How to Build a 13-Week Cash Flow Forecast, Step by Step

A 13-week cash flow forecast is a week-by-week projection of the cash coming into and going out of a business over the next quarter. It deals in actual bank movements, not accounting profit, which is why accountants, lenders and turnaround advisors rely on it when cash is tight. Thirteen weeks is long enough to see a shortfall coming and short enough to forecast with real detail: specific invoices, specific pay runs, specific bills.

This guide walks through building one from scratch, rolling it forward each week and reviewing variances so the forecast gets more reliable over time.

Why 13 weeks, and why weekly

A monthly forecast can hide a crisis. If a large payroll and a sales tax remittance land in the same week, and the biggest customer pays two weeks later, the month can look fine while the account goes into overdraft for ten days. Weekly buckets expose that timing. Thirteen weeks equals one quarter, so the forecast always covers at least one full cycle of monthly bills, a few pay periods and most quarterly obligations.

Step 1: Set up the structure

Use one column per week (week ending dates are clearest) and these rows, from top to bottom:

1. Opening cash balance: cash in the bank at the start of the week.
2. Receipts: customer collections, plus any other cash in (loan advances, owner injections, tax refunds you are confident about).
3. Disbursements: payroll, suppliers, rent and other fixed costs, sales tax and payroll remittances, loan payments, owner draws, capital purchases.
4. Net cash flow: receipts minus disbursements.
5. Closing cash balance: opening balance plus net cash flow. This becomes next week's opening balance.

Add a row for your minimum cash buffer, the balance you do not want to drop below, and any available line of credit. The point of the forecast is to see which weeks breach that buffer.

Step 2: Start from the real bank balance

Week 1's opening balance should be the actual balance across all operating bank accounts on the start date, reconciled for any cheques or payments already issued but not yet cleared. Do not use the balance sheet cash figure from the last month-end; it is already out of date.

Step 3: Forecast receipts

Receipts are where most forecasts go wrong, because invoices are not cash. Build them from the accounts receivable aging:

• Place each open invoice in the week you realistically expect payment, based on that customer's actual payment history, not the invoice terms.
• For sales not yet invoiced, estimate weekly sales, then apply your typical collection lag (for example, most cash arrives five to six weeks after the sale).
• Treat disputed or very old invoices as zero until there is a firm commitment.
• List non-operating receipts separately so they cannot mask a weak operating trend.

Step 4: Forecast disbursements

Disbursements are usually easier to predict, because most of them are committed. Work through this list:

• Payroll on its actual pay dates, including the employer's share of deductions and the remittance of source deductions on its due date.
• Suppliers from the accounts payable aging, in the week you intend to pay, not the week they are due.
• Fixed costs: rent, leases, insurance, software subscriptions, utilities.
• Sales tax (GST/HST, and QST in Quebec) and income tax instalments, using the business's own filing frequency and due dates. Confirm these with the CRA, Revenu Québec or your accountant.
• Debt service: loan principal and interest, equipment financing.
• One-off items: equipment purchases, annual fees, owner draws, bonuses.

Worked example (illustrative numbers)

Here are the first four weeks of a hypothetical small distributor's forecast. The business wants to keep at least $30,000 in the bank. All figures are made up for illustration.

Week 3 is the first warning sign: a sales tax remittance and a loan payment land together, and the balance falls to just above the buffer. Week 4 drops to $25,000, below it. With a month of notice, the owner has real options: chase two overdue invoices, move a large supplier payment by a week with the supplier's agreement, or arrange a temporary draw on the line of credit before it is urgent.

Step 5: Roll it forward every week

A 13-week forecast is rolling: each week, the oldest week drops off and a new week 13 is added. Pick a fixed day, for example Monday morning, and follow the same routine:

1. Replace last week's forecast with actuals from the bank.
2. Reset the opening balance to the real current bank balance.
3. Update receipts from the latest receivables aging and any customer conversations.
4. Update disbursements from the latest payables, payroll and new commitments.
5. Add a new week 13.
6. Flag any week where the closing balance falls below the buffer.

Step 6: Review variances

Compare last week's forecast with what actually happened, line by line. In the example, suppose week 1 customer receipts came in at $15,200 instead of $18,500, a $3,300 shortfall. The question is why. If one customer simply paid a week late, move the amount into week 2; it is a timing variance. If the customer is in difficulty, reduce or remove it; it is a permanent variance. Over time, note which lines are consistently optimistic (receipts usually are) and adjust your assumptions. A forecast that is reviewed this way becomes noticeably more reliable within a few weeks.

When to use a 13-week forecast

• Tight cash: the business is close to its overdraft limit or juggling which bills to pay.
• Lender conversations: banks often ask for a short-term cash forecast when renewing or increasing a credit facility, or when covenants are under pressure.
• Turnarounds and restructurings: it is the standard tool for showing whether a recovery plan keeps the business solvent week to week.
• Seasonal businesses: to plan for the low season before it arrives.
• Growth spurts: fast growth consumes cash through inventory and receivables, even when the business is profitable.

Common mistakes

• Using profit and loss figures instead of cash timing.
• Forgetting quarterly or annual items such as insurance, instalments or annual software renewals.
• Ignoring sales tax collected: it sits in the bank but is owed to the government.
• Building the forecast once and never rolling it forward.
• Skipping variance review, so the same errors repeat every week.

How CashflowCycles helps

CashflowCycles connects to your bank accounts through Plaid with read-only access and syncs transactions daily, so the opening balance and last week's actuals are already there. Its Obligation Radar lays out the next six weeks of committed payments, such as payroll, rent and remittances, against the cash on hand, and what-if scenarios show what happens to those weeks if a client leaves or revenue drops. For a longer view, forecasts run 3, 6 or 12 months ahead with a moving-average, linear-trend or polynomial model, each with base, best and worst cases. Alerts flag a shrinking cash runway before it becomes urgent. It works in English and French, with a 30-day free trial.