FinPod

Meeyeon [00:00:14 - 00:00:41]
Hi, everyone, and welcome to another episode of "What's New at CFI," the series where we talk quite literally about what's new at CFI. And today's topic is a good one. It's about a course. It's called "Financial Modeling Guidelines." And I've got the perfect person to talk about it today because he's the one who built it. We have Duncan McKeen, one of your favorite instructors, certainly my favorite instructor. He's our EVP of Financial Modeling at CFI. Welcome, Duncan.

Duncan McKeen [00:00:42 - 00:00:46]
Thanks so much, Meeyeon. I'm excited to be here. Thanks, everybody, for joining in as well.

Meeyeon [00:00:47 - 00:01:14]
And so this is, this one's interesting because it's not your typical course. It's not, "Let's build a model from roughly scratch," or "Let's improve this model." At a high level, this is a, this is quite a different course. I think it's really, really fun and interactive and interesting for everyone that wants to become the most polished professional when it comes to financial modeling. But Duncan, maybe you could give us a very high-level walkthrough of what this course is about and what the approach is. Yeah, definitely.

Duncan McKeen [00:01:15 - 00:02:14]
Thanks, Meeyeon. Well, I will say, first of all, this did not start as a course. It started as a document, like an e-book. It's about 90 to 100 pages long. And we came up with the idea of making it into a course a while ago. And the idea with making it into a course is basically that it's much more interactive and applied for everybody. And essentially, the way the course is designed is that we start with a model which really has just an awful structure. We start with a model which is essentially not followed any of our guidelines. And through the course, we effectively will be rearranging and moving things around in the model so that by the end of the course, we've brought it up to all of our highest possible industry standards and followed the guidelines the whole way through.

Meeyeon [00:02:15 - 00:03:05]
I love that because instead of starting from scratch, which is less realistic in an actual real-life job setting, you're basically being handed what sometimes we would call a legacy file, which can be a little bit of a mess because it'll be one associate touches it, one VP touches it, one analyst touches it. And before you know it, it becomes kind of a behemoth of a mess. And learning to clean that up is a very realistic task that you are going to be faced with as an analyst or associate time and time again. So such a real-world way to learn.

I have to ask something. This is kind of like personal experience. And you can be honest here. When you look back at your early days of building models, is there anything that you used to do that now makes you cringe a little?

Duncan McKeen [00:03:06 - 00:04:09]
Yes. Well, lots of things. First of all, what one that stands out was that you're working in equity research. You're always dealing with quarterly models, of course, because you're trying to like forecast, you know, like the earnings that the company is going to release in the next quarter. And so one of the things I used to do, and it's actually it's very common, you're going to encounter it, is I would build a model where you'd have quarter one, two, three, four, and then you would sum that into a year in like the fifth column. And that is not a great structure at all. You know, the better way to do that is to have one tab, which is only quarters. So you'd go Q1, two, three, four, and then you go into the next Q1, two, three, four all the way across. And then to get years, you'd have a separate tab where you'd usually use some if or some ifs function to basically some things in into years. But yeah, that's one that really stands out that makes me cringe that I personally used to do in the early days of equity research.

Meeyeon (00:04:10 - 00:04:15)
And so with that in mind, who do you think is going to get the most out of this course?

Duncan McKeen [00:04:17 - 00:05:35]
Oh, gosh, I think that anyone working with financial models will get a lot out of this course because before there was this course, someone would go through one of our modeling certifications. They would see our best-in-class models, but they wouldn't know exactly why they were built the way they were. But now, you know, with this course, somebody can go through this course and then they'll come away knowing exactly why we do things the way they do.

And so then they can take all of those principles and apply them to like those old legacy models that you referenced earlier, because as you mentioned, Meeyeon, and you were quite right about that, is that it's more likely that people are going to start with some existing model rather than building from scratch. And when they open up those existing models, it's probably fairly likely that they're going to be a bit of a disaster just because a lot of people haven't taken financial modeling training. So then they can come into those models, see what's wrong with them right away because they understand the guidelines and then start to apply those guidelines and fix the models.

Meeyeon [00:05:36 - 00:06:07]
And as people make their way through this course, they're going to learn what makes a good model. But just talking about the concept of what makes a good model, you've seen a lot of models over the years, probably more than most people see, like the average person per se, in a lifetime of models. So when a really good one lands on your screen, what do you think immediately stands out to you on, like, a first pass of what makes a good model?

Duncan McKeen [00:06:08 - 00:07:11]
Surely. Yeah, I would say that most of the models I see are usually a disaster only just because I also used to do a lot of consulting work. And if anybody's coming to you for financial modeling consulting, they already have a disaster on their hands that they need help with. But even apart from that, yeah, a lot of models are not built very well, just because of that, you know, a lot of people haven't taken formal training. But probably the first thing that stands out when you open up a good model is probably the formatting. It's the first thing. It's kind of like your first impression. But then, when you dig a little deeper into a good model, you'll start to see, "Wow, they've broken large formulas into small formulas. You know, that's really nice. Or they've gone through the effort of segregating everything into small little segments or schedules. You'll see that the model is printable, has nice output pages. So those are all kinds of nice things that you see. But definitely, formatting is the first thing you notice right off the bat.

Meeyeon [00:07:12 - 00:07:31]
And then on the flip side, it's funny because we actually had a conversation earlier in the week about something along this topic. But what do you think is, you know, one of the worst model that you've ever seen? And what is, what is like what characterizes what you think is like that bad?

Duncan McKeen [00:07:32 - 00:08:43]
Okay, I'm not going to mention any names here. No names. No names.There was a model that we got brought into review on a consulting mandate. And there was one formula in the model where, I mean, everybody on this podcast probably knows that, you know, you can obviously look up at the formula bar and, you know, see the formula. Well, when we looked up at the formula bar, we couldn't see the whole formula. So we opened up the formula bar. Most people would do that with their mouse. You can actually hit Control-Shift-U to open the formula bar up. And we hit Control+Shift+U and opened it up. So it filled the whole screen. That still wasn't the bottom of the formula. There was a scroll bar, which appeared. So it was like full screen. And now we're scrolling up and down, looking at this formula. And that was probably like the most extreme example of a really, really complex formula that desperately needed to be broken into its component parts. So that's probably the most extreme example for sure.

Meeyeon [00:8:45 - 00:10:53]
For me, this is not necessarily anything to do with like poor analysis or really anything significant, but it was so memorable for me. I'll never forget. There was this, and this was when I was working in DCM. There were certain files that a colleague of mine worked on. And he would do like 95% of the work and then kind of jam through the last bit. And in that last bit, what he ended up doing was putting hard-coded cells that he would hard-code and reference on the non-printable side of the view of the Excel spreadsheet. And then he would turn that into a white color font. So he would, if it's in the printable area, he would put in a white color font. If you put it on the other side, it was still in white. And so there would be certain times where you're auditing the sheet and you can't understand why these numbers don't tie.

Well, the cells that he's referencing look empty to you. But if you go Ctrl A and go in the whole thing and just turn all of the text into black, you would see that there are these like very, very random references. And so for me, that is one, it's like this is from 10 plus years ago, but it's so memorable because it was within the first three months of the job. And I just remember thinking, I was like, hey, why is this doing this? And my then colleague being like, oh, it's just because I turned this thing white. And I was like, "Oh my gosh, how would I ever know this if you didn't tell me? And his response was like, well, that's why I'm here. And so that's one of the things that I kind of learned. I was like, OK, like I'm never going to do this because it's extremely specific to the individual. It's very hard to pick up where you left off and colleagues have to take it. But it's it's a unique, Eric, let unique poor modeling practice that I've experienced and one, one that I kind of like to share because I really don't know anyone that's encountered that before.

Duncan McKeen [00:10:54 - 00:12:10]
Yeah. And it's, and that's a crazy story. And I've definitely seen sort of similar things where people are hiding using different techniques, white font being one of them. And for people on the podcast, you know, I mentioned DCM, which is, of course, debt capital markets. And one of the things that that story really just breaks one of the guidelines that we discuss, which is transparency. So, like, you know, everything in a financial model needs to be visible either when you're looking in Excel at the screen or when you go to print the model.

And one of the easiest ways to test if a model meets that transparency benchmark is simply to print it out. And you should be able to go from the top of the model all the way to the bottom and figure out how to calculate each line item. If it's fully transparent, you'll be able to do that. And if it's not, you'll get stuck at certain points along the way where somebody's buried too many details of a formula inside a cell. And you can't see enough to kind of get to the next step or the next line item. So but yeah, definitely, that's a great, you know, that's a great example of something that's...

Meeyeon [00:12:12 - 00:13:09]
Yeah. This is what I think is the most valuable and interesting part of the course, because it's not just about reinforcing the idea that, you know, you need to get all the numbers right. Everything needs to tie. But it's about building something that other people can open up your colleagues, your team members, understand, trust. and then additionally build a pond so that you can have something that is a reliable model that people can use and reuse in the future.

Real-world question: How has your own approach, do you think, to building models changed over the course of your career? Like, do you build differently now than you did 10 or 15 years ago? Like, for sure, for sure you do like today, for example, because we have things like, you know, Claude in Excel that or Claude in Excel that we never did before. But just overall, how has your approach changed, do you think, over your career, or has it maybe not really changed so much?

Duncan McKeen [00:13:10 - 00:14:56]
It's an interesting question. I think that maybe maybe there's I could answer it maybe in two parts in terms of the like, like the goal or the end product that we're trying to get to. I think that hasn't changed too much over the years. It's still aiming for always like the best in class product. The only small changes which have happened along the way have maybe been because of like advancements in Excel, for example, like, you know, like we can put checkboxes into models, you know, which came out not that long ago in Excel.

So, of course, we started to incorporate those into our models and use them a little bit. So that changed the end product a little bit. But the thing that has changed in this continuing to changing it will continue to change, sorry, is the way that we build or what we use to build. So we're starting to use AI a lot more to do a lot of the want to call it the labor-intensive modeling work. So I can do a first pass of a financial model, get you with a working draft to start with. If you have inherited, inherited it, a industry model, I can definitely do a first pass of an audit of that model and find some of the more obvious mistakes that might be in it.

But then, of course, we always will take over and do a full end-to-end audit done by a person to make sure that the model is error-free, and also often we're finding that the human also needs to often restructure the model to get it into the correct format that we talk about in our modeling guidelines course as well.

Meeyeon [00:14:57 - 00:15:05]
And for someone who goes through this course, what do you think they'll actually do differently afterwards? And what do you think will change for them?

Duncan McKeen [00:15:06 - 00:15:56]
I think that they will look at financial models differently. They will know why things are the way they are in our models, I think. And in terms of what will they do differently, I think they're after taking this course, they will be more likely to fix and improve models that they encounter in the workplace, I would say. You know, I think after this course, when they encounter models that aren't up to these standards, I'm hoping that it will bother them. You know, they will see a bad model. They'll be like, "Oh, wow, that really bothers me. It shouldn't be like that. And I know exactly now how to change it and I will change it. So I'm hoping that that's the outcome and that was the intention of the course.

Meeyeon [00:15:57 - 00:18:39]
Thinking back to this period of my life, I think for a lot of people who go through an investment banking analyst training program, that program, at least for me, and I think a lot of my friends who were in it, is a really monumental time in your career. I think that week of condensed learning that I did with instructors that came into the classroom really did set me up for skills that I built on for the rest of my career, and were so instrumental in me being able to build the skills that took me the rest of my career, and I still value today.

This course really is kind of like a condensed version, I think, part of the analyst training program. And what I hope changes for anyone that takes financial modeling guidelines and what the investment banking analyst training program changed for me was the confidence and the mindset that I went into opening any Excel file with. And so, for example, you know, prior I was hired off cycle, I had six months of working prior to the training program. And basically every day was like trauma where I would get these Excel files, I would open them, and I would be like, "Oh, am I ready for this? But what happened afterwards was that after I learned, you know, financial modeling guidelines and how to approach models, how to read them, I became a lot more confident and calm. When I was opening files, I wasn't afraid about what I would encounter. I wasn't stressed about the number of hours and time it would take me to figure out how to understand the model first and then do the actual task that I'm being asked to do.

This type of course significantly improved my confidence that I walked into every task with. It really allowed me to take what I thought was kind of like a mess of like, okay, like how do I improve things? How do I read this model? It took all of these kind of ideas that I've had and really structured them in a way where it's almost like if you were a craftsman, like a leather craftsman, you start with your working table and your bench, and you clean everything up. You have like all your tools on one side, all your leather on one side. And like that's kind of how I envision this helping students after they do take the course. Long-winded experience. But I don't know, I find that the whole journey, so to speak, of learning how to build audit, read financial models and do so confidently is a really meaningful experience for anyone that's working in finance.

Duncan McKeen [00:18:40 - 00:19:48]
Yeah, it definitely is, Meeyeon, for sure. And you were lucky enough to have that experience in person. And also, what I will say is that when Meeyeon went through that training program, at that time, there was no CFI offering that type of super high-quality training to people remotely and at a really reasonable price. You know, Meeyeon went through a very expensive training program there. And one of the things that we've tried to do at CFI is make that training literally available to anybody, anywhere, and at a very, very reasonable price. But one thing I maybe would add to that is that, you know, although that was really an amazing training program, I don't think they actually would have had a course like Financial Modeling Guidelines specifically. It would have had a lot of courses on financial modeling. You probably would have used really great templates, super well-designed templates, but not necessarily learned exactly why those templates were structured the way they were, which is, of course, what we'll cover in this guidelines course.

Meeyeon [00:19:50 - 00:20:30]
Yeah, for us, it was mostly kind of like, this is how the models that you will work with 90% of the time will look like, this is how it should look like, these are the general guides in terms of like formatting and presentations that you should aim for. And then, you know, when we had the time, it was kind of like the why that it was structured in certain ways, but that was the part that you were kind of, I think, expected to apply on your own in your role because there was only so much time for it. The last big question, if you could get more people to build just one habit when they're modeling, what would it be and why? I think I know your answer.

Duncan McKeen [00:20:31 - 00:20:32]
You do know my answer.

Meeyeon [00:20:32 - 00:20:33]
I know your answer.

Duncan McKeen [00:20:33 - 00:21:32]
Yeah, keyboard use for sure. 100%. It's like, and I wish somebody had convinced me to start using the keyboard younger, at a younger age, earlier in my career, like I wasted hundreds of hours. I probably wasted, you know, weeks or months of my life in lost productivity using the mouse in the early days. I think the reality is, if you're working in finance and you know that you're going to be working on financial modeling, just, you know, ditch the mouse immediately and start using the keyboard. You'll thank yourself later because it's just, it's so much more efficient, and it does not take that long. If you quit cold turkey and you're working in Excel or in models quite a bit, I want to say it only takes a week or two. And by saying it only takes a week or two, I mean, after that short amount of time, you'll get to the point where you're more comfortable and faster on the keyboard and then you'll never look back after that. 

Meeyeon [00:21:33 - 00:22:04]
Yeah. The idea of like, for example, think about the idea of selecting like a column and you're going from cell B6 to 64 and you're like, you know, scrolling and scrolling and you lose it. Oh my God, you have to go back. You know, like Control Shift arrow down. You're going to save a lot of time over the course of days, weeks, years.

I initially was very intimidated by it. But then after, you know, after the training program, I was like, okay, I see the value in this. I need to lean into this. But when it came to, you know, like answering emails and the stuff, I was still on my mouse.

Duncan McKeen [00:22:05 - 00:22:06]
Yeah, for sure.

Meeyeon [00:22:07 - 00:23:17]
Yeah. But overall, like I think there's a great there's great value in that. And no matter how good I think I get, we're still going to be encouraging the use of the keyboard is always going to help regardless. I will provide shortcuts.

Now, I hope that this has been an insightful episode for everyone listening. And if you haven't taken this course yet, Financial Modeling Guidelines is going to be your staple course, I think. Regardless of which certification, specialization, or general course you're interested in. If you are interested in a career in finance, you're going to need to use Excel. You're going to do data analysis and financial modeling guidelines. It's going to give you such a strong foundation and really allow you to present yourself in polished professional ways that are going to make you stand out from the crowd.

So, thank you so much, Duncan, for joining me, and thank you everyone for listening. If you haven't taken this course yet, I hope you will. And if you want to learn anything more about our little preferences and habits when it comes to Excel or financial modeling, feel free to follow us on LinkedIn on our personal profiles or at the CFI page. And until next time, everyone.