Why Understanding Data is a Critical FP&A Skill with Mark Proctor

Did you find out with the people in the teams that you're working with that it was actually your thought process starts in an entirely different place to somebody else's thought process? So when it comes to solving a problem a normal finance person will be like I'm going to copy and paste this in I'm going to use an x look up or a v look up and they might be like oh I'm using x look up now aren't I so advanced but the problem that they're solving is a data problem and yet they're trying to solve it too late in the process but you being a data person you're looking at going this is a data problem I need to solve this before it even gets into excel. Welcome to another episode of fpna on locked where finance meets strategy I'm your host Paul Barhurst aka the fpna guy in each week we bring you conversations and practical advice from thought leaders industry experts and practitioners who are reshaping the role of fpna in today's business world we'll uncover the strategies and experiences that separate good fpna from great and this week I'm thrilled to welcome onto our show mark proctor mark welcome to fpna on locked thank you very much Paul I've listened to a lot of these episodes so it's great to make it onto this podcast yeah as i say I know you've been on financial modelers corner I think what about a year year and a half ago now two years ago I think probably I forget so at this point I was figuring it out recently I think I'm closing in on 450 episodes that's a long between the four shows that I've had so yeah it's hard to keep track anymore I just know I think I had that person on that's kind of where I'm at is that anyone left do you haven't spoken to that gets to be a challenge sometimes but yes I've heard there's more than 500 people in fpna globally so mark why don't we start why don't you give our audience a little bit of a background about yourself you just want to kind of give them a little bit an overview of who you are and so I've worked in finance and accounting in various forms for the last 25 years I've started an audit then I went through project accounting technical accounting financial reporting fpna business partnering financial control so I've been across a lot of different areas of finance and I've worked in organizations of various sizes so small organizations all the way up to billions of turnover that I suspect what people probably most know me for is excel off the grid which in October will be 10 years since I started excel off the grid which started as just a blog the time so then I had just started a new role and I was working a lot of hours in that role I've used this phrase before that I didn't see my children I didn't see my wife and I barely saw daylight that was the kind of role it was my desk before the sun came up and leaving my desk after the sun had gone down so that was a lot of eventually I just decided I need to take control of this and bit my bits like to chip away automate my work and that's when I started excel off the grid to try and kind of share what I'd learn and help other people to do to do some of the things thankfully I had an incredible line manager at the time who took a bullet for me more than once while I was sorting all these things out he knew it was an impossible situation but bit by bit we got it into a scenario that was not just manageable it was a fantastic job that's how excel off the grid started as a resource to try and help other finance people to automate their work so they could hopefully spend more time with their kids and with their families sharing your story of survival so they can survive and actually be home before midnight so yeah we've all had those jobs where you're like will I ever see the light of day and those great on you're pretty quick and if you can automate a lot of that it makes it more manageable for sure unless I said that a great line manager who took a bullet for me several times when I didn't calculate certain numbers in the right way as we automated and things got better then all of that kind of went away and there's one time where he came back from a come with it's a budget or a forecast review meeting with the CEO and the CEO just made this comment he said I don't know what you've done but the transformation in the last few months in the quality of your reporting has been unbelievable now I wasn't in that meeting at that time but he passed it onto me and that was one of the most proudest moments I've ever had I've the changes we'd made were so evident that even the you know we were just a small division of this exceptionally large company but even the CEO was like this has changed significantly so that was a good moment those are some of the best moments when accomplish something that leadership notices that you know's made a difference because how often when you're automating a report cleaning up data are you thinking will anyone even ever see this and let's face it most of the time you know there is like a iceberg isn't it what they see is the piece at the top but there's a whole lot of ice going on underneath which is all the hard work just to try and get everything into the right shapes that little piece that's poking out actually provides the right value right it's like all the software demos they always show the pretty dashboards 95% of the work is underneath that but they make it sound like you press a button and you see beauty if only yes so before we jump into a little bit of a kind of more about your career I want to jump a little bit into that the new programming created I like to ask every guest this question so I'm going to see how you answer it if I ask you to define great FPNA how would you define it so great FPNA let's take a step back first now what is good FPNA okay let's start there there's like a benchmark so I think good FPNA is kind of creating information which is good enough that it helps people to or it helps influence people to make decisions to challenge decisions I think that's what good FPNA is like great FPNA is actually then not just providing that information but being involved in that being involved in those decisions being involved in challenging those decisions and not just being involved but actually highlighting where those decisions need to be made because easy to say oh yeah I was involved in this decision you might in the room where you did you actually influence the output of that so I think that's the that's where great FPNA comes in and when you start when FPNA teams start to highlight areas that people hadn't considered we need to consider this area because of this that's when it adds a huge amount of value and I will say that there was I've only ever met one person who I would say was exceptional FPNA this was somebody who was invited to meetings because they wanted his opinion on what they were about to do not because he was FPNA but because he had this he was but he had established his position so well in that company is like we want this person in the room because we respect their opinion and the insight they have about the company as a whole and I said I've only ever met one person who has who has done that but that is the when when you carry so much gravitas that people want you in the room just because of what you bring then I think that is the pinnacle of what we can expect from FPNA I agree with you I've had someone share a similar story to me and when they want you there like she had someone say basically we don't make an important decision without you in the room right that's where you want to get to and that's that's something to aspire to and hard to achieve I like that son of a when they want you there because actually it's kind of with FPNA if they only want you there when you're providing information then it's probably not great FPNA so that's probably quite a good problem isn't it when do your stakeholders want you there and the more they want you there the better you're doing as an FPNA yeah I think a great way to think of it if they want you in the room when important decisions are being made you know you're doing a good job because how often I'm sure you've had it I've had it for sure you get this know hey we made this decision go find it in the budget and you're just like well the money's not there we're not hitting the target but okay I'll go I'll go find it in the budget so I know you spent your career in accounting and finance and I heard a rumor and I'm gonna I'm gonna bring it up now I heard you to said you wanted to be in the accountant at the age of seven is this true I believe it's true and I think I I have a recollection of this so my parents tell me that at the age of seven I told them that I wanted to be an accountant and I think I even remember where we were when I told them this there's where we live yet so we had a dog and there was a quarry that you could walk around the planning crew we could you can walk your dog and I'm I'm pretty sure it was on that walk around there that I told my mom that I wanted to be an accountant I'm not even sure I knew what an accountant was at the time but it seemed to fit with my natural skillset and abilities anything to do with numbers and maths and those kind of things I was I was always on board yeah I don't know many seven-year-olds that know what accounting is let alone want to be one but I like it so you were always from a young age a big math person I will yes so whether I knew what I count I'm less face even when I started work as an accountant I didn't really know what an accountant was like in theory I knew you know I've been through the careers advice this mark you could you should consider accountancy this would fit with your skillset and I go yes okay that's that's why I was thinking too but until you start and you actually get on your first day and your first week and your first month oh this is what we do so yeah even even at the age of 21 I still I think it was a word rather than something that I really knew what it meant and it just takes experience and learning to handle that you you must have had similar experiences yeah so you know my background I didn't do accounting I went into business I worked for the government I went back to career at school did finance and finally ended up in F P&A and I really I mean yeah I knew F P&A did some budgeting but I really didn't know what it was nice to remember people are like well how did you get in F P&A well I was trying to get promoted there was a role I was doing a little bit of work related to it I knew the hiring manager I was able to convince him to hire me yeah like people think it's some grand plan like well you talk about F P&A you must have wanted I don't even know what it was in college let alone want to do it so I definitely wasn't thinking about it at seven enough but I do remember as a young teenager wanting to be a lawyer because I thought lawyers got to to argue and I liked arguing. - That's a good reason. - Yeah, and then I got older and said, what am I doing? thinking. Well, in my finance career, I've definitely done some arguing when I worked in on the more technical side, you know, what's capex and what's not capex. That gets into some fruity areas. So yeah, that was there was there was there was a lot of debates that going on. I was more than happy to take on that kind of discussion. Yeah, I had one where I had a business leader try to convince me you pick and choose what payments you recognize. I don't think that's how accounting works. So you definitely get some fun ones. So you recently launched a new program called the finance team accelerator. Tell us about it. How did this come about? The idea kind of happened in probably around the start of this year's a January 2026. And the so we've got we've got an online training platform which is a membership and people join that and they and there's lots of Excel training there power query power pivot dynamic arrays all the the core Excel skills that are now kind of required for the working well in Excel. But the I started to run some some live sessions as part of that. And the number of people who signed up to a live session compared with who watched the same course which had the same content uh it was interesting that people would sign up for live sessions and they would turn up to that. And that's part that's partly why I ran it. So it's an experiment to see when people are forced to take some kind of action. You know, you have to turn up to a live session or else you miss it. You know, how does that change behaviors? And the truth is that a lot more people turn up to those live sessions than I was expecting. So it's not what my goal of things I mentioned is really to try and help finance people save enough time that they can spend more time with their kids. That's that's the ultimate vision. So then it was like, well, how can I help finance people through live sessions rather than pre-recorded materials? Because there's loads of pre-recorded materials. There's there's stuff on YouTube. There's there's our courses. But when those things are available all the time, people don't always take action. You must need a deadline to commit to and say, yes, we're going to we're going to take this. So then when I first started to build this out, I thought, well, how do you make this work? Because there's such a huge amount of content that people could learn. So you've come about distilling it down further and further and further until it became like the caffeine bullet of what finance teams need to be able to succeed and influence their teams so that they can create the most benefit for themselves and the organizations they were for. So that was that was the goal. It's not about trying to teach everything in Excel, every technology, every piece. But how do you select which if you select certain tools and technologies and you put them together? They work in such a way that it solves such an in such a broad selection of problems. Actually, if we could pinpoint people at those ones, how do we solve those? What are the steps we go through? And then people taking action on that that that creates a huge amount of benefit because they have to be there and it's about and it's because it's so condensed down and it's so focused. It means that it's not how you can you've got to take these lots and lots of hours of courses, which as finance people, the number of times you get, oh, I haven't got time to take courses. I'm too busy to do this. I'm too busy to do that. So how do we make this manageable, break it down into small chunks and then make that knowledge stick so that it doesn't just it's not just training in terms of, oh, yeah, somebody learned something new, but it's how do you create transformation? So that's the that's the piece we really try to get to, which is why we're not it's not entirely focused just on Excel because there's lots of tools out there. Excel isn't the only tool, but the techniques, the principles, the methodologies, they exist outside of Excel. So let's understand those. Then we can apply them to our scenarios and then we can apply them through Excel or other tools, whatever they happen to be. But because we understand how things should work, that then creates an environment in which people can create fantastic processes because they understand what they're doing and why they're doing it. They're not just a one I need to do the next step and the next step and the next step is a holistic view that gives hopefully people what they need to get the most benefit in the shortest period of time. It feels like a couple of things kind of led to it. You know, first you mentioned you've done some experimentation and just the whole, hey, let's try some live stuff to see if it creates that sense of urgency, it gets people, you know, to complete the training because I would imagine with the online stuff that you have, lots of people start courses, not many finish if I had to guess. No, Paul, my course is all perfect. Everyone finishes all of mine. No, you're a liar, Mark. Now keep going. Oh, how good my sarcasm is coming through on the podcast there. There's nothing of with courses. So we've got a like a dynamic arrays course and it goes into quite a lot of detail. It starts off with your basic, but by the end, I think you've even taken that course. I have. Yes. By the end, you end up in lambdas and a whole load of other areas. Which are really interesting. Some, for some people, that is, that's what they need. That's where they are. But do you need to use that to get the most benefit? No. So let's focus on the bits that do that create enough headroom so that you can then actually enjoy your role. You can do, you know, you'll no longer just chasing around every budget and every forecast and every month end, just chasing your tail. Let's create some headroom so that, you know, from an FPNA perspective, you can spend more time with your, you know, business partner with your stakeholders because you've created that space where you're no longer just, oh, here's the next thing and the next thing and the next thing. You've got space to actually think about what's best for my stakeholders. How do I serve them? And then you can provide more and more value to them. I want to dig into that a little bit, a little bit more and we'll come back to the course. You know, first we hear a lot of this and I just want to get it out there. I call it the elephant in the room nowadays. But can EI just fix it all for me? AI cannot fix it all for you. There is a, but that doesn't mean we shouldn't use AI. I will say that. We should definitely be thinking about how we can use an leverage AI. But even yesterday, or even the day before, I had, I was using AI and it had a phrase, personally, I would, I would, I would choose this option. I thought personally, that's interesting. And at first, it struck me as does AI now think it's a person? And then I was like, that's not how AI works. It's been trained on enough data. It's got enough information that, probabilistically, the next word it should use is personally. It doesn't have an opinion, it doesn't have any sense of risk. And so therefore, it's, it's not going to solve it all for you because, because it can't. It doesn't actually know what the right answer is or isn't. And there's that, I'm not sure if you've seen that meme. There's a, it's like a, a gravestone and there's a, there's a robot standing next to the gravestone. It says, I'm like, I think my name Dave, Dave died from mushroom poisoning on this date. And then the robot's got a little speech, but we'll next to it. They said, oh, no, you're right, Dave. That mushroom is poisonous. If you'd like, I can give you recipes without poisonous mushrooms. And I think it's the, because there's no, there's no risk. And this story is of, you know, in the Amazon a few, a few months ago, their AI just deleted some production servers and other things because it doesn't, there's no consequences with it. When you've got something that's probable and then you've got no consequences, it's, it becomes quite a, actually, it's got a risky scenario. So my view is that actually, we should be using AI now to help us build the processes that mean we don't need to use AI afterwards. So once you build a robust, a robust process, doing then need to use AI after that. So we use it to build the process, but there's no AI in the process at the end. Now, you might say, but we can use AI to write commentary and other things yet. We can, and we, and in those scenarios, we, we can definitely think about using AI for those areas to add benefit, but it's not the key part of your process. You could write all of that commentary on your own without any AI. The AI is there to just help you with that. So that's what I think the, that's where we are with AI. And actually, we, it hasn't, we don't really know where it's going to be in six months time. Now, I don't mean in terms of technologically where it's going to be in six months time, but open AI, I think is, is, it should run out of cash based on its current run rates in like 2027. What's going to happen with open AI? We don't know. What's going to happen with token costs? We don't know. It may be that in six months time, the token costs are so high that end users are going to have to almost put in an application. This is what I want to use AI for. This is the process I want to use. So it may be that we're now in a honeymoon period that we won't have going forwards. So can AI do it all, even if it could, we're not, we're not in the phase where we're ready to go. Yes, we should definitely use it for everything because there's just so many risks, not just in AI, but in the AI business model, which are currently unknown. Today's sponsor is Linios. Most finance teams, I've talked to a bill something impressive in Excel. Real models with years of institutional knowledge sitting in the formulas, then the data underneath GoStell, the moment someone posts the journal entry, and somebody spends a morning rebuilding it. Linios provides reporting intelligence by connecting your ERP data directly to Excel. Reports refreshed with current data, your formula stay exactly as you built them and you can trace any number back to the transaction behind it. Everyone works from one version of the numbers including whoever is building the board deck. Your team moves from assembling reports to analyzing them when you leave it to Linios. Learn more at www.insightsoftware.com/linios/solutions/ reporting dash intelligence. That's www.insightsoftware.com/linios/solutions/ reporting dash intelligence. It's going to be interesting to watch and I think you make a good point of AI, you know, one of its strengths is it's great for helping you create processes that can run on its own or they take deterministic parts and maybe in you're using AI and some certain parts, but this idea of just turn it all over to AI doesn't seem the right approach to me. I'm giving time will obviously prove that but I'm with you. I think there's lots of situations where it's like, okay, how can we use it to help us create a robust automated deterministic process? Yeah, and I think that's at the moment because we are in this honeymoon period where there's so much investment going into AI that token cost is lower than it will be in the future. We will miss need to make the most of that now because we might not have that benefit in the future. Well, it'd be interesting to see what happens with token cost. I mean, you look at computer costs and they kept coming down and down and down. The models are getting cheaper at the same time. The cost is astronomical to support all these data centers and so it's going to and they're nowhere close to profitable. So it's going to be interesting to watch how that all comes out in the wash. So the minute I'd say use AI leverage AI but don't double down on it at the minute in terms of the baking into everything you do because there's just too many unknowns. Okay, fair enough. I can totally understand that. So the next question I have, you have the program, where is it that you think finance people are struggling? I mean, obviously there's plenty of training out there. There's plenty of tools. So why is it that so many people struggle to automate streamlined processes? What do you think the core problem is? I think the we go back to, I don't know, was it the 1400s when double entry was created or something like that? Yeah, I think it's 14. I think it's right. It's some of that. It's 1400s and everything was our shoe. I'll ask AI. Everything was written down on paper and that that continued for hundreds and hundreds of years, right? We have kind of ledger books of people writing down transactions and all these accounting principles have developed over time. And what's happened is even, I know Excel has been around for more than 40 years, but Excel came along and and finance people said, "Oh, fantastic. We can now, rather than just writing it down on paper, we can write it down on Excel," which is like digital paper. It will calculate things for us. And that gave us a huge amount of benefit. But ultimately, we still stuck with what are the, how do we think about finance? How do we lay out finance in a way that makes sense? And because of that, because of the way that's developed, finance was always a semi mixed presentation and data together. And you know, the cross tab layouts, where even in the ledger books, you'd have dates down the left in your general ledger account across the top. And then the numbers would be written in each column and then they'd be added up and moved on to the next page and added up on, moved on to the next page and so forth. And then all that, you know, Excel came along, it was just like an infinite piece of paper, but the methods that you didn't change. And in businesses where we worked, we looked to our line manager, how do we do this? And like, this is how we do this. And all this knowledge is passed on, you know, the techniques that we use is passed on via some kind of osmosis. It's just passed on from one person to next. And then you get training courses that are like, this is how you use the look up. But everyone's mindset starts off with, how can I use the look up in my current finance, you know, framework, which is this traditional cross tab layout? Until I say, I don't know, 10 years ago, 15 years ago, actually that's kind of where Excel stayed. It had its grid references and had the formulas that enabled you to treat Excel like digital paper. And it's all with the kind of the introduction of tables and power query and the data model and power paper that it then created these data tools directly into Excel. Now, I know that finance teams have had other tools available for, well, other finance tools have been available. But most finance people on their first day, it's like, here's Excel, go and use it. We've got these other systems over here, but Excel is where most finance people spend most of their time. So until, say, 15 years ago, 10 years ago, actually, we had to kind of treat it like digital paper. But now with these introduction of these new tools, we don't have to. We can adopt the data principles, which IT teams and data teams have had for, you know, they've had SQL databases and everything else. And we now get to adopt and use some of those principles. And that's where the benefit comes from, but that's also where I think finance teams struggle, because it's now thinking about everything in a different way, because finance principles were established over hundreds of years, you know, starting the 1400s, data principles have kind of been, I don't know, from the 50s, 60s maybe. So trying to bring those together actually creates a bit of a juxtaposition, because of the way that we think about solving problems and handle problems and how we work through it. And that's why I think finance teams has struggled, because we haven't really had the tools that existed, but not in our, in our world, because we've been predominantly looking at Excel as our, as our tool. And even when they do exist, we haven't had the line managers haven't been there, people to show you how to use it in the right way, the techniques that you should be using, the methods that you should be using. Even the concept of what, what data is, how it structures, you know, first normal form and data models and relationships, that hasn't really been there. And so that's why I think finance teams struggle, because historically that's, we've never been forced to think in that way. I think you had to have the nail on the head there is, you know, many finance people struggle with data and understanding how to work with data in a proper data structure, in a data base, the science of data, so to speak. We know how to work with data in a finance sense, but, you know, it's, people are getting better and better at that, as you have more and more people to have to have. But, you know, I was fortunate enough that I did a master of science and information management in my first role. I was called a finance analyst was really a data analyst. I wrote SQL and worked with the database and, you know, how to deal with all those type of things. So when I switched into FPNA role, that data side was a huge benefit, because I could do things that other people couldn't with the data, because I understood how to properly structure it and always understood, okay, first normal and duplicates and, because we've all, if you ever written SQL, I'm expecting a million back and you get back a billion, right? And you're like, hmm, I have a duplicate somewhere. This is the media at first thought you go to, but if you don't know that, then how do you know where to go? Did you find that the, with the paper you, you know, in the, in the teams that you're working with that it was actually your, the thought process starts in your process start in an entirely different place to somebody else's thought process. So when it comes to solving a problem, a normal finance person will be like, I'm going to copy and paste this in, I'm going to use them and X look up or a V look up, you know, and they might be like, oh, I'm using X look up now, aren't I so advanced? But the problem that they're solving is a data problem and yet they're trying to solve it too late in the process. But you being a data person, you're looking and I going, this is a data problem. I need to solve this before it even gets into Excel. Yeah, I'm almost always, I can't think of how many projects it was starting with the list of lookups and treating those lists. So I had a master list of my data. So now if I wanted to use power query or however I wanted to do it, I knew I had that list to work off. So I almost always went back to that data. How do I create kind of that master data? So to speak for whatever the project is. And I think one, because in FPNA you have, there are two kind of core disciplines. There is the financial modeling side and then you have the data reporting and analysis side. And the structurally they are completely different beasts. Today's episode is brought to you by our sponsor Linios. Linios brings live ERP data directly into Excel, allowing your team to build reports with fresh data. No still files in site. More on that in a few minutes. Yes, the way you build a great model versus the way you build a great data model, like a financial model versus a data model, are very different principles. One is position-based, and the other one is not position-based. And I think there's almost no limit if we elevate financial modeling to some degree. And therefore, we get people who take financial modeling principles and try and apply them to how we report on financials. And that then creates a similar kind of problem as to how the finance function has developed over time, and we have this cross-tabular view of how things should work. So those two principles are, and FPNA are there, but there's a lot that's covered about financial modeling, and there's actually less covered about how we report with data, because it's almost hidden behind, oh, that's a data team. And I say that the, what we do is that the context of the task is finance. Why are we doing this thing? Because we're working finance, it's a finance task. What's the content of a task? Well, that's the finance information, that's the financial record. But what's the process? Well, that's a data process, because how you get from your inputs to your inputs is not any longer a finance process, that is a data process, which is why we need to start thinking about data principles and how we work with those. Yeah, so I mean, I know that's something what you're trying to solve with your course is helping them understand some of these data principles, because modern Excel requires a better understanding than data, than traditional Excel of a 2D grid of rows and columns. That's very much a spreadsheet financial modeling way of thinking, data model, power query, dynamic arrays, require a different way of thinking. Would you agree? A hundred percent. And that's the challenge about how we try and change mindsets and how, once you can change that mindset, how do you make those changes early on in your process so that those data principles and all those that flow through naturally, so that, you know, people talk about automation, but the truth is that if you build, if you build with the right principles to start with, the automation is almost inherent because it flows to automatically, because the data principles we've got, the principles of how we calculate on data, the principles of how we create visualizations from those calculations, it flows through, which means that if we build things the right way, it almost by itself, it removes manual actions because we understand how things flow from one step to the next. Well, you know, funny enough, it also does kind of going back a little bit into our AI discussion. If you've cleaned that data, you know, that data first approach, it's much easier for AI to analyze it, you know, not necessarily build a process, but even just look at it and find things, because let's face a big data sets, AI is going to find patterns much quicker than we are. And if your data is normalized, your data is clean and it's easy for it to understand the structure, it's much more likely it's going to get a right answer than it's going to give you garbage that you're looking at going, okay, that was a waste of time. Exactly. Yes. If you provide it to something that is structured, that there are inherent rules as to how that comes together, I mean, AI has seen enough data sets, it's been trained on enough data sets to find out those things and it's expected, if it's looking at something that looks like data, it's expecting things in columns. And so therefore, it no, it's, it knows where it should be looking, but I say it knows the patterns that it applies, you know, it understands that the, it's pattern recognition is good enough that it can say, this is a, it was a the last thousand times, I'm 99% confident, it's A again or whatever, AI is a, you know, it's an awesome pattern matching machine, you know, and it's, you know, the fuzziest matches, it can still create something from it, but let's not give it something fuzzy, let's give it something that's, that it can easily understand. And then the probability of it getting a good response is higher. Yeah, because as soon as you give it fuzzy, you've created just many more permutations who use power query joins with the fuzzy, and you look at it and you go, you know, 60%, 70%, and you're like, yeah, I still got to figure out 50 of these or whatever, you know, because it doesn't, it's never quite as good as you expect. Yeah, I think that's a good, you know, good way to look at it. So something, I think you'll find interesting, I want to give your thought, I want to get, you know, talk a little more of this data first approach. So over the last few months, this AI has become more and more important. The people I've talked to who are using AI a lot have started to say one of the most important skills or the most important technical skill I'll find out if people need it. They often hear now is system thinking, design thinking, data, which is why I said that. Which I think is what they've always needed and also is what you need for Excel, but especially for modern Excel. So how do we, and I know that's what you're trying to accomplish in your program, but what advice do you give people when they're coming from that finance mindset? And they have no idea that cross tabulated data is not the way to store your data as a simple example, right? Summarize data if you want to call it that. Some people are probably like, what are you talking about when you say cross tabulated? But, you know, summarize where you, what you would expect to be in roses and columns, you're here, your dates or whatever. For anyone who's coming from, from that kind of world into, into this work, the ultimately data is the hub that is the central, the central place. Sometimes I'm training, I give, I give this example where when I first moved to the town where, where I now live, there's a, there's a roundabout in the US you don't have roundabouts to you. We don't have a lot. I have one not far from me, but we have a few. There's a key landmark, we'll call it that, there's a landmark which, which every time he came into the town, if I was visiting somebody, I'd be like, well, how do I get to your house? I would have to go, well, well, no, let's start from this roundabout. How do I get to your house from this roundabout? And the things, I could, from that roundabout, I could get to almost anywhere in the town. I had no idea how these things were interrelated, but I knew if I could get to that one place, I could go anywhere from there. That's kind of how we need to view data is that we've got all these inputs coming in. We need to get to data and from data, we can then get to anywhere. And because when we're trying to get to those places, we predominantly, you know, to carry on the road analogy, we're predominantly carrying on the major roads and then just turning off when we need to, which means what we build when we think about data is we're getting to the data and then we're building solid foundations going out, which means that everything just flows naturally from that one point. And that's where people are making that transition, or even trying to use AI or leverage AI is understanding how we structure data. So for an Excel user, that is, you should be using a table or the data model or something like that. And initially, it's going to feel uncomfortable because you would have these things called structure references. Now, what is this? I don't understand why can't I just use C3 and then C3 is just a coordinate. It doesn't mean anything. It's not, it's not the value. It's got no meaning. I often say that on, you know, in Excel, if you're just using a grid, it's disparate items placed near each other to imply meaning, which is what they are. We use the Gestalt to principles of how humans, we process information and we see that grid and we understand that there's relationships there, but Excel doesn't know there's relationships there. You know, the cell could be in C2 or it could be in D10, Excel doesn't care. But once something's in a data structure and it follows those rules, it then understands there should be relationships between down columns, there's relationships across rows. And that's where you almost have to place yourself into that to then gain the experience. And that's why we're kind of trying to, with this program, we're trying to get people into that point to understand that as much as possible because the, and even before we get to data, one of the simplest things that people can do is just naming files in the right way. I mean, you must have seen this pull, you get budget V1, budget V2, budget V3, budget V final, budget V final, budget V final, budget V final, use this one, budget V final, that's what you get, budget really final, yeah, yeah, yeah, yeah, you've seen that, you've posted on that about LinkedIn, and it's such a common pattern. And that's because we're not thinking about any of the workbooks that we're using as being sources of data. If you had, if you think about Amazon, imagine every time Amazon backed up their website, they created, oh, it's Amazon V1.com and then they backed it up and it came Amazon V2.com and then Amazon V3.com, people would have no idea where they're going, they don't know how to find it, they don't know which version they should be using, people are buying products from V1 that no longer exist and it would just create an enormous mess. And that's because the IT world has gone, that's not the right way to handle these kind of things. You want one place where everyone goes to and when we create a backup, when we archive something, that thing remains exactly where it is. But with files, for some reason, as fun as people, we've not taken the same approach, but we need to take the same approach. So it's something in that file path, that is the latest version. And that is, now I don't advocate linking files together, but even if you do, that would save you a whole load of hassle, because suddenly, you don't have to update those links. Equally, anyone trying to find that file knows exactly where it is, they know exactly where it's named, in two years time, they know where that file is going to be, what's it, and how it's named. And that's just even just thinking about if we work with data, where is data, how, not even about how it's structured, but where is it? It has a consistent name and a consistent path. And just even just taking that approach, we'll start to save a huge amount of time for, but people who are using a more traditional way of thinking about finance and how we work with files and various other things. - Good examples. And, you know what I think I was thinking of, we were talking about all the different versions, right? Anyone who's worked with data or has started using VS Code with AI and different things, right? Get help. It's just a repository where it's just saving a snapshot. It's the same name all the way through. You can go back and see all the version history. I mean, that is a much better approach than, oops, I named this one version four, and it's really version three, and John's now screwed. You know, 'cause I know I've had more than one thing where a broken link or I'm not using the latest file or whatever, and I was like, oh, these numbers don't look right, and you go in your research, oh, the refresh didn't work. You must have had it, maybe you didn't, but early on in your career, when you ended up with the cascade of spreadsheets, they're all linked together, and you end up, how deep do they go? And you end, you know, it can be like 10 layers deep just to add up this single number. And again, that comes down to this kind of a finance way of thinking, rather than the data way of thinking about things, but you must have had things like that. - Yeah, yeah. So I, on the show, when I would ask people on financial models corner of their horror story, I think someone told me that one that went, that opened up 47 different files to get back to the root of a link, because they'd gone through that many different files to get all the way back to the original source data. And I just, I was like, I would have pulled out my hair. - How can anyone have any confidence? - You can, the reality is it's wrong. - It's wrong if you have that many links. - But you just can't know whether any of that is right, because one of those things being linked to the wrong cell, and you know, if somebody goes into one of those workbooks and they just update, they just insert one row, suddenly everything's thrown out. And I'm sure that's happened to everyone at some point, and then someone has to go back and try and fix it, or even just fixing it creates more and more problems. It's just, yeah, it's a mess. - It is. So kind of stepping back, one more question I wanna ask here. If someone's listening, and they're like, this, you know, this makes sense, I need to learn data better. Where would you advise them to start? So would you, is it talking to somebody in IT? Is there a book, should they be taking a course? How do you start to have that data thinking? Like what do you think that, if you could give one piece of advice? - The one piece of advice I would give is, first normal form inside a table in Excel. So that means that, where you've got different months across the top, you don't want different months, you want a column and then a single column of values. So research what first normal form is and how it's structured. That is the first place. And then make sure that it's placed inside a table in Excel. That's the place to start, because then you've got something that you can work with and understand how we refer to data. And people will be like, but I want something from column C. It's like, column C isn't data isn't positional. - Precended doesn't exist for this purpose. - Yes, this column has a name. The row has a number. But equally, if you moved any of those rows, or if you moved any of those columns, if you moved any of those rows, that shouldn't make any difference to how well anything calculates from that. So this, almost this straight jacket of having to put things in, wrote in a structured rows and columns frees you from all of the issues which Excel often gives you. Because we work, 'cause if you work with standard cell references, you're dealing with exact positions. And when things move, it causes problems and risks and other things. But if you've got something in a table, something can move up and down a row, it doesn't matter, it can move across columns. It doesn't matter because data doesn't have a position. And that's the, I think that's the thing to really grasp, is that you can get first normal form in a table that will set you up as a first point and just start using it on everything you need to do, understand how it works, that will then build up that skill. And you have to force yourself, because it's a different way of thinking. But you'll think you're really good. - Great advice, and I will throw out there, I'm sure you have lots of resources to get to more of those. But one of the courses I have free on my website is Excel Tables. Just an intro. And one of the first thing I say is, for your table, here's what you want your data to look like. Here's what normalized, or if you want to call it first normal, data is. Columns and rows, what records should be complete, all those not sub-totals and go through all that. And that's where I start 'cause I think that's first thing you need to learn to really understand tables, especially coming from a data approach. So I'm totally with you. I think that's really good advice. So I want to ask one more question, and then I'm going to jump into some of the standard questions I have. So understanding data is really one of the keys as we've talked about to building robust processes. Why beyond data, anything else you think that really causes so many files, so many processes that default well-short of what's possible with Excel, with the Microsoft tech stack, with Google, with the tools that everybody has available. Regardless of what tools, why do you think other than data, is there another reason is that we're just too busy, we don't want to learn it, what's your thoughts? - One of the, and I realize I've given a few in that analogy today, I'm going to give you one more. - Sweet. - The four by 100 meters in the Olympics is, it's always the, it's towards the end, right? But it's always quite exciting, because it's not just about who's the fastest runner. You can have the fastest runner, but if you then pass the bat on across, smoothly, then your team will lose no matter what, even if you've got the fastest runner. And often people think about, they look at Excel and they become really good at, you know, really good at my thing, I use formulas for this, I use pivot tables for this, I use this feature. But what we need to think about is not how any individual feature works by itself, but how do we, from one feature, pass across the input into the next feature, and then the next feature, and then the next feature. And it's the passing of the bat on, you know, you start off with some raw files, or a download, or a database connection somewhere, you've got something that happens, and then you need to pass that across to the next stage, that's you passing across the bat on. So you need to pick the tools and the methods that are optimized to pass on to the next stage. Now, what that means is that there are various features in Excel which aren't particularly compatible with how that works. And now I know that a lot of people love pivot tables. They love pivot tables, but the pivot table is actually quite a simplistic tool for calculating various numbers, which is why so many people end up, they calculate a pivot table, then they have to copy and paste the results to somewhere else to then continue calculating on the results that they've just generated. You see this all the time, but there's tools in Excel, you know, dynamic arrays, power pivot, there's tools in Excel that have abolished the need to do that. But because people are so, you know, pivot tables are almost a tool that they go to. Now I'm not saying there's anything wrong with a pivot table, even a pivot table layout, but it's the calculation engine behind it. So we need to think about, so why don't people get the most out of Excel? Because then, because then they're not thinking about what's the next thing that I need to do? Or how do I optimize for the next thing that I need to do? Instead of going, right, I've solved this problem. What's my next problem? I'll solve this. I'll solve that. I'll solve that. And that's the way they've continued throughout their careers. Rumping it, I need to take, but what's a holistic approach? What tools do I need to solve this problem from start to finish? And how do they fit together? Unfortunately, that does take experience. Because the experience is that you have to been through that enough time to make those errors to realize that really wasn't the best option. But when you start to bring these things together, I think it then gives you a clearer path as to how you get these things out of Excel. You know where the capabilities of Excel ends. And then you go, you know, we're going to use parallel to make for this. We might use VBA for this. But you use the features which automatically pass information to the next stage, which does mean that in terms of, you know, you might have a financial model. Now, I don't see many financial models that have an output page. Now, I don't mean like a visual page, but a page which is just a table so that somebody else can easily pick up the data from there. If you've got a rolling forecast, you need to use potentially the outputs from your financial model back in your reporting. So let's create the output layout and we can easily then generate and pick up the values that we need for the next stage in our process. So all of that kind of adds together to create something that works from. from start to finish. I like the example of the 4x100. You could have great scales be the fastest sprinter, but if you can't pass it on and and each of those different things are for different stages as you pass that baton, so to speak. Even if someone might say, oh, this this this feature is quite slow. It might be slow, but it's definitely faster than you having to step in manually copy and paste and then do some other stuff. If it just run from start to finish, that's faster no matter what. All right, so I want to move on to our kind of standard sections. We just got a few minutes left here. First, we have an FPNA section and I have a couple questions I'm going to ask you on this one. If I asked you today, what is the number one technical skill that FPNA professionals need to master? What would you say? And I think I have a pretty good idea, but I'll let you say it. I'm going to let you guess on this one, because I think you probably know. I'm going to guess accounting. Yes, you are correct. They need to know technical accounting, capex, anything else. It's obviously data, isn't it? It just is. That is the number. I agree. It's the data, the system design. I think that all goes together, but yeah, that core foundational data. That makes you better in Excel. It makes you better with all the tools you use. It doesn't matter what tool it is. All the tools you use, including AI, you will use more efficiently. Because, Paul, I think you've got you've got a background in TM1 as well, haven't you? If you've used TM1 and I used TM1 for 10 years in my career. That's things that to TM1 and other tools, they're all based on these FPNA tools based on these principles, which we can use ourselves in Excel right now. That's what these tools provide us. Why don't we just start using the right principles? The principles that all these tools are using already. Cubes and reference points. Yeah, we can go on forever on that. What about the soft skill or the human skill? I think there's two. I think there's curiosity. Because without curiosity, you're just going through processes. You're just running it and you're not investing what you're doing or what the company is doing or what your stakeholders care about. So curiosity is the first one and the second one, I think, is influence. Because if you've got curiosity, and you can. Now, it doesn't have to be stakeholders, you have to do this, but even just getting a seat at the table, even convincing people that you have something to add takes various levels of influencing skills around, you're just noise. So I think curiosity of what's going on, what do we need to do, how can we improve this, what do my stakeholders need to know, and influence will get you a long way. I think those are definitely two solid answers. So I like it. All right, so this is just a total fun one because I want to see how you answer it. If Excel removed one feature tomorrow, which one would cause you the most panic? This might come slightly left field. I think it's probably custom number formats. I wasn't expecting that answer, I'm not going to lie. I mean, tables, yeah, we could go back to the old way. If power crews removed, yeah, we could use VBA, we could do. There are other things that we could do. But if custom number formats were removed, suddenly you've got numbers that are completely unreadable and almost no way of really resolving how they should be displayed. So everything else, we could revert back to poor things that we rubbish ways of doing stuff. But custom number formats, I think are pretty critical. I can't say I ever thought of that, and that's the first time I've been given that answer. So I like it. Thank you, Mark. All right, so here's the get to know you. Just two questions we're going to do then we'll wrap up. What's one thing you do today, or that you use AI for in your personal or work life, you can go either one, that you can't live without, that you couldn't go back to doing the old way. Normally, my wife and I play this game, and it's called Hyalower. Now, and it's Hyalower than Tom Selik. I don't know why Tom Selik came into this ever, but he's sick before. So if we see a tall person on TV, we'll say Hyalower, and then we go on to like AI or somewhere else and find out how tall is this person? Are they Hyalower? Now you could say, well, why don't you just go on Wikipedia? And I find there's a Wikipedia can be changed by anyone at any point. So AI actually gives you a, it does it enough of a broader research, but in a short period of time, it will give me an answer that I'm reasonably comfortable with. So in my personal life, it's predominantly around fun things, rather than around something that, I mean, I could go to the roots, but most of it is fun, rather than anything of any real seriousness. So I will never think of Tom Selik the same now. Next time you see a tall person on TV, is they Hyalower than Tom Selik? And I have no idea why it was Tom Selik. I have no idea why that was established, but that's just, that's just the way things develop over time. All right. So what's the hobby you have that would surprise most people? Most people don't know about you. Because I'm involved in another podcast, a lot of people I play guitar, so I do get on a regular basis, but one hobby that people might not know, and this is kind of an old hobby, because once you get kids, it becomes difficult to find the time for this, but I used to make furniture out of junk wood. So old wood, I used to collect from places, and then I used to make furniture. Some of it we've got in our house, so made toy boxes and bedside tables and dressing tables, all of that from random pieces of wood that I collected from, who knows where. So you just need enough of it to get, although it sounds like it's junk modeling, and this is, it looks like trash, it actually looks quite reasonable. Nice. Well, you'll have to show me a picture sometime. I've got to see some of this junk. It smells off his junk, but once I've finished, it looks all right. I want to see the finish, so I can go, yeah, I can see both in there. It was, and it is. All right. So last thing, if people want to learn more about your program, get in touch with you, it's kind of yours to just share a little bit about that. I'll let you take it wherever you want. So for people to get in touch with me, the best place is probably on, or just to follow me in general, it is on LinkedIn. So I think my will be LinkedIn, whatever the extension is, and then Mark, Stephen Proctor is the handle. I think that's the term they use. He wants to find me there. So you can find me on LinkedIn. In terms of programs, we've got finance team accelerator.com, which is our new cohort training program. If people want to check that out and go through some of the things we've talked about today, and transform your finance team, that's finance team accelerator.com. More general excel academy, which is if you want to deep dive into features of Excel that really make you a master of Excel, then that's over at exceloffthegrid.com. So finance team accelerator.com for the pure finance caffeine billet and exceloffthegrid.com for more general excel skills. Oh, perfect. Well, thank you for joining Mark. Appreciate carving an hour out of your evening. Because I know it's a little later for you being in the UK, but thanks for coming on the show and chatting and as much as I joke, the number one technical skill is accounting. I am 100% with you. It's data. I really think the more we understand the data, the better we'll be able to transform our processes, which allowed two things, as you mentioned, getting home earlier, but also supporting the business from that influencing perspective, which is really the ultimate goal of why we're trying to learn the data and get better with the technical tools. If you're running from, you know, if you are one of those people who is not seeing daylight very often, the chances are it's because you're wrangling with spreadsheet, that's where the time suck is, which means you're not really giving your take hold as a great service. Because you're too busy just trying to get, you know, as we mentioned, I'm not sure if you mentioned this before, the podcast started, but I mentioned about the iceberg, you're too busy just trying to get the stuff under the water. You're not really helping your stakeholders. So you've got to break through that because data is almost, it's the base level of business partnering because you need to get to that point so that you can then help your stakeholders with the data that you've got. So do yourself a favor, get better with the data so you can be a better business partner. Exactly. All right. Well, thank you for joining us again, Mark. Really appreciate it. Thank you, Paul. And yeah, thank you for doing this podcast. It's been a joy among various dog walks to listen to you discussing all these topics. So yeah, thank you very much. It's great to be here. Thanks. That's it for today's episode of FPNA Unlocked. If you enjoy FPNA Unlocked, please take a moment to leave a five star rating and review. It's the best way to support the FPNA guy and help more FPNA professionals discover the show. Remember, you can earn CPE credit for this episode by visiting earmarkcpe.com downloading the app and completing the quiz. If you need continuing education credits for the FPAC certification, complete the quiz and reach out to me directly. Thanks for listening. I'm Paul Barnhurst, the FPNA guy, and I'll see you next time.