Fintrix Consulting LLC

Fractional CFO

Financial forecasting for small businesses, step by step

November 18, 2026 · 10 min read

← All articles
Colleagues at a meeting table reviewing a bar chart, one pointing with a pen, another using a calculator
Photo via Pexels

Quick answer

A financial forecast is your best current estimate of where revenue, costs and cash will land over the next 12 months, rebuilt from a few drivers (like clients × average fee, or headcount × cost per person) and updated every month with real numbers. Unlike a budget, it changes as you learn. Build three versions (base, best, worst) and you can see a cash problem months before it reaches the bank.

You signed two new clients last month, your team is stretched, and a strong candidate just asked when you can make an offer. The P&L looks healthy. The bank balance looks fine today. But "fine today" isn't the question. The question is if you'll still be fine in September, after the new person's salary, two quarterly tax payments and a slow summer.

That's what a forecast answers: a living 12-month view of where the money is going, built from numbers you already have. Below is what goes into one, how to build it from simple drivers, and a worked example where a services firm tests a hire against three scenarios.

Is a forecast the same thing as a budget?

No.

A budget is a target. You set it once a year, it says what you plan to earn and spend, and you compare actual results against it. If you haven't built one, start with how to build a business budget. We won't repeat that here.

A forecast is a prediction. It says what you now expect to happen, given everything you know today. When a client leaves in March, the budget stays the same (that's the point of a budget) but the forecast changes the same week.

Budget vs forecast

The plan

Budget

  • Set once a year, usually before the year starts
  • Answers: what are we aiming for?
  • Stays fixed so you can measure against it
  • Mostly a P&L view: revenue and expenses

The best current guess

Forecast

  • Updated every month with real results
  • Answers: where are we actually heading?
  • Rolls forward, always looking 12 months out
  • Revenue, costs and cash, linked together

You want both: the budget tells you if you're on plan, the forecast tells you what to do about it.

What should a small business forecast?

Three things, and they have to be connected.

Revenue. What you'll bill or sell each month. The hardest to predict, and usually forecast too high.

Costs. Payroll, contractors, rent, software. Easier than revenue, because you already know most of them.

Cash. What actually lands in and leaves the bank account, and when. This is where the first two meet reality. A sale in March might be cash in May. A tax payment that isn't an expense on your P&L still empties the account. If that gap is new to you, cash flow vs profit walks through it.

The link is what makes the forecast useful. Change one driver (say, a client starts two months late) and revenue, profit and cash should all move. Three tabs that don't talk to each other will be wrong the first time anything changes.

What is driver-based forecasting, in plain words?

It means you don't type in revenue as a number. You type in the reasons for the number, and let the spreadsheet multiply.

  • Revenue = number of clients × average monthly fee
  • Revenue = orders × average order value (for a shop)
  • Revenue = billable hours × utilization × hourly rate (for a firm that bills time)
  • Payroll = headcount × fully loaded cost per person

Why bother? "Revenue will be $96,000 in August" is a guess. "We'll have 16 clients at about $6,000 each" is something you can check against your pipeline, and when you're wrong you'll know which part was wrong.

Keep it small. Five to ten drivers is plenty for most small businesses. Honestly, a forecast with 40 inputs is a forecast nobody updates.

How does a 12-month rolling forecast work?

A rolling forecast always looks 12 months ahead. At the end of each month you do three things:

  1. Replace the month that just ended with actual results from your closed books.
  2. Revisit the drivers for the months ahead. Did the pipeline change? Did someone resign? Did a price increase stick?
  3. Add a new month at the end, so you're still looking a full year out.

Update it monthly, right after the books are closed. Weekly is too often for a 12-month view (short-term cash is the job of a 13-week cash flow forecast). Quarterly is too slow: you notice problems two months late. The SBA's business plan guidance makes a similar point for new businesses, suggesting quarterly or even monthly projections for the first year.

A worked example: can a services firm afford one more hire?

Say you run an 8-person marketing agency. It's December, and you're building a forecast for next year to decide whether to hire a senior designer in April.

Where things stand:

  • 14 retainer clients at an average of $6,000 a month, so $84,000 billed monthly
  • Clients pay about 30 days after the invoice, so this month's billing is next month's cash
  • Payroll of $60,000 a month (8 people), overhead of $12,000 (rent, software, insurance)
  • Owner draws of $6,000 a month, plus $9,000 estimated tax payments in January, April, June and September
  • $60,000 in the bank on January 1, and you never want to drop below $40,000 (your cash floor)

The new designer would cost about $8,000 a month fully loaded (salary, payroll taxes, benefits). For how to work out that number for your own hire, see can you afford your next hire?. The bet is that the extra capacity lets you take on more clients over the year.

The only thing that changes between scenarios is when the new clients arrive. Costs are identical in all three.

ScenarioNew clients from the hireLowest cash balanceCash on December 31Profit for the year
Best caseOne each in May, June, August and October (four total)$51,000 (June)$138,000$210,000
Base caseOne each in June, August and October (three total)$45,000 (June)$96,000$162,000
Worst caseOnly two, in September and November$30,000 (September)$48,000$108,000
No hireNone (team has no capacity)$57,000 (January)$96,000$144,000

Month-end cash with the April hire: base vs worst case

Hypothetical 8-person marketing agency, $60,000 starting cash

  • Base case
  • Worst case
$0$25k$50k$75k$100kCash floor $40,000Jan · Base case: $57,000Feb · Base case: $63,000Mar · Base case: $69,000Apr · Base case: $58,000May · Base case: $56,000Jun · Base case: $45,000Jul · Base case: $49,000Aug · Base case: $53,000Sep · Base case: $54,000Oct · Base case: $64,000Nov · Base case: $80,000Dec · Base case: $96,000Base caseJan · Worst case: $57,000Feb · Worst case: $63,000Mar · Worst case: $69,000Apr · Worst case: $58,000May · Worst case: $56,000Jun · Worst case: $45,000Jul · Worst case: $43,000Aug · Worst case: $41,000Sep · Worst case: $30,000Oct · Worst case: $34,000Nov · Worst case: $38,000Dec · Worst case: $48,000Worst caseJanFebMarAprMayJunJulAugSepOctNovDec

Hypothetical example. Cash in each month equals the previous month's billing; cash out is payroll, overhead, owner draws and estimated tax payments.

Show the numbers
 Base caseWorst case
Jan$57,000$57,000
Feb$63,000$63,000
Mar$69,000$69,000
Apr$58,000$58,000
May$56,000$56,000
Jun$45,000$45,000
Jul$49,000$43,000
Aug$53,000$41,000
Sep$54,000$30,000
Oct$64,000$34,000
Nov$80,000$38,000
Dec$96,000$48,000

What the example tells you

The base case works, but June is tight. June has the hire's third salary month, a tax payment, and cash still arriving at the old 14-client level, because the June client's first invoice isn't paid until July. Cash bottoms out at $45,000. Above the floor, not by much.

The worst case breaks the floor in September. If clients are slow to sign, cash drops to $30,000 in September and stays under $40,000 through November. Not a crisis, but the kind of squeeze that leads to late vendor payments.

Look at the base case year-end. It finishes at $96,000, exactly where the no-hire path ends. So the hire costs you nothing in year-end cash in the base case, and you go into the following year with three more clients and a bigger team. That's the real argument for hiring, and you can't see it on the P&L alone.

In the base case, revenue is $1,098,000, payroll $792,000 ($60,000 × 12, plus $8,000 × 9 months) and overhead $144,000, so profit is $162,000. Owner draws ($72,000) and tax payments ($36,000) come out of cash but aren't business expenses on the P&L, which is exactly why profit and the bank balance tell different stories.

Turning the worst case into a plan

Here, the owner could decide now: if the 15th client hasn't signed by the end of July, skip the August and September owner draws. That puts $12,000 back, and the worst case's lowest point becomes $42,000 in September, above the floor. Or open a line of credit in spring, while the numbers look good.

What do you need before you can forecast?

Mostly, clean monthly numbers. A forecast built on messy books just predicts the mess.

Inputs for a 12-month forecast

From your books

  • At least 12 months of closed, reconciled monthly P&Ls
  • A current balance sheet (cash, receivables, payables, loans)
  • Accounts receivable aging, so you know how fast customers really pay
  • Payroll reports by person, with employer taxes and benefits

From the business

  • The sales pipeline: who's likely to sign, and when
  • Known changes: price increases, lost clients, planned hires
  • Debt payments and their dates
  • Planned owner draws and estimated tax payments

Decisions to make once

  • Your cash floor (the lowest balance you're comfortable with)
  • Your five to ten drivers
  • Who updates the forecast, and on which day each month

Missing the first group? Fix that first. A few months of properly closed books beat two years of numbers nobody reconciled.

Common forecasting mistakes

The hockey stick. Revenue flat for years, then the forecast shows 40% growth starting next month. If nothing in the business is changing, the numbers shouldn't either. Base growth on drivers you can point to: signed contracts, a real pipeline.

Forgetting when cash moves. Forecasting revenue in the month you bill it, and cash in the same month. If customers pay in 45 days, that's a six-week gap your forecast has to show. In our example, a one-month lag alone is what makes June the low point.

Building it once and never touching it again. A January forecast is a history document by May, and nobody will trust it.

Leaving out taxes and owner draws. Neither shows up as an expense on a pass-through business's P&L, so they're easy to miss. Both take real cash. The IRS divides the year into four estimated tax payment periods, with payments generally due April 15, June 15, September 15 and January 15. Put each one in the month it's due.

When should you get help with forecasting?

Many owners should start with a spreadsheet. Our free cash flow forecast tool is a reasonable first draft if you'd rather not start from a blank sheet.

It's worth bringing someone in when:

  • you're making a decision you can't easily undo, like a key hire, a lease or a loan
  • you're talking to a bank or investor who'll ask how you built your numbers
  • the forecast keeps missing and you can't tell which driver is wrong
  • your books aren't closed monthly, so there are no actuals to forecast from

This is a core part of what a fractional CFO does: building the model with you, keeping it updated each month after the close, and helping you decide what to do with what it shows. If that would help, here's how our fractional CFO service works at Fintrix Consulting.

Frequently asked questions

How far ahead should a small business forecast?

Twelve months, rolled forward every month, covers most decisions. Add a 13-week cash forecast if cash is tight, and a three to five year view only if a lender, investor or buyer asks for one.

How accurate should a forecast be?

It won't be exact, and it doesn't need to be. What matters is that misses are explained. If revenue came in 10% under forecast, you should be able to say which driver was off (fewer clients, lower fees, late payments) and adjust it.

Can I forecast if my business is brand new?

Yes, but lean on drivers instead of history: how many customers you can realistically reach, what you'll charge, and what it costs to serve them. Update monthly as real numbers replace guesses.

What's a sensible cash floor?

It depends on how predictable your cash is. A common starting point is enough to cover two to four weeks of payroll and fixed costs. Businesses with lumpy, seasonal or slow-paying revenue usually want more.

The bottom line

A forecast is not a promise. It's your best current guess, built from a handful of drivers and corrected every month with real numbers. Forecast revenue, costs and cash together, run a worst case alongside the base case, and decide in advance what you'll do if the worst case starts to happen. Do that, and decisions like a new hire stop being a leap of faith and start being a calculation.

This article is general information, not advice for your situation. Rules change and details matter, so talk to us or your CPA before acting on it.

Have a question about this article?

Ask us. A real person will reply by email within one business day. Your question isn’t published.

Protected by reCAPTCHA. Google’s Privacy Policy and Terms apply.

Ready for books you can trust?

Book a free 30-minute consultation. We'll review where your finances stand today and show you exactly how we'd help.