Enter your opening balance and the money moving in and out each week. The table below rebuilds as you type and tells you the one thing a profit and loss statement never will: the first week you run out of cash.
Cash goes negative in week 10
The week beginning Oct 5, 2026, ending at -$14,014.
Cash figures, not accrual. A sale counts in the week the money arrives, and a bill counts in the week you pay it. Amounts are rounded to whole dollars.
What is in the bank on Aug 3, 2026, across every operating account. Exclude money you cannot spend, such as a restricted deposit.
Collections, deposits, loan draws, anything landing in the account.
Payroll, vendors, rent, debt service, taxes, owner draws.
Weeks start on Monday. This window runs Aug 3, 2026 to Nov 1, 2026.
| Week | Week of | Receipts | Disbursements | Net change | Ending balance |
|---|---|---|---|---|---|
| Opening balance | $82,040 | ||||
| Week 1 | Aug 3 | $23,914 | $59,710 | -$35,796 | $46,244 |
| Week 2 | Aug 10 | $31,986 | $8,570 | $23,416 | $69,660 |
| Week 3 | Aug 17 | $23,914 | $44,930 | -$21,016 | $48,644 |
| Week 4 | Aug 24 | $33,064 | $6,430 | $26,634 | $75,278 |
| Week 5 | Aug 31 | $23,914 | $59,710 | -$35,796 | $39,482 |
| Week 6 | Sep 7 | $31,986 | $8,570 | $23,416 | $62,898 |
| Week 7 | Sep 14 | $23,914 | $44,930 | -$21,016 | $41,882 |
| Week 8 | Sep 21 | $23,914 | $6,430 | $17,484 | $59,366 |
| Week 9 | Sep 28 | $23,914 | $59,710 | -$35,796 | $23,570 |
| Week 10 | Oct 5 | $31,986 | $69,570 | -$37,584 | -$14,014 |
| Week 11 | Oct 12 | $23,914 | $44,930 | -$21,016 | -$35,030 |
| Week 12 | Oct 19 | $23,914 | $6,430 | $17,484 | -$17,546 |
| Week 13 | Oct 26 | $23,914 | $44,930 | -$21,016 | -$38,562 |
| 13-week total | $344,248 | $464,850 | -$120,602 | -$38,562 | |
This page is yours to keep using. If you would rather the forecast built itself from your bank feed and your QuickBooks invoices, and told you every week where the estimate was wrong, that is TreoCast. $49 a month for one entity, $99 for multiple, with a 14-day money-back guarantee.
A 13-week cash flow forecast lists the cash you expect to receive and pay out, week by week, for one quarter. It is a bank balance projection, not a profit and loss statement. Revenue you have earned but not collected does not appear until the week the money lands. A cost you accrued in March shows up in the week the payment clears.
That distinction is the whole reason the format exists. A business can be profitable on paper and still miss payroll, because profit and cash arrive on different days. This is the report that tells you which day.
Thirteen weeks is a quarter. It is long enough to catch the payments that actually break small businesses, the quarterly estimated tax, the annual insurance renewal, the software bill nobody remembered. It is short enough that you can still forecast each week from real information rather than a trend line. Past a quarter, weekly precision turns into invention and a monthly forecast serves you better.
The format came out of restructuring work, where a lender wants to see exactly when a borrower runs out of room. It has since become ordinary practice for any business managing against a cash constraint, which in a tight year is most of them.
Read the ending balance column first, top to bottom, and find the lowest number. That is your real cushion. Not the balance in the bank this morning, and not the balance at week 13.
Then read the net change column. A week that swings hard negative is usually payroll or one large payment. If the same swing repeats every other week, that is payroll and it is not a problem. If one week is far worse than its neighbours, something lumpy is landing there and it is worth opening the calendar to check the date.
If the calculator names a shortfall week, that is the first week your ending balance drops below zero on these assumptions. Treat it as a deadline rather than a prediction.
There are four levers, roughly in the order most businesses reach for them.
Pull receipts forward. Invoice earlier in the cycle, ask your three largest customers to pay on terms rather than on habit, or offer a small settlement discount to the ones who always take an extra fortnight.
Push payments back. Move a vendor run by a week, ask for terms on a large purchase, or hold discretionary spending that currently sits inside the shortfall week.
Fund the lumpy item. An estimated tax payment or an insurance renewal usually cannot move. Knowing eight weeks out that it lands in a thin week is normally enough time to set the cash aside for it.
Raise the floor. Draw on a line of credit, or arrange one before you need it. Lenders respond far better to a dated forecast than to a phone call in the week it matters.
Then rebuild the forecast with the change in it and check that the shortfall week actually moved. Most fixes shift the problem rather than solve it, and the table will tell you which one you have.
Roll the forecast forward every week and compare last week's estimate to what actually cleared the bank. The variance is the useful part. A forecast that is never checked against reality drifts within a month, and a forecast built once and filed is worse than no forecast at all, because it feels like knowledge.
Yes. There is no email wall, no signup, and no download. The whole projection is computed in your browser, so your figures are never sent to TreoCast or anyone else. Your inputs are saved in your own browser storage so a refresh does not lose your work, and clearing your browser data removes them.
Every 4 weeks steps a fixed 28 days, so an item starting in week 1 lands in weeks 1, 5, 9 and 13. Monthly steps by calendar month from the same date, so it lands in whichever week that date falls in. Use every 4 weeks for things tied to a pay cycle and monthly for rent, insurance and subscriptions billed on a date.
Cash. A 13-week forecast projects your bank balance, so an invoice counts in the week you expect to be paid, not the week you raised it, and a bill counts in the week the payment clears. If you enter accrual numbers the totals will look right and the weekly timing will be wrong, which defeats the exercise.
The first three or four weeks are usually close, because most of the cash movement is already committed. Accuracy falls off after that as collections timing becomes a judgement call. The point is not to be right to the dollar, it is to know roughly when you get tight and to have enough notice to do something about it.
You can, and plenty of finance teams do. A spreadsheet becomes a chore when the forecast has to roll forward every week, actuals have to be pasted in, and each version has to be reconciled to the last one. That weekly maintenance is what TreoCast automates. This calculator is the arithmetic, free and unencumbered.
Your inputs are kept in your own browser so they survive a refresh, and the reset button restores the example scenario. There is no account, no server-side save and no shareable link, because adding those would mean asking you to sign up. For saved scenarios, versioned plans and multi-entity forecasts, that is what the paid product does.
TreoCast builds the same 13 weeks from your bank feed and your QuickBooks invoices, rolls it forward every week, and shows you where last week's estimate was wrong. $49/month for a single entity, $99/month across entities, with a 14-day money-back guarantee.
The calculator above stays free and open, with or without an account.