Forward planning and cash control that puts you in charge.

A budget is about profit — what you expect to earn and spend over the year. A cash forecast is about timing — when the money actually moves. The two diverge constantly: you can invoice £40,000 in March, book the profit, and not see the money until June while the VAT on it is due in April. A business can be profitable on the budget and still run out of cash, which is the single most common way otherwise sound businesses fail. You need both, built from the same underlying figures, and the forecast is the one that stops surprises.

It is quoted as a fixed fee for the build plus a fixed monthly amount to keep it current, agreed in writing before anything starts. The build is priced on how many revenue streams and cost lines there are and how clean twelve months of bank data is; the ongoing figure covers loading actuals each month, rolling the forecast forward and running scenarios when a decision comes up. You can take the build alone, but a forecast written once and never updated is fiction within a quarter, so we will tell you honestly if that is what you are buying.

Thirteen weeks in detail for cash, twelve months at a higher level for the budget. Thirteen weeks is far enough to see a VAT quarter, a Corporation Tax payment and a seasonal dip coming, and close enough that the assumptions are still real. Beyond twelve months you are planning rather than forecasting, which is a different exercise — see business planning. The forecast rolls: every month the actuals go in and another month is added at the far end, so you always have the same horizon rather than a shrinking one.

Monthly for most businesses, weekly where cash is tight or the business is seasonal. A stale forecast is worse than none, because people still trust it — the number on the screen carries the same authority whether it was updated last week or last March. Updating means loading real bank and ledger figures against the forecast, seeing where the variance is and why, and rolling forward. That variance is often the most useful output: customers paying eleven days later than their terms is invisible in a profit and loss and obvious in a forecast.

Then it has done its job, and you have options you would not have had three days out. The usual levers, in the order they are worth trying: tighten collections and chase the aged debt that is already earned, move discretionary spend out of the pinch week, talk to HMRC early about a Time to Pay arrangement if a tax payment is the issue, and only then look at funding. All of them are easier and cheaper three months ahead. Approaching a lender in a week when you are already at the overdraft limit is the most expensive version of every one of these conversations.

They answer different questions. Management accounts tell you what happened and whether you made money. The forecast tells you what is coming and whether you can pay for it. Businesses that fail rarely do so because nobody knew last month's profit; they fail because nobody saw the week where payroll, VAT and a supplier all landed together. That said, they are built from the same underlying data and are much cheaper to run together than separately, which is why most clients who have one end up with both.

You can, and plenty of owners do it well. What tends to go wrong is not the arithmetic but the maintenance: the model is built in one enthusiastic weekend, updated twice, and abandoned by March, at which point it is more dangerous than no forecast at all. The other common failure is building it from the profit and loss rather than from real payment behaviour, so it assumes customers pay on terms and ignores VAT quarters and tax dates. If you want to keep doing it yourself, we will happily review your model once and tell you what it is missing.