George Mount is an Excel MVP and author of Modern Data Analytics in Excel (he describes it as a guide to becoming a data analyst in Excel). His latest book, Advancing into Analytics: From Excel to Python and R could be subtitled “becoming a data scientist in Excel. As founder and CEO of Stringfest Analytics, he provides analytics education and upskilling including works with finance departments at the top ten banks. In this episode he answers how someone in FP&A – killing it in Excel – can go further in their career while using Excel as home base.
In this episode:
- Two books and the ROI you get as a finance professional from reading it
- Using Excel as your home base for FP&A
- When to use Excel vs Python
- citizen data scientists (or citizen data analysts) in Python
- Using low code/no code tools
- Excel, copilot and Python as a new “trinity” for FP&A
- Getting your data house in order before getting to AI
- A surprising Excel favorite befitting an MVP
Connect with George Mount on LinkedIn: https://www.linkedin.com/in/gjmount/
Further Reading
George Mount: Modern Data Analytics in Excel: Using Power Query, Power Pivot, and More for Enhanced Data Analytics
George Mount: Advancing into Analytics: From Excel to Python and R
Full transcript
Glenn Hopper:
Welcome to FP&A Today, I’m your host, Glenn Hopper. Today I’m excited to welcome George Mount to the show. George is the founder and CEO of String Fest Analytics, a consulting firm specializing in analytics, education and upskilling. He’s an expert in helping individuals and organizations excel at analytics, particularly in bridging the gap between Excel and more advanced tools like Python and R. George is the author of two highly regarded books on the topic, advancing into analytics from Excel to Python and R.
And the recently released Modern Data Analytics in Excel using Power Query, power Pivot, and more for enhanced data analytics. His work has been instrumental in guiding Excel users and finance professionals into the world of data science and analytics. With a background spanning economics, finance, and information systems, George brings a wealth of knowledge to the data and education space. He’s a regular speaker and blogger on data analysis and workforce development. He’s also a two-time recipient of the Microsoft MVP award for his exceptional technical expertise and community contributions. Today we’ll be diving into George’s insights on the evolving landscape of data skills for finance professionals, the role of Excel and the age of data science, and how FP&A teams can stay ahead of the curve in an increasingly automated AI driven world. George, welcome to the show.
George Mount:
Thank you, Glenn. Thank you for that thorough introduction. Looking forward to, to speaking more about these topics.
Glenn Hopper:
Yeah, I was, you know, I was actually looking at that before the show and I thought, well, I need to shorten this down. And I thought there’s, what, what am I gonna take out? It’s all gold. You’re doing great work out there, <laugh>.
George Mount:
So yeah, thank you. Thank you. Appreciate the plug for the two books. I’m sure we’ll talk about those more.
Glenn Hopper:
Yeah, and actually let’s dive right into it. ’cause Your, your first book came out the year that my book did. So in 2021, advancing into Analytics came in and it, the goal of that was to help excel, Excel users make the leap into data science. And you and I have had had a very similar mission on this, and I think we probably came to it in about the same way. But I’d love to hear kind of what inspired you to write that book and, and, and sort of realize that you needed to get out there to find a way to bridge that gap for finance professionals.
George Mount:
Sure. Yeah. So for those of you watching, here’s the visual aid of the first book. It came out in 2021. It it’s not too long. I try to keep my books brief. And the reason for that is I just want the biggest ROI for your time because I know reading books is not easy. Writing books is not easy. And in this particular case, I wanted the quickest A to B jump from people using Excel into getting into more advanced techniques and methods and tools like Python and r I was in a graduate program where I was doing a lot of heavy statistical quantitative modeling and stuff like that. And, you know, I was really starting to hit the brink of what felt comfortable to do in Excel. And, you know, I had thought about learning Python and r just couldn’t really get my teeth on it. I didn’t feel like there was a great set of resources for it. And I just thought, well, let me, let me write the books. So have it structured in a way, again, that it takes that knowledge that you have in Excel. Because if you are coming from Excel, you know a lot about data, you know a lot about analytics. So how can you take that information and get into new techniques and tools as quickly as possible? And that was my goal for that book.
Glenn Hopper:
So, <laugh>, so first off, your mention of, of keeping the book short, I think I’m, now, I’m worried. So I’ve got my, my next book is coming out in, in a, in a few weeks. And my first book was around 40,000 words, and my next one is like a hundred thousand. So it’s gonna be like a 300 and something page book. And I I worry now because, you know, when we’re writing about this stuff, we go, we go pretty deep Mm-Hmm. <Affirmative> into, into the details, and it’s, people have to want to get there. But I mean, I think once you sort of see the way data scientists are, are working, like I, I think the goal with your books is similar to the goal in mine. It’s like, let me broaden your horizon and, you know, show you this new world. Well, you know, Excel is great and Excel, honestly, even with a huge data set, because I can quickly get around it and it’s just so user-friendly. That’s the first place I’m going is is in Excel.
George Mount:
Sure.
Glenn Hopper:
And I still use Excel every day. And actually, to be fair, I probably in the past month have opened Excel twice as much as I’ve done anything in Python, you know, so <laugh>. But it, but that said, when you kind of see the other side, you start to realize, I could replicate this, I could do things more efficiently. I’ve got more ability, you know, but there is that, that learning curve. So I’m wondering, as you have talked to people about your book, as you’ve done your training, how has the reception been, you know, to the idea of Excel users who’ve gotten, you know, there’s people who are do awesome things in Excel, and now you’re saying, well, what if you learn Python and r you know, how, how do people respond when they realize I’ve gotta learn a new computer language, a new way of interacting with data? And, and have you seen, you know, so your book came out in 21. Have you seen since then, especially with Python in Excel now, are you seeing like attitudes and adoption change since in 21 when you first, when the book first came out?
George Mount:
Yeah, that was the big watershed moment, I think, for acceptance in Python. And with Excel in particular. Once it was available in Python, I really felt like that was the stamp of approval where people started to take it seriously. Because before then I got attitudes ranging from confusion to almost like outright hostility. And I remember presenting at meetups about interfacing Python with Excel and hearing people say things like, well, why do I need Python? Everything I need is in Excel, just not, not interested. So that’s been a, a big development. And I’m seeing as this progresses, you know, I’ve been doing a lot of those challenges you might see on LinkedIn where people post a data set and, Hey, you need to do X, Y, and Z. How are you gonna do that? I’ve been doing them in Python. And I think it’s really opening people’s eyes to the fact that this isn’t always that challenging, right? I mean, yeah, you can write a huge long function with eight parentheses scattered here and there at eights, not many <laugh>, way more than that. Or, you know, Python sometimes is really easy on the eye. So I think that just seeing it in action inside of Excel has really lowered that barrier of adoption, really got people thinking and, and brought it out of this area where it seemed like this outside quantity and into something that’s part of the larger Excel universe.
Glenn Hopper:
And interestingly, and we talked about this before the show, but you know, talking about broadening your horizon, going into R and and Python, and I always think r is like the gateway drug to Python, you know? Yeah. ’cause it’s maybe a a little more user friendly, but you start to quickly kind of run up against its limitations and then, you know, it’s easier to bridge that gap over to over to Python. But, you know, <laugh>, so what we were talking about before the show, in your latest book, you kind of go back to basics or you, you bring it all back home with Modern Data Analytics and Excel. And, and here you’re focusing more on empowering data analysts with like, you know, power Query, power Pivot and, and the other Excel features. So, you know, not to completely contradict what we were just saying, but what made you kind of decide to go back to Excel for this book?
George Mount:
Yeah, I think for a few reasons. So, so this book came out in 2021 or 2021. It was on the heels of me doing advanced statistics and things like that. And then I got back outta school into the workforce and into it was a good reminder that not everybody’s there. Not everybody needs those skills, but there are a whole lot of people who aren’t familiar with those tools that you mentioned. Like Power Query, power pivot, other things like dynamic arrays, all these things that have come out in Excel. So the way I I like to think about it is that this book is meant to help you become a data analyst in Excel, right? And this is your book for becoming a data scientist in Excel. And while they are emphasizing different things, there’s all, there is a lot of crossover. I was actually just working on a post earlier today about all the different modern Excel tools that are used for Python and Excel to even work.
So things like link data types, dynamic arrays pictures inside of cells, right? There are all these new tools that have come out that Python and Excel actually needs to function. So it really is kind of building on the shoulders of modern and Excel. So in a way, you know, this book is a good one that will help you learn more because this book came out before Python and Excel ever happened. So I don’t talk about that specifically, but, you know, Python is Python, so it’s still stuff that, that’s good to know. If you wanna get into that,
Glenn Hopper:
I’m sure you’ve thought about this a lot, and this is kind of what prompted me to go back and, and write another book this year, is thinking about when I was trying to make that transition, not having like a single text where I could figure out the path I need to go down. And I think the combination of your books does a really good job of, you know, different doorways into the, the same space of being able to work with data better. But, you know, for someone who’s just starting on this journey, so maybe they’ve, you know, they’re working in Excel and now they’re in Tableau or Power bi and they’re starting to do a little more with data and maybe, you know, even visualizations stuff that they weren’t doing before. But they’re starting to get kind of the concept of joins and maybe, you know, venturing into sql. But what’s, what to make that leap. I mean, I’m, I’m thinking of like incremental baby steps. What’s the typical, if someone came to you and said, Hey, you know, I’m in FP&A, I’ve been doing this for five years, I’m killing it in Excel, but I, I want to go beyond that. What’s the typical learning progression you’d recommend for someone in that spot?
George Mount:
The way I would think about this is you wanna milk Excel for all it’s worth. You definitely wanna get into Power query, power pivot, all that stuff. I would say focus on the interface and the things that you can do with the click GUI interface in Power Query before you worry too much about programming power Pivot. You are gonna need to know some programming there, but, you know, just get well-rounded in that data analyst core for Excel. And then from there, you know, I think about this like these three legs of data professions in a way, right? You have your data analytics, have your data science, and you have your data engineering. What do you want to lean into the most? And we can almost think of like business intelligence as kind of an offshoot of data analytics because you already have the basics of data analysis down with all these tools in Excel.
So, you know, if you’re really into like, predictive analytics, machine learning, well that’s data science, right? So you’re gonna be getting into Python, you can build basic Python scripts in Excel now. So you’re gonna stay in Excel. If you want to get into bi, probably gonna be more of a Power BI thing over there. So, but again, you’re building on what you already know with Dax and you could get into MCO and stuff like that. So, you know, what I’m trying to do is like scaffold you into just say what you know from Excel and then building. And then last but not least, there’s data engineering, which is kind of a newer thing, especially in Excel. That’s where you get into your power platforms. So getting into like power automate office scripts, you might think about even learning like JavaScript for example. ’cause There’s a lot more that’s being done in Excel with JavaScript.
Now that’s definitely at the point where people are resistant and hostile toward JavaScript, just like Python was a few years ago. But that’ll probably change. So that’s the way I think about it. And I’m really sticking like inside the Excel universe. ’cause That’s what I know the best and that’s what I can give the best advice on. But I think if you think of Excel as like your home base, and then you think of, okay, is this data engineering data science or data analysis or bi, right? How do I kind of like play between those fields?
Glenn Hopper:
I think about this a lot. You know, there, there’s that expression. If the only tool you have is a hammer, everything becomes a nail. And I think about the way, you know, the financial people use Excel, and then the way that other business people use Excel. And the other day I found myself, I was making a list of, of just a straight text list, but I was doing it in Excel because I was gonna have to sort it alphabetically, <laugh>. And I thought, now I’ve turned into the guy that’s just doing everything in Excel weirdly. And, you know, this is just a list, you know, of like, you know, 50 items or whatever. And but you know, before I made the jump into doing, you know, even the, the little bit of coding that I do, you, you can figure out ways to do, I mean, you know, barring the, the data limitations, but you can figure out ways to do so much in Excel. I mean, it’s such a versatile platform. And I’m not even talking about using Python in Excel. I mean, just the basic Excel. So if I’m someone who’s like, trying to understand why I need to move beyond that, like help me draw the line between, you know, I can do this in Excel and this in Python, what situations does it make sense to, to work in Excel versus this is really better suited for Python.
George Mount:
So if you’re looking at your work and you are noticing places where Excel is just like forcing you to get hacky or forcing you to point and click, and there’s really no way to build this in just A to B automated straight shot. Like those are things that Python might step in for. So for example, you’re looking at building some type of a plot. Maybe you wanna look at small multiples, which is a pretty common data visualization practice, right? Where you’re gonna break this plot down by different categories. It, it lets you, you take in those differences a lot easier than necessarily maybe putting all those categories on the same plot where it gets busy. That’s not really something that excel’s cut out for. And I see people build small multiples, but it’s a lot of effort. And there’s really no way to write a reproducible script that, hey, maybe you want your coworker or your coworker said, Hey, how do you, how do I do small multiples?
Well, you know, and then you send ’em to a blog post where it’s like, well click this, but actually not that you need to go here and hit the dropdown and, and all this, like that kind of hacky stuff. I think if you’re seeing that, that’s a good place for Python, right? Because you’re writing everything in code, but you’re not writing code like a developer. You’re writing code like what we would call a citizen data scientists or citizen data analysts. These are packages that were built for people without formal programming experience. So a lot of the stuff you’re gonna do, you’re just using functions just like you do in Excel. You know, you’ve written an X lookup function before, you know, there are different arguments, some are optional, some are implied, that kind of thing. This is the same thing. So when you organize to that point, like plotting I mentioned is a good one. I, I would say time series is another good one. Everybody, you know, dates and Excel are kind of the butt of a lot of internet jokes. So things where you want to do more advanced time series analysis, whether that’s moving averages or sampling the data at different time frequencies or even getting into more advanced things like building forecasts and things like that. I would say those are two good ones to focus on.
Glenn Hopper:
I love that, that you were talking about the citizen development. That’s the only kind of dev, I couldn’t imagine any of my code ever being in like a production environment. If it’s not in a Jupyter Notebook or a collab project, I’m like, <laugh> and, or if an actual coder looked at my code <laugh> like, what are you doing here, man, <laugh>. And on that note, so another confession I’ve really leaned in this year because I felt like I needed to understand the fundamentals, but I finally started leaning into kind of the low code, no code tools. And I wish, I think we were talking before the show, you know, years ago when I first got deep into analytics, I kind of, I went Tableau because it seemed like the easier route to go. And I, I know I could have done both, but there’s, you know, there’s only so many hours in the day and I really didn’t dive into Power bi, but I’m seeing now with just across the Microsoft universe, like what people are doing with it.
And I feel like, eh, I probably went with the wrong horse on that one. But at this point though, there are so many other tools and I, and I guess Power BI does have some, some of this drag and drop functionality. I don’t, are you messing around with any of the low code no code tools that are out there and like, how do you see these tools kind of fitting into the toolkit for Excel focused financial analysts? Because I’m, I dunno if you’ve used like, the ones I’ve, RapidMiner is probably the one I’ve done the most in lately. Dataiku is another one. But where do you see those tools kind of fitting in with Excel and Python and all that? Or, or have you even messed around with any of those?
George Mount:
I’m probably most familiar with the power platform and power automate power apps. Seems like that. I think there’s definitely opportunity for finance people there. Whether you are trying to build workflows or applications, you know, maybe you’ve been tasked with setting up some kind of approval process for budgeting or expenses or whatever. Or you wanna set up some kind of an app that, you know, lets people track, I don’t know, inventory expenses or whatever it is. You know, thinking about Excel right now, it’s every Excel workbook. It’s kind of its own thing, right? And we don’t really have a good way of tying them into emails or workflows or projects or to do tasks or stuff like that. You know, using it as a like approval process in Excel is kind of a nightmare, right? Trying to track things. And again, Excel is the database is not, not a good connotation.
But I think with power platform and power apps, you know, you’re not necessarily gonna use Excel as that database, but you are gonna have the opportunity to use Excel as like one piece of your wider stack. So you know, if people are looking at building those kind of in-house websites and, and apps and things like that to help with those kind of, you know, day-to-day operations that’s more so on like the development and engineering side that, you know, you’re not gonna do a lot of predictive analytics and machine learning there. But I’ve been getting into that slowly. I think that’s a pretty exciting opportunity. I think, you know, Excel add-ins right now, they’re kind of in a, at an impasse where, you know, I try not to use them. I know a lot of people do use add-ins or, or ask me wanna use them.
There’s probably gonna be a big shift there where Excel’s really been pushing things being built on a unified platform, which they chose to be JavaScripts. You have office scripts as I know a lot of VBA developers are gonna be out for blood, but if I say it’s an alternative or a modern replacement for VBA in a lot of ways, it is. We can go into the basics. Yes, there are always few cases for VBA, just like there are always cases for cold ball and fortran. But I mean, come on for most people, I think office scripts will be a good idea to learn. So those are all really cool ways to get into, and that’s, I mean, that really veers off of low-code, no-code. But, you know, thinking about all these citizen developer data engineering tools that the power platform offers is a, is a cool place for people there in finance to go.
Glenn Hopper:
When you were talking about add-ins, you got me thinking about copilot and we’re really waiting to see how copilot it becomes fully integrated in, into Excel. I mean, I know like there’s the, what the, the copilot finance package, which is really limited
George Mount:
Yes,
Glenn Hopper:
Right now, but I mean, as they, as they roll out new features and, and anyone who listens to the show knows I can’t go a single episode without mentioning generative ai, so I’m gonna <laugh> Sure, I’ll, I’ll go down the road here, but maybe, you know, so copilot in Excel has a way to go, but I, you know, you can kind of see the writing on the wall and you can see what it’s gonna do in the, in the coming years. But already GitHub copilot is out there and it’s like, I, I think about when I was taking my, like CS 50 and when I was, you know, first learning Python, I had that book Python in a Day. I don’t know if you remember that one, <laugh>. Yeah. but, you know, all the stuff that i, I went through and how bad my code was and like forgetting to close loops and just all the stupid mistakes I made. Yeah. How much easier it would be if you actually had the GitHub copilot back then. Sure. Like, I’m, I’m really jealous of you know, people who were just learning or anyone who works in it all day to, to have that. I don’t know how much you’ve messed around with GitHub co-pilot, but what, I mean, how much is that changing coding for either experience coders or just people starting out?
George Mount:
Yeah, I’ve done a lot with GitHub co-pilot, but I have worked with copilot in Excel. I certainly see the, the overlap there. You know, I think a good point to be made with Python is that for a variety of reasons, it is a great tool for communicating with generative ai. Again, not necessarily as a machine learning specialist, but even just you know, general Excel or, or data analyst, right? Often the results you’re gonna get once you speak in Python, Python is a programming language, right? So it kind of speaks a similar language to the AI models. So whether you are looking to get randomized data or you’re looking to build a specific plot, right? Python is a really good kind of common lingua franca sort of thing for generative ai. So we’ll see how that works with, with copilot. There’s a little bit that I’ve been able to coax Excel copilot to do with Python. But, you know, the way I think about this is that Excel, copilot and Python really do form this like new Trinity for data analysis, right? Where you’re using Excel as your canvas, you’re using generative AI and Python to get the answers that you want, and then you’re using Excel as that way to really present it to the user.
Glenn Hopper:
Fp and a today is brought to you by Data Rails. The world’s number one fp and a solution Data Rails is the artificial intelligence powered financial planning and analysis platform built for Excel users. That’s right, you can stay in Excel, but instead of facing hell for every budget month end close or forecast, you can enjoy a paradise of data consolidation, advanced visualization reporting and AI capabilities, plus game changing insights, giving you instant answers and your story created in seconds. Find out why more than a thousand finance teams use Data Rails to uncover their company’s real story. Don’t replace Excel, embrace Excel, learn more@datarails.com.
As you’re talking about all the, all the tools and, and the plugins and everything. I’m, I don’t want to just, you know, have a, a Microsoft commercial, however, <laugh> I thinking about as we expand our environment, just that’s one thing right now that in from, you know, power Automate and Power BI and, and Excel and all the integrations across, even the way that Chat GPT is being integrated in, I mean, that is a pretty nice ecosystem. And there are so many things that plug into, and, you know, data rails included that plug into and, and work well with Excel. I guess if I’m in that, you know, I’m, I’m looking to expand. I mean, really it’s, as we’re talking, I’m thinking if I’m looking to expand the tools that I’m working with, it makes a lot of sense to stay in the Microsoft universe and, and go through. I mean, is there, do you think it, that the logical first step is to go from Excel to start doing some stuff in Power bi? What’s sort of the prioritization there as you’re moving through the technologies?
George Mount:
Yeah. I think I would go back to this idea of you know, data science versus bi versus data engineering and looking at the prospects inside of your organization outside, you know, market statistics and things like that. There are plenty of other tools and vendors on the market so it, it is good to be able to situate how they all work together. Yeah, I think in general, you know, we’re getting to a point where there’s not such a line in the sand between things like proprietary versus open source. That was many years ago that that line really blurred. I think one of the lines that’s blurring now is this idea of low no code versus coating because of generative ai. Really anybody can write code might not be good, but you can write it. And this idea that it has to be low, no code is, is kind of falling by the wayside. So yeah, really situating, you know, where do we, where do you wanna really plant your flag in the sand of data analytics and data science and all that stuff. And then just looking at the landscape and, and seeing what works best as individual results may vary. Like they say,
Glenn Hopper:
You know, another thing I was thinking as just we’re sort of trying to put together this roadmap of how to move beyond and, and take on these new skills. You and I both had the a advantage of starting out using R Sure. While we were in a program in, in school, and, you know, r did help. It was the first kind of, since writing, and I’m dating myself here, but since writing my basic programs in the eighties, yeah. You know, r was my first free introduction back. I get, you know, HTML monkey kind of stuff and, and a little bit of JavaScript before that. But, you know I, but now though, if I weren’t in school, like if you’re in a statistics class and you’re you know, r makes sense it, it’s easy to use and everybody’s doing the same thing. But if I’m already in the working world, is there any reason for anyone to learn r Like, I, I don’t, I think I hid the icon of R Studio on my laptop, you know, I don’t even,
George Mount:
Yeah.
Glenn Hopper:
I can’t remember the last time I, I, I’ve opened that. Is there a reason for people to learn r at this point? Or should you go just dive straight into Python?
George Mount:
I think some of that depends on, on industry dynamics. There’s certain places like a lot of academic research or biomedicine and, and engineering and things like that. There is still some, if you look at you know, some of the really quantitative trading places, a lot of them actually still use R for research, right? And then they put stuff into Python for production. But, so there are, you know, you might find certain pockets where r still prevalent, but I think with Python and Excel, that was really the, the knockout punch for, for Python to be the, the place for Excel users in general to go for better or worse. I actually really like r I, I like it as a language. If I could do my work my way, I would probably honestly pick up RA lot more often than, than I would now. But, you know, just personally, I don’t really get a lot of requests for it anymore in terms of like training and consulting. So the market’s kind of spoken to me too, that it’s not not as, as big of a force as it had been.
Glenn Hopper:
Yeah. Yeah. Makes sense. I don’t want to beat a dead horse here, but I’m really looking at it, it does feel like there’s a bit of a, a sea change. And what I always think is, you know, FP&A folks, I feel like we were the original BI people. Like we, we were doing BI in Excel before there was, you know, before all this blew up. And then when machine learning came along, like, and may I, I think it’s the amount of data that machine learning, that marketing and sales had versus kind of the limited GL data that we had. And we had to figure out like how to get the data first. But I do feel like sales and marketing, like jumped over the finance side of the house using all this cool new, you know, machine learning algorithms for classification and, and regression and you know, and all the prediction that they were using.
And, but now there are, I think with generative ai, maybe opening some doors and making data science a little more accessible because you don’t necessarily have to, I mean, you don’t have to know coding as well to use the Python anymore. And, you know, with like chat GPTs data analyst, it can happen under the hood and you don’t even know what’s going on with Python. But seeing now where more and more is gonna be automated by technology, if I know I’m an FP&\a person, my career path, I love FP&\a, I wanna stay doing this, and, you know, maybe that’s ultimately a, a path to the chief financial officer seat. What do you think? Like, where should my focus be? Should I, how much do I have to lean into becoming a data scientist or a data engineer or, you know, just increasing my bi chops and, and thinking about as we get access to more data, that data science, like, it’s kinda like in finance, you have to know the difference between, you know, net income and EBITDA and, and gross margin.
And you know, things that if you’re not, if you don’t have the domain expertise, you don’t even know the questions to ask. And then data science, if you don’t know the basics of it, you’re gonna get yourself into trouble by using some wrong model or, or trying to forecast on something you don’t understand. So, I mean, where should someone who knows FP&A is what I’m doing now and what I wanna de do for my career? Like how do they focus what they need to stay on top of to be very good at that?
George Mount:
I think monitoring search trends is a great way to do that. One thing that I’ve noticed is that there does seem to be a stepping back where if you think about a few years ago, data scientists was the sexiest job of the century. Everybody wanted to get into machine learning and big data and, and all that stuff. And, and now it’s, well, hold on, right? Like, is all this data worth collecting if it’s not the right data is just hiring a bunch of data scientists and throwing them at, at poor quality data going to solve your problems? So I think the market is kind of correcting for that. The next thing that is happening with I, what I’m seeing with generative AI is that, okay so generative AI can just start building our dashboards and our analyses and you know, we, we don’t really need data analysts anymore.
So I think that’s probably a good place to really fight the fight right now where, okay sure we have generative AI and copilot that doesn’t mean that, you know, we should just set this loose in among the organization. We don’t need data specialists anymore, because then I think you’re gonna get the bloat of, right, like all these dashboards that nobody looks at, or, you know, having these data scientists, building these models that nobody can use because they don’t really do anything. So that’s probably a really good place to be right now, which is as I’m thinking about it, right, how do you as the Excel person become like the source of reason and like the kind of guide of adopter of adoptin for generative ai right. Starting with that data quality problem, which maybe we can finally get right. Now that there’s so much attention on data and analytics. So starting there, getting that data quality right and then, you know, doing whatever makes sense, what you like, what your organization is looking for, whether that’s building predictive models or doing business intelligence just depends on, on different, different things. But yeah, getting, getting generative AI and data quality right, is probably a good place to start.
Glenn Hopper:
Yeah, I mean, data is the foundation, and I think especially if you’re not at enterprise level in your business, the amount of data you have, you know, it’s not as, it’s, it’s not as much to work with. I mean, you can pull you, you know, you can pull in macroeconomic factors and all this external stuff, but getting, you know, clean data. I work a lot of times with businesses that are in say, the 10 to $15 million range, and, and you know, they talk about where they want to go with ai, and you get in there and you look at it and you think, w we gotta address a, as an accounting before you can even <laugh> right? Move on to AI because, you know, we gotta go and clean up their chart of accounts and we’ve gotta get their, their data straight. And they don’t even have that, that foundation. So, and it’s, you know, that’s the, unless you have been, I know we’ve been talking about digital transformation for like 30 years now, but unless you’ve truly done it and you’ve kind of stepped up the data maturity scale, it’s very hard to make AI work. And especially if you’ve got a, a limited data set and it’s not clean data, or you have
George Mount:
Yeah.
Glenn Hopper:
Still have data silos, which you see all the time. Yeah. I mean, there’s, you know, systems not talking to each other and stuff. Living in Excel is a database and all that <laugh>. But I think, I mean, it’s sort of like, you know, the bell’s ringing now if you haven’t gotten your data house in order. Yes. And where we are with ai, you’re, I mean, <laugh>, you’re, you’re gonna start being left behind if, if it hadn’t already happened. Yeah. And, you know, thinking about <laugh> being left behind, apparently you don’t sleep because you’re already, we did talk before the show that you’re already working on your next book and with ai, you know, and I think we both kind of are in the same place where you see how quickly AI is moving and it’s like, we need to get out there and, and, and talk about this. So tell me a little bit about your next book. And I, I guess, so it’s what Python ai data engineering skills for Excel users. What’s the focus of it? Who’s audience is it? Will people need to have read your first two before this one makes sense? Tell, walk me through what you’re, what you’re doing now.
George Mount:
Okay, so yes, it is somewhat set up to be a trilogy right? You have your data scientist book, you have your data analytics book. Number three is gonna be more of a data engineer machine learning kind of book. When I wrote Advancing in Analytics that was pre Python and Excel I even have a chapter on Python using Python to automate Excel in the second book. So I do wanna update all that. I want to get into copilot, I want to get into ai, I wanna get into the basics of Power Automate. So it really is you know, starting with the premise of, okay, it’s great that we’re talking about Power Query. And yes, I love Power Query, but it’s almost like 15 years old at this point. I mean, like, come on, let’s, what else is out there? And what are these AI skills that are in some ways, you know, evergreen so that I know people are reluctant in a lot of cases to use copilot in Excel.
It is, it does present in some ways a catch 22 where the only people who are gonna get benefits of copilot in Excel are people who don’t know enough about Excel <laugh> to use Copilot. So I get that there’s a lot of resistance, but I mean to say that there’s no place for AI in finance data Excel, like that’s not, we gotta meet somewhere in the middle. So what I’m gonna try to do with this book is like, here’s some foundational things that even if you’re not in love with Copilot yet, it’s gonna get better. Here’s why. Because this is how AI works. Here are things that you should know about how to structure your data, how to add, you know, how to troubleshoot things that excel’s already using that are AI that actually work really well, right? Like even flash fill is AI as pattern recognition doing like fuzzy matching building forecast, right?
There are a lot of things that Excel’s doing right now in a much narrower sense that’s still ai. So let me show you all those things and get you thinking about, you know, how AI works, how does AI learn from data? How does AI need things structured to work at its best? So there’s gonna be a lot of that. And then yes, getting more into, okay, so now that you know Excel pretty well and you’re writing scripts with Python and you know, you’ve got your workflows down in Power Query. Okay, how can we start automating stuff both inside and outside of Excel with power automate office scripts and things like that, and really trying to excel into like a modern development platform, really, right? In a way that VBA just can’t do. Sorry. Again, every time I mention that, I just think of angry VBA developers. I, you know, don’t wanna offend anybody, but it’s not really a cloud first tool. That’s the way things are for better or worse now, right? We use the cloud for pretty much everything and VBA does not use the cloud. So you do the math on that one, and I’ll be talking about office scripts.
Glenn Hopper:
Yeah. And not to knock VBA ’cause I get the same feedback that you do, but you talked about what power Query is now 15 years old. How old is VBA a at this point,
George Mount:
Right, is,
Glenn Hopper:
Yeah, <laugh>
George Mount:
I mean, Python’s been around for a while. I don’t, you know, age in some ways the longer a tool exists and has staying power, that could be a good thing, but that’s only if it is adaptable to Right. The current environment. Yeah. And, and you know, Python was able to do that, and Python’s been a really great tool for web development, for example. You know, can you version control, right? You can build modules and packages with Python you know, VBA for better or worse, just isn’t set up for a lot of those things. So you know, I think that that gives an opportunity to try some new stuff. And that’s what I’m gonna write about in the book, and people can like it or not. But, you know, I will say that I was talking about Python in Excel way from the get go, you know, so we’ll see what happens with the next trend, right?
Glenn Hopper:
Yep. Yep. And so do you have a timeline on the book yet of when we can expect it? Or is it early stages? What, how, how are you sitting
George Mount:
With it now? Probably sometime next year. I don’t have a great timeline. There’s a, a lot of other things I’m working on.
Glenn Hopper:
I wanna hit on that for a minute because we did dive straight into, into your, your books and, and the, the approach that you’ve taken. And I, you know, how it applies to FP&A folks, but you also are a consultant with, with StringFest. So tell me a little bit about your consulting work and, and kind of the, the size of the companies and sort of where they are and the data maturity level and what you’re coming in and doing and how much you work with finance teams in in that kind of work versus the sales and marketing and, and other teams.
George Mount:
Most of my clients are in, I guess I would say three buckets. I do work with a lot of operations, types of teams who work in manufacturing and, you know, have a lot, you know, they’re working with data from ERPs or other planning systems that they’re getting into Excel. They just wanna build some basic data cleaning and analysis workflows. So those tend to be more like medium, smaller businesses. I also do some work because I do have a background in like academic research. I was in a PhD program, so I, I do get work from academic organizations, nonprofits, and those are the people that are actually still using R for research. And they’re in the same kind of place where this is very, very small data, often because they’ve collected it themselves or they’ve gotten it through, you know, a third party policy nonprofit or things like that.
And they’re doing the same kinds of things. They want to be able to report on this data, clean it efficiently, so they’re on more of a research heavy kind of a, a path there. When it comes to banking and finance, that’s a good portion of my audience, a good portion of my clients. A lot of those really are at, at large scale. So we’re talking, you know, very large, top 10, top 30 kinds of banks. And these are, these are groups that at a high level, they’re thinking, how do we plan for our talent, right? How do we develop this talent and how do we stay ahead of it? So what are the things that people coming in need to know about Excel, about data analytics, and where do we go from there? So how do we, you know, keep that learning pathway and plan for what’s coming next?
So those are the, those are the three main clients that, that I serve. A lot of it’s through workshops. I do some, you know, like learning pathway development and things like that, but I love training. That’s a big part of what I do. I don’t really do a lot of Excel consulting in the sense of just building somebody’s workbook and sending it over. I tend to be a lot more pedagogical in terms of, you know, walking people through what they need now, what they’re gonna need later, and then really building that path along with them.
Glenn Hopper:
With everything you’ve got going on, I’m surprised you’re able to find enough hours in the day <laugh> to do everything. But outside of, of, of writing the books and, and doing the, the consulting and, and all the work that you’ve done around this, what’s, what’s something that you know, may on the personal side maybe that most people don’t know about you or something that we couldn’t find just by googling you?
George Mount:
Well, it is on my website, but I don’t know how many people find it. The origin of StringFest has some different meanings. The most obvious to, I think, more technical minded people would be strings as in, you know, computer science. I do a lot of writing and that on, on tech, but it does have another meaning, which is a string est as in musical instruments. I’ve been playing the violin since, I think I was like eight maybe in the guitar since I was like 13. So it’s been a long time. I mean, people ask how many years I’ve played these instruments. At this point I can’t do the math. So I do that a lot. I’ve been trying to do it more. You know so music is like a big part of my life. I also have, I think I have it here, so I won the king Oscar Sardine Haiku writing contest. This was pre generative ai, so I had to write this haiku all by myself, and they sent me this lovely pewter viking longship that I get to have at my desk now. And that was from eating king Oscar Sardine. So that is my literary claim to fame.
Glenn Hopper:
That is, that is beautiful. That’s awesome. <Laugh>. so I’m guessing if you’re, if you’re writing haikus and, and you play instruments, do you write music as well?
George Mount:
I’d love to do more of it. You know, that’s on my list to get into songwriting. I’ve never tried it myself. I, it seems like it’s something I could handle, but it’s one of those, it’s like learning a language or learning python, right? Where if, if you don’t see that learning pathway, it’s really hard to know where to start.
Glenn Hopper:
Yeah, I think so. The trick is when they turn one of your books into a movie, big, short style, you know? Really? Yeah. Who would’ve thought Big Short would be a movie, one of your books has turned into a movie. You could do the score, you could have <laugh>. That’s right. Yeah. Very exciting. Analytics score. <Laugh>.
George Mount:
Yeah. Or, you know, at least play, at least play in the orchestra performing it.
Glenn Hopper:
Yeah, we ask everybody, and I’m always curious to see, and we have some kind of go-to answers that most people give. And then there’s some people, it’s funny, like if I talk to a CFO who’s been a CFO for 20 years, and I ask ’em this question mostly they’re like, I haven’t, I, I don’t, I don’t know, because people hand me things in Excel, I don’t get in there. But the question we ask everyone, I’m sure you know it’s coming, is what is your favorite Excel function and why?
George Mount:
Yeah. I’m gonna off offroad a little bit here because my favorite Excel function isn’t technically a function, but I love the spill operator, right? If anybody’s played around with the little pound sign that you can put by that lets you refer to a dynamic array, so you’re not referring to an individual cell, you’re actually referring to the whole array. What I love about this is that really becomes your gateway between Python and Excel, because anything you write in Python, if you bring those results into Excel values, right? Python is an object oriented programming languages. So it stores things in objects, as in Excel. We don’t really have objects. We just, you know what, you see what you get. So you want to convert your Python objects into Excel values. What that spill operator lets you do is take the results of that, those Excel values that were generated from Python, and then you can refer to them elsewhere.
So, you know, let’s say you build some kind of a model in Python, you wanna feed that into some kind of a chart or, you know, whatever downstream analysis you wanna do in Excel, that’s really that link that’s gonna let you, you know, build between Excel and Python and vice versa. So that’s, that’s the one. I guess the opposite. One of that would be the PYY function that lets you take Excel data and bring it into Python. But I really like, so I guess that would be the more vanilla, like that’s actually a function. But I was thinking more of that spill operator.
Glenn Hopper:
It’s funny because I think about all the, you know, index match I think is probably our most popular response there. But as you know, as the abilities and as Excel just morphs and changes over time and more people get the skills that you, that you’re teaching in your books, I, I see in the future, in the next, over the next few years, we’re gonna start getting more and more responses like what you just gave rather than the kind of the traditional old school Excel stuff. Because it’s gonna be integrated and people are gonna work through and have different ways to do it. And certainly your books are a, a great start down the road of, of having that level of understanding. We’ll put in the show notes, we’ll put the links to your books and everything. But if anybody wants to get in touch with you about StringFest or anything else you have going on, what’s the best way for people to get in touch with?
George Mount:
Definitely yeah, so string fest analytics.com is my website. You can get in touch with me there, you can follow it, subscribe to the newsletter. I post pretty often on all topics, data analytics and Excel. I’m also very, very active on LinkedIn. I post basically every day there, so people are welcome to, to follow me there. Speaking of LinkedIn, I do have a couple of LinkedIn learning courses on AI and Excel. There’s one on copilot and one on just miscellaneous AI tools in Excel. So if you are connecting on LinkedIn, you may as well hop over to that LinkedIn learning page and check out those courses. So those are the two major ways that, that I stay active in the community.
Glenn Hopper:
Great. Great. Well, George, I really appreciate having you on the show. Love your books. Keep up the great work. Can’t can’t wait to see the next one.
George Mount:
Awesome. Thanks very much Glen.