Key takeaways
- A 13-week forecast covers one quarter, which is far enough ahead to act and close enough to be accurate.
- Forecast receipts by when customers actually pay, not by the due date printed on the invoice.
- Review the forecast every week, replace the forecast with actuals, and roll one more week onto the end.
- The forecast exists to trigger decisions, so every negative week needs a named action beside it.
A 13-week cash flow forecast is a simple spreadsheet that shows, week by week, how much cash will come in, how much will go out and what the bank balance will be at the end of each week for the next quarter. It is the single most useful finance document a growing company can own, and most of the BC companies we meet do not have one.
We like 13 weeks because it is long enough to see a payroll crunch or a quiet season coming, and short enough that the numbers stay honest. Annual budgets drift. Weekly cash does not lie. Here is how we build one with an owner in a single afternoon.
Start with what is already known
Open the sheet with today's bank balance across every operating account, less any cheques or transfers that have not cleared. That is your opening cash. Then list everything you are already committed to paying over the next quarter. Payroll dates and amounts, rent, lease payments, loan and line of credit payments, insurance, software subscriptions and the remittance dates your CPA has given you for payroll deductions and sales tax.
Next, print the accounts receivable aging and the accounts payable aging from your accounting system. These two reports are the raw material for the first four or five weeks of the forecast, because most of what will happen in that window has already been invoiced.
Lay out the sheet
Columns are weeks, starting with the current week. Rows are grouped so that the sheet reads like a bank statement. Keep it to one tab and one page wide so it can be read on a laptop in a meeting.
- Opening cash balance
- Receipts: collections from existing invoices, collections from sales not yet invoiced, deposits and retainers, other receipts such as refunds or loan advances
- Payments: payroll and source deductions, suppliers and subcontractors, rent and occupancy, loan and lease payments, sales tax remittances, owner draws, other
- Net cash movement for the week
- Closing cash balance, which becomes the next week's opening balance
- Available credit on the operating line, shown separately so you never confuse borrowed room with cash
Forecast receipts by behaviour, not by due date
This is where most forecasts go wrong. An invoice with net 30 terms does not arrive in 30 days. It arrives when that customer habitually pays. Go through the receivable list customer by customer and place each invoice in the week you genuinely expect the money, based on how that customer paid the last three or four times. If a customer is in dispute, put the invoice in a separate line at the bottom and do not count it.
For sales that have not been invoiced yet, only include work that is signed or scheduled. Anything in the pipeline but not committed goes on a memo line below the forecast, so you can see the gap between what is booked and what you hope to book. That gap is the sales conversation for the week.
Forecast payments by commitment
Payments are easier because you control them. Fixed costs go in on their actual dates. Variable costs such as materials, subcontractors or freight are tied to the jobs or orders in the receipts section, so when a job moves, its costs move with it. Discretionary spending, which includes new hires, equipment and marketing, gets its own row so the owner can see exactly what could be deferred in a tight week.
Confirm remittance dates and amounts for payroll deductions and sales tax with your CPA or bookkeeper. These are not optional and they are not your money, so they always sit in the forecast at full value on the correct date.
Review it every week
Pick a fixed time, usually Monday morning, and spend twenty minutes. Replace last week's forecast with what actually happened, note the variance and ask why. Then add a new week 13 at the end. Over two or three months the forecast becomes noticeably more accurate because you learn which customers slip and which costs you forget.
Every week where the closing balance falls below the minimum you have set needs an action written next to it. Chase a specific invoice, ask a supplier for an extra two weeks, delay a purchase, draw on the line or push a hire. The forecast is a decision tool, not a report.
Mistakes we see most often
- Forecasting receipts on invoice due dates instead of customer payment habits
- Leaving out sales tax and payroll remittances because they are not operating costs
- Counting the undrawn line of credit as cash
- Forecasting in months, which hides a payroll that lands two days before a big receipt
- Building the sheet once and never updating it with actuals
- Keeping the forecast private from the people who affect it, such as the person who sends invoices