How to Build a 13-Week Cash Flow Forecast

Profitable companies still run out of cash. It happens when a large customer pays late, when an import L/C matures in the same week as payroll, or when a festival bonus arrives at the low point of the collection cycle. The annual budget will not warn you, because it is built monthly and usually on an accrual basis. The bank balance will not warn you either, because it only shows today.

The 13-week cash flow forecast sits between the two. It covers one quarter — long enough to see problems coming, short enough to forecast with real data — and it is updated every week. Treasury teams, lenders and turnaround advisers rely on it for good reason. This guide explains how to build one, how to run the weekly cycle, and how to use it to make decisions.

1. Why 13 weeks?

Thirteen weeks is one quarter. Within that window most cash flows are already known or highly predictable: invoices have been issued, purchase orders placed, salaries fixed, loan schedules agreed and L/C maturity dates set. Beyond about three months, the forecast relies increasingly on assumptions and the monthly budget becomes the better tool.

Weekly granularity matters because cash problems are rarely monthly problems. A month can end with a healthy balance and still contain a week where the account would have gone overdrawn. Payroll, tax deposits, utility bills and supplier payment runs all have fixed days; collections arrive unevenly. Only a weekly view shows the trough.

2. The structure of the forecast

Use the direct method: forecast actual receipts and payments, not profit adjusted for working-capital movements. Each line should reflect a real cash movement that someone in the business can confirm.

Keep the line items stable from week to week so trends can be compared. Twelve to twenty lines is usually enough; more detail can sit in supporting tabs.

3. Forecasting receipts: the hardest part

Collections are where most forecasts go wrong, because they depend on customer behaviour. Avoid the shortcut of assuming every invoice is paid on its due date. Instead, forecast collections from the receivables ledger customer by customer for the largest balances, and use historical payment patterns for the rest.

• Top customers: forecast invoice by invoice, using each customer’s actual average days late, not the contractual terms.
• Export proceeds: align with shipment dates, document submission, buyer payment terms and the bank’s realisation timeline.
• Smaller customers: apply a collection curve (for example, 40% in the month after invoicing, 45% in the second month, 10% in the third, 5% doubtful), based on the last six to twelve months.
• New sales: include invoices not yet raised only for the later weeks, based on the confirmed order book.

4. Forecasting payments

Payments are more controllable. Start with the fixed obligations — payroll, statutory deposits, loan instalments, L/C maturities and rent — and place them in the exact week they fall due. Then add supplier payments from the payables ledger and open purchase orders, using the payment terms actually practised. Finally, add discretionary payments such as capital expenditure, which management can defer if needed.

Tagging each payment as committed or discretionary is one of the most useful habits in treasury. When a shortfall appears, the discretionary list is the first source of options.

5. A worked example

The chart below shows an illustrative 13-week forecast for a manufacturing exporter. Opening cash is BDT 120 million, and management has set a minimum operating buffer of BDT 60 million.

Weeks 4 and 6 contain the pressure points: a large import L/C settlement and the salary-plus-festival-bonus run, at the same time as export proceeds slow because shipments were delayed. Without a weekly forecast, the team would see this only when the bank balance fell. With it, they have time to act — for example by agreeing with suppliers to move part of the week-4 payment run, drawing on an unused working-capital facility for three weeks, following up two overdue export bills, and deferring a non-urgent capital purchase.

6. Run it as a weekly cycle

A 13-week forecast is a process, not a spreadsheet. A practical weekly rhythm looks like this:

1. Monday: load last week’s actual bank balances and actual receipts and payments.
2. Monday–Tuesday: compare actual with last week’s forecast line by line and record the reasons for significant differences.
3. Tuesday: roll the forecast forward — drop the completed week, add a new week 13 — and update inputs from Sales, Procurement, HR and Commercial.
4. Wednesday: review with the finance head; agree actions for any week below the buffer.
5. Thursday: share a one-page summary with leadership: closing cash, lowest point, headroom and actions.

7. Measure and improve accuracy

Forecast accuracy improves only when it is measured. Track the variance between forecast and actual for the first forecast week (where accuracy should be high) and for week four or five (which shows how well the team anticipates). Analyse misses by line: if customer collections are consistently 15% lower than forecast, the collection assumptions need to change, not just the numbers for one week.

8. Turning the forecast into decisions

The forecast earns its place when it changes what the business does. Typical decisions it supports include timing supplier payment runs, deciding when to draw or repay short-term facilities, negotiating payment terms, prioritising collections, phasing capital expenditure and deciding whether to accept a large order that needs working capital upfront. For lenders, a well-run 13-week forecast is also strong evidence of financial control, which can help in facility renewals.

The forecast also connects directly to working capital. If cash is persistently tight, the answer is rarely just more borrowing; it is usually faster collections, better inventory turns or better-aligned supplier terms. My guide to working capital and the cash conversion cycle covers those levers, and the article on receivables ageing and DSO shows how to improve the collections side.

9. Common mistakes

• Using the monthly budget divided by four or five instead of real weekly flows.
• Assuming customers pay on contractual terms rather than actual behaviour.
• Ignoring L/C maturities, bank charges and foreign-exchange conversion timing.
• Building the forecast once and not rolling it weekly.
• Not reconciling opening cash to bank statements.
• Presenting a long spreadsheet to leadership instead of the lowest point, the headroom and the actions.

Frequently asked questions

What is the difference between a 13-week cash flow forecast and a cash budget?

A cash budget is usually monthly and covers the full year, often derived from the profit budget. A 13-week forecast is weekly, uses actual receipts and payments data, and is rolled forward every week to manage short-term liquidity.

Should the 13-week forecast use the direct or indirect method?

The direct method. It forecasts actual cash receipts and payments line by line, which makes it easier to verify with the business and more accurate for short-term liquidity decisions.

How much minimum cash buffer should a company keep?

It depends on the volatility of cash flows, payroll size, facility availability and lender requirements. Many companies set the buffer to cover at least one payroll cycle plus committed short-term obligations, and review it as the business changes.

Conclusion

The 13-week cash flow forecast is one of the highest-value tools a finance team can run. It is not complex — a structured direct-method forecast, real data from the ledgers and the business, and a disciplined weekly cycle. What it gives leadership is time: time to negotiate, to collect, to defer and to arrange funding before a shortfall becomes a crisis.

Written by Liakat Hossain

Finance and accounts professional with experience in FP&A, SAP FICO, Power BI, internal control, reporting, audit, tax, VAT and export-oriented business operations.

Author profile · Editorial policy

This article is original educational content based on professional finance practice. Real-project examples are anonymized, and scenario figures are identified as model outputs rather than guaranteed results.