You open a spreadsheet at the end of the quarter, stare at the empty cell where next quarter’s revenue is supposed to go, and type in a hopeful round number that feels ambitious but not insane. Then you call it a forecast.
I have done exactly that more times than I would like to admit. But a real forecast is not a wish dressed up in a spreadsheet. It is a structured estimate built out of your own history, carrying a range and a set of assumptions you can actually defend. I run my own small businesses, and I spent years pulling next year’s number out of the ceiling because that was the only method anyone ever showed me. This is not a modelling lecture, it is the plain habit I now use to turn what I already know about my business into a forecast that is humble enough to be useful.
What a revenue forecast actually is
A revenue forecast is a reasoned estimate of the money your business will bring in over some future stretch, built from the patterns already sitting in your past data, and stated as a range rather than a single confident point.
Notice what it is not. It is not a target you are promising to hit. It is not a goal you are motivating yourself with. It is not a guarantee of anything at all. A forecast is what the numbers expect, which is a very different animal from what you happen to want.
Start with the run rate
The simplest honest forecast you can build is the run rate, and for a lot of small businesses it is enough to start. Take a recent stretch of revenue that felt fairly normal, say the last three steady months, average them into a typical month, and project that flat across the months ahead.
If those three months brought in four thousand, five thousand, and four thousand five hundred dollars, your run rate is around four thousand five hundred a month, and your first honest forecast is simply that number repeating. Treat every figure here as illustration, not a benchmark to copy.
Projecting flat feels almost lazy, and that is exactly why it is a good default. Most solo businesses are not gliding along a smooth compounding curve, they are bumping around a level that drifts slowly, and a flat run rate respects that honestly instead of inventing momentum that is not there. Start flat, then adjust it deliberately for things you have real reason to believe.
Bottom up beats top down
There are broadly two ways to build a forecast. The top down way looks at your total revenue line, sees a trend, and extends it forward. The bottom up way ignores the total at first and instead forecasts the handful of drivers underneath it, then multiplies them back together.
Top down is faster, but bottom up is the one that actually teaches you something, because it forces you to say out loud which lever you expect to move. When the two disagree, that disagreement is a gift, and it usually means one of your assumptions needs a harder look.
Here is the heart of the bottom up method: your revenue is almost always a small multiplication of a few plain numbers. A shop might be visitors, times the share who buy, times the average order value. A subscription might be customers, times the price they pay each month. Forecast each driver from its own history, then multiply. The beauty is that every guess becomes visible, so if you inflate your conversion rate, you can see exactly where the optimism entered.
Add growth without fooling yourself
A flat line is rarely the whole story. The safe way to add a trend is to measure your own recent growth rate, month over month or better yet year over year, and use that actual figure rather than the one you are hoping for.
If your revenue has genuinely climbed around five percent a month lately, you can lean on that, gently. But decay it as you look further out, because no small business grows at the same clip forever, and a growth rate that is real today almost always cools as you scale. An honest trend fades toward flat, it does not sprint off the top of the chart.
And here is the trap that catches almost everyone. Take a healthy looking monthly growth rate, say seven percent, apply it to twelve straight months, and the compounding will hand you a number that more than doubles your business inside a year. It looks thrilling and it is very nearly always a fantasy. Whenever a forecast shows your revenue exploding, the first thing to suspect is not a boom, it is a growth rate you let run unchecked for far too long.
Respect your seasonality
Many businesses breathe on a yearly cycle, and ignoring that will fool you in both directions. If you sell more in December and less in January, then comparing this January to last December and calling the drop a decline is simply reading the season as a trend.
The fix is to compare like with like, this January against last January, so the yearly rhythm cancels out and any real underlying change can show itself. One year of history is the bare minimum to even see your seasonality, and two is far better before you trust its shape.
Forecast recurring revenue on its own
If any of your revenue recurs, forecast it separately from the one off sales, because the two behave nothing alike. Subscription revenue is wonderfully predictable: you start with the customers you already have, subtract the ones you expect to churn, and add the new ones you can reasonably win.
That retained base is the closest thing to a solid floor a small business ever gets. One off sales, by contrast, start from zero every month and swing hard, so lumping them in with your steady subscriptions just contaminates the calm part with the noisy part.
Always forecast a range
If you take only one habit from this, take this one: always forecast a range, never a single point. Build a low case, an expected case, and a high case, side by side, because reality does not land on your favourite number, it lands somewhere inside a spread.
| Case | What it assumes | What it is for |
|---|---|---|
| Low | A quiet stretch, weak months | Planning so a slow patch does not sink you |
| Expected | Your honest run rate and trend | Your working plan for the period |
| High | A strong stretch, things break your way | Being ready to serve demand, not banking on it |
A single figure is wrong the instant you type it, and worse, it hides how uncertain you actually are. A range is honest about the fog, and it lets you plan for the quiet month without betting the business on the good one arriving.
Where does the width of that range come from? Your own past variability, not a feeling. If your monthly revenue has been swinging wildly, your future is genuinely uncertain and the range has to be wide to be honest. If your revenue has been calm and steady, you have earned a tighter range. This is exactly why the spread of your data matters, the same standard deviation and fat tails I have written about before.
Write your assumptions down, then check them
Every forecast should be anchored to assumptions you have actually written down, in plain words, right next to the numbers. What price are you assuming, what churn rate, what conversion, how many new customers a month.
A forecast without its assumptions is just a number floating in space, impossible to argue with and impossible to fix. A forecast with its assumptions is a little model, and when one assumption breaks, you know exactly which cell to change instead of starting over.
A forecast you never check is worthless, so the most important step comes after the month is over. Put your forecast and your actual result side by side and measure the gap. Do that a few months running and a pattern will jump out, and it is very often the same pattern: that you run consistently optimistic. Once you know your own bias, you can correct for it.
Two mistakes worth naming
Do not build your forecast on your best month. When you cherry pick the single monster stretch and assume it is the new normal, you plan your whole business on a peak that was probably a fluke. Use a representative stretch instead, or lean on your median month rather than the average, so a couple of freak weeks do not drag your baseline somewhere no ordinary month will ever reach.
And keep your forecast and your target in separate columns. A target is what you want to achieve, a motivating line in the sand. A forecast is what your data honestly expects, wishes removed. They are allowed to be different, and the danger is planning your spending against the target while pretending it was a forecast. That is how an ambitious quarter quietly turns into a cash squeeze.
A cash caution, not financial advice
One honest caution, and this is a flag, not financial advice. A forecast of revenue is not a forecast of cash in the bank, because the two run on different clocks. You can book a sale in March and not actually be paid until May, and a business can look healthy on its revenue line while running tight on the money it can actually spend.
So if the reason you are forecasting is to know whether you can make a big purchase, remember to account for when customers really pay, not just when the sale is recorded. Booked and banked are not the same thing.
The honest limit
Be honest about how far out you can really see. A forecast a month or two ahead can be reasonably tight. A forecast a full year out is a wide, soft guess, and pretending otherwise just sets you up to feel betrayed later. Match your confidence to your horizon, and make it a rolling forecast you revisit every month.
A forecast does not create a single dollar of revenue, it does not remove one ounce of real uncertainty, and being confident about a number has never once made it come true. All a good forecast really buys you is a structured, updatable expectation and the humility of a range. Every metric and method like this one is explained in plain English at dataresearchanalysiscollection.com, so you can read it again slowly with your own revenue history open. No hype, no promises about your results, just the idea explained until you can build the first honest version for your own business.
Get new guides and videos first — join the Telegram channel.