John King, Senior Manager, Analytics & Insights, Private Brands at Walmart reveals how the largest US retailer makes use of a “firehose” of data. This amounts (according to some estimates) to 2.5 petabytes of unstructured data from 1 million customers every hour.
Walmart’s FP&A unit ensures this potentially limitless information provide instant business decisions for the US retail giant.
First John explains his path to being an FP&A leader, starting as a Geographic Information Systems (GIS analyst) at Tradewind Energy in Kansas.
Here his love and appreciation for technology – which plays a central role at Walmart- was born. John’s passion for technology saw him thrive at his roles at two of Walmart’s FP&A divisions: Realty Market Strategy, and now Private Brands. Private brands – a good that is manufactured for and sold by Walmart competing with brand-name products is a huge source of data and revenue. In fact,18 of Walmart’s private brands do more than $1 billion in sales and its largest name, Great Value does more than $27 billion a year globally.
In this interview John speaks about the intersection of FP&A and business-decisions at Walmart and his career.
- His passion for technology and the tech stack used at Walmart
- The core metrics which Walmart judges FP&A on and what other retailers can learn
- How to get insights for the business “today or tomorrow” through dimensional modeling
- How to keep focus on the most important data in the face of potentially limitless consumer data
- As both a Python and Excel expert, whether FP&A leaders need to know Python, Excel or both?
- The importance of taking logic and analysis to the data – rather than the other way around
- The most important advice for anyone starting in FP&A to succeed
Paul Barnhurst:
Hello, everyone. Welcome to FP&A Today I am your host, Paul Barnhurst aka the FP&A Guy, and you are listening to FP&A Today. FP&A Today is brought to you by Datarails, financial planning and analysis platform for Excel users. Every week, we will welcome a leader from the world of financial planning and analysis and discuss some of the biggest stories and challenges in the world of FP&A. We will provide you with actionable advice about financial planning and analysis today. This is going to be your go-to resource for everything FP&A, I am thrilled to welcome today’s guest on the show, John King, John, welcome to the show.
John King:
Hi Paul. Thanks for having me.
Paul Barnhurst:
And we’re really excited to have you, so I’ll give a little bit about John and then we’ll give him the opportunity to introduce himself a little more. So John went to school in Kansas. He earned his undergraduate in geographic information systems and analysis. And then he also later on earned an MBA in Finance from the University of Kansas. He started his career working in the energy sector as a GIS analyst, and then after grad school, he went to work for Walmart. He works in their FP&A department, and he’s currently located in Bentonville Arkansas. So John, why don’t you maybe just go ahead and tell us a little bit about your background and you know, how you made that switch to finance from geographic information systems.
John King:
Yeah. So thanks again, Paul, for the introduction. Um, and so I did start in information systems and so, um, in particular geographic information systems. And so what that does is it marries up basically geometry that you may think of as like Google Maps, right? . And so in my early career, uh, I used that type of tools to do analysis into glaciers and glacier research and parlayed that into work in, like you said, renewable energy. And so that was a nascent industry at the time. You know, the tooling and the thinking about that, the engineering was all brand new. And so we were, we were, uh, plowing some new ground, as you might say. And so, I got to work with a really interesting group that was very forward thinking and, and an innovator in that space.
I started, I think there were 30 or 31 people, so you could call it a startup. And so technology was at the core of what was going on there. And I was very much part of a team that was bringing a lot of information about how you designed those projects very carefully, make sure the community’s included, make sure that we’re taking into consideration, even things that regulations didn’t require to make sure we were dotting our i’s and crossing our Ts. And so, you know, after a number of years of gathering that type of information we found that we could help inform some of the economics of those projects, right? And do a better job of selecting those that are both operationally the best, but also, uh, those that were financially productive. Right? And so mirroring up of those two things became an important part of how the company moved forward.
John King:
And so seeing that, and, you know, having conversations with folks across the aisle, on our finance team, I thought that was a really cool way to take my career. And so I had no idea that it would take me to Walmart at the time, but, um, went to school at night, right. As I was, I was building those systems doing, uh, those types of things. Graduated, I think in 2019 and then very quickly began looking for you know, another place to kind of spread my wings, to see where we could make an impact. And so here we are.
Paul Barnhurst:
Cool. No, I appreciate you sharing that. And, and that makes sense. And I could see how you know, as you’re working on projects and trying to understand that operational financial, you could become interested in that financial aspect and find that there could be, you know, a natural fit there to make that switch into finance. So that’s interesting. I wouldn’t have thought of it that way, but I could see that, you know, something you had mentioned is, you know, we’ve kind of been chatting back and forth and understand is you’ve always been interested in technology from a young age, you know, and obviously the tech stack is talked a lot about in finance, you hear about digital transformations, you know, technology all the time. So maybe talk a little bit about technology and how it shapes what you do in your work today.
John King:
Oh, wow. This is a huge topic. Right. And there’s no way that I can cover it. And so from my perspective technology is designed to be an enabler, right? And so at least in my mind, there’s this delineation between sort of technology as a tool and technology as a toy, right? And I think a lot of times we get carried away with technology as a toy, we see the flashing lights, we see, you know, all kinds of stuff. It gives you like really neat displays. But when the rubber really meets the road, if you know, it’s being paid for, uh, it’s an investment, right. It’s designed to move the business forward. And so, I’ve really thought about that a lot and how to use technology as a tool. And so behind that, I thought about a lot of things, like if I adopt the technology, I want it to be around in 10 years.
John King:
Right. And so, I learned pretty standard tooling. So SQL is perennial. It will always be there in terms of fetching data from operational systems in terms of preparing analysis and, and serving it up to an application so that people can see it or view it, which is much different than I think a lot of folks in FP&A work today. And I think that’s the conversation that digital transformation really tries to address right now from a lot of different angles. But I think at the core of a lot of it is how do we work as teams in a way that is we can communicate and also work towards a common goal. I know I’ll just give an example. If we’re all working on a similar project, right? Maybe it’s, you know, the business model’s stable, we’re trying to build out our perspective on what the financial productivity of our company, our department or project is.
That should be a relatively stable thing. And so over time, what we wanna do is build up a set of functionality to do that. And if each person’s working in their isolated work stream, it’s very hard to combine efforts. And so I’ve, I’ve borrowed a number of ideas from the software development world, right? Usually versioned, you know, workflows, agile methodologies, sprint planning, those types of things to have a little bit longer term view on what are we building for the future and how do we improve it today? And so around those ideas, you know, the actual technologies mostly fall into them. And so, you know, whatever company’s developing it, or whether it’s in the cloud or on prem, those conversations are more detailed, but eventually we have to get to the question of what does this do for the business? And so, I’ve tried to angle all of my technology learnings towards that. And that’s, I think everybody should do that. It’s, that’s going through a transformation, uh, digital transformation inside their business, or as a consultant or whatever is asking that question. Like, what does this do for the business?
Paul Barnhurst:
You know, I love leading with that question. You know, what does it do with the business? Because it’s easy to get lost in the bells and whistles, or think how it will make your job easier, which is great. But really the idea is to benefit the business as a whole and to make sure it’s adding value. And I, I like the point you mentioned of, hey, is this something I’m going to be able to use in 10 years, maybe it’s five years, whatever that timeframe is, but thinking more than just the immediate need, you know, you gave the example of SQL. And I learned SQL at the beginning of my career. I started out as writing reports as a financial analyst and did a lot of SQL for a couple years. And it was really valuable when I moved into my first FP& a role cause my boss would always call me, Hey, can you get access?
Can you pull this data? Cause he knew I was the only one that knew the system had access to it still and could go in and pull my own reports and he found it invaluable. So it’s, you know, it’s been a good, good tool for me is having that ability to get to the data without having to always go to somebody. So I’m a, I’m a fan of SQL, you know, I’ve used Power Query a lot as well, but having some kind of tool that you can get access to that data versus always just having to have it given to you. Yeah. So, Alrighty. I would imagine, you know, as you talked about tech, I can imagine that, you know, Walmart with just the sheer amount of data they have, has to have, you know, some pretty, pretty powerful tools that they use to look through that data. So maybe, can you talk a little bit about, you know, kind of some of those tools and things you use beyond Excel? I mean obviously SQL being one, but maybe some others.
John King:
So what is interesting is, as you say, SQL, maybe called SQL really is the workhorse of, a lot of the analysis that we do now. There are other tools in our stack, but the data is stored in a way that we need to access it using that language. And so I’ll also give you another idea to, you know, the listeners, another idea to work with, which is, you know, if, if you’ve got analysis you need to do on maybe five gigabytes of data for sure. That can come down to your local machine. But if you’ve got 500 gigs terabyte, 10 terabytes, a hundred terabytes of data, you have to do something with it’s just not possible. And so I’ll say it like this: data because of the size physics apply, meaning like you cannot move it, it’s too big to move.
John King:
And so instead what we do is we take the logic and the analysis to the data rather than the other way around. Right. Leave it where it is, bring a small 10 kilobyte file to the database, bring back a small result set that is in the format that you need to do further analysis or to present it. It really is a huge simplification of that problem. And so beyond that, the stack that we use is, you know, let me divide that up also a little bit further. And typically you have a data layer, a logic layer and a presentation layer, and it’s really useful to break those things up and design, whatever you know, interactive dashboards or charts or tables that you’re going to present to your stakeholders or to make decisions or for the analysis. It’s really important to break it down to those three steps.
John King:
And so for the most part the storage layer and our logic layer are SQL, right? And then the presentation just happens to be different BI business intelligence, visualization tools, because we’re talking about volumes of data here. And the point is to democratize and to make simple, the understanding of what it is that we’re looking at here, whether it’s revenue or costs or something else. Right. And so those BI tools have been made relatively simple. They connect almost every data source. You can think of, uh, whether it is a flat file on your PC or it is a database on a cloud hosted somewhere else. A lot of those tools say Power BI or Tableau are probably the most popular ones, and do a really good job of bringing those things together. And so, um, really that’s the majority of the data stack.
John King:
And because we keep it simple, there’s a lot of folks out there that can use that data stack or that tech stack. And so there’s, there’s a good pool of talent that can do it. And so you don’t end up with some very, very specific set of tooling that you have to hire someone who’s a consultant to do it, or just the talent’s not available maybe locally and you don’t want someone working remote. And so there’s, there’s a lot of thinking that’s gone into picking tools that are somewhat ubiquitous often free. Right. And that, that plays nice across an entire ecosystem of different vendors.
So that’s the stack that we use. It works, it works at scale. And I’ll say it this way, like, uh, if you think of the data that’s being generated, a lot of it just across, across the board is being generated by things like sensors or automatically at a point of sale or an eCommerce checkout site. It’s not people. And so that generates a fire hose of data. I mean, it’s a tremendous amount. And so you do, you have to fight fire with fire. You can’t, you can’t come with a toothpick to a gunfight, so to speak, right. And so that’s what we’re doing is, is basically using the same tools that create the data in order to analyze and combine and present the results.
Paul Barnhurst:
No, and, and that makes a lot of sense in your point of right you are dealing with huge volumes of data. You often have to go to the data, you have to know the layers, you wanna build something that you can hire the right people. So, you know, having a tech stack that scales, that’s understandable, that’s agile, you know, a lot of those, those principles, that all makes sense. And yeah, I can imagine there’s some large, you know, data lakes, databases, data, warehouses, whatever. They may be probably a little bit of everything, which is, you know, pretty standard depending on what you need. So that, that all makes sense to me. One thing, you know, is that I, you had mentioned, you’ve also used Python some in your career. So maybe talk a little bit about how that’s helped you and, you know, would you recommend, is that something you’d recommend to the average, you know FP&A professional, cause there’s a lot of debate out there right now of, Hey, should I learn Python? Should I work on getting better at Excel? And everybody has a different opinion. So kind of, what’s your take there?
John King:
I’ll say this Python can do everything that Excel can do. I’ll just say that out loud. But I think you need to understand what the goal is with learning a tool like Python, right? And there’s others out there. The reason I choose Python is the same reason I choose Excel is it’s almost ubiquitous whether it’s web development, whether it’s, you know, creating or calling on APIs to, you know, send or receive data over the internet or is used to create scripts for making analysis better, or making visual charts to publish onto a webpage. There’s a lot that you can do with it. There’s a tremendous amount that you can do with it. And then there’s some other benefits to it. Like it’s, you know, much easier for, uh, teams to collaborate on text files than it is for teams to collaborate on a proprietary data format, like an XLSX for Excel.
John King:
Now there are some things that Microsoft is doing to allow Excel to be hosted and say one drive and multiple people to edit. But at the end of the day, you can only edit one cell at a time. Right. And so, I mean, you can do it quickly, you can drag, but there’s a lot of hard work. And so, where Excel kind of goes, you know, an inch deep and a mile wide Python should be used for problems that are an inch wide and a mile deep, right. If you’ve got something very specific that happens in a very repetitive fashion or it’s something that needs to access resources on a lot of different say servers in order to get them to communicate, it’s a great tool. And with all of that said I don’t know that I can make a recommendation because each person’s situation is very unique, right.
John King:
But Python is a very powerful tool. It also changes the way, say you learn it, say, say you spend four to six months, you know, a couple hours a day, reading some books, doing some tutorials and some examples finding some data sets and playing with them on your own, trying just some functions or whatever it will change the way you think about how you handle your information. It simply will. And I think that that sort of regimented approach will bring a lot of clarity to how you maybe design your financial models or how you think about combining data sets in order to present them, uh, for a loader analysis. And so I would say, absolutely learn it right. We’re in the 21st century, it’s 2022 today. Uh, those types of programming languages aren’t going away also learning won’t hurt you. Right.
John King:
And so having a new perspective on how you may do it a different way or another approach is never a bad thing. So I would say at the very minimum pick up, pick up a book, right? On Python, hopefully it’s one, that’s less specific. You know, there’s some, uh, books out there about Python as an object oriented language. I would say read that it’s going to be tricky. It’s not going to be something that you read, you know, as a page turner but those ideas will come into play into how you think about structuring models, how you think about sourcing data and combining it for analysis later and building more sustainable pipelines or flows. If you are working with large data sets.
Paul Barnhurst:
Now thanks, that’s helpful. And I like, at the beginning, you said, you know, you can’t provide a recommendation for everybody. It’s a decision they have to make. Right. And, and I knew that when I asked you the question, but I wanted to hear your thoughts about it, because you know, I’ve toyed with learning R and Python and have never fully jumped in, played with it a little bit. You know, obviously I know SQL or SQL, whatever we wanna call it, right. Tomato, tomato. I’ve learned, you know, how databases work. I’ve pulled data from databases. I’ve done some BI with different tools and having that understanding of how data works. So just, you know, taking away programming for a minute, but data structure has changed the way I worked in Excel. It’s really helped me. So I recommend to anyone that they at least understand how data works, understand data structure.
If you go on and learn a programming language, great. I think SQL is a good place to start. It’s an easier one, but you know, whether you choose R, whether you choose Python, whether you choose to become an expert in Excel, for me, understanding data, understanding the limitations of the different systems and how to really work with it is invaluable. And once you really start to understand that, whether you learn the other tools or not, you start to see the value of them and can see why you would learn Python. Like I totally get it, but where I’m at in my career right now, I just don’t have a use for it. You know, I wish 20 years ago, I would’ve learned it when I was first learning, you know, SQL or 15 years ago or whenever it was. Right. Sure.
And so I can totally see the value. And I usually like to tell people start first by really learning how to use Excel and become good at it. Because I feel so many people don’t become good at the tool that we primarily use and they think they need to learn all these other things. And I’m like, okay, you know, figure out how to be good at what you’re doing. And if that requires Python learn Python, if that requires becoming really good at Excel, become really good, but understand how data works, figure out your tool set and develop your skills. And that’s usually kind of how I will present it, but I think Python’s a great one to learn. And I don’t think anyone would be hurt by choosing to learn it. Right? No, there’s no downside in spending the time.
[DataRails ad]
Paul Barnhurst:
So, you know, working with Walmart, obviously we’ve talked a little bit about that, you know, huge company, one of the biggest here in the US obviously employs a lot of people. And you know, one thing I understand is you, you know, participate some with private brands. And so maybe, can you talk a little bit about private brands, how retail companies manage private brands with other brands and you know, a little bit of some of maybe the metrics and how you think about that from a finance perspective.
John King:
Sure. So I guess first let’s talk about what are private brands?. Just for those that aren’t maybe as familiar. So, a private brand just first and foremost is a brand that a retailer potentially manufacturer owns in house under that banner products are developed and then, you know, manufactured based on the positioning of whatever that brand is. So they manage it internally, like, you know, an external brand that a national brand manages. And so they carry a value proposition just like any other. And sometimes that’s a value proposition, meaning, you know, the cost is relatively low relative to what you get. Some others may be more aspirational, you know, depending on what price point you’re trying to target or who your customer is that you’re looking for, or you’re trying to address their needs. A lot of retailers use private brands.
John King:
Sometimes we don’t even know, you know, uh, really big companies like Amazon, Target, Kroger all have a very deep book of private brands. And traditionally, uh, there have been thought of to be a way to squeeze a lot of the expenses or extra costs out of delivering a really, you know, a, a, a good product to the shelf to keep costs down and to compete on that playing field with, with others. And so in the last, probably 10 or 12 years, we’ve seen a lot of private brand development across, you know, a lot of different retail segments or sectors. And that’s what got me interested and involved. And so in particular, you know, I tend to look at sort of a macro view sometimes of the US economy. I’m just interested in what’s happening.
John King:
And the grocery market is a really big one, right? And so I became very interested in why are grocery prices falling? Right? We were actually in a deflationary environment for about eight or 10 years. And it was because of the introduction of among other things, right. Increased supply chain, whatever. But a lot of it had to do with the introduction and the proliferation penetration of private brands into a lot of these retail stores. Talk about just the grocery side that also happens on the general merchandise side. And so, you know, first and foremost, that’s, that’s what private brands are. That’s what they do. It’s kind of the effect that they have in general. Sometimes they are used as a competitive tool, sometimes not, but just in general, when you think about retailing, you think about merchandise mix. And so in different time periods, in different locations, at different, you know, seasons you can think about, you know, you’ve got certain categories like sports, right.
You may have a baseball glove that is a national brand, $110. Something like that recognized brand good quality, you know, it is, you can trust it. Um, and then, you know, there’s a private brand, almost exactly the same product. Maybe even the same manufacturer priced it $59.99 or $49.99 set visually just below or just above it on the shelf or online. And so, um, that’s the, that’s kind of the lever that you’re able to pull when you swing a bat as big as say, a Walmart or an Amazon or a Kroger, you can pull those levers because, the purchases are so large that, you know, it can, it can be a third or half of the production out of a particular factory or an entire organization manufacturing for you. And so the leverage there is really big, um, in terms of the metrics, uh, I’ll, I’ll just kind of cover this at a high level.
John King:
Right? Sure. And so mostly what we’re looking at is working capital productivity, right? And so this is an FP&A podcast. The people listening to this are used to looking at all the way from revenue down to net profit and the cash in between. Mostly what we focus on is turning inventory into revenue, right? . And because Walmart’s a cash business, as soon as you make the trade it’s cash, now there’s some returns, right? There’s some other accounts in there, but for the most part, we’re a cash business. And so that, you know, if you look at the total balance sheet, we’ve got a negative, sorry, the, the income statement we’ve got a negative working capital position and why is that right? It’s because that we take 30 days to pay or 60 or 90 days to pay, but we move that merchandise through our stores and e-commerce sites faster than that.
And so playing that game we need to do a really good job of managing that merchandise because it is our working capital. And so there’s a number of different things that go into that. In addition to that, understanding who our customer is, understanding the white or gray spaces that we can address and, and filling that with, uh, a compelling product that’s on the shelves or online for them to purchase is the other aspect of it. Now there’s a lot that goes into fulfilling that we measure things, you know, about the supply chain. We measure things about commodity prices. We measure things regarding, you know, foreign exchange rates. So there’s a lot that goes into this, right? It’s a global business. It really is. But at the end of the day, it all comes down to managing that working capital position. And how do we make sure that that inventory is turning over very quickly? You know, because that’s really the name of the game
Paul Barnhurst:
Now. And that makes a lot of sense. Thanks for the explanation and sharing a couple, you know, big brands out there, you know, Kroger obviously does it a lot, like you mentioned, Walmart is one that does it, you mentioned Target, you know, Amazon, the one I shop at a lot and think about is Costco, Costco, right. You know, everybody’s heard of Kirkland their brand there and you know, the working capital, it gives you some greater control there. There’s often, you know, some, you may have, you have more leverage on the price when it’s your own private brand. And so I could, I could see where there’s some real benefits and I could see why you mentioned, you know, from a finance standpoint, really understanding that inventory, turn, understanding that working capital, how it’s flowing through all the logistics and just all the dollars that are along that process.
Like you said, you know, in a business that’s retail, such as Walmart and most retail businesses, you know, at point of sale it’s cash, right? Yeah. You mentioned there’s trades and yes, there could be a day for the credit card to, you know, that actual cash hit the bank account or whatever, but it’s pretty much cash at the time of sale for practical purposes. So that makes a lot of sense. So, you know, one thing I’d like to talk a little bit about is, you know, understand you use dimensional modeling in some of the work you do in your job. Could you maybe talk to our audience a little bit about what is, you know, kind of dimensional or multidimensional modeling and you know, why do you find that important?
John King:
Sure. And so, um, I’ll just refer very quickly back to our initial discussion about the tech stack, right? And so we’re dealing with volumes of data that require that we use basically industrial computing systems, right? Uh, cloud hosted computing systems that scale to whatever level you need. And so in order to do the types of analysis that when you do to make reasonable, you know, insights for the business, we have to organize that data in a particular way. Right? And so the dimensional model, which I follow what’s called the Kimball method, there’s others. The dimensional model is a way to break down different types of information or data into one facts and two dimensions. And so a fact is something that like a transaction, right? Let’s just use a sale for example, or an expense, something, you know, a cash flow.
There’s a lot of others that you can do. You know, when you get into the eCommerce world or the online world, there’s a lot that goes into how people browse a webpage, for example, and measuring those things can be used as facts, but let’s just stick to a sale right now for simplicity. And so if all we have is a huge ledger, a huge table of sales, we may want to categorize or group those things into different taxonomies, right? Maybe it’s a department or maybe it’s a category. Maybe it’s a type of outlet like is this online or in brick and mortar, or there’s a lot of other ways that we can cut the data. And so defining those taxonomies are called dimensions. And so inside our data model, we can ask a lot of very, very specific questions about that data about just that sales data to get into some really fine grained business points of view so that we can go take action because that’s ultimately what we want to do.
And because that merchandise turns so fast and because we’ve got other inventory coming, we need to, we need to be able to get to those insights today or tomorrow we can’t wait a week or two weeks to get to it. And so that dimensional model, I think for folks that are coming from the Excel world, or maybe the finance role in general, um, an example may be to take that same sales data and merge in some invoice information or some invoice details from a different event system, a different table. And you may be able to analyze something like your customers, like days to pay. So you can tell who are your best customers, but you can also tell who takes the longest to pay. And you may do a benchmark about that credit quality of a customer that you are serving right now.
And so just as a simple example that I think kind of hits across industries across all of finance, like doing something like that and to be able to have insights into that on a running basis will be really important to me. So, in Excel you do something like a V Lookup to merge those two tables and then maybe a pivot table to categorize them. And so that’s really all that’s happening. It’s just happening at a huge scale. Right. So just to bring everything back down to earth so that I, so we don’t get lost in the clouds.
Paul Barnhurst:
No pun intended, right. Cloud software in the clouds.
John King:
And so, I mean, I, I do, I do think it’s really important. I don’t think it’s for everybody, right? Uh, there are certain situations where it works. The one thing that I would say is, it’s more likely to work inside of a mature company, right? I’m not saying don’t do it, uh, if, if you’re starting new, if you, if you start fresh and, and do it that way, uh, that can be terrific, right. If you just build it from scratch with those ideas in mind, however, sometimes the systems and the data maturity just isn’t there in order to actually have access to the information that you need to build that model. Sometimes the amount of work it takes to do that is more than one person. Sometimes it’s as much as 10 or 15 people to centralize that information before you can even use it for this.
And so early in my career, I did a lot of onboarding for third party data a lot, and there was a whole process to make that data go into a particular central repository. So we could learn from it and make decisions with it. Moving to Walmart I was very deliberate to choose a large multinational because they have the resources, they’ve already identified their data as an asset, and they’ve already begun collecting in a way that’s accessible to the associates in general. And so, um, I wanted to be on the other side of the table so that I could use it rather than collect it. And so dimensional modeling is like Python, right? It’s something that if you want to go deeper and you really want to understand what may go into a digital transformation, it is 100% essential. It is really the only way to get to something like that beyond an off the shelf solution that comes and sits on top which has its own pitfalls.
Paul Barnhurst:
Sure. Now, thank you for sharing that. And I like to think of, you know, as you talked about that great example, sharing Kimball and facts and measures. You know, I think of it it’s understanding the data and how to model it, how to structure in such a way that it scales. And in some companies all you’ll have to do is access it and knowing SQL can allow you to access it. In other places, you may be involved at some level in helping figure out what that looks like. You know, FP&A as a general rule doesn’t necessarily have to get into that day to day, but there are plenty of roles where you can. You know there’s financial systems, roles in FP&A and different things. One of the five roles we talk about, I think it’s, well, it’s seven now, but there’s really five with an FP&A Trends report.
Paul Barnhurst:
One of the roles that they called an architect and, you know, bringing in some, in some ways, almost like a data architect, because you deal with so much data in FP&A. Everything’s converging, and they’re wanting you to analyze it all and being able to understand how to structure it so that you can easily access it. And so that it makes sense is important. And I like your analysis of VLookups. It’s why when I teach Excel, one of the main things I like to teach people is a little bit about tables and tabular data, how to structure your data to a proper analysis in Excel. Now that can be different of how to structure it in a data warehouse or a data lake, or, you know, depending on what system you have, but understanding how to structure data so that you can analyze, you know, large volumes of data more easily is becoming more and more important in the world we live in. Because data’s only increasing. It’s not gonna decrease anytime soon. Yeah. So that’s some, you know, really good advice for people. So, you know, changing subjects here, well kind of a little bit the same, but along the same vein actually. So, you know, obviously you deal with large amounts of data, you know, how do you keep yourself focused on, you know, the key pieces of data? How do you go about thinking about taking something that’s a big set of data and making sure you’re focusing on what’s important.
John King:
Um, and, and this is a really important conversation, I think, and, and it gets beyond the technical systems, the actual insights and really gets to the heart of what I call stakeholder relationships.
Paul Barnhurst:
Yep.
John King:
Who, who am I developing this solution for? And what decisions are they gonna make? Hard stop. Right. And so there’s no way to get to that from a keyboard in front of a screen there just isn’t right. And so we have to, uh, you know, lead our teams and develop processes that encourage sitting down with stakeholders and people who can make decisions and allocate resources and give direction, right. We need to sit down and learn what is your business and what’s working? What, what could be working better? Right. And how do we get to that? And so there’s a, there’s more than a bit of interpretation that goes on here. And so the homework that you should do as an FP&A, or insights person, is to understand all the information that you have access to, generally the formats it sits in, had access it before you ever start engaging stakeholders so that you know what it is that you have to offer.
John King:
If they say I need X, Y, and Z, it’s very helpful to be able to say we have X and Y. We’ll have to go find Z and then be able to give a timeline and to be able to give a level of effort associated with that. And so, you know, even though I’m inside, uh, of a large company, I tend to think of myself as an internal consultant. And part of the reason that I do that is because it makes me think of those folks as clients. Right. And I certainly don’t wanna lose a client, right. Especially if they’re a good one or if they’re a strategic partner and we hear a lot about partnering in finance and so doing your homework upfront, having that relationship, it needs to be initialized, I think or, approached from the analysis side, uh, to offer those leaders in the business, the insights that they need, and then begin iterating feedback loops, build prototypes, ask more questions, come up with new ideas.
John King:
Have you thought about it this way? Right. And so staying focused on a key metric, I think you have to know who is interested in which ones. Right? And so, as personally, as I’m designing or creating metrics, I do it per stakeholder. And then I create something called perspectives, right? And so they’re entry points into those analyses that are easy to navigate, right? You wouldn’t build a website or something like that without having your top navigation page be simple to find where you’re going about and the settings and whatever it is that you need, the blog, you just wouldn’t do that. And so the same logic applies there. And so having a simplified tech stack, right, that’s mostly SQL, that’s mostly simple BI tools that can publish to a web page allows all those things to come together. So we can approach a stakeholder and we can think just about what data we have? What information can we pull and how do we deliver the metrics to the business that they need in a way that’s intuitive to them. And it also gives them a single access point to find those analytics. Otherwise we’re going back and forth email, PDFs versions. It gets very complicated, very fast. And so all of that in the spirit of delivering quickly managing expectations and giving them the stakeholders the insights that they need today and tomorrow, that’s, that’s really what it boils down to.
Paul Barnhurst:
I really appreciate how you focused. You started the conversation there on getting out from under your keyboard, getting, you know, get out from the desk, talking to, and knowing the people you work with, your stakeholders, your business partners, your clients, your customers, whatever you want to call them the people you support and really understanding their needs. Because how can, you know, what key pieces of data you should focus on, if you don’t understand what problems they’re facing and where they need your help. So, you know, it’s funny, we’ve talked a lot about technology in this episode, we’ve talked about, you know, Python and SQL and tech stack and modeling and different things. But at the end of the day, all that is so we can better serve our business. We can better serve our partners. And so if we don’t go that last mile, so to speak of developing that relationship, a lot of it’s for naught, we can’t be the internal consultant, as you said that we need to be.
Paul Barnhurst:
So I, I loved how you brought that back to relationships, cuz they’re so critical to everything we do, especially in finance, being the central hub and seeing so much of the business. If we’re not good partners, we can’t, you know, do the value creation the business needs us to do. So I really appreciated the answer and there’s a lot of great advice shared there. So next, you know, next thing I’d like to ask you here is you look back at your career so far. What accomplishment are you most proud of? So say something you’d share in a job interview if they asked, Hey, what are you most proud of from your career? What would that be?
John King:
Sure. Uh, if it were more of a manager behavioral type question, I think the thing that I’ve really done well especially as I made the transition from a startup that grew from 30 people and then over six or seven years to about 110. Right. So still, uh, I still knew everybody’s first name. I’ll say that. There was a particular way of working there right. There was a level of expectation. There was a level of visibility. Right. And those ways of working and those relationship building skills, uh, I brought with me to a large multinational, right inside a, a division or department that’s a thousand people or 2000 people alone. The ways of working were different. Right. And so bringing that agile, simple, agile methodology and sharing that and installing or embedding that into how, um, insights analytics, you know, uh, are, are put forward that I would say is probably my, my biggest accomplishment.
John King:
If we really zoom out, there’s a lot of things that have come under that heading. Um, I would say that’s probably the biggest thing because it includes a lot of other people besides myself, right? And it comes down to earning trust of not only people that you’re delivering products or, you know, uh, deliverables to, but also earning the trust and working, being a good working partner of people who you’re working along the side and making them smarter and better and less frustrated. And there’s, you know, the list goes on. And so having a very clear plan on how to move forward, bringing those ways of working into the teams that I’ve worked on at Walmart, I’d say it’s probably my biggest accomplishment so far. That’s the behavioral side. There is one small technical piece that I am very proud of.
John King:
And I did this earlier in my career and it actually taught me a lot. It took longer than it should have, but you know, moving into the information systems world where everything’s stored in a database, if we’re gonna go from one or two analysts to five or 10 or 15 or 20, we need to solve the problem of what’s called the race condition. If someone updates information and someone else’s updates, disagree, how do you handle that? Right. And so not just once or twice a day, but say a thousand times a day for years and then information needs to be right, because huge decisions are being made on them. And so I read about 15 lines of Python and SQL combined that makes it, it just is bashed against it. It gets used day after day, still does in that system in order to separate out whose edits are going in, uh, what updates are happening and if there’s any conflict we can reconcile those. And so that allowed the analytics team there Tradewind to scale from three to 20 to 22 to 25 included some contractors that were remote. A lot of different things happened, but it was 12 or 15 lines of code that enabled us to grow in that way. And we simply couldn’t have done it otherwise.
Paul Barnhurst:
No, thank you for sharing. And that’s, I, I loved, uh, how you talked about enabling, you know, people to work and things to scale. Because that’s always something to be proud of as an accomplishment when you help enable, you know, and kind of talking a little bit about enabling, you know, obviously our podcast here is sponsored by datarails and they’re big fans of Excel is they’ve built a, you know, financial planning platform that helps enable Excel that helps address some of those challenges. We’ve talked about a little bit, that version control, that change control, right? Because we all, we’ve all felt the frustration of 10 people trying to work in Excel at the same time or 10 different versions of the file going around. And you’re just like, what’s the right number and you get all done and nobody recognizes the number like, well, that’s not the right number. Well, that’s what was in file number 12. Well, it should have been file 14 Right? And so they help address a lot of those things. So we’re big fans of Excel. So, you know, kind of stepping back to this tool that I think everybody in finance uses is Excel. What’s your favorite Excel function?
John King:
Oh, this is easy. This is the easiest question you’ve asked me all day. I’d say combined pivot tables and pivot charts real easy to get a perspective on those categories. How are we performing based on those categories, maybe over a period of time, right. Month by month, week by week. It is an amazing set of tools.
Paul Barnhurst:
Yeah, no, I’m a, I’m a big fan of the pivot table and you, you love how easy it is to sum, summarize and aggregate data. It’s amazing what you can do with a, with a good pivot table so that not, not surprised. I think you’re the third or fourth person who said that. That’s a good one. So I’m definitely, I’m there with you. It’s a great, it’s a great tool in Excel. So now we’ll get a little more personal here. What’s something that not many people know about you, something they wouldn’t find online, maybe something unique about you.
John King:
Sure. So I live in Bentonville, which is Walmart’s, you know, headquarters city. And around here there’s a lot of really, really well maintained bike trails. And so when I moved here, I’d heard of mountain biking, but when I, when I finally settled in, I decided to get a mountain bike and I actually live about a two minute ride from one of the, one of the best parks in town here. And so five o’clock five 30. If it’s cool enough in the summer, it gets hot here. It’s cool enough, I’ll go, I’ll go take a ride. There’s a lot of groups that, uh, will do night rides. There’s some fun stuff. So I like to be active. I’ve reached an age where I can’t run as much as I used to. I used to. And so the knees and hips just don’t allow me to do that anymore. But I can ride with the same intensity and then honestly, it’s a lot more fun. So I’ve been, I’ve been really leaning into that and just enjoying it. Um, all that Bentonville has to offer over the last probably two years.
Paul Barnhurst:
Great. No mountain, biking’s a fun one. I’ve done a little bit of that. Myself. Not a ton. I’m still doing the running, but I understand the knees and the, the other parts as you get older, your body. Yeah. Speed. All of it. It’s not as easy as it once was. I can say that for sure. So last, last question here, you know, we’ve really enjoyed having you on the show. Lots of great advice. I’m excited for our audience to get the opportunity to, you know, listen to what you have to offer, but on that note of what you have to offer, if you could give one piece of advice for someone starting out in FP&A today what would that be?
John King:
Um, can I do two?
Paul Barnhurst:
Mm, yeah, sure. Why not?
John King:
Okay. I’ll, I’ll do two. So the, so the first is keep learning, never stop learning. And you know, finance is a particular business that you can learn forever and still never know it all. It is. It is infinitely deep. And so continue learning even aspects of it that you don’t think apply to whatever it is that you’re doing. They do. They will. And so those perspectives I think will lead to a much more fulfilling career and also knowing what opportunities and options are available as you navigate the next, you know, say it’s 10 to 30 years or whatever your career is, to have perspective on where you’d like to be, right. And to be able to, to speak intelligently and have the conversations you need to with the right people to do that. So that’s, that’s what I would say is, uh, advice number one, the, the second piece of advice is no matter how sophisticated or powerful technology becomes people still make decisions. Hard stop. That’s it.
Paul Barnhurst:
No, I, that, that’s a, I, I love the second one there. The hard stop. I mean, that sums it up. People make decisions. Yep. So you, you need to remember that the technology doesn’t make the decision. Well, John, I really appreciate you taking the time today, being on the show with us, it’s been fun. Getting to know you and learning a little bit more about, you know, your background, what you’ve done at Walmart. And I think our audience will really enjoy listening to this. I know I’ve enjoyed talking with you, so thank you for being on the show.
John King:
Great, Paul, likewise, have a good day.
Paul Barnhurst:
You too. Thanks.