about proposing very concrete problems to bring to the Joint Statistical Meetings and say, “We come from this other planet called ‘Compliance.’ We have these types of problems. We have these types of data. How can you help us?”

Zach Coseglia: All right, Patrick—this has been such a good, amazing conversation You’ve identified so many Better Ways: the use of data analytics, the importance of curiosity. But now, we want to have just a little bit of fun—I think we prepared you for this. We’re a very human-centered organization here at R&G Insights Lab, so we want to put the human at the center of the conclusion of each episode of our podcast. We’ve created our own Proust questionnaire—for me, it was very much inspired by James Lipton and Inside the Actors Studio, Vanity Fair and a whole bunch of others. So, we have some questions to ask you to get to know you a little bit better. Question number one—you get to choose which one you want to answer, so either: A) If you could wake up tomorrow having gained any one quality or ability, what would it be? Or, B) Is there a quality about yourself that you’re currently working to improve, and if so, what is it?

Patrick McGowan: I guess I’d say A), and can I say the Vulcan mind trick or Jedi mind trick? That’s one I would absolutely like to make use of.

Zach Coseglia: We are so on the same page because everyone else says, “I’m going to give something that could actually happen.” Although, I think me manipulating time and you getting the Vulcan mind trick is about as likely as Hui singing opera.

Patrick McGowan: Mr. Spock would’ve made a great compliance professional in some ways. He’d get his way in the organization.

Hui Chen: Question number two is: Who is your favorite mentor? Or, who do you wish you could be mentored by?

Patrick McGowan: It’s someone who is deceased now, but Luis Alvarez. Maybe people don’t know Luis Alvarez, but he was a physicist who was awarded the Nobel Prize in physics. I just remember the first time I encountered him—I was reading the book The Cuckoo’s Egg by Clifford Stoll. I’ve read that book a couple times—once 30 years ago and once about 10 years ago. Clifford Stoll starts with that he’s fallen in the cracks in the academic space a little bit—he describes that at UC Berkeley, he gets stuck trying to solve a $0.75 accounting error. It ultimately leads to an epic journey and discovering that a hacker was using their computers to break into the Department of Defense at the time, and he helps unravel it. But at one point, he’s frustrated with this problem—no one else cares about it. He’s sitting at a table with Luis Alvarez, this paragon of science, and he’s complaining and venting that, “Not enough money. No one cares. I’m off track on my career.” And he sets him straight in a paragraph that Clifford Stoll accounts, where he says, “Forget about all that other stuff. You’re never going to have enough money. No one’s ever going to care. You’ve got to run faster than this other guy.” It was just the kind of tough love of someone who has profound achievements in their rearview mirror that gives them a point of reference and the perspective on, “You’re wrong about that.” I think that’s what you want out of a mentor—it’s not your champion; it’s not your wartime consigliere—it’s someone who is going to help you in critical decisions at times like that and steer you right, where you are clouded by your own biases and lack of perspective, and someone else who has that and takes an interest in you.

Zach Coseglia: Amazing. What is the best job you’ve ever had?

Patrick McGowan: It’s got to be when I was in the Marines. There’s probably nothing else that has had quite that sense of camaraderie, shared purpose, and my love for our country. It was certainly not the easiest job I ever had or the most glorious, but it was awesome. What a great organization and great people—and people that are in your life for the rest of your life, and people that really understand taking things to that next level, pushing your limits, and the shared sense of mission objective. So, yes, absolutely the best job I’ve had.

Hui Chen: What is your favorite thing to do?

Patrick McGowan: In the autumn, in deer season, out in the woods—whether or not you actually see any deer on that particular day—just having a direct connection with nature and just the smell, sounds, and tranquility of being in the woods like that. I’ve had really great experiences with one of my old marine buddies going out in the morning on a brisk November day and coming back to his cabin, having some country cooking, camaraderie, and fellowship of all that—that’s pretty awesome.

Zach Coseglia: This is a related question, which is, “What’s your favorite place?”

Patrick McGowan: Sometimes, I also like when you have that opportunity to have a great evening out and appreciate the arts and maybe go to a nice upscale dinner, but then go see the symphony, or something like that, or the theater. I really have enjoyed over the years things like Tanglewood, the Boston Symphony in the Berkshires—that outdoor concert is just incredibly fun. So, seeing people that have, again, a dedication to excellence, in my mind, in a way that the live performances of those masterworks and stuff like that is just something to behold.

Hui Chen: What makes you proud?

Patrick McGowan: Professionally speaking, I would have to say, recently, some of the people that I’ve worked with in this business, and especially some of the ones that I’d recruited or pied-pipered onto my little team at my last company that have gone on to do other things that are just remarkable and seeing them develop and take on new levels of leadership, capabilities, and new solutions. I’ve got to say that is just awesome, to see stuff like that take place.

Zach Coseglia: That’s really great. We’re going to go from that, the depth of that response, to the shallowest response on this quiz, which is: What email sign-off do you use most frequently use?

Patrick McGowan: If it’s someone like a fellow military personnel or a marine, I use “Semper Fi” a lot, but most of the time, I use “Best” or sometimes “Cheers”—those are the most common ones for me.

Zach Coseglia: And finally: What word would you use to describe your day so far?

Patrick McGowan: Earnestly, I’d have to say “inspiring,” because I love talking about these things, especially with Zach and Hui, who are people that are similarly interested in this and get juiced about exploring new solutions. That’s the most fun part of this line of work. My job is like any job I’ve ever had—you may not love every task related to the job, but you may love your career and your role. And it’s when you get to do the most exciting parts of it and think about, “What are the possibilities here, and what could we be doing?” To me, that’s an exciting element of this.

Zach Coseglia: You’ve been so generous with your time, Patrick—thank you so much. Thank you all for tuning in to the Better Way? podcast and exploring all of these better ways with us. For more information about this or anything else that’s happening with R&G Insights Lab, please visit our website at www.ropesgray.com/rginsightslab. You can also subscribe to the series wherever you regularly listen to podcasts, including on Apple and Spotify. And, if you have thoughts about what we talked about today, the work the Lab does, or just have ideas for Better Ways we should explore, please don’t hesitate to reach out—we’d love to hear from you. Thanks again for listening.