AI, Excel and the Changing Role of Financial Modelers with Danielle Stein Fairhurst

AI is already building financial models, and that is quite terrifying to me because as a modeler who I've always kind of prided myself on my technical prowess, to now know that AI can just do it like that is quite discombobulating.

What's your biggest pet peeve that you see in financial models? The thing that drives you crazy?

I think it's kind of—

Today we're joined by Danielle Stein Fairhurst, the principal financial modeling specialist of Plum Solutions, an internationally recognized financial modeling trainer, speaker, and master. Also the author of Financial Modeling in Excel for Dummies and a Microsoft Excel MVP.

If your job is to just copy and paste or to do what you're told, you know, where you're not actually creating or thinking through strategically, then yes, you should be worried. The best thing that we can do is kind of, I think it's important That we just sit down and think strategically about what we want rather than outsourcing all of your creativity and all of your thinking to AI.

If you were to go back and audit every model you've built over your career, is there a theme, story, something it would tell you?

It happens sometimes.

People bring— Financial Modelers Corner is the world's premier modeling podcast. It is brought to you Financial Modeling Institute, the world's leading financial modeling accreditation organization. Welcome to another episode of Financial Modelers Corner. I'm your host, Paul Barnhurst, aka the FP&A Guy. In this podcast, we talk all about the art and science of financial modeling with distinguished financial modelers from around the globe. The Financial Modelers Corner podcast is brought to you by the Financial Modeling Institute. FMI offers the most respected accreditations in financial modeling, and that's why I completed the Advanced Financial Modeler. And our guest this week happens to be a Master Financial Modeler. I'm thrilled to welcome on the show Danielle Stein Fairhurst. How you doing, Danielle?

I'm very well. Nice to see you, Paul.

Yeah, it's been a while. It's good to see you. I know you've been up to a lot. So it's been— I think the last time you were on the show was the Great Debate.

Yeah. Oh, wow. That was a while ago. Yeah.

Yes. I think with Lance, you and— I think you and Lance had one of the questions. I think you and I got to discuss whether modellers was with the apostrophe S or S apostrophe for fun. Of course.

Whether it's multiple modellers or one modeller.

I know, I know. We had a little fun there.

And with 2 L's. Modellers is spelt with 2 L's, of course.

Yes. I don't remember who battled that one. I had someone from America and someone from Europe battle that one. So they could argue whether it's one or two, but that was a fun event. So I think that was the last time we had you on the show. So excited to have you. Let me share a little bit about Danielle's background and then we'll jump right into our questions. So Danielle is a Sydney-based financial modeling specialist with 2 L's and a corporate trainer. She is a Microsoft Excel MVP, a Master Financial Modeler from FMI. I think there's what, 8 in the world now?

Yeah.

She's a LinkedIn Learning instructor and the author of Using Excel for Business and Financial Modeling and Financial Modeling in Excel for Dummies. She's a big advocate for women in finance. She founded the Women in Financial Modeling Meetup group and runs other groups and forums, including a LinkedIn group with over 220,000 members. She is also a member of the Financial Modeling Global Leadership Council. Love the background. I think there's all things modeling in there.

Yes, everywhere.

Ever wish you could get career advice from the best financial modelers in the world? Well, now you can. That's why we created the Career Corner. Let's go ahead and see what today's guest has to say. So I'm curious, when you were coming out of school, did you have any idea that financial modeling could be your career path?

I had never heard of it. Honestly, I did not know that modeling even existed. Absolutely not.

Not surprised. I didn't know it was a career. I mean, yes, I knew that you built models from finance classes, but that was about it. So I'm not surprised to hear that.

Yeah, I just studied finance and just kind of got into it that way and then discovered how wonderful Excel is. And then, you know, that mix of Excel and finance was like, well, obviously financial modeling. took it from there. So it's just an absolutely wonderful career.

So did it kind of start with the love of Excel and then you combined the finance and started doing modeling? Is that kind of how it happened?

Yeah. Well, having the finance background and I worked in investment banking in London and then just sort of started using Excel and was like, well, financial modeling is obviously the mix of the two.

Did you start your career in London then? Is that where you—

Yeah. Well, I studied here, but then I traveled, of course, as most young Australians do, ended up in London and worked in investment banking Glad to get off outside the island for a while. Yeah, to another island. Yeah, I was there for about 3 years and just kind of started my career there.

I'd love to get your thoughts. What would you say to young people, you know, people coming to college early in their career, they're worried that AI will make the financial modeling job obsolete?

That's a big question. Financial modeling job, I mean, with any jobs, I think if your job is to just copy and paste or to do what you're told or to look that up and do this formula and, you know, where you're not actually creating creating or thinking through strategically, then yes, you should be worried. But, you know, I think the best thing that we can do is kind of stay up to date and see what's possible, what's available, keep experimenting and kind of know what's there because I think there's still a lot of opportunity for financial modelers. In fact, I think there's more opportunities than there were before because the boring part of your job is now being done by AI, you know, so we get to do all the other stuff. So yeah, I still think it's a big change, but I still think there's a lot of opportunities there.

And I agree with you. I definitely think there's opportunity. And someone else made a good point, as you know, people are like, well, is it going to eliminate investment banking? And does anyone need to model? And someone said, you got to remember there's a supply and demand side to this. It was Matan Feldman who runs Wall Street Prep. And what he said is, He goes, okay, so you could do the models twice as quick. So in theory, you need half as many people. At the same time, he's like, there's a lot of people that are going to be able to afford modeling and afford investment making that may not have before. They can go down market. So he's like, you got to remember there's a supply and demand and it will all even out. Yes, there may be some job loss. AI is going to take some in pretty much every profession that's business-wise, but there'll be jobs created. And I thought it was a really good point because I think everybody worries about one side of it. And he's like, you got to remember it also drives down cost, which changes who can use certain products.

And modeling becomes more available. Uh, and so, you know, as an expert modeler, you then become more in demand.

Yeah. And so I think that I'm, I'm with you that any job, if you're just following instructions and it's task-oriented, you should be worried. You need to make sure you're adding strategic value.

Yeah.

So what advice would you give to someone, you know, kind of maybe they're in finance, they're early in their career, like maybe I really want to be a financial modeler, any advice you'd offer them?

Well, as I said, to keep learning and to find out what's out there because it is changing so quickly. And the last thing you want to do is turn up at a job and, you know, talk about things that are out of date because somebody will always go, hey, there's a new way to do that. Didn't you know? And so I think, you know, just keeping up with what's happening and to keep learning. And there's just so many fantastic resources that you can use to learn all about these different tools and techniques. I think that's probably the best thing to do. And of course, to learn how to use AI inside the processes of building a financial model. I think being across that is important.

No doubt you have to nowadays figure out how to use AI in your job. And if you're modeling, that's really how to use AI in Excel. And you may use it outside as well, but Excel is going to be a big part of that. So I usually ask guests their horror story about modeling, but I've asked you that before, so I mixed it up here. What's your biggest pet peeve that you see in financial models? The thing that drives you crazy?

Okay. I think it's kind of The same as like at home, you know, in the house when people don't pick up after themselves. They don't tidy up after themselves. Drives me nuts. And in the same, when you can see when someone's been using your model and they've saved it in the middle of the document. It's like you're out in the middle of nowhere in sheet number 5. You know, just to kind of like save it as you would want someone to look at it, or they've put it onto some weird scenario that's never going to happen. You know, so when you've finished using a model, go Ctrl+Home on each page. We call it a beauty save. Yeah. So Ctrl+Home, go to A1, and then go to the front page, reset everything back to the way you'd want to open it so that when you open up the model next, you can see what the heck you're looking at. That drives me crazy.

Funny enough, I, uh, had sent out a file and I usually try to do that, you know, make sure every sheet's up at the top and It's a file I sent out for a class today and I opened it up and it was like on sheet 5 and I was like, dope, like mad at myself. So I know what you're talking about because I try to make sure it's clean and you open to the front sheet. And so I could understand that one. I want to guess you have a tidy home.

A bit. No, no, no. Don't look at my background here. It's below the camera. There's stuff everywhere.

Ah, so you trick us. I like it. I do the same. See, it looks all clean, but if I showed you my desk, it would be a big mess. So, all right, it's been— like I said, it's been a couple years, we mentioned at the beginning, since you've been on the show. So maybe talk a little bit about what you've been up to. Any big— any kind of, you know, big things that happened, or just kind of in general how, how it's been going in modeling for you?

Yeah, uh, just busy. I think, I think when the first time I was on the show, we were talking about my books, which was a long time ago. So no changes there. So those books are still Um, out there. But I've, uh, I think since then I've recorded a few LinkedIn Learning courses, which has been great.

Yes, yeah, I've heard about that.

Great experience. Yeah, yeah, love that. And at the moment, I'm actually just, uh, revamping my own online courses, which I've had for a long time. Some of the— they were all recorded in a studio, and the studio parts are fantastic, except of course my, my hairs. I've got like a micro fringe, and I look I look a little bit different, but what I've said is still really relevant. And so what I've had to do is go through and edit them and just kind of replace the screen share because the version of Excel is a previous version. So I'm just updating it to make sure it's all kind of with the latest tools. So that's what's been keeping me busy lately.

Yeah, it's amazing how much time it can take to re-record and redo courses. I have a few that are old and I'm like, I know I need to redo them and I just, where do I find the time sometimes is the challenge.

Well, I've kind of, yeah, this, this month I've just really sat down and made myself do it. And AI is really helpful for that because being able to, it's amazing, you know, you run, you know, put it into, uh, into the recorder and you can actually edit so much, uh, more easily than it used to be a really time-consuming process. And it's a lot easier than it used to be.

100%. I love, I have a, I use Camtasia and their software Audiate where I just open it up and edit the transcript and it edits the Yeah, it's like that. So a lot of, a lot of the tools have that now. They're great, aren't they?

Yeah.

So I remember the first time I used it, the transcripts were terrible, and now they're really good. Like, you know, they're 99% accurate. There's words they get wrong, especially the unique finance words. Like, it almost never gets FP&A right, but that doesn't surprise me, you know. But that's just a click and change all of them at once, maybe twice, because it spelled them slightly different or whatever.

Yeah. So I'm just running it through the, like this, like a lot of the auto subtitles are really off. And I've just found if I run that again, just do it again, it just improves it a lot.

It's funny, you know, you mentioned AI and I've had a lot of people say just almost on any complex task, ask it to check itself 2 or 3 times. And by that 3rd time, it's usually quite good.

Yeah, it was one the other day I was doing with them, because a lot of it is with instructions. And with the newer version of the model, I'd inserted a couple of rows. And of course, all the instructions were out. And so I just asked AI to, you know, to check it. And I was like, okay, great. And, you know, that's all good. And then I said, how about we just, you know, run it again? And it found all these errors the second time. It's like, what happened? Like, why didn't you not Why did you not pick those? Oh, we missed them the first time. So we ran it through, had to run it through like 3 or 4 times to actually find all of the places where the instructions and the Excel file were not matching. Um, and a lot of it's like that, having to ask it multiple times before it gets it right.

When the model goes to the board, it's your name that is on it. The advanced financial modeler is how you prove the work I took the exam. It sharpened my modeling skills and taught me how to be a better financial modeler. I invite you to join me and start the journey to becoming an AFM today. Go to fminstitute.com/podcast. That is fminstitute.com/podcast. Podcast and use code podcast for 15% off. Yep. Or what I'll do, you know, certain things is you, you run it with one and then you bounce it off the other.

Yeah.

You know, ask Claude something and then ask ChatGPT to check Claude's work. Yeah. It's a weird world we live in sometimes. This is the first session where I've been doing this training for, you know, 3, 4 years. And we have an exercise we do on day 1, and immediately they turn to AI for it. And we've been waiting because we kind of encourage them afterwards. We show AI, see how long it would take you. I kept telling my instructors, it's only a matter of time until they use AI to do the exercise. This class, they used it for everything. And what it's allowed us to do is focus on what we really want them to learn instead of them rushing to try, because it's shortened the time of the exercise when they need to focus on presentation or, or different things. You know, it's, it's really in many ways enhanced the learning because they're doing it to learn. They're not just doing it to do the exercise, if that makes sense.

Yeah. Yeah.

So it's been interesting to watch. Yeah. All right. So re-recording all the courses. Um, and also let's see, I know one thing, this isn't modeling, but I want to ask about this. You recently did a pilgrimage in Europe. Talk a little bit about that journey. I think it was about 500 miles you were saying. So tell our audience about that. Yeah.

It was amazing. Absolutely amazing. Just 34 days walking about 20 to 25 kilometers. Sorry, I don't know what that is in miles, but a lot. Like basically walking half the day. It was absolutely incredible. I loved it. Everyone says it's a life-changing experience, and I was like, well, I hope not because I kind of like my life. I like the way things are. And coming back, I mean, of course, everything's kind of gone back to the way it was, but it was just wonderful to take that amount of time out and just really, really reflect. And I bought this tiny, tiny, tiny little laptop to take with me because you really, really light. You wear the same clothes every single day, which is very freeing, and to take the laptop With me because I just felt really insecure without having it with me. You know, what if somebody needs a model? You know, what if I need to— and I did actually have a fairly complex proposal to write at one point. But other than that, I didn't use the laptop as much as I thought I would. But it was kind of like a security, like having that laptop with me the entire time.

What was maybe kind of the key takeaway from the whole pilgrimage? Like what Was you kind of finished it, like, as you reflect on it, what was your kind of maybe takeaway from it?

I don't know. I just loved it. I just, I guess you, you know yourself quite well. Uh, and I know that I love traveling and I love sort of being out of my comfort zone, which definitely was. Uh, yeah, I lost a few toenails for sure. And, uh, yeah, it was pain. Like, it was just, yeah, quite It was a very freeing experience, I think. Yeah, I don't know any other sort of takeaways. And then you sort of come back and things are as they were, but I realized how much I love what I do and that's great. Nothing really changed from it.

Yeah, I like it. You said it was freeing, as you called it. You recognized how much you like what you do. You like challenging yourself. Any other kind of big events? Any other challenges coming up?

Oh, I know. I feel a bit like I want to organize something because I, we look forward to that for such a long time. Uh, and yeah, I don't know, maybe go to, uh, got a few trips coming up. So yeah. Yeah. Looking forward to a few things.

I'm glad it was, you know, such a rewarding experience. That's what you always hope.

Yeah. Yeah, it certainly was.

All right. So kind of back to modeling. Obviously, I think everybody, whether it's modeling anywhere in business, we've talked already about it a little bit, AI. Right? It's just, it's the word of the year. It feels like the word of the decade. You hear about every day. So how are you thinking about AI and financial modeling? What's kind of your take? Where are you at?

Well, I was hoping that you wouldn't remember, but when we did the great debate that you reminded me about earlier, one of the questions was, can AI ever build a financial model? And I said, no, absolutely not.

Yes, I think it was you and Lance Rubin, if I remember right. He was on the opposite side.

How long ago was that? It was not that long.

18 months, maybe?

Yeah, completely wrong. I mean, it is happening. AI is already building financial models. And that is quite terrifying to me because as a modeler who I've always kind of prided myself on my technical prowess, you know, that I'm really good at the modeling and To now know that AI can just do it like that is quite discombobulating. So it's taken a bit of, you know, I've really had to lean in and embrace it and make sure that I'm using it because it would be very easy to just go, la la la la la, don't want to hear about it. It's not happening. Uh, but it is happening and we have to come to grips with it and figure out how we're going to use it because we have to use it.

Yeah, it's pretty crazy how quickly developed this year. It really, when I think it was Opus 4.7 came out, we took a giant leap forward around December, January, late, late, very late last year, beginning of this year. And ever since then, it's, it's been a different world. That was, I still remember in Schnorr. When we did our episode where we tested 4.8, he's like, everything's changed. You know, and Ian's not one for hyperbole. Very, you know, very level set, very, and he was, you could tell he was shocked, like, wow, this is actually good. Still has its issues, but it's not like before where you're just looking at it going, why would anyone use this?

Yeah, I really love that series because it, uh, you know, it just really, it was kind of in real time as things were progressing. To see how we can build a financial model using AI.

And it was, you know, it's amazing to me to just think back to the first episode to where we are. It just, sometimes things move so fast, it just blows my mind at times. Like, how did we go from there to there? And it's exciting. But yeah, that was a really fun series because I think we picked just the right time to really start getting in and show and see that change.

Yeah. As you started, I don't think you realized Just how much you were going to achieve.

And I think we all admitted that like 6 months, you know, 6, 8 months later, we're like, did we think we'd be here? And we were all like, no. Like our opinions changed from day 1 to now. You know, we still try to do once a month an episode and our, now we're, the things we're testing is more how far can we push it? Not can it balance the balance sheet every time, or can it get it in the right row? Which was early on some of the challenges. So, I mean, what do you think? You mentioned, you know, really proud of technical prowess. And I think all of us, you know, a lot of modelers are. I think that's a hard part, not just for you. I think many modelers struggle with that. But how, how are you thinking about AI now? Are you using it some in your modeling? How do you think about it in teaching? Kind of give more your thoughts on where it, where it fits in what you're doing.

Yeah, I just use it wherever I can, uh, as just take As with any kind of learning, to take every opportunity to use it where possible. And it's very easy to kind of get set in your ways and go, yep, this is the way I've always done it, and just keep doing it the way I've always done it. But it's usually, no, stop and think, is there now an easier way of doing this, or is there a different way of doing it? And that is where you really learn, is to try it a different way.

What's the biggest thing you've learned so far when it comes to Excel modeling and AI?

Yeah, I guess just how much it can do. Yeah, it's quite— and every day, every day I'm learning new things all the time. Like, yeah, just amazing. And then sometimes it'll do something and you're like, this should be easy. Why can't you do it? And it just creates this absolute rubbish. And Yeah. And then sometimes you go, it does stuff so quickly, then you just can't believe it. So, um, I think the thing that's interesting is not only is it faster, I mean, yes, we're saving time, but we go deeper. You know, that's, that's something that surprised me is I've kind of realized that, that yes, yes, yes, I can do all everything that I did before quicker. Well, once you get the hang of it, it always takes me a bit longer with experimenting, but once you kind of get the hang of it, yes, it's quicker, but you can actually get more than what you did before. So the analysis, the insights, all of that is further and more detailed. And just learning, and you have to— it's a very personal thing. And just experimenting with it and working out How, um, you know, how your relationship with it and how you can get the best out of it, um, working together as a copilot, of course.

Nice Microsoft plug. Well done.

Throw that in there. And, and that's how it should be, is that you're kind of working together to get the best out of both of you.

I love how you said the best out of both. And there is that trial and error. I mean, the first time you do it, it may take a lot longer, but as you start to figure it out, there's definitely areas It can help quite a bit. And I'd love to know, is there an area where you're using AI? It doesn't have to be modeling or Excel, but just where you couldn't go back, where you're like, oh my goodness, I used to have to do this manually.

Yeah, actually a lot of the admin tasks in my business now, I use AI. Like, I just wouldn't— I would find it difficult to not have that anymore. It does scare me though sometimes when, like, when I think, oh, I've got to like do something or I've got to think about something or I've got to write something. And I, the first thing I do is turn to AI and I think, hang on a minute, just hang on a minute. Just think it through, use your head for a second and think through what you want from this process before you immediately turn to AI or start opening Excel or start jumping into the tools. I think it's important that we just sit down and think strategically about what we want rather than outsourcing all of your creativity and all of your thinking to AI.

There is a difference between editing a template and building a model from scratch. The Advanced Financial Modeler tests whether you can actually build a model from scratch. What are you waiting for? Get started on your AFM journey today. Go to fminstitute.com/podcast. That is fminstitute.com/podcast. And use code podcast for 15% off. 100%. I think that's a huge challenge everybody's struggling with. There's times when you're like, Am I just delegating everything to AI? Like almost like, am I becoming stupider in a sense? You know, because I'm just letting AI be my brain. And so sometimes like, no, no, just stop and think through a few. You can use AI to help you, but at least get yourself started and get your brain working versus AI, AI, AI, because it's so easy.

Yes, exactly. We, and I do, I do think it's making us dumber. I do feel like we are not using our brains as much.

I agree. I, you know, I've said that in the long run, AI will make society dumber as a whole. I hate to say that, but I think it's true. You know, I'd like to think it's not, but we'll see. Time will tell. All right. So, you know, you do, you do a lot of training. You also help people build models. So you have your paying clients, you have your, your training. Are there any areas where You know, when you're training people to build models versus paying clients, you see kind of a, you know, kind of that disagreement or different identities where, hey, this technique you use on your client work, but you would never teach in a course, you know, those types of things. Or sometimes you're like, here's the best practice. You teach all this in a course and then you go to a client and you're like, oh wow, that's not best practice.

Yeah, absolutely.

Yeah.

Yeah. And I think you probably have that problem as well. So yeah, definitely. Like when I first started the business, I thought, oh, I'm going to do I do consulting and then if I do training, like it'll be the same people. But generally it's not. Like the people who want to be trained and the people who are like, can you just build me a model, are completely different types of people. And very rarely do you get the same people. Although there is the odd training client who'll be like, oh, can you just build it for me? Like, whereas I generally kind of go, here's how you do it. And, you know, I'll kind of build it along with them. So that they can, it's their model and they can take it over. But often they're just like, I don't care, could you just do it for me? That's, and that does happen often. But as you mentioned, yes, definitely. So when you're in a training course, you have all these sort of training examples and they're all sort of sanitized and they work perfectly each time and they showcase whatever technique that I want to show them. And, you know, it's not messy, it's not real world. I mean, it's a little bit messy, but it's kind of like, I kind of am in control and I know exactly—

Yeah, you control the environment.

Yeah, of course. Whereas when you're in the real world, you end up, you know, you just got to get the job done and you don't want to sort of be really pedantic about having everything best practice because, you know, it's not always practical. Sometimes, you know, when training, I say, you know, formula should be no longer than your thumb. It should, you know, try to keep your formula as short as possible and you get into a client model. And you've built a Lambda or something that's huge and terribly complicated. And it's important then, depending on who the client is, but generally I try to sort of make sure they understand the technique and why I've done it.

I think the formula rule is probably one of the big ones. There's a few others, you know, just the complexity, right? You always tell, always keep it simple, make it, you know, really straightforward. But the real world rarely is. One of the best advice I heard was from a guest. He goes, make— when working with clients, he goes, make the model as simple as possible because your client or your, you know, your manager, whoever you're building it for, they will complicate it for you. So if you start out making it complex, you're going to end up with a monster.

Exactly. And I mean, AI is getting better, but it used to— like, the, the solutions that would come up with are just this long and It doesn't need to be that complex. So part of the prompt, I always say, you know, make it as simple as possible, as short as possible. Just get straight to the solution. You know, don't, don't sort of add all of this extra stuff in there.

Yeah. Kind of over-engineers, over-builds. I had one where it used a LET formula to reference a cell on another sheet.

Oh, wow. Okay.

It took me a while to figure out what it was doing. And I realized it was just referencing like C7 on sheet 1.

Mm. Okay.

And they used the LET formula to do that.

Okay. Wow.

That was probably the worst I saw where I was like, what made it think that was a good idea? Like it just kind of, you know, couldn't understand what was going on. So I, I know exactly what you're talking about. Uh, we definitely saw that a lot in the mod squad that we talked about earlier. All right. So I want to talk for a minute, you know, in your training, one of the things you do is help prepare people for the AFM, for the CFM, you know, prepare them for those tests.

Uh, AFM. We don't do—

Just AFM. You don't do CFM.

AFM.

Yeah. That's right. So, any advice you'd offer to people who are preparing for the test?

Yeah. Yeah. So, we do coaching. Like, I have a prerecorded prep course, but then we also do coaching as they kind of get closer to the exam, as well as the mock exams that help them get ready because it's just practice really, you know, just practice and practice and practice. We do drills, like over, like it's about speed and about having things happen as quickly as possible. So, it's Just certainly with the AFM, the CFM is different because you've got lots of different types of models. So you are not as, you can't be as well prepared, but with the AFM, it's exactly the same structure. The inputs are different, but the situation, you always have to build a 3-way model. You always have a debt schedule. You always have a revolver. You know, everything, there's very few unknowns. In that situation. And so it's just practice, practice, practice, and getting your sort of muscle memory with the keyboard so that you can do it as quickly as possible. I think that's probably the best advice is to practice because it's such a practical exam. It's one of the only modeling exams that is just so practical. So you know that if someone's got the AFM, that they can actually build a model in the time.

100%. It's, it's one of those few tests that it, like I said, very practical, very much you're doing the job versus is it A, is it B, is it C, is it D? You know, you're trying to figure out an answer and it's, it's speed. If you put in the time, there's no reason you shouldn't pass the AFM. That's what I tell people. If you've, if you've built enough of the models, you've practiced, you know your shortcuts, You know your, your statements well and how to build each schedule. Nothing should surprise you. May you get a little bit of a curveball with some kind of capacity thing or whatever? Sure, there'll be a little bit of a curveball, but you have plenty of time if you practiced to figure it out.

Yeah, well, you have to finish it, but you need to be able to finish it within 3 and a half hours so that you've got that extra half an hour just in case something goes wrong. And you know, if you're ready. Like if you can either do it or you can't. So you do it on a, you know, we do the mock exam 2 weeks prior to the date. And, you know, and then the pass rate is really high because you know that you're not going to sit it unless you know that you're ready. Yeah. And I was going to say that you also don't want to go into it less prepared. Like you want to get the most out of the experience. So if you're kind of flying by the seat of your pants, am I going to pass? Am I not? It kind of defeats the purpose. So you're better to sort of go in knowing and to take your time preparing as well. You have up to a year from when you sign up that you've got to actually take it. And it's good to take at least, at least a couple of months to really like make sure that it's in there. Because if you just prepare for something very quickly and then sit the exam very quickly, it doesn't It doesn't always stick.

No, totally agree. I wish I would've spent more time studying, but I remember when you mentioned giving yourself time, I got done with the basic model, my balance sheet didn't balance, and I spent like 2 or 3 minutes trying to figure out why. I'm like, forget it for right now. Get everything else done. Make the executive summary and the scenarios, make those all match, and then come back to you if you have time left over. Because I don't want to spend the whole time, never find it, and the thing's in disarray. Exactly.

And I know I failed. pass if with a half-finished model, there's no way you can pass. But if everything is perfect and you made the tiniest little mistake and so your balance sheet isn't 100%, maybe you'll get through. Yeah.

And that's what Ian said. He goes, we'll pass some people if everything else is good and it's a simple mistake, you know? And then I had about 15, 20 minutes left or maybe half an hour when I got all done and I started going through it and I just put a, I think I had a negative in the wrong place or parentheses in the wrong place. It was just something simple where I'd messed up a formula and I just wasn't catching it when I first looked, trying to be quick because you're all nervous. And yeah, giving yourself that time is huge in case something like that happens because it's easy to do in an exam.

Yeah. And so you found it just to— I did.

I found it with about 10 minutes left. Now I'm like, okay, I feel good. If I don't pass, I'll be shocked. Yeah. Before that, I was like, am I gonna fail? Like, this was— and I was running Financial Modelers Corner at the time. So that would have been fun to come on the show and tell everybody, I failed my AFM. Let me talk about modeling now.

Yeah, yeah, exactly, exactly. It's, um, I was quite nervous. I think, you know, you know that you've passed, like you'll, you kind of, you've either done it or you haven't basically when you come out of it.

All right. So dynamic arrays, where are you at with dynamic arrays in financial modeling? A lot of different opinions out there. I know you've shared some things on LinkedIn, so just like your thoughts in general.

I love dynamic arrays and I really think that they are made for modeling. You know, like the whole point of modeling or one of the principles of best practice in financial modeling is to have consistency of formulas. There's basically less that can go wrong if you have a consistent block of formulas and that's essentially what a dynamic array does. You have the formula is in the corner and everything just spills across and down and so there's very little risk. But then there is, there are other risks that are involved with dynamic arrays. And that's the thing that's hard. And so I struggle with the idea of having a model that is completely 100% dynamic because, you know, things like the corkscrew are a lot more, you know, with the, with, I'm talking about the debt schedule, for example, where you've got an opening and a closing and an opening and a closing. And that can be quite Quite difficult to do with a dynamic array. But I mean, it's, you know, once you're used to it, it's not that hard. But that was actually my latest LinkedIn Learning course was actually building one from scratch, like a full model using dynamic arrays. But again, it's one of those learning sanitized examples. It's not a real model because I don't think I have been able to build a real model in the wild. That is not, that is fully dynamic, uh, because there's always going to be the odd thing that, that just is much harder to do and make it dynamic. Like you may end up making it like a really long and complicated formula just for the purpose of being dynamic. And sometimes that's not always practical. So I do struggle with it. Um, and I do try to make them dynamic wherever possible. Because I really feel also like that's the future of Excel. Like pretty much every new function that Microsoft has brought out is dynamic. Like you have to, you have to be using dynamic arrays in order to use the function. So it's kind of the future of Excel is dynamic arrays.

FP&A Guy here. One of the certifications I'm most proud of is the Advanced Financial Modeler, the AFM from FMI Institute. Why am I so proud of it? Because it taught me how to be a better modeler. I didn't just take a test. I had to demonstrate that I could build a fully integrated 3-statement model, a skill that is invaluable for financial modelers. It's one of the certifications I'm most proud of, and I encourage you to earn it as well. So go ahead and join me. Get started on your journey to becoming an AFM charterholder. Go to fminstitute.com/podcast and use code PODCAST for 15% off. Yeah, I mean, if you look over since the engine came out of the 50, 60 functions they've added since what, 2020 or whenever it was, '19, Dynamic Array came out, 90% of them are dynamic functions. There's the occasional one that comes out that's not, but it's usually solving a very specific problem.

Mm-hmm.

So I'm totally with you. I think it's where we're heading. And so do you think we'll hit that point where most models are built dynamically? Kind of, where's your thoughts on that? Because there's still some real challenges, just like you mentioned, the debt corkscrew and circular references and things that require complex lambdas or stuff that the average user just doesn't want to have to deal with. It doesn't mean they can't learn it. They just don't want to deal with that level of complexity.

Exactly. Exactly. And I'm not sure that it's really necessary for the average user to be fully dynamic. You know, I think people are just trying to get the job, whatever gets the job done the fastest. Sometimes that's a dynamic array and sometimes it's not. So, I mean, I think we're moving towards that. There are a few things that I think need to happen, like the formatting, for example. You know, the fact that when you spill—

Microsoft.

Yeah. Yeah. They spill the formula down and the formatting doesn't come with it. So, because of course, when you're copying and pasting, you're copying the formatting, but when you spill, the formatting doesn't happen. So something there like that's always quite, quite annoying having to fiddle about with the formatting.

100%. You'll laugh at this. Mark Proctor told him, he goes, well, can't you just, you know, make it so it spills with it? And the engineer's like, do you know how complex that was? Said something like that to him. And Mark's like, that's not my problem. You guys got to figure that out. Like, I don't care how complex it is. I'd love to have it. I think it was Mark. It might've been someone else, but it was one of the MVPs. I think it was Mark who said that.

I kind of love that. Sounds like Mark.

Yeah. Doesn't it sound like something he might say? All right. So this is a fun question I wanted to ask you. If you were to go back and audit every model you've built over your career, is there a theme, story, something it would tell you? Do you think?

It happens sometimes. People bring or say send me models that are years old and oh my gosh, I think, oh my goodness, I can't believe you're still using that model. And there's, you know, there's a much easier way of doing that now.

Yeah.

Yeah. I think like all of my models, I think anyone who's kind of seen me on LinkedIn, like I often post a lot of models and I have like a very specific Formatting that I use. Like, I think, you know, that I'm a little bit of a pedant. Yeah. So I like things to be—

I wasn't going to say it.

Yeah. You know that, you know. And so having the formatting, like, looking nice, I think is important. And so a lot of the time I probably spend too much time with client, like, making sure that the colors match the logo of the client, for example. example, you know, stuff like that, making it, making it look— because people do actually trust your numbers more if you use formatting, make it look professional.

So 100%, if I open a model and it looks really clean and professional, you automatically have a higher level of trust than you open it even if it's completely right, it looks messy.

Exactly, exactly. So yeah, I'm totally with you. I mean, AI, now we have that kind of layer of polish that it may— like, it looks good. And so you kind of have that inherent trust of it and you realize it's complete garbage. So that's, that's, that's kind of, that's kind of scary.

100% agree. I think we're— and we're on one of the worst spots in the sense of if it's total garbage, it makes it easier. It's when it's 95, 97, 98% right and you start getting lax because it's mostly right. And so you get lazy in your checking. And some people are going to get bit. I have no doubt there's going to be a big, just like there's always been with Excel and everybody will blame the tool or you'll blame AI, but there'll be some big, you know, $100 million or billion dollar loss due to something being done by AI and somebody not checking it. We all know it's coming.

Absolutely. Yeah.

It's probably out there and they're just, some haven't wanted to share it yet. So instead of asking you a favorite shortcut, since I've done that on the show before, your favorite dynamic array function?

Oh yeah, yeah, probably the one that my absolute go-to is the SEQUENCE. You know, we just, it's such a useful function. We just, particularly with timing and dates and things like that in modeling, you know, you're always having to look at timelines and the SEQUENCE function does that really well. The UNIQUE is also super useful. Yeah, just loads of those, those really, really simple functions that can just, that you can use in your model, especially the SEQUENCE. Love that one.

So I have to ask on SEQUENCE, when it first came out, did you look at it and go, well, what's really the purpose of this? Because I did. Like, where would I use that? Were you the same way?

Yeah, I just thought, oh, that's handy. Like, I can use it for lists. Like 1, 2, 3, 4, 5, you know, oh yeah, you know, and it took a while before I was like, wow, like this is actually really useful. I can use it in modeling. You know, I was competing in the Financial Modeling World Cup at the time when it first came out. And I was like, oh, I can do like, you know, 12 months across, you know, for 50 years and just do that very, very quickly. And, you know, it's just a really fast way of adding some timelines.

Well, I love how simple it is for like a sensitivity analysis. I just create lists and, you know, dropdowns and Got your price and COGS. You could be like, oh, I want to look at price across 15 different scenarios or 10 of the, of the cost. And instead of dragging anything, it's just, it's so simple. And when you show people, I can do it in training and I'm like, it's basically one function and a formula. Now the function changes, you just got to learn SEQUENCE and then you're building your formula in the middle of that. And people are just amazed. Like, is it really that easy? And so that's one of the things I love, one of my favorite to use SEQUENCE. For. So I'm with you. It's, yeah, at first I was like, I guess I might use it a little bit. I guess it replaces ROWS. You know, when you use ROWS or COLUMNS to increment.

That's kind of what I thought at first.

But it's way, yeah, 100% agree. It's way better now, especially when we understand dates and the more you use it, the more you realize, oh wow, this is a really versatile function. Rapid fire incoming. 20 questions. Yes or no answers. No hedging allowed. Circular references, VBA, the future of Excel, and more. Let's find out what our guest really thinks. Circular references in models?

No, no, definitely not.

I got that earlier today too. VBA?

Nah.

What about lambdas in our models?

If necessary, yes. Okay.

External workbook links?

Not, not if I can help it.

Should modelers primarily use keyboard shortcuts?

Yes.

Okay. Should models always be print ready?

Yes, but I will elaborate on that. All right.

We'll let you elaborate on the end. Are merge cells ever acceptable in a model?

Never, never, never, never.

You, I get such a variance on that. My favorite answer was somebody said, it goes, My saying I say around the audience, you around the offices, use merge cells, you go to hell. And so that was probably the most forward answer I got. I'm like, where do you stand on that again? So not surprised. Okay, should financial modelers learn Python in Excel?

Yes, but no. Yes, but I don't.

Okay, what about Power Query?

Absolutely yes.

How about Power BI?

Yeah. Yeah, useful.

So should every financial modeler be able to build a fully integrated 3-statement model?

Yeah.

Will Excel ever die? The eternal question.

Of course not. Never.

I already know the answer to the next one. I almost got ready to take that question. Have you used AI to help you build a model? Your answer is yes there. What financial statement is most important for you? Income statement, balance sheet, or cash flow?

Well, I tend to spend more time on the P&L. So P&L, probably.

Okay. Your favorite LLM, which one do you like to use?

Claude, mostly.

I'm notifying Microsoft.

No, I'm kidding.

I'm actually Claude as well. That's my favorite. So we're both on the same boat. Okay. If you could pick only one for all your models, you could either run sensitivity analysis or scenario analysis. Which one would you pick?

Scenarios are more detailed, I suppose. Suppose. So yeah, scenarios.

Do you believe financial models are the number one corporate decision-making tool?

From where I sit, yes.

Okay, fair enough. We get all kinds of answers on that one. What's your favorite lookup function? Which one do you turn to the most?

I like the XLOOKUP.

I'm with you. That's usually my go-to. That or Power Query, which isn't a lookup function, but—

Of course.

All right, so as we wrap up here, let me— you wanted to elaborate. I think you said print ready.

Yeah, on the print That is something that I find surprising about the FMI qualification is that you have to make sure that everything is printable. I guess some people must use it, but I don't know anyone who prints a model out, but it's apparently important. But I suppose it's like, yes, you do need to do it if you want to pass the exam, but I don't think anybody else would print a model. I don't know. Do you ever print models, Paul? No.

I'll send in a note telling them it's overrated.

Yeah. Yeah. But maybe it's important for certain industries. I don't know.

Yeah. And certain bosses sometimes, I guess, still need it, but I'm with you. I mean, many of the models I built, especially FP&A models, we had tons of cost centers and stuff. It just wasn't practical that nobody was printing those models.

Yeah. But if you want to pass the exam, you have to like hit all the marks and that's one of them in the FMR.

Yeah. Yep. That makes sense. Anything else you want to elaborate on there before we wrap up?

No, that's good. I think I kind of got my opinions in there.

Yeah. Any final thoughts you want to share?

Always nice to chat to you, Paul. A really interesting conversation. Thank you.

Thank you. Enjoyed it. And if people want to learn more about you, where should they go? Do you want to learn more about your offering?

Probably I spend most of my time on LinkedIn. So just looking me up there and my website is plumsolutions.com.au, which is, uh, Yeah, we'll throw both those in the show notes.

I know you're on LinkedIn a lot. We cross paths pretty regularly. Appreciate all you've done in modeling, and I've enjoyed following you now. I think it's been 7, 8 years, probably since I first saw you on LinkedIn, I think. So it's been a while now, and I've always appreciated your work. So thanks for coming on the show.

Thanks, Paul.

Financial Modelers Corner was brought to you by the Financial Modeling Institute. This year I completed the Advanced Financial Modeler certification, and it made me a better financial modeler. What are you waiting for? Visit FMI at www.fminstitute.com/podcast and use code PODCAST to save 15% when you enroll in one of the accreditations today.