Reports
A cash flow forecast for a small job, without a spreadsheet degree
Every job has a week where the most of your own money is buried in it. The forecast’s only real job is to name that week and that number before you sign, because a job you cannot afford to finance is a bad job at any margin, and you find out cheapest on paper.
QScope Team·27 March 2026·5 min read
A forecast is two lines of dates and one subtraction, and the discipline is entirely about using real dates. The wishful version, valuations counted as income the month they are earned, is not a forecast, it is the profit illusion with columns.
The receipts line: when money lands, not when it is earned
Start from the valuation schedule and push every sum through the machinery it must actually survive: work done in March, valued early April, due and payable by the final date your contract sets, so landing late April at best, the four numbers give the exact lag. Then apply honesty: deduct the retention from every receipt, it does not land with the rest and comes back on its own slow clock, and if this client’s record is paying a fortnight late, forecast the fortnight. A deposit or advance, where you have agreed one, goes in on its date too, with its recovery thinning later receipts.
The payments line: when money leaves, which is earlier than you think
Labour leaves weekly from week one. Materials leave on supplier terms, big packages sometimes on pro forma, before the work they belong to earns anything. Subcontractors leave on their own subcontract dates, which do not wait for your client. Plant, fuel, and the job’s share of running the business leave monthly regardless. Put each on its real date, and resist netting anything: the whole point is the timing gap, and netting hides it.
Read one number: the deepest point
Run the cumulative difference week by week. It will go negative early and stay negative for months, that is the shape of construction, and somewhere near the middle it bottoms out. That trough is the working capital this job demands from you. If it is £38,000 and you have £15,000 of headroom, the job as structured will break you at week eleven while being perfectly profitable on paper. And now, before signing, you can restructure it: a materials deposit, stage payments moved earlier or made monthly (a right on longer Act jobs, a negotiation elsewhere), the big package supplier paid directly, the start date shifted past another job’s final account.
Keep it alive, ten minutes a month
The signing-day forecast decays immediately: payments drift, variations add unpriced weeks. Each valuation cycle, update actuals against forecast and re-read the trough. The forecast tells you where the account is going; its sibling the cash position tells you where it is, and the pair together is the whole cash management a small firm needs. Nothing here changes with a homeowner client, cash arithmetic is client-blind, except that with no statutory payment floor (section 106) the receipt dates are purely whatever your contract says, which makes forecasting before signing more important, not less.
What to do this week
1. Build the two lines for the next job you are pricing, before the quote goes out, and read the trough.
2. If the trough is bigger than your headroom, restructure the payment shape now, in the quote, where it is a term rather than a favour.
3. On the live job, do the ten-minute monthly update in the same sitting as the valuation.
Where the information stops
If every forecast you run shows a trough beyond your means, the problem is not the forecasting, it is capitalisation, and the conversation about overdrafts, invoice finance or simply smaller jobs belongs with your accountant before the next tender, not after the next crisis.