The 13-Week Cash Flow Forecast: A Practical Guide for Mid-Sized Companies

Why thirteen weeks

A quarter of weekly detail is the horizon where cash management is still actionable. It is far enough ahead to renegotiate terms, delay a purchase, accelerate collections or draw on a facility, and near enough that the inputs — the receivables ledger, the payables ledger, the payroll calendar — are real rather than assumed.

It is also the horizon at which a profitable business discovers it is illiquid. Profit is measured monthly on accruals; cash arrives and leaves on specific days. A 13-week view is what makes that difference visible before it becomes urgent.

Structure the model correctly

• Weekly buckets with explicit week-ending dates, not four-week months.
• Opening cash for week one taken from the bank position, reconciled — not from the balance sheet cash line if they differ.
• Receipts and payments on a cash basis, categorised into a small number of lines management recognises.
• Closing cash for each week carried forward as the next week's opening balance.
• A separate line for facility headroom, so the forecast shows liquidity rather than only bank balance.

Keep the category list short. A 13-week forecast with fifty payment lines is maintained badly; one with twelve is maintained weekly.

Deriving receipts

Start from the receivables ageing rather than from revenue. Each open invoice has a due date and a customer whose payment behaviour is observable, and the forecast should place expected cash in the week the customer actually pays rather than the week the invoice falls due.

Derived DSO — receivables divided by daily revenue over a comparable period — is useful as a sanity check and for forecasting collections on sales not yet invoiced. Where a few large customers dominate, forecast them individually and apply the derived pattern only to the remaining population.

• Open invoices phased on observed payment behaviour by customer.
• Future sales converted to cash using derived DSO for the non-concentrated population.
• Known one-offs — deposits, milestone payments, tax refunds, asset disposals — placed in the specific week expected.
• A stated assumption on bad debt or dispute for anything materially overdue, rather than carrying it at full value.

Deriving payments

Payments split into three sources. The payables ledger provides committed invoices with due dates. The purchasing and production plan provides commitments not yet invoiced — often the largest source of surprise in inventory-carrying businesses. The fixed calendar provides payroll, tax, rent, insurance, debt service and dividends, which are predictable to the day and should never be spread evenly.

Derived DPO — payables divided by daily purchases — again serves as a check on the aggregate rather than as the primary driver. If the forecast implies a DPO your suppliers have never accepted, the forecast is wrong.

Finding the pinch points

The output that matters is not the closing balance in week thirteen. It is the lowest point in the period and the weeks where headroom falls below the level the business is comfortable operating at.

Read the pinch points alongside their causes: a payroll week coinciding with a quarterly tax payment, a seasonal inventory build, or a large customer whose payment consistently arrives a fortnight late. Each has a different remedy, and identifying it eight weeks out is the entire point of the exercise.

• Minimum weekly cash and minimum headroom across the horizon, highlighted explicitly.
• The specific weeks where a covenant or internal floor would be breached.
• Sensitivity on the two or three assumptions that actually move the low point — usually collection timing and a single large payment.
• A short list of available levers with the lead time each requires.

Governance: the cadence that makes it work

A 13-week forecast that is rebuilt monthly is a report. One that rolls weekly is a control. Each week, the model rolls forward one week, actuals replace the forecast for the week just closed, and the variance is reviewed line by line — not to assign blame, but because the variance is where the forecasting assumptions get corrected.

• One named owner responsible for the weekly update, with a defined backup.
• A fixed update day, immediately after the bank reconciliation.
• A short weekly variance review: what did not arrive, what was not paid, and why.
• Assumptions documented in the model itself so the reader can challenge them without asking.
• A monthly reconciliation between the cash forecast and the P&L forecast, so the two do not drift apart.

Common failure modes

• Forecasting receipts on invoice due dates rather than observed payment behaviour.
• Spreading payroll, tax and debt service evenly instead of placing them on their actual dates.
• Omitting purchase commitments that are placed but not yet invoiced.
• Building the model at a level of detail nobody can maintain weekly.
• Never comparing forecast to actual, which leaves the same estimation error in place every cycle.