A 13-week cash flow forecast projects your cash position one week at a time across the next quarter. It uses the direct method: receipts in, disbursements out, and for each week the opening cash plus receipts less disbursements gives the closing cash. The owner rolls it forward each week and compares last week's forecast with what the bank recorded.
In Rever, Cash Pilot builds a 13-week cash forecast, Rever's name for the 13-week cash flow forecast, from the same reconciled receivables, payables and bank activity the close runs on. The decision on what to pay, and when, stays with your team.
The 13 week cash flow, defined
The 13-week cash flow model is a grid. The columns are weeks, thirteen of them, with last week's actuals beside them. Opening cash heads each column, receipts and disbursements follow, and net cash flow and closing cash sit at the foot. The illustration below shows four weeks in thousands of dollars. Each column stays a projection until the week ends and you replace it with the bank's actuals.
| Illustration, $ thousands | Week 1 | Week 2 | Week 3 | Week 4 |
|---|---|---|---|---|
| Opening cash | 500 | 450 | 500 | 650 |
| Receipts | 300 | 250 | 400 | 200 |
| Disbursements | (350) | (200) | (250) | (450) |
| Net cash flow | (50) | 50 | 150 | (250) |
| Closing cash | 450 | 500 | 650 | 400 |
To build one, work down the rows in this order:
- 1. Opening cash. Start week one from the balance the bank reports.
- 2. Receipts. Load open AR by customer, in the week each customer tends to pay, and add other inflows you can name, such as an asset sale or a draw on a credit line.
- 3. Payroll. Place each payroll and benefits run in its calendar week.
- 4. Vendor payments. Load approved AP by the payment run that will release it, with rent and other fixed payments.
- 5. Debt service, taxes and capital spending. Add each in the week it falls due.
- 6. Net and closing cash. Net the receipts against the disbursements and carry closing cash forward as the next week's opening cash.
- 7. Minimum balance. Compare each week's closing cash with the minimum balance or the borrowing headroom, and flag any week that falls short.
The direct method shows the cash itself. An indirect forecast starts from net income and adjusts for working capital, which suits a monthly or annual plan. A controller deciding whether to release this week's payment run needs the receipts and payments themselves.
Thirteen weeks is one quarter. Weekly buckets show what a monthly view averages away: the payroll week, the week a large customer pays, the week a quarterly tax payment leaves. A quarter reaches past the next quarter-end, where a lender tests covenants if you have one. Beyond thirteen weeks, fewer receipts and payments trace to open invoices, payroll calendars or signed commitments, and you are estimating.
One person owns the forecast. In a company without a treasury team that is the controller or the CFO, with inputs from AP, AR and payroll. A lender or a sponsor may ask to see it weekly. The weekly refresh has four steps: drop the week that ended, load its actual receipts and disbursements, add a new thirteenth week, and explain any variance above the threshold you set.
Timing breaks first
A 13-week forecast can get the quarter's totals right and the weeks wrong. The owner decides when to pay from the weeks, so a forecast with the right total and the wrong weeks leads to the wrong payment run.
Stale AP timing. The forecast assumes AP pays each vendor invoice on its due date. AP pays on its own run: it holds some invoices for approval and releases the rest in a batch the week cash allows. A month later the disbursement rows describe a schedule AP stopped following weeks ago.
Receipts assumed on the due date. Customers pay late, or pay several invoices in one lump with a deduction taken. A receipts row built from due dates runs ahead of the bank, and the owner finds out when the bank balance comes in under the forecast; the accounts receivable reconciliation then shows the difference as short payments and unapplied cash.
Intercompany sweeps. In a group with several entities, cash moves between accounts on a schedule or by hand, and the intercompany reconciliation is where both sides should meet. In a forecast built account by account, a sweep shows as a receipt in one entity and a payment in another, and whoever consolidates can count it twice or miss it.
The three failures share a cause: the forecast's assumptions drift away from the transactions your ERP and bank record. The owner spends the weekly roll re-tying actuals by hand before the variance means anything, and the variance is the only test the assumptions get. A 13-week forecast stays useful while its actuals tie to matched transactions and a person chose its payment timing.
Cash Pilot's part, and your approval
Cash Pilot reconciles opening cash to bank and book balances, separating restricted funds and unsettled items. It then builds each week from source obligations and ties opening cash plus the week's movements to closing cash. Matcher keeps the bank-to-book reconciliation current as transactions land, so opening cash and settled items match the bank.
Each week Cash Pilot compares the forecast with actual cash and explains the misses, with Flux Analyst tracing the differences to their drivers. It also recommends collection and working-capital actions, ranked by their effect on cash.
No agent moves money on its own. Collection commitments, term changes and money movements go to your Inbox, where your approval thresholds apply and material amounts take two approvers. Sentinel reviews risky cash movements, and the payment run goes out on timing a person approved. The cash forecasting page shows the product side.
A month with Rever shows Cash Pilot beside the daily matching, and the month-end close checklist shows where the forecast roll sits in the close.
The first step is the zero-day scan, the first pass over your connected records once read-only access is in place. It applies the relevant reconciliations, verifications and controls to the in-scope population and reveals initial findings with the evidence behind them, and for cash forecasting it builds the first 13-week view from reconciled records. Its unexplained bank differences are the items that make a forecast's actuals hard to trust.
