From Gut to GPS: How to Make Financial Projections for Your Business

From gut to GPS

Today we're going from gut to GPS: how to look at the future of your business with clarity and confidence, and feel like you know what you're doing when you think about where you're headed. A lot of owners feel like there's so much going on they can't wrap their heads around the future, let alone what happened last year. We'll look at the confidence formula and exactly what goes into projections. Some of this may feel overwhelming. That's fine. Pick the one thing that makes the most sense to focus on next, rather than leaving with a list of everything and going back into paralysis.

Wouldn't it be incredible to know how much cash is likely to be in your bank account a year from now? Or what happens to your cash if you make the next hire, raise your own salary, pay down a debt, or make an investment? Most owners have no model for thinking about that. And even better if it's visual. Yesterday I walked a client through their model, hit two buttons, and a chart came up. I asked what story it told, and in five seconds he said his business looked really good a year from now.

Every business comes down to two levers: margins and cash timing. A good CFO or finance team adapts the model to your business specifically.

The confidence formula

After working with hundreds of business owners, what they want most is confidence: I know what needs to be done, I know how to do it, and I trust myself to do it. Confidence comes from clarity, times margin, times buffer, divided by complexity. Clarity means I know where I am, where I'm going, and how to get there. Margin means having extra left over every month, in both time and money. Buffer means reserves, so a bad month doesn't sink you. And more complexity means less confidence. If I know what to focus on and have systems guiding my decisions, my confidence goes up. Which of those are you most lacking?

Forecast vs. projection

I minored in meteorology, so here's the weather version. A forecast tells you what's likely over the next week or ten days based on current conditions. A projection tells you that Orlando in August will be hot, from historical averages. A forecast is short term, usually one to three months of weekly cash flow, and can be highly accurate, but it's volatile if you don't understand the variables. A projection is long term, one to two-plus years, based on assumptions and trends.

For a forecast, you need four things: how much is in the bank, how much you owe, including payroll, overhead, card payments, and lines of credit, how much you're owed, and how much you're likely to sell. And for each one, when. Most owners stay on top of the bank balance, but don't know what they owe when. Checking the bank account gives you a false sense of security. It's one slice of the picture.

Your accounts receivable might be recurring revenue for a software company, purchase orders for a manufacturer, or insurance billing for a physical therapy practice. Don't take your accounting software's receivables summary at face value. If there's old receivables in the 60, 90, or 91-plus day buckets, you don't know when that's coming in. We assign a score to each one for when it's likely to come in and how likely it is. We also automate as much as possible, because a manual forecast blows up every time you make a decision that wasn't in it. Expenses come out on the days they actually leave the account: loan payments, direct labor, overhead labor, distributions. Then the things you control in a cash crunch, like whether to pay down the line of credit this week or push a supplier a week, are manual.

Can AI do this for me?

It will try, and it will look really convincing. I put a similar scenario into ChatGPT and it gave me a wall of numbers. There's some directional correctness, but how useful is it to stare at as an owner? It makes a ton of assumptions, and we don't know what they are or how accurate they are. It's only as good as the data you give it, and about 80% of the clients we work with have some degree of messiness in their books. Feed it wildly off numbers and it takes them as truth. I love AI and use it behind the scenes all the time, and I've been trying to make it do financial modeling for years. But for these decisions, it tends to be wildly confident in highly volatile variables, and it can't transfer understanding. Right now we still need human conversations, with AI as a tool and a lot of double-checking. It's like a really smart five-year-old proudly showing you its tied shoes. Did you superglue those?

Step 1: gross margins

List your top 10 or fewer buckets of income: a service line, or a category of products. We have a candle company with 50 scents, but we group them into wholesale and direct to consumer. Then ask how much it takes to produce a single sale of each: materials, direct labor, maybe marketing, sales commissions, card processing fees, shipping. Most owners are fairly confident in the cost of goods on their P&L, but haven't factored in marketing, commissions, processing, or the equipment it took to build things in the first place.

Gross margin is what you sell it for minus what it costs to deliver, divided by what you sell it for. Sell something for $100 that costs $40 to deliver, and that's $60 of gross profit, a 60% gross margin. Without this, projections fall apart. Owners say they'll grow 10% or 30% this year, but if the projection doesn't scale up the costs that come with more sales, it's wrong. For a home services business, that might be roofs, siding, and fencing. For manufacturing, materials, direct labor, and equipment costs spread over five years. For trucking, driver time, fuel, repairs and maintenance, the trucks themselves, and dump fees.

Step 2: overhead and debt

Look at labor overhead, like admin and office staff; monthly overhead, like rent, electricity, and insurance; and non-monthly planned costs, like quarterly insurance or annual licenses. Most projections average expenses and forget the seasonal variability, so the projection looks great when a certain month is really going to be tight. If you give bonuses in December, plan for it.

You need clean books to do this well, and many businesses have cleanup to do. Even owners doing a million or more often have a bookkeeper split between payroll, bills, and the books, or an outside firm with 15 other clients. It starts with taking ownership: if something you delegated isn't right, you need to address it. If the books aren't clean, there's still plenty we can do with bank access or your point-of-sale data, which is also a great check on the books. This morning we finished a business's margins down to the dollar without touching their books.

It takes time. But shutting down your business, lying awake at 2 a.m., or telling an employee they no longer have a job takes a lot more. Don't wait until this is screaming for your attention and you're deciding which vendor you can't pay so you can make payroll.

Then debt. You pay interest, which may show on your P&L, and principal, which never does. So your P&L can say you're profitable while the money sneaks out the back door to debt principal, especially with merchant cash advances paid weekly. For every card, line of credit, and loan, figure out how much goes to principal each month, or each week if you're in a crunch. In one example, $16,000 a month was going to principal that didn't show on the P&L.

Q&A: two different businesses

If you have two separate companies, you usually want two sets of books. If they operate the same way in two locations, you might combine them for analysis. If they're very different, like landscaping and a beauty business, do each one separately. Start with whichever is easiest, or whichever makes or costs you the most, and build the muscle memory. Then combine the projections to see how you're doing overall. We have a client with six very different businesses combined so the owner's pay stays consistent even when one business has a down month.

Step 3: check your trends

Look at history by revenue stream, by month, ideally for two years or more, so you see real trends. Lawn care might be slow in January. Beauty products might spike in November and December. For lumpy businesses like home building, look per project instead. You want to know: do I have seasonality, am I growing or shrinking, and what recent changes are affecting me?

Seeing it laid out changes decisions. If your slow season, when you're burning cash, runs July through January, you set a target for how much needs to be in reserve by June. And if you run a trucking business and ask whether to buy a new truck in August, the start of your slow season, the answer is obvious once you see it. It's just hard to see the forest for the trees when you're in the weeds.

Putting it together

Know your gross margins, map your overhead and debt, and check your trends. Then it's simple: what's in the bank to start, what comes in, what goes out, and what should be in the bank each month for the next year. Draw an “oh crap” line, the balance below which you won't sleep well. Then you can see, for example, that December and especially April will be tight, and decide what to do about it. It's not a perfect crystal ball, but it's the most likely reality based on your business.

Then you can test what-ifs. Say you want to hire someone at $5,000 a month. Add it, and you're fine until September. But if that person will bring in $15,000 more revenue a month starting four months after they're hired, add that, and you can see whether it works. That's the confidence to hire, invest, or pay yourself a certain amount, because you know what's likely to be in the bank six months from now.

For a construction client, billing happens on an accrual basis but collections happen differently, so we model how much comes in within 30 days, 30 to 60, 60 to 90, and 90-plus, and direct costs that come in a month later. Then we can adjust the year's revenue target down to something more realistic and immediately see the effect: they'll be a little tight for a few months, still above their line, and fine after that.