Maintaining a Strong Data Foundation to Drive Business Results
Introduction
Michael Daehne: Hey y’all, welcome to Inflect. I’m Michael Daehne, and on today’s show I’m joined by my friend and former colleague Justin Nguyen. Justin is currently a senior data leader at the Home Depot and has previously served in data leadership roles at organizations like Chewy and 7-Eleven.
On today’s show, Justin talks about how data engineering is at a critical inflection point, about the importance of organizational culture and intellectual curiosity in the teams he’s led, and about some of his favorite AI hacks for parenting. I hope you enjoy our conversation.
Opening & Background
Michael Daehne: Hey, Justin, how’re you doing?
Justin Nguyen: Hey, Michael, I’m doing well.
MD: It’s great to have you on the podcast. Thanks for taking time. Glad to connect and be able to share some stories and experiences today. And obviously you and I go a ways back and have worked together, but for our listeners who may not know you, will you start by telling us a little bit more about yourself, your career journey, and your current role?
JN: Sure. So first, thanks for having me on. It’s always good to connect with old friends and colleagues.
My name is Justin Nguyen. I’m a data professional. Gosh, my journey is a long one, so I’m just going to hit the highlights. Formal education was in engineering. I worked for a little bit as a software engineer and a data scientist, did a little bit of consulting.
Fast forward several years. I worked at 7-Eleven. I was a senior director and head of enterprise data, supporting all of their data initiatives across the enterprise. And then, after that, I joined Chewy. At Chewy, I also led data engineering and analytics, supporting mostly the customer experience and customer service organizations.
And then, as of about a month ago, I joined the Home Depot as a technology director and leading enterprise data and analytics, focused very heavily on data engineering and our data warehouse. So yeah, I’m again happy to be on and I’m looking forward to the discussion.
MD: Yeah, that’s awesome. I feel like your career journey is paralleling my life. You know I’ve always been a 7-Eleven customer. We got a dog around the time you went to Chewy. We bought a house around the time you went to Home Depot. So, I think I like I’m following along doing my best to support your employers.
JN: Yeah, and I’m doing my best to support your lifestyle.
MD: There you go.
Data Engineering’s Inflection Point
Michael Daehne: Well, you mentioned data engineering and as you know, there are so many podcast conversations out there right now about AI and data science kind of tends to get the glory. It’s a little, a little bit more sexy, I suppose, than data engineering.
But you’ve talked recently about how data engineering is at an inflection point of its own. And I think you and I have both obviously seen how critical that is to the broader AI puzzle.
I’d love to hear more about your perspective on, on the role of data engineering and maybe the evolution in that space.
Justin Nguyen: Yeah, certainly you know, data engineering and data science and AI, they, they go hand in hand. I used to be a data scientist myself, and it’s kind of funny now that data science is getting all the attention, I’m moving away from the spotlight and more into data engineering, but, like you said, data engineering is critical.
You don’t really have data science or AI without the engineering, without the data that’s feeding it and tuning it and training it. I think data engineering is at an inflection point. It’s an area that doesn’t always get the, quite the spotlight, get the sizzle that AI does.
But we’re at a point now, I think, where the field has matured quite a bit and it’s no longer about going forth and conquering. Especially compared to some of the more innovative areas like AI, you know, data engineering, it’s established. It’s been there for a while. There’s a lot of tools and processes in place.
It’s no longer about going forth and conquering the unknown as much as it is kind of maintaining and establishing and governing and managing. And I think that’s actually the harder thing to do, you know, when you’re going out and you’re building and you’re connecting all of these end points together.
You know, certainly that’s exciting and fun, but you know, the harder thing, and I think where you kind of start to see things fall is when you step away from the empire building and go into this empire management, empire governing mode, where you’re trying to track and manage, keep things efficient and effective, also keep up with all the changing dynamics of everything upstream and downstream of you.
I think that’s the harder thing to do. And also, just something that’s not necessarily a recipe that anyone’s figured out yet. It’s a new field, but it’s not too new. And there’s still a lot of challenges in that space that everyone’s still figuring out.
MD: Yeah. And I think you hit on this idea of management process, data governance, kind of whatever buzzword you want in that space. But I find that is, that’s challenging because it’s less tangible than saying, ‘Hey, we’re going to build a data pipeline,’ right. Or we’re going to build a connection from source system to data platform or whatever the case may be.
So, it’s kind of hard to wrap your head around, but as you noted, so often that’s the harder problem to solve, but the more value adding problem, if you can solve it. To ensure that your data is high fidelity and trusted and you have built, built an empire that’ll last instead of just building an empire.
JN: Exactly. It’s multifaceted. It’s not just an engineering problem where you need things like monitoring and alerting and defensive programming. There’s also a lot of processes involved with how things get changed and if things even get changed, if there’s some kind of approval workflow or, you know, governance structure in place.
There’s certainly a lot of aspects. There’s the business partnership, right? I mean, data is such a unique entity in that it moves from business to business, and then also moves upstream and downstream. I mean, you have transactional systems that you collect data from, and then you engineer it all together. Create these data sets that could be consumed by business users downstream.
And just the web of endpoints and nodes that you have to connect, and when that empire is already built, trying to keep track of all of those moving dots, not just connecting them, but understanding them, understanding when they change and why they change and how you have to adapt, that to me is the inflection point.
MD: Yeah.
Key Enablers of Data Transformation Initiatives
Michael Daehne: And I think, building on this a little bit, something that I’ve always admired about you and obviously enjoyed when we worked together is you have this deep technical expertise, you know, engineering education and brilliant technologist, but you’ve always had this understanding and appreciation for kind of the business side of it and what organizations have to do from a people and process side to get the most value out of any type of technology initiative, certainly on the data side.
And so, I’m curious as you look at the different organizations you’ve been a part of, if there are a few key enablers or common themes you’ve seen that have helped drive the success of large-scale data transformation initiatives.
Justin Nguyen: It goes back to what I said about how data is just interconnected into so many areas. It’s one of the reasons why I like working in data. You know, not only do you get to learn about marketing and merchandising and supply chain, learn about all these different business units, you’re also able to contribute in all of these areas as well.
In terms of the key enablers, it comes down to how well you understand everything upstream and downstream of you. Typically, in large organizations especially, it’s hard to keep up and understand all of these upstream transactional systems that can happen in your supply chain, distribution centers, in your marketing apps that are touching all these CRM applications. It’s hard to understand what’s happening and what data they’re actually capturing and then being able to understand everything downstream of you.
My business users, you know, merchandising, my merchants, what’s keeping them awake at night? Are they worried about pricing? Are they worried about selection? Are they worried about cost of goods sold? Supply chain, are they worried about last mile delivery, you know, late deliveries, never delivered, like all of these different types of dynamics that are happening.
I think, especially in a large-scale data transformation, you need to be able to understand everything upstream, everything downstream, and connect the inputs and the outputs so that you’re able to create that value.
Understanding Business Functions and Drivers
Michael Daehne: Something interesting in your answer there around getting to know the business and getting to understand the drivers of the business and like the different functions.
When you come into a new organization, whether it was in your consulting days or in the work you’ve done at places like Chewy and Home Depot, how do you approach that business ramp up? Like, do you get into the data first and use that to understand where the business value is, or do you come at it from the other angle? I’d just love to hear more about how you, you learn different businesses.
Justin Nguyen: Yeah, I think it’s a tale of two worlds. I think first, you know, there is the business side of it where you have to understand what’s keeping them awake at night. What are the metrics that they’re looking at every morning and that they’re tracking to and that they have to make some basis point improvement or delta by the end of the quarter or the end of the year.
And then you also need to understand your own wheelhouse and your own data footprint. What are your key data artifacts, your key data sets that you’re delivering to the business? How are they being used? How frequently are they being used? What are the key executive reports or key decisions, key strategic initiatives that they’re feeding and fueling?
And then working from there. Especially in large enterprises, you know, when you’re the data warehouse, or the data ecosystem, you have a lot of different forces acting on you. Whether it’s all these different business units. And I mentioned marketing, merchandising, supply chain. Also things like finance, and FP&A, and legal, and accounting. Trying to understand where are they coming from. What kind of altitude or speed do they need to operate at and then understanding your own wheelhouse and then figuring out the gaps from there and that’s what you attack.
The Importance of Organizational Culture
Michael Daehne: So, the other thing, Justin, that you and I have talked a lot about in some of our offline conversations as we’ve been catching up over the last few years is the role of culture in organizations and the importance of positive cultures as something that kind of creates tailwinds for any of these hard change initiatives or large technology projects.
I’d love to hear if you have a war story or a pearl of wisdom or observation about how strong cultures may help or weak cultures may hinder the success of a large change initiative.
Justin Nguyen: Typically, when people think of culture and what drives culture, the finger’s usually pointed towards leadership and, you know, the leadership defines the culture. From what I’ve seen and from what I’ve learned, including from leaders that you and I have both worked with, is that culture is defined by identity.
And so, when you get up in the morning and you ask yourself, Hey, I work at fill-in-the-blank. What does that mean to you and how do you, how do you identify with that? And when I say identify, it’s who you are today and who you want to become the next day. What are you working towards? And so I think that’s really what drives culture.
Where you can find challenge with changing cultures is, How they identify is how they’ve always wanted to identify and how they’ve, you know, always done work there. You know, it’s like, hey, this is, this is how I’ve always done things. This is how we’ve always done it here. And typically the places that identify with the hey, this is how it’s always been, are also the places where the culture doesn’t change either. And so, I think that’s where you focus first is, how do you define the identity of your organization or your team? And why they show up to work at, you know, fill-in-the-blank versus a competitor.
MD: Yeah, absolutely.
Seeking out Intellectual Curiosity
Michael Daehne: A big part of your role at these organizations is building up strong teams. Are there things you look for as you’re interviewing new team members to come on board from the lens of culture that are, that are important or might be kind of deal breaker type things for you.
Justin Nguyen: I like people who are very, very curious and have to understand things. I don’t want to say to the point where it keeps them awake at night, but to the point where it does bother them.
I’ll give you an example. One of my favorite interview questions for data scientists is what’s an eigenvector and why should I care?
[For the rest of us: https://en.wikipedia.org/wiki/Eigenvalues_and_eigenvectors]
And it seems like a very specific question, but it tests your knowledge on a couple levels. First, you can very quickly identify a bona fide data scientist if they know what an eigenvector is, and especially if they can communicate it. The next part of the question is why should I care? That tells you if they understand how to apply their knowledge. That’s critical, especially in data sciences. You know, your knowledge is only useful if you can apply it. And so, yeah, I think in interviews and, going back to the whole building an identity and the curiosity component, making sure that the people that you’re surrounding yourself with have that curiosity to the point where they cannot just understand it, solve the problem, but then be able to apply it in a way that’s useful.
MD: Yeah, I love that. I love that focus on curiosity. I also will just tell our listeners, if you also just Googled eigenvector, you’re not alone. That’s what I was doing while Justin was giving his answer. So this is why I am not a data scientist, but I’m glad no one’s ever asked me that question in an interview.
I think that’s a great, a great way to think about kind of curiosity and how that, how that may show itself in the workplace and maybe be indicative of folks that are approaching things with fresh eyes and a genuine desire to seek to understand, which is so valuable in any line of work, but certainly the work that you do.
Advice for Executives around Data Initiatives
Michael Daehne: One more question and then I’ve got a few fun ones. Having been through a lot of these large data initiatives and seeing a lot of different things, I’m curious if you have any other tips or tricks for executives out there, maybe those that are a little more functional and they may not be as deep in the data, kind of CEO, CFO, COO type, any tips or tricks you might have for them as their organizations are embarking on a big data initiative.
Justin Nguyen: Yes, so I think what I said about working in data and being able to connect the dots between everything upstream and everything downstream, I think that also applies on both sides, not just on the data slash IT side, but also on the business side as well.
The question I get a lot from C Suite is, what’s the value of data? It comes down to the inputs and the outputs. If I pay X dollars amount towards a capital budget or an OSG&A budget, whatever budget it is, now what’s the output I get out of it? That’s a great question. I don’t think that’s purely on the data leader or the technology leader to answer. I think it also requires an understanding of data from C Suite as well to not just understand, okay, well, I get reporting and analytics and insights, but also understand like what is upstream, right?
Like I want insights where I want data on this particular aspect of my business, but what data is available, I think is a fair question. And I think that also helps kind of set reasonable expectations and just understand like what’s the value. Also, what’s the limitations of what my data can do.
MD: Yeah, love that. I think that’s a great way to frame up the understanding like we were talking about before, making sure there’s mutual understanding of not only the work to be done, but where the opportunity is from a business value perspective.
JN: Mm hmm.
AI Hacks for Parenting
Michael Daehne: Okay, we’ve gotten through those questions and we’re now to the question I really wanted to ask. Because you’ve shown me some cool stuff on this front.
We both have young children. I want to hear about your AI hacks for parenting.
Justin Nguyen: I always need more hacks for parenting, so any one of your listeners, if they have hacks for parenting, please send them to me.
On the AI side, I’ve been using AI for hacks for forever, since you and I have known each other, whether I’m out of beer in my fridge or if I need to walk my dog. AI is very integrated into just how I try to make my own life a little bit easier.
A couple things I do on the parenting side. I have a toddler. She is wonderful. She’s also strong willed sometimes and doesn’t want to brush her teeth. And so, I’ve got a gen AI application that generates images. It basically uses Midjourney and Stable Diffusion.
And I have this application that shows pictures of Elsa brushing her teeth ‘cause she’s a huge fan of Frozen. And if Elsa is doing it, she will do it. It doesn’t matter how much her father begs, but if she sees the picture of Elsa doing it, she’ll do it.
We like to have movie posters hanging up in our home theater and so, I also put her into some of these movie posters and they’re hanging up on our walls and so, there’s a Star Wars themed one where she’s holding a lightsaber and there’s like a space battle going on in the sky and, so anyway, there’s a whole bunch of stuff, especially on the art side that I’ve used as a parenting hack.
MD: That’s awesome. Well, I do think it says something about how we’re getting older and the phases of our lives are changing that when you and I first got to know each other, you were using AI to figure out when the fridge needed a beer replenishment, and now you’re using it for parenting hacks.
So, we’ve come a long way.
JN: Equal impact, yeah.
MD: There you go. And yeah, with your permission, we might need to get the Elsa brushing her teeth picture and add that to the blog post for this podcast.
JN: Oh yeah, I have a whole gallery of Elsa with healthy habits. Eating vegetables, brushing teeth, yes, everything,
MD: Oh, that’s awesome.
Lightning Round
Michael Daehne: Well, hey, before I let you go, just a few lightning round questions.
So the first one, what’s the best piece of advice you’ve ever received?
Justin Nguyen: I think the best advice I’ve gotten had to do with things that I didn’t necessarily believe I could do. But I had a really great mentor who told me I could do it. And if they didn’t tell me directly, they put me in a situation or gave me an opportunity to prove to myself that I could do it.
So those are, those are kind of the best advice and equally the worst advice. The things that I could do and I had, you know, a boss or someone telling me that I couldn’t do it. Um, but yeah, those I’ll call them inflection points in my career. I think those really stick out to me because they, they certainly altered the path of my career and what I believed I was capable of.
MD: Yeah, absolutely. And those are the things that I think we, we often as managers and leaders take for granted of what we are implicitly telling our team members by the positions we put them in, or don’t put them in about what they can and can’t do. So, I love that you highlighted that.
What about outside of work? What’s your favorite thing to do?
JN: Favorite thing to do is being a dad. Every minute I get. You know, obviously it’s not always rainbows and butterflies, but even when it’s not that, I just, I cherish every moment. I’m in kind of a golden age right now because the next 10 years, my, my kids will want to hang out with me and spend time with me, they think dad, like hanging out with dad is the best thing ever, which is only true for 10 more years and then we’ll see after that.
MD: Oh yeah, no doubt. Well, I’m only five months into the journey, but I would have to agree with you about it being the best.
What about best book you’ve read or podcast you’ve listened to recently? Other than Inflect, which I know you listen to religiously.
JN: Yes, I do. The best book I would have to say, I had a book recommended to me. It wasn’t recent. I probably read it about two or three years ago. It’s called Loonshots. It’s about these just crazy innovative ideas that had no shot of being successful, but then somehow became wildly successful.
And so it’s obviously a book, which is a big investment. I recommend reading at least the first chapter. The first chapter is on Vannevar Bush and how he redefined our military research during World War II and just is really the blueprint for me when I think about some of the things we talked about earlier, about building a great culture, about engineering and how to build a very strong engineering culture. I think of Vennevar Bush in the first chapter of Loonshots.
MD: Nice. Great recommendation. I’ll have to check it out.
And my last question, what do you think you’d be doing if you had not pursued a career in data and analytics?
JN: Stay-at-home-Dad. Trophy husband. Yeah, I’m still aspiring to do that. My wife tells me it’s not feasible, but both of my kids seem pretty athletic, so we’ll see.
MD: There you go. Well, hey, I hope your one-year-old little linebacker continues developing and can play for Georgia Tech someday.
JN: Go Jackets. Yep.
MD: There you go. Big win this weekend.
Closing
Michael Daehne: Well, hey, Justin, thank you so much. I know you’re very busy. I’m so glad we’re able to, to make time for this conversation. I know our listeners are going to enjoy it.
Justin Nguyen: Awesome. Thanks, Michael. It’s been a pleasure.
Mentioned
Loonshots: How to Nurture the Crazy Ideas That Win Wars, Cure Diseases, and Transform Industries (https://www.bahcall.com/book/)