Financial Modeling for Startups

There are all sorts of founders — and that’s a good thing. A disruptive business’s strengths lie in its differences, and those differences grow from the unique perspectives of its founders and leaders. Same-y people create same-y businesses.

But we shouldn’t write off same-y people entirely, they’ve got one distinct advantage over the innovators — rock solid organization and communication. It turns out that the more a business differs from the norm, the more challenging it is to operate. Every decision is a brand new direction to coordinate, and then it needs to be communicated to those who don’t necessarily share the same vision.

You may not be a numbers person, and even if you are spreadsheets still probably don’t excite you much. But reducing your vision to numbers is a critical exercise in distilling your ideas into something more objective that can be analyzed. Objectives and opportunities that can be quantified can be communicated, compared, and measured.

And of course dollars are one very important type of number. It helps to know what you’re looking to accomplish when you start to lay them out.

Why Build a Financial Model?

Many people have a preconceived notion of what a financial model is, and that includes many of the people that build them. It’s a mess of rows and columns, looks something like an income statement, and extends on into the future, month-by-month or year-by-year. But just like with businesses, starting with the same set of objectives every time isn’t the path to creating the ideal solution. Better to pay a bit more attention to what the point of this exercise even is.

Dollars aren’t the only type of number that matters to your business. Depending on what you do and what you want to do, there may be any number of other quantitative factors to consider. As an example, a bicycle business developing a new design needs to engineer the frame first. Does the configuration under consideration support the required physical stresses when made with the desired materials?

Just like the physical world, the financial realm has its own immutable laws. A financial model is just the tool we use for our feasibility test against those laws, where success is measured in dollars rather than joules or amperes. If you’ve developed even the simplest business math — “can I sell this for more than my costs plus shipping?” — you’ve already started along this path.

So a model doesn’t have to be complex to “count” (and anyone who says it does isn’t actually listening for your actual needs). But it should go through some evolution as you begin to consider the various purposes it may serve, and the different audiences it could reach.

Projections: Building a Foundation

A projection is, generally speaking, built in advance of a new venture, be it a brand new company or a novel endeavor for an existing business. This is where businesses first begin to translate ideas into dollars.

A version of this projection model appears in almost every seed investor pitch-deck you’ll see. It looks something like this:

The primary audience for this type of model is prospective investors, and the goal is telling a story with numbers. The ratio of intangible ideas to tangible numbers here will still be relatively high, and certain execution assumptions will need to be taken on faith: can you really sell that many units? Open that many locations? There’s a reason early stage businesses focus so much on refining their message, because good presentation creates confidence in execution.

Yet despite the intangibility of certain assumptions, there are still a few key things that a projection can be objectively judged on.

The first is completeness: the main difference between a real projection model and the back-of-the-envelope math we discussed earlier is that a projection aims to build a more complete picture of the business. So, if your model is missing any key costs or considerations, it will cause observers to lose faith in the financial story. It’s worth it to talk to someone with experience as you’re developing this part of your pitch.

The second is detail, and more is not always better. Choosing the appropriate level of granularity can often seem more art than science, but generally speaking you should only break out details you can reasonably estimate. If you’re not certain of your marketing costs per new user, far better to have one round assumption you can justify with industry averages, than a detailed amount broken into several categories. This creates the illusion of precision, and makes the model more difficult to maintain and update. Investors trust founders who know where their unknowns are.

Finally — presentation, and this goes equally for every other type of model. Numbers aren’t just numbers, and simple math clearly presented always beats complex math delivered inscrutably. If you’re presented with something that looks like an arcane tumbleweed of digits and horizontal lines, it’s not because you’re not a numbers person, it’s not a model, it’s a spreadsheet.

Forecasting: Aligning Objectives

A projection takes execution on faith, a forecast seeks to validate it. You’re probably familiar with the concept of SMART goals (specific, measurable, attainable, relevant, time-based). A forecast applies this concept to your business, both in terms of setting the goals and ensuring accountability to the numbers.

One might say that while a projection describes the idea of a business, a forecast represents your business as it actually is. A projection says you’ll need 2.5 employees for every 1,000 monthly orders. A forecast tells you that you’re opening a distribution center in Hoboken, NJ with your technician, Janice, and one more hire with $60k budgeted.

In general, your business grows into the need for a forecast as it evolves from theoretical to practical. But you shouldn’t think of this as simply a “mature” projection. The audience for this document is much more internal, management team members and perhaps close investors or board members. The purpose of a forecast isn’t to tell a story, one which might go 3–5 years into the future. The purpose of a forecast is to outline a plan for execution.

A forecast has different success criteria than a projection model. Most importantly, a forecast needs to communicate with whatever accounting or reporting practices you have for performance. Outlining key performance targets is only useful if you can hold yourself accountable to these goals by comparing actual performance with expectations. This means that for all the key drivers in your forecast (conversions, new locations, repeat customers, etc.), you’ll want to devise a means of tracking.

Part of the success of this document is the process around it. Your forecast should grow steadily more accurate over time, as repeated measurement of key drivers focuses you down on what matters to your business. The businesses that most effectively raise successive rounds of capital are those who have identified their own win conditions, and can put investor money to work in service of those objectives most efficiently (more on that later).

Different parts of your business become tangible at different times. You may begin by only being able to forecast part of the company, with the rest (future sales, etc.) remaining a rough projection. What’s important is that those parts that can be forecast be executable. You may recognize that something that seemed plausible (5x growth in 18 months) as a projection may be challenging as a forecast (hiring 12 good salespeople in 3 months).

Again, recognizing the unknowns is important. You should de-emphasize those elements that aren’t yet ready for a more detailed forecast. You’ll need to update this document regularly, and going into great detail into an aspect of the business that pivots entirely (and then does so again) means spending too much time in a spreadsheet and not enough time executing. Many founders go wrong by using a forecast where they should have been working with a looser projection.

Budgeting: Approaching Precision

Now, there will come a time, probably early on, when “recognizing unknowns” will seem like an irresponsible lack of precision. When you’re worried about your bank account hitting zero, or where you’re trying to improve the efficiency of the capital you’ve raised, what you’re looking for is a budget.

A forecast describes goals for coming periods, a budget outlines expectations. You can think of it as a details-obsessed forecast with anxiety. If a forecast says you have $4k set aside for “rent and occupancy” this month, a budget might know that Janice requested a $500 electrical inspection on the 18th. Where a forecast should be roughly correct on average, a budget should be as correct as possible, always. In the startup world, budgets usually focus on spends, sometimes ignoring income entirely (i.e. the most conservative view of cash needs).

The target audience here is your operational team, letting them know how much money they have available to them to execute on their goals, making sure you don’t run out of cash anywhere, and that upper management is aware of the needs (and spending habits) of operational staff.

Ideally, your budget will communicate with your forecast model, taking the more detailed information and totaling it up to a summary level. You can even use your budget vs. actual comparison to rate your budgeting success, teams that are on top of their future spends should rarely be surprised by unexpected burn, outside of genuinely unforeseen circumstances.

Generally speaking, the most successful budget is user-friendly enough to maintain consistently. It can be a difficult habit to get into, but one that generally pays dividends around capital efficiency (and remember the importance of efficiency for raising successive rounds of capital). Not everyone needs a budget, and not for every part of their business, but if your level of spending ever surprises you, it can be a good practice.

Conclusion

Our goal here has been to give you a broad picture of the uses of a financial model, and some terminology to apply. These terms aren’t set in stone, and you shouldn’t be surprised to find other uses of them, but having a firm conception of what a model can do and why will help you determine what your specific needs are, even if you’re hardly a financial guru.

Not everybody needs all of these documents. Some won’t need any. The best model is the one you can actually use, and the number one cause of model-death is growing too big to maintain due to detail-obsession. Focus on your audience. Focus on the needed information. The more specifically you can isolate what you need out of a model (or models), the more effectively you can develop a simple, elegant document that tells you what you need to know, and nothing you don’t.

This article was written by Ben Coleman and can also be read here.