How to do financial projections for a startup

A lot of companies trying to fill out the GWiz™ financial projections don't have financial projections, or at least not good ones. Since I try to be efficient, I have decided to lay it out simply and easily in this one short article...

NOT!

OK - so that was just a ridiculous statement. Financial projections are anything but simple. However, I generally do a first cut at a simple projection before I even start to think about creating and growing a company. Here are the questions I start with:

Who sells what to whom and how?

"Nothing happens till someone sells something" - actual origin is debatable - Henry Ford? Watson Sr. (IBM)? Oh well. Let's be clear. Selling things is not magic or free.

Some (one or more) human being(s) do(es) something. What do they do? How long does it take? What does it cost? Including their time and all of the other people that get paid. Make a list, make sure you check the prices for everything, check the list with someone who has done this before to make sure you didn't miss anything. Then multiply by 2 - because you almost certainly missed a whole lot of things - and underestimated how many NO responses it took to get to the "YES" responses you see.

To do this, you create a sales sieve (some call it a funnel). For details watch our free video "The Sieve". Then read next month's article on unit economics and projections.

Who does what to fulfill and how?

You sold it, and whatever it is, you have to provide it (unless you are a fruadster). This is called fulfillment.

Some (one or more) human being(s) do(es) something. What do they do? How long does it take? What does it cost? Including their time and all of the other people that get paid. Make a list, make sure you check the prices for everything, check the list with someone who has done this before to make sure you didn't miss anything. Then multiply by 2 - because you almost certainly missed a whole lot of things. Did you include the cost of electricity? Waste disposal? Drinking water? Indeed all of these things and many more are required for people to do their jobs in most cases... How much of the average work day is spent in the bathroom, walking from here to there, rebooting a computer, waiting for an email to get processed, and let's not forget thinking! Yes - people actually have to spend time thinking to be productive. And that goes for every job.

What does it all cost and what is left?

Now you have to learn how to do addition, and likely multiplication. Yes - I know - you have a spreadsheet for that. But almost every financial projection spreadsheet I have ever seen has lots of errors. And that includes user errors, modeling errors, calculation errors, errors resulting from changes not well controlled and thought out, etc. So I do simple calculations that ultimately disagree with the spreadsheets and find that my addition was usually right and the spreadsheet was usually wrong.

And organize your calculations. Make lists of things and then walk through the actual activities in your mind, and in a model. Actually literally walk through a room and see everything and what it is connected to, touching, near, and try to do every task involved and make sure you aren't missing anything. Make a list and check it twice. Then put numbers on every step and thing involved, and what they depend on and they depend on and they depend on. Organize these things in terms of time, material, wear and tear, the function they perform in the bigger picture, and list them ultimately under function and nature. Do they recur, or does it cost only once? By the way, nothing costs only once because nothing lasts forever. Even something like product development normally has to continue in order to deal with changes over time, improvements, competition, and ultimately the next product introduced into the market.

Don't forget that everything you figure out how to do has to be taught to others to do in your place. What does it take to teach and learn these things, not by you, but by someone you pay to do it? And how often does a worker go on to something else and require retraining for their new tasks and for the person who replaces them in their old task?

How soon does what happen?

All of this takes time. And it all has to be orchestrated by you and those who work for / with you conducting the business. In order to understand the projections, you need to understand cash flow, and that has to do with time. Indeed relatively minor changes in the timing of a market entry can change cash flow needs by large amounts, and this changes investment amounts and growth rates and equity and valuation. This is critical for projections, not only as you begin, but as you operate, and you update them to reflect realities on the ground.

The same sort of thing can happen with other parameters that can change, and thus projections should be understood with respect to variation of parameters within identified bounds related to the certainty of the information upon which the projections are based. Properly done projections include the basis (assumptions) and their justifications, along with expected variations in those assumptions and the effects of those changes on the projected outcomes.

As time goes on and real numbers start to appear, the projected assumptions and uncertainty around them should then also be adapted to get more and more accurate projections. All of this of course in the context of the fact that it can all change because of external events. And all of this drives pivots BEFORE bad things happen.

How soon has to do with the plane hitting the ground. Unless you are very skilled at business management and operations, or very skilled at getting more money from investors, companies fail when they run out of cash. And projections should be telling you that this is going to happen in a defined time frame. The time available gives you time to adapt, and of course you should be projecting the effects of these adaptations to make sure they actually improve the situation. In other words, you should be betting with a stacked deck wherever possible.

The stuff you don't (yet) know

If you have never actually done it, how can you go through it and estimate it? You either have to learn how to do it or learn what it takes from someone who has. So start asking questions and visiting locations. Ask them not only what they do and how they do it, but how they learned to do it and how long it took them. Try doing parts of it, but not the dangerous or damaging parts. Leave those to the pros. But learn what it takes and include all of this in costs and in your mental model of what has to be done.

This will never end, as the business will keep changing with time. So you need to keep learning if you are going to successfully run a company, especially a company that is growing and changing with time.

A call to action

Building financial projections is something we help almost every company do at some point. Even companies that just come to present at our demo days get a very brief review of their projections, and many make improvements even in this limited process. How do we do this? It starts with the process of preparing for your GWiz™ for presentation at one of our Demo Days:

Go To Angel

As you do it we will help identify some of your potential problems and correct to help you accelerate your company.

In summary

Don't just guess - project...

Copyright(c) Fred Cohen, 2021 - All Rights Reserved