What it means
Every forecast is built from three ingredients: a base of known facts, a set of drivers that move the numbers, and explicit assumptions about those drivers. The known facts might be a signed contract book or a lease schedule, while the drivers might be customer numbers, price and churn.
The quality of a forecast depends almost entirely on whether those assumptions are written down and tested. Forecasts matter because most business decisions are commitments made before the money arrives.
Hiring, ordering stock, signing a lease and taking on debt all rely on a view of the next 12 months. A forecast that is materially wrong does not just embarrass the finance team; it causes real overtrading or unnecessary cuts.
The two dominant methods are top-down and bottom-up. Top-down starts with a market or a growth rate and works inwards, which is quick but easily detached from operational reality.
Bottom-up builds from units, customers and headcount, which is slower but far easier to challenge line by line. Good practice now favours the rolling forecast, where the horizon always extends the same number of periods ahead rather than stopping at the financial year end.
A rolling 12-month or 18-month view keeps management looking at the same distance all year, instead of forecasting only six weeks ahead each March. Most organisations pair it with scenario analysis, producing a base, upside and downside rather than a single number.
The nuance that catches people out is that accuracy is not the only measure of a useful forecast. A forecast that is consistently 5% high is more useful than one that is randomly wrong by 3% either way, because a known bias can be corrected.
Tracking forecast error over time, and understanding which assumptions cause it, matters more than any individual prediction.
In practice
Real-world examples.
Example
A wholesale distributor forecasts cash weekly for the next 13 weeks, driven by the debtor ledger and confirmed purchase orders. The exercise reveals a three-week trough in April, so the finance director arranges an overdraft extension in January rather than in a panic in spring.
Example
A restaurant group forecasts staffing costs from covers per hour rather than from last year's payroll. When a new site underperforms in its first month, the model automatically reduces the hours planned for the following month rather than locking in the original assumption.
Example
A hardware startup builds three demand scenarios before committing to a production run of 40,000 units. The downside case shows unsold stock tying up $1,200,000, so the founders split the order into two batches and accept a slightly higher unit cost.
Formula
Calculation
Forecast for next period = current period x (1 + growth rate). For a multi-period forecast, compound the growth rate.
A subscription business ends the first quarter with revenue of $2,000,000 and, based on its pipeline and historic renewal rates, assumes 5% growth per quarter for the rest of the year. Q2 = $2,000,000 x 1.05 = $2,100,000. Q3 = $2,100,000 x 1.05 = $2,205,000. Q4 = $2,205,000 x 1.05 = $2,315,250. Full-year forecast = $2,000,000 + $2,100,000 + $2,205,000 + $2,315,250 = $8,620,250. A downside scenario at 2% quarterly growth gives $2,040,000, $2,080,800 and $2,122,416 for the three quarters, a full year of $8,243,216, which is $377,034 lower. That gap is what the business must be able to absorb without breaching a covenant or running out of cash.Case study
Seen in the real world.
Pellworth Instruments is an illustrative, wholly fictional maker of laboratory equipment. For years it ran a single annual budget, approved each December and defended for the following twelve months regardless of what happened in the market.
In one particular year a large export order slipped from June to the following January. The budget was never changed, so every monthly report showed a growing unfavourable variance, and by autumn the management team had stopped reading the pack because everyone knew the target was unreachable. Decisions were being made on instinct while the formal numbers were ignored.
Pellworth moved to a rolling 12-month forecast, refreshed monthly, with the budget retained separately as the accountability target. The forecast showed the slipped order landing in January, which let the board hold a hiring plan rather than cancel it, and gave the bank an honest view three months before the covenant test. In this fictional case nothing about the underlying business changed; what changed was that management had a number they believed.
Watch out
Common mistakes.
- Confusing a forecast with a budget, then either quietly rewriting the target to match performance or refusing to update the expectation when facts have clearly changed.
- Forecasting profit but not cash, which hides the timing of receipts and payments and is how profitable businesses run out of money.
- Building a single-point forecast with no scenarios, so nobody knows how much room the business has before a covenant, an overdraft or a payroll becomes a problem.
Questions
People also ask.
How far ahead should a business forecast?
Cash is usually forecast 13 weeks ahead in detail, while profit and loss forecasts commonly run 12 to 18 months on a rolling basis.
What is a reasonable level of accuracy?
It depends on the business, but many companies aim to land within 5% on revenue for the current quarter and accept much wider ranges further out.
Who should own the forecast?
Finance should own the process and the model, while the operational managers who control the drivers should own the assumptions, otherwise nobody feels accountable for the outcome.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%