Starting Up & Right: Conversations with a Startup CFO - Episode 01 - Deep Dive: The Financial Model


RUNTIME: 31:34

FULL TRANSCRIPT

Trae Nickelson: Hi, there. Welcome back to Starting Up & Right: Conversations with a Startup CFO. I'm Trae Nickelson, your co-host, and with me as promised, startup CFO, Ryan Keating. Ryan, thanks for joining again today.

Ryan Keating: Thank you.

Trae: In the last episode, we talked at a very high level about financial modeling. I want to take this episode - it’s such an essential tool, such an important ingredient for a startup success especially in the finance side of things - I want to really dive in in this episode and talk a little bit more about the financial model. Let me play the role of a Startup founder, a CEO. I've got my new company. I'm just getting started and I'm coming to your office, and you're going to talk to me now about financial modeling and why I need it.

I'm an engineer. I've got a great idea, I've got the skills to build it, I'm ready to get head down and code. You bring up financial modeling, my question is what is a financial model? Why do I need this?

Ryan: This is not an atypical situation for us. Most of our clients do come in with a very solid technical background. They've got a product, they know what they want to bring to market, but they're faced with having to go raise money. The financial model is, in my opinion, one of the core tools for that. The financial model is used primarily at this stage that you're describing to be a fundraising tool.

I think of when you fundraise, there's really three things you need to have. There's the investor presentation, there's the executive summary, and there's the financial model. A financial model is really the numerical representation of your business plan. It's what investors want to see to understand, one, how big is this opportunity, two, how does the team and the founders think about how they're going to attack this, and three, how much money do you need and what are you going to spend it on.

Trae: I know you've worked elbow to elbow with some of the most active venture capitalists in Silicon Valley. When you put yourself in their shoes and they're evaluating an investment, and evaluating a founder, they're pitching to you as a founder - If they show up with this model, if they show up with a well-thought-out bottoms-up model with some detail obviously that they thought about, does it affect your first impression?

Ryan: Absolutely. Granted, I'm a little biased, I'm a real fan of the financial model and its role in fundraising, its role in setting investor's expectations, but what it shows me is that this team has thought through not just bringing something to market, the invention, the product but they've understood or at least they've thought through how they're going to make a business out of it, and it's a difference. Having a piece of technology that gets funded and making a business out of it are not the same thing.

The financial model really shows the investor that you've thought through things like... Do I need a sales team for this? Is this all going to be sold through a Google AdWords buy? What amount of churn do I expect? The investors want to know that you are realistic in thinking how you're going to take their money and put it to work. That's really what the model shows, is this is my plan, this is my roadmap to put to use the proceeds that you're about to invest in me. 

And it's going to change. You're going to have to make adjustments but you need to show that group, that audience that you have thought through the steps that it's going to take to get from where you're at today to what you're promising them you'll deliver on their investment.

Trae: That's good. That's a good point. I've heard you say several times there's two basic purposes for the financial model, and you mentioned the first of that, there's fundraising. You've got to paint that picture to the investors and then afterwards, it becomes the measuring stick for how you're doing towards what you said you're going to do. 

Ryan: That's a good point. Really when you think about a model, it needs to serve two different, very unique purposes from each other. One is that fundraising picture that we talked about. You need to go into an investor and show there's a market here, there's an opportunity, we have a plan on how to attack it and we know where we're going to spend your money. We know what our hiring plan is, we know how to get to market, we know what channels we're going to use.

The other side of the model, which is, honestly, when it comes down to it, probably even more impactful for a startup, is using the model to actually have a very near term, granular roadmap of when can you make operational hires? When do you need to change assumptions? Really looking month to month about what can we afford to do? What's the next step we can take? Almost more importantly, what happens if we can't afford to take that next step right now?

Trae: Those two roles, the fundraising, and then the monthly operational role that it plays, and your planning each month, let's dig into those a little bit later in this. Let me first back up and just ask some global questions about the model, perhaps some different variations or complexities of what a model needs to be? What is the model? What is the format? What is the deliverable? What is the file you send me?

Ryan: For the most part, it's Excel. Google Sheets, who we're seeing, obviously come into play a bit more.

Trae: A spreadsheet.

Ryan: Yes, it's a spreadsheet. There are third-party tools that are out there that have cloud-based FP&A, financial planning and analysis. For the clients we work with, that are early, that are fundraising, that are executing against a very defined cash runway, those tools are further out. Those are when you have an FP&A department of three or four people. So it's spreadsheets. When we build a model, it's a deliverable in a spreadsheet, either Excel or Google Sheets.

Trae: I'm starting from scratch, I've got nothing as far as a model, I've just got my idea. You've convinced me I need a model. What am I looking at here in terms of cost, effort? Guide me through what the next few days are going to look like and what I need to budget?

Ryan: Sure. We've been doing this a long time. We've got a really good process around creating a model. It's a pretty iterative process that we'll go through with our clients to really press them to make sure that all the details have been captured and thought about. We don't expect them to come in knowing these things. Part of what we do is we impress upon them the things that we need to understand.

We'll flesh that out in what initially is usually a half a day long, sit down brainstorming session, where we'll go through several almost painful questions because we're trying so hard to understand and translate their ideas into an Excel formula. We have to know, what are the parameters? What are the exceptions? What are the use cases? Then we'll model that in so that at the end of the day, the goal is you have a tool.

A tool that you can use to make assumptions and adjustments to… What if I change my price point? What if I have a slower ramp? What if my cost of acquisition is $1.50 versus $1? We'll make sure that we identify those sensitive assumptions to the model, and we'll flesh those out so that there are actual knobs and adjustments that you can make in this, basically, what becomes a tool for you to execute against.

Trae: It sounds like that's a very detailed process, extremely valuable, you're going to force me as the founder to think about some things that I perhaps had not known even to think about. Cost, hours and what's the team look like on your end? Who am I sitting down with in the room helping me with this?

Ryan: That's a good question. The team that we bring in is a CFO coupled with a financial analyst. The reason is that we want to make sure that you're not paying the CFO to do all the heavy lifting in the spreadsheets. It's a really nice pair. They come in, they work together, the CFO is providing guidance, is helping read the tea leaves, understand the output, and the analyst is really building the formulas and building the relationships. That process can take, we see these come out quite a bit between, between say 40 and 60 hours of time.

Trae: Nice. I assume you've done this a few times. You're probably not starting from scratch completely each time? You've got some templates and best practices?

Ryan: Absolutely, you're right. It's a financial model. We don't start from scratch; we have a great template that we use. You can imagine the idea of hiring somebody in a particular month at a particular salary, that doesn't need to be built out every time. That's just stuff that should be in the model. How do you handle a receivable? When does it turn into cash? That's already built. We have to put in the right assumptions.

Where the majority of our time goes, in terms of customizing this template, is entirely around the revenue model. That's where we really like to start from scratch. Especially at a startup phase, where you don't have 5, 10 years of history to understand how it works and what your churn is and what your lifetime value is, we need to come up with all this.

We like to start from the beginning on revenue, and really capture the specific revenue drivers for each company. To your point, things like travel expenses, salaries, when do taxes get accrued? That stuff is already built into the model and so we don't recreate that.

Trae: As you're sitting down with me as the founder, and you're pulling these details, you're making me think of things I haven't thought of, we're modeling revenue, I guess you're challenging me to figure out how are we're achieving these revenue goals, right?

Ryan: That's a good point. It's very common that when we start working with companies, they will have thought out how much revenue they hope to generate. Even putting together like, we're going to sell so many units, at so much per unit, and it's going to grow up into the right over five years. One of the things that we see is missed a lot is have you thought about the appropriate level of spend or investment to support that revenue growth? Depending on your business model, it could be very simple as have you put in enough of a marketing budget?

We'll use metrics, cost of acquisition. There has to be a certain acceptable level, like, are you going to be able to drive consumer adoption at $1 per install of your app? Or does the industry for these types of solutions more run at $4 to $5? per install? So challenging the entrepreneurs. Have you thought about your marketing spend? Do you have enough people? If it's an enterprise sales solution, do you have a big enough sales force? Are their quotas reasonable when you think about what you're asking each salesperson to produce on a quarterly or annual basis?

Really forcing them to show in the model that they have thought through what it takes to generate this revenue growth. Not just showing that there's a market and there's an opportunity, but how much is it going to cost and how do you go about attacking that opportunity?

Trae: Right. I've heard you very often use the term “sparring partner”, that the CFO sometimes acts as a sparring partner for the CEO, and you guys get to hash things out. I guess, that ties in, you're working on the complexity of the revenue tab, that kind of ties in to another pretty complex or very crucial tab, or section of the model, which is the headcount or the hiring plan. Talk about how that can ripple effect.

Ryan: It's interesting, especially the stage of companies we generally work with. They're early stage. Oftentimes, when a client of ours raises a round of funding, the number one use of proceeds is to hire. Understanding who you need to hire to build the technology, who you need to hire to actually sell the technology, even who you need to hire to support the technology once it's in consumers' hands. Think of a subscription where you need customer support there to help people actually use the product or engage with the product.

The hiring plan is really crucial to support the expectation around getting into market with a product, growing that product's adoption, even keeping churn levels at a certain point. If you don't have customer support, how could you expect to have a reasonable 5% to 10% of your churn rate? We really push on these and make our clients really think about who do they need in this business to support not only the technology creation but what often gets thought about less, the revenue growth? It comes down to headcount. Headcount is usually the biggest use of proceeds that we see early stage when people are trying to bring a product to market.

Trae: As I'm thinking of hiring this all-star team, I guess you guide me in salary expectations, keep it competitive. It's a competitive labor market out there. Is that something that--

Ryan: Absolutely. Startups have two forms of currency and we always emphasize this. You have cash and you have equity - options. Your access to those will change as you move through funding rounds, or as you become more successful. Early on, a company usually has more access to equity than they do cash, or they maybe are trying to protect cash a little bit more and they use equity a little more strongly.

We will advise and it's based on the stage that you're at, it's based on the funding that you've received, how should you be pulling people in? What mix of cash and stock makes sense at the stage that you're at. You can imagine a company that's at say a series seed or a Series A, you're going to get a bigger piece of equity.

You imagine a company who's just about to go public. You may get very, very little equity but they have cash. They might be paying market or even above market rates on the salary side to bring in the right talent. A startup usually has to have a large equity component to attract the right people early on in their business model and their funding model.

Trae: I'm going to sit down with you and build this model. We're going to flesh this out. Who do I need to bring with me? What are your suggestions on, do I fill the room? Who do I bring with me to help me build this model for my team?

Ryan: It really depends on how many stakeholders you have around the table. The real answer is the stakeholders need to be involved in this, because when we finish this model, whether the audience is the investor or the board, we are basically signing up for certain progress. We're signing up to hit certain milestones. It's not uncommon that we will see a founder, or the real technical founding group project out the demand for this product, for this service, for this solution.

Then once we get the marketing or salesperson involved, they're like, "Whoa, I don't know if we can do that. You're underestimating the marketing costs or you don't have enough salespeople." So all those stakeholders need to buy into this because ultimately at the end of the day, this is what we are promising. This is what we are saying we are going to do with your money, Mr. or Mrs. Investor or Mr. or Mrs. Board Member. This is the progress we're going to show you. Having those stakeholders involved and actually contribute to the process to get their buy-in, is really important.

Trae: When we're building the model, we're also setting up milestones and expectations for the investor. Talk about how to set up those milestones, what should be in my mind as a founder in establishing milestones through this model.

Ryan: At the end of the day, investors are buying into the progress that you're telling them you're going to make. I always think of it as our goal when we're talking to investors is we need them to take a leap of faith. We need them to get on-board with what we're trying to accomplish. The shorter we can make that leap, the better. One of the ways that we do that is we show them very clearly milestones that we expect to hit and how we're going to get there. We have the right people, we have the right spend, we're going to hit these milestones. That's a very clear way of communicating how we're going to build value.

What that does, however, which I prefer, that puts the task on us to now deliver. The risk of putting out a milestone is that you don't hit it. There's a lot of execution that has to happen between each milestone, especially in a startup, there's so much that's being projected and there's so much that can be out of our control. The timing of the hires. Are they the right hires? The results from the beta testing. Regulation that we need to get through, if you have a piece of hardware that needs some certifications. There's a lot of steps that have to happen to hit milestones. What we do is we put out those milestones to investors and it's almost a way that they will be able to measure us.

One of the downsides is if you put out those milestones and you don't hit them, you don't execute against them. I always make sure that we put out stuff that this is something that we hopefully can not only hit but exceed, as opposed to put out a milestone that is a reach and we come up a little short. Setting expectations is important. Most important is that when you set them that you can execute and reach them.

The worst-case scenario that you can imagine is you're a startup that raises money to deliver on a certain set of milestones and you don't hit those milestones. Then what the reality is, is you're going back to likely that same investor group and saying, "Hey, I told you we could do this but it didn't quite work out. We need more money to do the same thing we told you we would do last time, to finish it." That's never a good position to be in as an entrepreneur talking to your investors. It's a very weak position.

What you want to do is set milestones and you want to hit them obviously. What that does is that sets you up for that next round of financing on an increased valuation with new investors coming in, and your existing investors wanting to participate in the round, that's the ideal scenario.

Trae: Okay, so we've sat down, we've built this model, it's very pragmatic, as a founder, I'm now confident and comfortable with the milestones. I've taken it out, I've raised some money, I've got my seed and my series A, thank you very much we're done with the model, right?

Ryan: No. It's interesting, this is where the model shifts its purpose and usefulness. Initially, as we talked about, a model really has two purposes for a startup. It's fund raising, but then even tactically and operationally it is a roadmap for the startup to execute against. The model is never done. It's something that needs to be tended to. You need to make sure that you are adding the actual results to the model. Based on those results, you need to make sure that you are adjusting your outlook.

An interesting example of where you could see this is very simply if you have a startup that's projecting to hire certain key people over a period of time. Let's say one of those key people is a head architect, head engineer. You could imagine if that hire, that person comes in say two months later than you projected, chances are it's going to have an impact on your go-to-market, your revenue ramp, your ability to build out the team.

Looking at those impacts, and how they affect your outlook is something that needs to be done honestly, on a monthly basis, especially in the near term. We often find our clients, once they receive outside funding, generally venture capital, they tend to have a monthly board cadence where they're meeting with their board every month. In those meetings, we always impress upon our clients, you need to go in there with a model, that's not still showing that you're going to be in the market in November, let's say, from a model you presented two, three months ago, if now the reality is it's going to be January because things shifted. You have to go into that board meeting every time with the most recent expectation because of how sensitive things like your cash runway are.

Trae: You mentioned monthly, so this is a monthly routine now that I'm going to use this model. Let's dive into that a little bit more. What is that actual process of plugging? I've got projections from a model, I've communicated those projections or my expectations to my milestones, and now we're measuring, it's end of the month. What's that process like?

Ryan: What we will do is once the accounting is closed, we'll take those actual accounting results for the month and we'll drop them into the model. The first step is we'd compare: what did we think we were going to do, so our projections, versus what did we actually do? So budget versus actual, pretty simple idea. We'll go down and say, "Okay, we thought we were going to spend 100,000 and we spent 120,000." Okay, why and where?

Sometimes it's very simple, like, "Hey, we spent more money on travel this month because there was a--" (This is 2019 example. Pre-COVID.) "There was a conference that we wanted to go to and we set more people." Okay, well, that's a one time jump in travel expense. Versus, "Well, geez, we really thought that we could hire this person for 150. She ended up costing 200." Okay, that needs to change in the model because if you don't make that adjustment, it's going to be wrong every single month.

The idea is, as you look at your budget versus actual and where you need to change to go forward, you change it. Where it's just a one-time delta from your projection, you explain it to the board, to whoever the audience is, but you don't necessarily need to change the model. Figuring out when you look at budget versus actual, what are the things that are changing our path versus this is just a one time we'll capture it, we'll explain it but it doesn't necessarily need to change our thinking.

Trae: Got you. We're looking at that delta each time and we're figuring out, are our assumptions wrong? Do we need to change the model going forward? Or was it just that we didn't execute this month somehow according to plan?

Ryan: We'll have clients where they're a consumer-facing application and they might have an assumption that their monthly churn rate is going to be 2%; 24% annually, 2% a month, because they have a monthly recurring revenue. The results may come back that they are consistently at 3%. If that's the case, the model needs to change to reflect that because it has an impact on everything that happens beyond this point. Those types of deltas, when we look at them and understand them we determine, "Okay, is this something that needs to be incorporated in the model going forward?"

Oftentimes, the answer is yes especially when you're so early and all of your assumptions are guesses. They're well-informed guesses. We've researched them. We've looked at examples, but at the end of the day until we actually know “what is our churn rate?” - then it is an assumption. As we start to understand those things like churn rate or cost of acquisition or lifetime value, the model needs to capture that most current and relevant thinking.

Trae: This becomes a big part of the board meeting because I'm a founder trying to explain all this to the board?

Ryan: Yes. For the financial update part of the board meeting, it is the financial update. It is “how are we tracking towards what we set the expectations. Is this impacting our cash needs? Do we need to raise more? Sooner, later? Are we actually getting the traction we thought? Are these sales channels developing?” All this stuff comes out when you're looking at the financial metrics and you're comparing it to what you thought you would do versus what you did.

In my opinion, for the financial update, this budget versus actual and understanding how it impacts our expectations going forward, that is a significant part of what we need to communicate in every board meeting.

Trae: Now as a founder this sounds a little, frankly, scary to me. I understand I've set goals and if I'm not meeting those goals, do I need to be afraid in this board meeting? How do I address this?

Ryan: That's a good question. I always try to work with our founders and we do. Most of our founders, it's the first time they've taken other people's money. We do have quite a few repeat entrepreneurs but an element of what we do is coaching, and I always try to work with the founders that the board is your best asset if you can use it properly. Once they're on your cap table, once they're in that board meeting, they are probably your biggest advocates.

They want nothing more than for you to succeed. Don't treat them like you're going up against a parole board, and you're looking to keep your job every 30 days. Use them, reach out to them, engage them in this process. They have a lot of experience. One of the best pieces of advice I had from a VC a long time ago was nothing new should be discussed in a board meeting. The board meeting is where you talk about how to take action. So communicate throughout the month. When things come up, reach out to them. Don't treat them as if every 30 days you're going to get basically a go or no go on if you get to maintain as the CEO.

That board can be very valuable. We always encourage people to communicate. Don't be afraid. Even if you realize, "Look, the metric we're going for is not the right metric." Don't blindly keep your head down and figure out how to engineer a result that's going to show that you can hit this. Go to your board and say, "This is the wrong target, let's adjust this." They don't want you to fail. If you've set out a path that isn't the right path, they will very quickly want to work with you to adjust that path.

Trae: So the model becomes a crucial part of communicating with my board. How do I use it ongoing or monthly to communicate with my team?

Ryan: Especially a team that has different stakeholders. Let's say, you have a Head of Engineering, you have Head of Marketing, you have Head of Sales. Looking at the model on a regular basis pulls in and coordinates the different efforts. You can imagine a scenario where maybe engineering is having a hard time hiring and staying on pace to launch their product when they thought they would be able to launch version one of the product.

That has a huge impact on the marketing department; when they need to start spending, when they need to start hiring. Huge impact on sales; when do they need to have their team in place? Without the ability to look at the actual results and make adjustments to your forward-looking expectations, these teams are operating and executing against outdated roadmaps, and it could be very spendy to do that.

You don't want to have a marketing team in place and sit in there if the product isn't ready. You don't want a bunch of trained salespeople carrying quotas if the product is not in the market. That coordination, by looking at the model among other things, but I think the model is a really good tool to see this visually, you can see it's implications. If something happens in engineering, it has an implication on sales. If marketing is not able to, because the product, not able to bring in as many consumer sign-ups as is projected, that has an impact on sales, that has an impact on other parts of the business, so that this coordination is necessary.

Part of doing this budget versus actual on maintaining a model is to make sure all of the internal stakeholders have set their expectations accordingly. It's a very flexible endpoint that we're shooting for and we need to know when to zig and when to zag. Looking at the results in the financial model, really gives you this input and to understand where can I make adjustments because of what happened last month? What do I need to adjust this month? The model's a great tool to see all that laid out.

Trae: All right, good stuff. Ryan, thanks again for explaining the financial model. I know just from working with you over the past few years what an important and powerful tool that is for a CEO. So founders, CEOs, if you don't have a solid financial model that's really challenging you, really setting that roadmap, you need that. Thanks again, Ryan. We'll talk soon.