The bank's email asks for a 13-week cash-flow forecast by the end of next week. You have an annual budget, a monthly income statement, and a general sense that the fourth quarter is tight. None of those is what the banker asked for, and the banker knows it. The request is a test of whether you can see your own cash 90 days out, and the answer is about to be visible.
A 13-week cash-flow forecast is the single most useful document a company between $3M and $50M can produce, and it is one of the least often produced. This article explains what it is, how it is structured, how to build one from the records you already have, what a sample layout looks like, why lenders and turnaround professionals insist on it, and the mistakes that turn a good tool into a bad spreadsheet.
What it is, and what it is not
A 13-week cash-flow forecast is a direct forecast of cash receipts and cash disbursements, by week, for the next 13 weeks, starting from today's actual bank balance. Thirteen weeks is one quarter. It is long enough to see the next payroll crunch, the quarterly tax payment, the insurance renewal, and the seasonal dip, and short enough that the numbers can be built from real invoices and real bills rather than from assumptions.
It is not a budget. A budget is an annual plan, built on accrual accounting, that tells you what you intend to earn and spend. The 13-week forecast is built on cash, ignores revenue recognition entirely, and tells you what will actually be in the bank on a given Friday. It is also not an income statement projection. Profit does not appear on it anywhere. A company can be profitable in every column and out of cash in week seven, which is precisely why the document exists.
The structure
The layout is the same in almost every company that uses one, because the logic is the same. Each week is a column. The rows run from beginning cash through receipts and disbursements to ending cash, with the line of credit handled below the ending balance so that operating cash and borrowing are visible separately.
- Beginning cash. The actual bank balance for week one, and the prior week's ending cash for every week after.
- Receipts. Customer collections, broken into categories that make sense for the business: by major customer, by job, or by division. Plus other receipts such as tax refunds, asset sales, or owner contributions, listed separately so they are not mistaken for operations.
- Disbursements. Payroll and payroll taxes, on the actual pay dates. Vendors and subcontractors. Rent. Debt service, principal and interest, on the actual due dates. Insurance. Sales and income taxes. Capital expenditures. Owner distributions. Each on its own row.
- Net cash flow. Receipts minus disbursements for the week.
- Ending cash before borrowing. Beginning cash plus net cash flow.
- Line of credit activity. Draws and repayments, then the resulting line balance and remaining availability.
- Ending cash. The number that matters, and the number the bank will compare to the actual balance 13 weeks from now.
The forecast also carries a minimum cash line, the amount below which you cannot operate comfortably, so that any week where ending cash drops under the minimum is visible at a glance. A week in the red, seven weeks out, is not a crisis. It is the reason the forecast was built. A week in the red on Thursday is a crisis, and it is the reason the forecast was not.
A sample layout
The table below shows the first four weeks of an illustrative forecast for an example service contractor. The figures are invented for illustration and are not a client's data. In a working forecast, the columns continue through week 13, and a final column totals the period.
| Line | Week 1 | Week 2 | Week 3 | Week 4 |
|---|---|---|---|---|
| Beginning cash | 184,000 | 96,500 | 213,000 | 111,000 |
| Customer collections | 142,000 | 310,000 | 98,000 | 265,000 |
| Other receipts | 0 | 0 | 12,000 | 0 |
| Total receipts | 142,000 | 310,000 | 110,000 | 265,000 |
| Payroll and payroll taxes | (118,000) | 0 | (121,000) | 0 |
| Vendors and subcontractors | (84,000) | (142,000) | (61,000) | (158,000) |
| Rent | (22,500) | 0 | 0 | 0 |
| Debt service | 0 | (31,000) | 0 | 0 |
| Insurance | 0 | (18,000) | 0 | 0 |
| Taxes | 0 | 0 | (28,000) | 0 |
| Capital expenditures | 0 | 0 | 0 | (45,000) |
| Owner distributions | (5,000) | (2,500) | (2,000) | (5,000) |
| Total disbursements | (229,500) | (193,500) | (212,000) | (208,000) |
| Net cash flow | (87,500) | 116,500 | (102,000) | 57,000 |
| Ending cash before borrowing | 96,500 | 213,000 | 111,000 | 168,000 |
| Line draw / (repayment) | 0 | 0 | 0 | 0 |
| Ending cash | 96,500 | 213,000 | 111,000 | 168,000 |
| Minimum cash | 100,000 | 100,000 | 100,000 | 100,000 |
| Line availability | 350,000 | 350,000 | 350,000 | 350,000 |
Even four weeks tell a story. Week one ends $3,500 under the minimum because payroll, rent, and a vendor run land in the same week as a light collection week. Nothing is wrong. The company has $350,000 of line availability and a large collection week coming. But the owner now knows on Monday of week one that a small draw or a phone call to the largest customer will keep the balance above the floor, instead of finding out from the bank on Friday afternoon.
How to build one
The first version takes a few days if the underlying records are reasonable. It takes longer if the agings are not reliable, which is itself useful information. The build has six steps.
- Start from the bank. Week one beginning cash is the actual reconciled bank balance, not the general ledger cash account. If the two differ materially, reconcile before doing anything else.
- Schedule receipts from the receivables aging. Take every open invoice and assign it to the week you expect it to be collected, based on that customer's actual payment history rather than the invoice terms. Then add expected collections from work not yet invoiced, using the billing calendar. For a contractor, that means the pay application schedule and retainage release dates.
- Schedule disbursements from the payables aging and the calendar. Every open bill gets a planned payment week. Then layer in the recurring items that are not in payables yet: payroll on its actual dates, debt service, rent, insurance, and taxes on their due dates. Pull last year's bank statements for the quarter to catch anything you forgot.
- Add the known one-time items. Equipment purchases, bonuses, distributions, deposits on new leases, and the settlement everyone has stopped mentioning.
- Compute ending cash by week and compare to the minimum. Identify the low weeks and decide what to do about them while there are still weeks to do it in.
- Roll it every Monday. Replace week one's forecast with actual results, note the variance by line, drop the completed week, add week 14, and update every remaining week for what you learned. The forecast is a living document. A 13-week forecast built once is a 13-week snapshot.
Why the weekly cadence matters
Cash moves weekly. Payroll is weekly or biweekly. Vendor runs are weekly. Customers pay on their own weekly check runs. A monthly forecast averages all of that into one number and hides the fact that the month's cash comes in during the last 10 days while the payroll goes out during the first 10. The week is the smallest unit at which the forecast is still buildable from real records and the largest unit at which it is still useful.
The Monday update is where the value lives. Comparing last week's forecast to last week's actual, line by line, is how you learn that the largest customer really pays in 52 days, not 30, and that the vendor run is always $20,000 heavier than the aging suggests. After six or eight weeks of variance review, the forecast becomes accurate, because it has been corrected by reality every week. The first version is always wrong. The tenth version is usually right within a few percent.
Why lenders ask for it
A bank asks for a 13-week forecast for two reasons, and the second one is the one to pay attention to. The first is that the bank wants to know whether you can service the debt for the next quarter, and the forecast answers that directly. The second is that the bank wants to know whether you know. A company that can produce a credible weekly forecast in three days is a company with management in control of its cash. A company that needs three weeks and sends a monthly budget instead has told the bank something too.
In a workout, a forbearance, or a bankruptcy, the 13-week forecast is not optional. It becomes the operating document. The lender or the court approves a budget by week and by line, actual results are reported against it, and variances beyond a set threshold require explanation. In our experience, the companies that survive those situations are the ones that had a working forecast before the lender asked, because the discipline of the weekly roll is what surfaces the problem while it is still fixable.
The mistakes that ruin it
A 13-week forecast fails in a small number of predictable ways. Each of them is easy to avoid and easy to fall into.
- Forecasting collections from the sales forecast instead of the receivables aging. Revenue is not cash. Collections come from invoices, on the dates customers actually pay.
- Using invoice terms instead of payment history. Net 30 is what the invoice says. Day 47 is when the check arrives. Use the history.
- Starting from the general ledger cash balance rather than the bank. If they differ, the forecast starts wrong and stays wrong.
- Forgetting the lumpy items. Quarterly taxes, annual insurance, the equipment note balloon, the year-end bonus. Last year's bank statements are the checklist.
- Building it once. A forecast that is not rolled weekly stops being a forecast after the first Friday.
- Skipping the variance review. The point of comparing forecast to actual is to fix the assumptions. A forecast that is rolled without being corrected simply carries its errors forward.
- Treating it as a budget. The forecast is not a spending limit. It is a picture of what will happen if nothing changes, so that something can change.
- Optimism in the receipts. Every forecast that fails, fails on the receipts line. Disbursements are easy to predict because they are in your control. Receipts are not, and they should be forecast conservatively and reviewed weekly.
The most common failure is the quietest one: the forecast gets built for the bank, sent to the bank, and never opened again. The bank was not the audience. Friday was.
Where a fractional or interim CFO fits
Building and running the 13-week cash-flow forecast is core work for a fractional or interim CFO, and it is usually the first deliverable of an engagement because everything else depends on it. The CFO builds the first version from the bank, the agings, and the calendar, runs the Monday roll and the variance review for enough cycles to make the numbers reliable, and then either keeps running it as part of an ongoing arrangement or trains the Controller or the owner to run it and documents the process so it survives the handoff.
For a company in a cash crunch, the forecast is the tool a turnaround CFO uses to decide which vendors get paid this week, which customers get called, and how large a line draw is needed, and it is the document the lender will require in any forbearance discussion. For a company that is merely growing fast, it is what sizes the line of credit before the growth arrives instead of after. The cash-flow consulting page on this site describes the broader work. The forecast is where it starts, and it fits on one page.