Mastering SaaS Models: Derek Baker's insights on building robust SaaS models
Show Notes
In this episode, Paul Barnhurst is joined by Derek Baker, who is drawn to really hard problems. Derek discovered early in his career that he loves startups and building companies. He began his career working for SaaS startups building out their core FP&A processes. For many of the companies he worked with, he was building their very first financial model to help them operate the business and make capital allocation decisions. He found that every business - regardless of stage - can improve performance with excellent FP&A.
Today, he works at Neighbor.com, a Series B startup, building the “Airbnb” of personal storage. Building the FP&A function for a marketplace has been the biggest challenge of his career, and he is loving every minute of it.
Listen to this episode as Derek shares:
How he got into financial modeling for SaaS companies
His advice for building robust SaaS Models.
The importance of understanding a business to build better Models.
Everything about SaaS and SaaS Companies.
His position on controversial modeling issues, including circular references, dynamic arrays, modeling standards, AI in modeling, and more.
Quotes:
“The only way you're really going to learn how to build a model is by building it. You can do all the study you want, but at the end of the day, it's like anything, without practice, it's just theory.”
“The better you know an industry, the better you know the operations of the business, the easier it is to build a model.”
“[What separates good FP&A from great FP&A] I think it's 1- business acumen and 2- business partnering. Anyone can build a model in Excel and can learn those technical skills. But using those technical skills to help the business and partner with the business to help them understand the business better and grow the business is exactly what FP&A should be doing.”
Sign up for the Advanced Financial Modeler Accreditation or FMI Fundamentals Today and receive 15% off by using the special show code ‘Podcast’.
Visit www.fminstitute.com/podcast and use code Podcast to save 15% when you register.
Go to https://earmarkcpe.com, download the app, take the quiz and you can receive CPE credit.
Follow Derek:
LinkedIn - https://www.linkedin.com/in/derek-d-baker/
Website - https://community.plan-buddies.com/home
Follow Paul:
Website - https://www.thefpandaguy.com/
LinkedIn - https://www.linkedin.com/in/thefpandaguy/
TikTok - https://www.tiktok.com/@thefpandaguy
YouTube - https://www.youtube.com/@thefpaguy8376
Follow Financial Modeler's Corner
LinkedIn Page- https://www.linkedin.com/company/financial-modeler-s-corner/?viewAsMember=true
Newsletter - Subscribe on LinkedIn- https://www.linkedin.com/build-relation/newsletter-follow?entityUrn=7079020077076905984
In today’s episode:
(00:22) Intro;
(00:45) Welcoming Derek;
(01:00) The worst Financial Model Derek has ever seen;
(02:00) Key takeaways from the worst Financial Model;
(02:41) Derek’s background;
(04:41) Experience and Challenges while building the first Financial Model;
(07:20) What Derek thinks was missing from Finance Education in school?;
(10:13) What is it that Derek likes about SaaS and SaaS companies?;
(12:31) Key inputs in SaaS Revenue Models;
(15:08) What is the first tool that SaaS Startups must invest in?;
(16:40 - 17:25) Validate your Financial Modeling Skills with FMI’s Accreditation Program (ad);
(17:26) Building a Three Statement Model - SaaS vs other Industries;
(19:00) Building Deferred Revenue;
(21:56) How to get started with building SaaS Models?;
(24:05) How often does Derek use Dynamic Arrays?;
(27:20) One thing that helped Derek in building Financial Models;
(29:20) Technical advice that can help people become better Modelers;
(32:26) Derek’s favorite Excel Shortcut;
(32:58) Rapid Fire;
(36:09) Nugget;
(36:55) Connect with Derek;
(37:21) Outro;
Full Show Transcript
Host: Paul Barnhurst
Welcome to Financial Modeler’s Corner, where we discuss the art and science of financial modeling with your host Paul Barnhurst. Financial Modeler’s Corner is sponsored by Financial Modeling Institute.
Welcome to Financial Modelers Corner. I am your host, Paul Barnhurst. This podcast is where we talk all about the art and science of financial modeling with distinguished financial modelers from around the globe. The Financial Modelers podcast is brought to you by Financial Modeling Institute. FMI offers the most respected accreditations in financial modeling.
I'm thrilled to welcome this week's guest on the show. Derek, welcome to the show!
Guest: Derek Baker
Thanks. Glad to be here!
Host: Paul Barnhurst
Great. Well, I have Derek Baker here with me, and we're going to talk about SaaS. But before we get to that and SaaS modeling, we're going to start with a question we ask everybody.
Tell me about the worst financial model you've ever seen.
Guest: Derek Baker
Yeah, the worst financial model I've ever seen is probably one that I've created. Sometimes I have to go back and look at the old models I've built for earlier companies, and I'm always appalled at some of the bad habits that I had earlier in my career when I was first getting started with financial modeling. Some of those are hard codes within formulas, and those are my pet peeve today. I would never want to put a hard code within a formula, but when I find them in old models, it just makes me cringe a little bit and sad that I ever made that mistake. So, the worst models I've ever seen are ones that I've built myself.
Host: Paul Barnhurst
I've built some awful ones. One of my favorite. I still remember as someone on the show mentioned how they had built a model and they had a second model, and they had to take a number from the first model and input it into the second model, and they'd literally, it was so big, it would crash their machine to open both of them. So they were telling me they'd open the one, take a picture, and then type the number in the next one. Yeah, that's pretty bad. He's like, I would never do that now.
What were the key takeaways from those early models? You mentioned never hard coding. Any other kind of key learnings you took from those first few models you built?
Guest: Derek Baker
Yeah. So never hard coding. I think another important one is simple formulas, which I maybe have too complex formulas at times today, but back then, I would put huge nested ifs into formulas, and I would put a lot of logic in multiple formulas when I could have break those out into multiple lines and have different flags that made the formula easier to read, and audit. That as well, is something that I've gotten better at over the years.
Host: Paul Barnhurst
Yeah, you and me both. I still have times they get more complex than I'd like, but it really is about keeping it simple when you can. So thank you, that's helpful.
So why don't you tell our audience a little bit about yourself and your background?
Guest: Derek Baker
Yeah. So I currently work at a company called Neighbor Storage. It's like Airbnb, but for storage. I did finance internships. I studied finance at BYU and did many internships throughout college. One of the first internships I did was at a company called InGeneron, which is a biotech startup. And that was where I built my first financial model, and my project for the summer was to build a financial model in Excel and then translate that financial model into Power BI, which this was back in 2020, and I think just in the last year or so, I've been seeing more and more FPNA professionals talking about building modeling tools within Power BI. And so he was way ahead of the curve with this idea of building something in Power BI.
So that was my first experience, getting into three SIT and modeling and having to build something in Power BI meant that you have to understand a three SIT and model inside and out. And so I learned a lot about modeling over that summer. And then I took all those experiences to a SaaS startup called Savvy Legal and built them a lot of the same tools that I built for in InGeneron, and I worked at Savvy Legal on and off throughout college. I would leave for a summer, go do an internship somewhere else, then come back, and while I was in school, I'd work for them again.
And I eventually went to a company called Client Book, which was my first real job out of college, and they were another vertical SaaS company selling tools to jewelry stores and furniture stores, things like that. I was their first finance hire, and so I did a lot of their FP&A, just getting started, built their financial model, and helped them to raise a little bit of equity and venture debt. By the end of my time there, I started to realize that I was taking on a lot of responsibility that I didn't feel qualified for, and I wanted to learn from somebody who had been and gone through a lot of those experiences I hadn't been to yet. And so that's when I found Neighbor and went to go learn from the CFO at Neighbor, and it's been a great experience there, building out their initial FP&A processes and budget versus actual meetings and their budgeting and re-forecasting processes. That's what I've been doing for the last year and a half or so.
Host: Paul Barnhurst
Great, and I appreciate you giving a little bit of your background and sounds like you got a lot of internships, a lot of great experience. You mentioned that first company you're at, the biotech. So it sounds like that was your first three statement model you built, financial model. What was that experience like, trying to build the first one in the real world, so to speak, versus anything you may have done in school, or what were maybe some of the biggest challenges, and just walk us through that.
Guest: Derek Baker
Yeah, so at that point, I was just barely getting started in my finance major, and so I'd taken a basic accounting class and a basic finance class, hadn't really gotten into the modeling yet. And my boss was a CFO at InGeneron, and he was an investment banker before that, and he just gave me a book. He said, here's this book, we use it to train investment bankers at Goldman Sachs, just follow this book, read it, and then go and pick a public company and build a three-statement model for them. And so he didn't really like, he was very hands-off, he just gave me the book and said, go do it. And it was very painful. It took a week to build a very simple three-statement model for I chose Apple, and I was trying so hard to get the balance sheet to balance. And I think where I was getting most caught up was the depreciation side of things, like making sure the depreciation tied to the balance sheet. And I just remember getting on the call with him and saying, I can't get my balance sheet to balance, and it took him five minutes to figure out what I'd done wrong and he fixed it and it balanced, and everything was great.
So it was a very painful process. But having to learn things the hard way like that and with very little instruction set up a really good foundation for me because I never made that same mistake again in my future three-statement models.
Host: Paul Barnhurst
Sounds like a great experience, I mean, sometimes being thrown in the deep end is the best way to learn. Not always, but a lot of times it forces you to figure out things that you wouldn't be able to if you just always had someone giving you the answer right away and can totally relate to going to somebody else, and they'd look at it and go, here's your issue. So it sounds like the biggest challenge you faced was learning to build that model, making sure the balance sheet balanced the typical things that today would have been fairly straightforward.
Guest: Derek Baker
Yeah, and I think that's going to be true for the things that I'm struggling with today. Like, in five years, I'm going to look back and be like, oh, that was such an easy problem. And then you move on to harder, more complex problems. So just the natural evolution of your career.
Host: Paul Barnhurst
100% agree. And I once heard, it's not that the nature of the task becomes easier, that is what's happened, it's really the fact that you've learned it better, right? So you can do it quicker. It's still the same task, same complexity to it, but it's much easier for you now just because you've learned. I like writing Excel formulas, formulas that I would have struggled to write five years ago now might be simple.
So in school, what do you wish they had taught you about three-statement modeling? Maybe, hat do you think was missing from kind of, some of the finance education you got?
Guest: Derek Baker
It's a good question. In my finance program at BYU, we did very basic P&L modeling that was part of that core finance program. I wish that was more complex, I wish the three-statement model was actually part of that corporate finance class that was part of the core program. I didn't learn three-statement modeling in school until I took an elective for a private equity and venture capital class, and that's where I was first taught in school how to build a three-statement model. And by that point, I already knew how to build one. That was the first thing. I wish they would just teach three-statement modeling at all as part of the core finance program. But I think what was missing from that three-state model that we built in my elective class was how to operationalize it, and that was kind of not the point of the class. The point of the class was to evaluate an investment by a private equity or venture capital firm. Those models, they're not living, breathing models like they are in FP&A. In FP&A, you have to be able to roll forward your financial models, usually on a monthly basis or at least on a quarterly basis. And having to rebuild your financial model from scratch, like a private equity or investment banker does, isn't feasible for FP&A. And so I had to learn a lot of stuff on my own, how to make these models dynamic and able to roll forward and easy to update.
That was something that definitely wasn't taught in school and in any of the classes I took.
Host: Paul Barnhurst
Yeah. And there's a real difference between building a project model, a model where you're evaluating something, what you think of as a one time, and a model that's operational that you can use going forward. I remember building a few models for some M&A we did, and then you trying to take that, and how do you make sure you can use that for a forecast, trying to think through. Because they're just different. And so the more experience you have on both sides, the easier that becomes because you can start thinking about all those little things as you're building it to go, okay, it needs this and this to be dynamic when I'm done, because almost no matter what project model, operational, forecasting, whatever the model is, they're almost always going to want it to be dynamic, when you're all done, they're going to want to use it for something else. And if you don't build it that way, it's really painful.
Guest: Derek Baker
Sometimes it's like fitting a square peg in a round hole, like for a private equity or a venture capital investment, it's all about at this point in time, what do we believe is possible, but in an FP& model, an operational model, it's about how do we create the future, like, how can we use this as a tool to drive and steer the business, which is not the purpose of a private equity or venture capital investment.
Host: Paul Barnhurst
Yep. And so there's definitely times you just got to build a completely different model. There are times you can use them and there's times when, hey, it just makes more sense, they're different enough, and the purpose. Would agree with you on that.
So, you know, you've worked for several SaaS companies. You mentioned Neighbor, Client Book, Savvy Legal. What has interested you most about SaaS? What is it that you like about SaaS and SaaS companies?
Guest: Derek Baker
Well, I'll make one small correction. Neighbor is SaaS like in the sense they have recurring revenue, but they're actually on marketplace. So it's actually a very different model than SaaS. But what initially drew me to SaaS was opportunity startups. They give you a lot of leeway to learn and to grow within the business, and I love that in startups I get to do everything. I am not just on one product of one business unit and going really deep on this one area of the business. I have to understand the whole business across functions, across products, across departments, because I'm the only one that's doing FP&A. I really like that opportunity of startups.
And then, SaaS companies in general, the reason why I stay there is because the recurring revenue makes the business very compelling. We see that a lot of the high-burning startups that went public in the last five years or so, they are now very profitable today because of this recurring revenue. They are able to turn off a lot of that sales and marketing, go to market motion and become very profitable and throw off a lot of cash flow, and I think that's the promise of SaaS startups, is that maybe you're high-burning, high-growth in the beginning, but in the future you're going to be highly profitable and throw off a lot of cash flow and there's a lot of opportunity there. So, I enjoy tech mainly though because of the opportunity and all the different areas that you get to touch in startups, that you don't really get that opportunity in large companies.
Host: Paul Barnhurst
Yeah, for sure. With the startup, you get to see everything. Having run my own business now for two years, I've done a lot more marketing and advertising and all kinds of things that I really didn't appreciate working in large corporations. So I can totally relate to what you're saying there.
One other thing you mentioned when you talk about recurring revenue, right? The dream everybody wants now is that recurring revenue, it's so much like an annuity. You got to put money in upfront, but then it pays out kind of into perpetuity. And that's the idea. If you can keep that churn low and continue to expand those existing customers, it's a great model, right? It's all about churn and how much it costs to get them. Can you make it work long term? But it's a beautiful model.
So talking about SaaS, I'm curious, when you build a SaaS revenue model, let's talk what are some of the key inputs that you're looking for? How do you start, because I'm guessing you start with the revenue and then build the rest of the model.
Guest: Derek Baker
Yeah, I definitely start with revenue. So if I were walking to a company and I didn't know anything about them, I would ask for the export of all their subscriptions that they've ever had active and canceled. And the first thing I would do is I would create a renewal schedule with their active subscriptions, and then I would do some analytics on what is their churn rate like in the past, I’d do some analytics on what is their expansion, their net revenue retention. I would start with just the subscriptions that they've had, because those are the key inputs on the revenue side of things. In a SaaS business, you build what's called a revenue waterfall, where you start with opening monthly recurring revenue plus new monthly recurring revenue, which is through the go-to-market motion plus expansion, monthly recurring revenue minus any churned, like lost business, minus contraction. And so having that subscriptions export, that's accurate and is up to date, is really important for making that MRR waterfall.
So that's how I start with any SaaS business, is build the MRR waterfall and forecast those individual components because that's going to drive the revenue in the business.
Host: Paul Barnhurst
And something you mentioned, you said, accurate. How often is it accurate when you first get the data? Do you find the data is often pretty messy?
Guest: Derek Baker
Yeah, it depends on the stage. Oftentimes they don't have CRMs at really early-stage startups, if they're just raising a series A, for example, their CRM probably isn't set up the way FP&A needs to do those MRR calculations. If they have a good billing platform, like Maxio, for example, which is expensive, so a lot of companies don't have that until later on, then it gets really easy. Maxio is one of the best SaaS billing software out there.
So it depends. It depends on what they're using for billing and what they're using for their CRM. I consulted a startup that had someone manually send out invoices every year on the renewal date, and they didn't have a billing software. I don't know why they're doing that. I think they just didn't know better and their CRM was not set up in the right way. And so it was a very manual process, sitting down with a founder and lining up all their customers and trying to figure out, okay, when do they start, when do they churn? And luckily they were small, slowly had, I don't know, 100 customers or so. But it was painful. Investing in those tools early matters a lot in SaaS startups.
Host: Paul Barnhurst
So speaking to that kind of a SaaS startup, what do you think is the first tool they should be investing in? Is it the CRM, is it the billing platform? And how early, you know, kind of from your perspective, I get there's no exact right answer.
Guest: Derek Baker
I would say from my perspective, the billing platform should be, especially because it's so easy, like just go use Stripe or something. Stripe has great subscription management tools. There's no reason to be using Quickbooks invoices if you're a SaaS startup. So I think the first thing you should do is get a billing software, because a billing software that is built for doing subscription management will allow you to do all those SaaS analytics that I've mentioned so far.
Host: Paul Barnhurst
With most of those companies, I'm guessing, do you still find even if they have a billing platform or CRM, unless they have something like Maxio, I'm guessing you're almost always doing those SaaS analytics, export to Excel and doing them in Excel.
Guest: Derek Baker
Yeah, when I've built them at companies, I did them in Power BI because I'm comfortable with Power BI and you can automate a lot of that stuff with Power BI and do more complex analytics there, better visuals as well. So that's my preference, is Power BI. But the first step for me would always be Excel because it's easy to see the data right in front of your face and to see how the calculations work. As a proof of concept, I think Excel always makes the most sense for those SaaS analytics.
Host: Paul Barnhurst
Excel can't be beat for a proof of concept, I agree with you. And then, okay, I figured it out, I've worked through the nuances and the data, now let me go see if I can build it in a BI tool, if that's the route you want to go, because there's definitely advantages. I get it, the visuals, the automation, a lot of things that you don't have in Excel.
Commercial break:
In today's business world, financial modeling skills are more important than ever. With Financial Modeling Institute's Advanced Financial Modeler Accreditation Program, you can become recognized as an expert in the field by validating your financial modeling skills. Join the Financial Modeling Institute's community of top financial modelers, gain access to extensive learning resources, and attain the prestigious Advanced Financial Modeler Accreditation.
Visit www.fminstitute.com/podcast and use code PODCAST to save 15% when you register.
Host: Paul Barnhurst
So I'm curious beyond kind of the revenue, and I know that's a big part, the whole subscription. What other differences do you see with SaaS maybe versus other industries when it comes to building your three-statement model? What are some of those other key differences?
Guest: Derek Baker
Yeah, so SaaS is very asset-light. You don't have inventory or fixed assets, so you don't have depreciation really either. I mean, you may have a little bit, you know you have to buy computers and tables and things like that, but it's very asset-light. So Capex is really not a big part of my SaaS models, except for new laptops for employees or refreshes. But again, those are very small components of the model. There are usually no liabilities at these companies except for, I should say, no long-term liabilities. They don't have debt most of the time. Sometimes you'll get into a venture debt or revenue-based financing, which makes it a little bit more complex, but oftentimes it's just straight-up equity, which makes it very easy to model on the liability side.
Another big difference is, I think most companies have some deferred revenue of some sort, but that's very important in SaaS companies because of annual subscription billing. You have to understand how deferred revenue is going to change over time. And that's another reason why starting with a subscriptions table is so important when making a SaaS financial model, because that'll inform you what your deferred revenue balance should be today and how that's going to be recognized until the next renewal.
So those are some key differences between most businesses and SaaS businesses.
Host: Paul Barnhurst
So I'm curious, a couple of questions it leads me to on that. First, you talked about the deferred revenue. How do you think about building that? How do you go about understanding that? Because I know there's billings, there's bookings, there's revenue, and over the long term, they should all equal the same, but there's a timing difference between all of those. How do you kind of manage that? Any kind of thoughts or advice there? How do you think about that?
Guest: Derek Baker
Yeah, I think the first step is to understand what's going on in the business, understand the operations behind it. And this is where it gets, I'm even getting too much into the FP&A, rather than the financial modeling side of things, but what I would do before even touching Excel is go talk to the salespeople and ask them, when do you consider ARR to be booked? And it's usually when they sign the contract. And then what's the step after that? Do you send it off to the accounting team, or how does it go from being a bookings to collected payment? So bookings to cash collections, usually there's some kind of delay there because if you have payment terms, no salesperson is like withholding a contract or trying to keep payment from happening. But oftentimes it's those payment terms which are usually 30 days for most SaaS companies I've seen. So contract signed, then you get over to accounting, they usually send the invoice pretty quick, and then you have to wait 30 days. And then from there, once the cash is collected, that's when, usually you start the revenue recognition, when the cash is collected.But it really just depends on the company's policy.
And also if you have any implementation fees, that also is different as well. Those can be recognized over a couple of months or however long the implementation period is. And sometimes you don't even start the recognized revenue of the subscription until the implementation is over. So understanding, and this is where I like having a CPA around because I don't have the accounting background, understanding how the revenue recognition is handled in your company is really important.
And then you go into the french model. And so when I build it into a french model, I build everything around the go-to-market motion. So a sales rep closes that, gets that contract signed, and that's a booking, and then I'll have some kind of offset of a couple of months, however long it takes to go from a booking to starting that revenue recognition. And I've seen a lot of professionals do a bookings waterfall where they like break everything down by cohort and then forecast out the revenue recognition by month. I think it's a nice way to visualize it. I don't build it that way because it's creating 30 lines for where you only need one line. So I just build formulas to do sumifs and things like that to build the bookings or the revenue recognition on its own line, so I don't have to have a bunch of lines mudding up my financial model. But it is a nice way to visualize it that way.
Host: Paul Barnhurst
Yeah, no, there's definitely pros and cons. I think the waterfall is good for people to see that visual of that deferred. I've had some where you build offset ramps to manage all that stuff. There's a lot of different ways you can do it, for sure.
But, next question I have for you, if we have someone listening and let's say they want to learn more about SaaS and kind of building a SaaS model, how would you advise they get started? Say they're not working in a SaaS industry today, they're in school, whatever, but they're really interested in SaaS and would like to learn more about building a model.
Guest: Derek Baker
I would say go to The Saas CFO.com. That's probably where I learned the most. There's lots of other SaaS people out there that talk about SaaS finance, but that's probably where I've learned the most about SaaS finance and building SaaS models and also just from LinkedIn. Follow Paul Barnhurst on LinkedIn. He doesn't talk about SaaS as much as others, but you'll learn a lot of good financial modeling tips and then just start building something. If you're not in a SaaS company right now and you know that's where you want to end up, go look at Salesforce's 10K and understand what they're thinking about the business. Start building a little basic three-state model for them just to get some experience.
But I think that there's just so much content out there that you can read and templates you can download that really can help you learn those things pretty quickly.
Host: Paul Barnhurst
I agree with you. I mean, content is just growing so fast today. If there's something you want to learn, you just need to put in the effort and you can find it. I think Ben Murray is great. I know him well. He's the SaaS CFO, I've got to have him on FP&A Today earlier if anyone wants to listen to it, we had a podcast where we talked to him. I encourage it. His website is a great resource. LinkedIn is a good one, like you mentioned. Also, I love the idea of going out and finding a 10K or a 10Q and just studying the company and trying to build a model. The only way you're really going to learn how to build a model is by building it. You can do all the study you want, but at the end of the day, it's like anything, without practice, it's just theory. It's like basketball. I can read a book about basketball, but if want to get better at basketball, I got to go out there and shoot and practice and work at it.
Guest: Derek Baker
I think also what helps is finding groups of people where you can ask them questions as well.
Host: Paul Barnhurst
Yes, but also, you know, you got the Financial Modeling Institute, the community there, they have a lot of great content. So there's Financial Modeling World Cup, there's so many great ways today, it's just a matter of putting in the time. It really is in my view.
So I want to ask a question. For someone who's relatively young in their career and who grew up, what I want to say, with more of a modern Excel. Myself, you know, dynamic arrays were something I didn't really learn till later in my career. Yes, I knew about arrays, I'm not sure if you ever had to deal with Control+Shift+Enter in your career, but I dealt with that a lot. That's how you used to have to deal with dynamic arrays and put a little squiggly bracket around the end of it, which meant it was an array and you couldn't do anything with it. I'm curious, as someone who's grew up with those much more familiar with dynamic arrays, how often do you use them in your modeling and in your work?
Guest: Derek Baker
I wish I used them more. I just barely started dabbling with Lambda and Let to build my own custom dynamic array functions, within Neighbor's financial model. I think one of the barriers for me in using dynamic arrays in a model is it feels like it has to be all or nothing. If the whole thing isn't a dynamic array, then it doesn't make sense. And I don't know if that's true. Maybe using a combination of dynamic arrays and normal Excel formulas is fine, but in my mind that's very intimidating that I have to build the whole thing with dynamic arrays. And that's pretty intensive and would require a lot of reengineering of the model and a lot of learning on my part. So I just haven't approached, I haven't tried to do that yet.
Host: Paul Barnhurst
Yeah, and you're not alone. I've seen some people that have built a full model, but it gets really difficult. And part of that is you can't do an array of an array the way they built the engine. And there are some other limitations, particularly around corkscrews. You either got to use multiplication matrices or some kind of complex Lambda. Craig Hatmaker has tried to solve that with some of his 5G component stuff he's building for depreciation schedules and other things. So yeah, I'm with you, right now it's really hard to be 100% dynamic arrays. I built a simple Capex one that there is one formula that's technically not dynamic, but I was able to take it out far enough that nobody would ever know. It's all within the range, so everything works dynamically. But I never figured out how to get that one formula dynamic and it drove me crazy, and I finally gave up. So, I totally get it, it's a real challenge. I think dynamic arrays are exciting, but in modeling we're still trying to figure out how do we get to make a full model dynamic.
Guest: Derek Baker
I have built a fully dynamic financial model for my own personal real estate investing financial model. So I think simple models are really good for that, and I actually, the real estate model makes so much sense because the timing of everything, like how long you hold the asset, matters a lot in the potential returns and things like that. And so when the years become a driver of the financial model, I think is when dynamic arrays become very useful, where you can say that I'm going to hold this for five years or ten years, then you want the model to dynamically update across those time periods. So I've been able to do it successfully, but on like a DCF model, not on an operational financial model.
Host: Paul Barnhurst
Yeah, I did it on an investment one as well, there is one thing that wasn't quite dynamic, but yeah, it can be done. I've seen some do it on a little more complex stuff, but I think there's still a few things that need to be solved before it becomes more common, cause it's still very uncommon. So that's helpful.
So I'm curious, you know, over your career so far, I know you're fairly early in it, but what is the number one thing you have learned that helps you when building financial models?
Guest: Derek Baker
Something I'm still learning, but I think I'm getting better at is slowing down to speed up. Earlier in my career, I was so eager to just get in Excel and build the model that I would miss things like some really important parts to how the business operates. And then you go and you present it to the founder or whoever, your bosses and they're like, well, the business doesn't actually work this way and so then you have to go and rebuild it again. I think slowing down and asking a lot of questions and really understanding the operations that you're trying to model within the business is really important. Before you get in Excel and start building. You want your financial model to match the way that the business thinks about how they run the business, and if it doesn't match, then they're not going to use it. So getting the operations down and understanding how the business talks about the operations and then trying to apply that in the financial model is something that has really helped me in building effective financial models for startups.
Host: Paul Barnhurst
Yeah, I think it's always true. The better you know an industry, the better you know the operations of the business, the easier it is to build a model. Obviously it varies some, but if you want to do SaaS, especially if you want to work in corporate finance where you're working for one company, you have to understand those operations. It's really important. Even people working in private equity, I'll say one of the most important things is understanding the business. You get to become a better modeler once you understand the business. So I say even beyond that, if you're building something where it's a one-time, you still need to know some operations, you may not need to go deep, but I think no matter what model you're building, the more you know about the business, the better off you're going to be. It's kind of like when you're building a budget, the more you understand your incomes and your expenses, the easier it's to build your own personal budget. Right? You don't know where the money is going, it's going to be a little harder. You don't understand all the inputs, so great point there.
So the next one, kind of shifting gears, we talked about learning the business, but let's talk a little bit more on the modeling side. What's maybe a technical piece of advice you could offer that would help people become a better modeler? Maybe something you've learned that you do now, every time when you're building a model that you think makes a difference.
Guest: Derek Baker
I think there are some basic things like hard codes are blue, linked cells from other sheets are green, formulas are black, like those types of things are maybe just more design than technical. But one of the things that's really leveled up my modeling recently is modeling the flows of different things. So on the balance sheet, for example, we at Neighbor collect from the renter and then pay the host 30 days later to mitigate our risk on refunds and things like that. And so when we used to just model our host payable is what we call this account, we used to just model it as a percent of our total GMV, which is fine, I mean, it's not wrong, like it is driven by GMV, so modeling as a percent of our GMV is not a wrong way to do it, but we learned a lot more when we modeled the flows of our host payable, where beginning host payable plus new GMV that we've collected, minus host payouts that we've paid to our hosts, equals ending host payouts. We just learned a lot about where are we missing our collections, are we paying out hosts that we never actually collected money on? How do we fix that? So modeling the flows for that specific balance sheet account helped a lot.
We have another account that I did this for, which is our property protection plan. That's kind of like AirCover, if you know how that works on Airbnb, it's like an insurance-like product. So we have reserves for our property protection plan, and we take a certain amount of our revenue every month and we put it in those reserves, and then we use those reserves when we actually pay out on a property protection plan claim. And we had never modeled the flows of that before, mainly because it was just really new, and so I did a cohort analysis to see at what times are we most likely to get a claim from a property protection plan renter and then model out the flows of the cash flows from that PPP reserve account. So you understand a lot more about the business when you model out those flows, rather than just doing something simple like a percent of revenue or a percent of expenses. When you model the flows, you understand more aspects of the business, I think.
Host: Paul Barnhurst
100% agree. When you start modeling out, exactly where does cash go? Okay, starts here, then this happens. Then this. Okay, so when do we pay this out and you start modeling all those flows, and you can often realize where the leakage is. Like, as you mentioned, the one example, okay, are we paying out where we're not collecting, or why are collections off? Almost always follow that flow out and you will find out more.
Now, obviously, percentage isn't wrong and it's often how we build models, especially high level, but when you're getting into operational and you're getting more into an FP&A role, or a traditional corporate finance role, where you're also trying to drive value beyond just building the model to find areas where issues may be, that flow can become really valuable. So I think that's great advice there. I appreciate that.
And then next one here, what's your favorite Excel shortcut?
Guest: Derek Baker
I am a big fan of shortcuts. I'm a no-mouse guy, but lately I've been really appreciative of Excel's Control+shift+v, which is paste values.
Host: Paul Barnhurst
Yeah, I use Alt+SV, which you can also do.
Guest: Derek Baker
Which I used to use, but Control+shift+v is just like a little bit faster. And it was something that I think Excel just recently implemented.
Host: Paul Barnhurst
Yeah, they did. They added that one recently. I haven't switched, but I love the paste values. I use that all the time, so I'm with you on that shortcut.
All right, Rapid-Fire section, are you ready?
Guest: Derek Baker
Yeah.
Host: Paul Barnhurst
All right. So how this works, you get no more than 10 seconds to answer. You can't tell me “It depends”. We have roughly about ten of these we're going to go through, and at the end you can elaborate on one of your answers if you want, but you have to pick a side. All right, here we go.
Circular or no circular references?
Guest: Derek Baker
No for most financial models.
Host: Paul Barnhurst
VBA or no VBA.
Guest: Derek Baker
No VBA for financial models.
Host: Paul Barnhurst
Horizontal or vertical?
Guest: Derek Baker
Vertical.
Host: Paul Barnhurst
Dynamic arrays, yes or no?
Guest: Derek Baker
Yes.
Host: Paul Barnhurst
External workbook links, yes or, oh, no.
Guest: Derek Baker
Oh no, no external links.
Host: Paul Barnhurst
Named ranges or no named ranges?
Guest: Derek Baker
Yes, but make sure you delete unused named ranges when you're done.
Host: Paul Barnhurst
Good advice. Do you follow any formal standards, like a standards board for your modeling? Yes or no?
Guest: Derek Baker
No, but I do follow, like the blue, green, black color, cell colors.
Host: Paul Barnhurst
Sure you follow some guidelines.
Will Excel ever die?
Guest: Derek Baker
No, I don't think so.
Host: Paul Barnhurst
Will AI build the models for us in the future?
Guest: Derek Baker
Yes. I'm not sure how long it'll be before they build them by themselves.
Host: Paul Barnhurst
Fair enough. Should you use sheet cell protection in your models?
Guest: Derek Baker
No, if it's going to an investor, yes, if it's going to a nonfinance person.
Host: Paul Barnhurst
Okay, fair enough.
Do you believe financial models are the number one corporate decision-making tool?
Guest: Derek Baker
Yeah, I do think so.
Host: Paul Barnhurst
And then, what is your lookup function of choice? Choose, Vlookup, Index Match or Xlookup.
Guest: Derek Baker
I'm in a bit of an identity crisis right now. I was an Index Match purist, but I found myself needing to use an Index X Match recently and realized I should just use Xlookup. So I'm in between.
Host: Paul Barnhurst
So you're torn. You're going through a transition.
Guest: Derek Baker
Yes, exactly.
Host: Paul Barnhurst
Well, if you need to talk about that, let me know. We're here to help you at the Financial modeler's Corner.
All right, was there any of those you wanted to elaborate on your answer?
Guest: Derek Baker
I want to elaborate on circular references because when my boss listens to this, he's going to be like, you liar, you use circular reference all the time. I don't think that the model itself, especially in tech and SaaS companies, you don't have debt and you don't have to do any weird average balance stuff that creates the circular reference in most debt-heavy companies. So there should be no circular references in the model. But I love using circular references to compare scenarios. So if you ever heard of a “self-referencing if “, where you say, if the scenario equals base, then give me the result else reference that on cell. And then the same thing would be, if the scenario equals stretch, give me the result else give me that on cell.
So you can compare scenarios side by side with a self-referencing if statement. And I love using those to look at different scenarios side by side.
Host: Paul Barnhurst
I have not done that. You'll have to show me that, that's interesting. I think I get what you're saying because of the editor of and you're comparing them, but I'd love to see that sometime. So I'll have to reach out to you on that one.
Okay, so we're going to go ahead and close here. We're just out of time. We have two more questions for you.
If you could give one piece of advice to our audience that would help them be a better financial modeler, what's your advice?
Guest: Derek Baker
Get closer to the business.
Host: Paul Barnhurst
I had a feeling that was coming, based on our conversation.
Guest: Derek Baker
I was recently asked what separates good FP&A from great FP&A, and I think it's 1- business acumen and 2- business partnering. Anyone can build a model in Excel and can learn those technical skills. But using those technical skills to help the business and partner with the business to help them understand the business better and grow the business is exactly what FP&A should be doing and how a financial model should be serving the business leaders that you serve. So, get closer to the business and understand it better is the best advice for building better financial models.
Host: Paul Barnhurst
And if our audience wants to learn more about you or get in touch with you, what is the best way for them to do that?
Guest: Derek Baker
Yeah, you can find me on LinkedIn. Derek Baker, I think there's like hundreds of Derek Bakers, so go to Neighbor.com first and then find me through where people are.
Host: Paul Barnhurst
All right, and we'll put it in the show notes for those who are interested in getting a hold of you. But thank you for carving out a few minutes for us this morning, Derek, and we're excited to release this here to the audience and appreciate you sharing your knowledge with us.
So thanks for joining me.
Guest: Derek Baker
Thanks for having me.
Financial Modeler's Corner was brought to you by Financial Modeling Institute. Visit FMI at www.fminstitute.com/podcast and use code PODCAST to save 15% when you enroll in one of their accreditations today.