How to Build and Run a Rolling 13-Week Cash Flow Forecast

Your income statement tells you whether you made money. It does not tell you whether you can make payroll in six weeks. A 13-week cash flow forecast does.

Thirteen weeks is one quarter. It is short enough to forecast in detail and long enough to act on what you see. Lenders and boards often ask for one when cash is tight. It is just as useful when cash is fine.

Direct versus indirect method

There are two ways to forecast cash.

• Indirect method: starts with net income, then adjusts for non-cash items and working capital. It suits monthly and annual forecasts.
• Direct method: lists expected cash receipts and payments, week by week. It suits a 13-week forecast.

Use the direct method. Your team can check each line against a real invoice, payroll run or loan schedule. That makes it easier to trust and easier to correct.

The structure

A 13-week forecast has four blocks. Each column is one week, usually ending on Friday.

1. Opening cash: the bank balance at the start of the week.
2. Receipts: cash coming in.
3. Disbursements: cash going out.
4. Closing cash: opening cash plus receipts minus disbursements. This becomes next week’s opening cash.

Add a line for your minimum cash buffer and one for available credit. You want to spot any week that dips below your buffer at a glance.

Building the receipts

Customer collections

Collections are usually the largest and least certain line. Build them from your accounts receivable (AR) aging. Go invoice by invoice for your biggest customers.

For each open invoice, forecast the week you expect payment. Base it on how that customer actually pays, not on the due date. If a customer on 30-day terms usually pays in 45 days, forecast 45.

DSO helps here. Days sales outstanding (DSO) is the average number of days customers take to pay. Calculate it by customer if you can.

Future sales

Invoices not yet issued also turn into cash within 13 weeks. Forecast sales by week, then apply your typical collection lag. With 45-day collections, sales from roughly the first six weeks will land inside the window.

Other receipts

Include loan drawdowns, equity, tax refunds and asset sales. Only add them when the timing is confirmed.

Building the disbursements

Group disbursements by how predictable they are.

Payroll needs extra care. Map each pay date to its week. If you pay biweekly, watch for months with three pay dates. Add employer taxes and benefits on the dates they actually leave the bank.

For payables, use your current terms and habits. Days payable outstanding (DPO) is the average number of days you take to pay suppliers. If you pay a key supplier in 30 days, don’t forecast 60 to make the numbers look better.

A simple worked example

The figures below are hypothetical. They show how one week flows.

Assume opening cash in week 1 is $420,000. The company expects:

• Collections of $185,000 from open invoices
• Payroll and benefits of $140,000
• Supplier payments of $95,000
• Rent and other fixed costs of $30,000

Closing cash for week 1 is $420,000 + $185,000 − $140,000 − $95,000 − $30,000. That equals $340,000, which becomes week 2’s opening cash. If the minimum buffer is $250,000, week 1 is fine. You then look for any later week that drops below it.

The weekly roll

A 13-week forecast is not a one-time project. Its value comes from running it every week. Each week:

1. Replace the oldest week with actuals. Pull receipts and payments from the bank.
2. Add a new week 13. The forecast always looks a full quarter ahead.
3. Update the remaining weeks. Refresh the AR and AP aging, payroll and known one-offs.
4. Review variances. Compare last week’s forecast to what happened.
5. Decide actions. Agree what to chase, delay or fund.

Set a fixed day and hour for this cycle. Monday morning works well, using the prior Friday’s bank balance. Once the template is stable, the cycle should take two to three hours.

Variance to actual

Variance analysis is how the forecast gets better. For each line, compare forecast to actual for the week just ended.

Sort differences into two types:

• Timing: the cash moved, just in a different week. A customer paid on Monday instead of Friday.
• Permanent: the amount itself was wrong. A customer short-paid, or an expense came in higher.

Timing differences usually even out. Permanent differences mean an assumption needs to change. Track forecast accuracy over time, especially for collections. If week-ahead collections keep missing by 20%, fix the method, not just the number.

The forecast is only as useful as the weekly discipline behind it.

Common mistakes

• Forecasting collections on due dates rather than real payment behavior.
• Forgetting quarterly and annual payments, such as insurance or estimated taxes.
• Mixing accrual figures from the income statement into a cash forecast.
• Building it in a spreadsheet only one person understands.
• Updating the forecast but never comparing it to actuals.
• Leaving out a minimum cash buffer, so a thin week looks fine until it arrives.

Give the forecast one owner, usually the controller or finance manager. Others supply inputs, but one person signs off each week.

Where to start

1. Pull your AR and AP aging, plus every fixed payment due in the next quarter.
2. Build weeks 1 to 13 in a simple template with opening cash, receipts, disbursements and closing cash.
3. Pick a weekly review slot and run your first variance check next week.

If you would like a second view on your cash forecast, MYGrowth offers a free 30-minute finance health check.