Capital, cash and fundraising · Answered by Golden Section from more than 400 B2B software companies observed
Build it on the direct method, week by week: opening cash, every receipt and disbursement in the week it will actually clear, then closing cash. Forecast receipts from open invoices and renewal dates adjusted for how each customer really pays, not from revenue recognized, and put payroll, vendor renewals, taxes and debt service on their real dates. The standard is 100% accuracy on expense and a revenue line that comes in under what actually arrives. The weekly view is vital whenever cash is near your minimum, and it sits beneath the longer forecast rather than replacing it. Give it to your controller or CFO to keep evergreen, rolled forward every week against actuals, and start from your receivables aging and contract register.
Run a 13-week view whenever cash is near your minimum, the monthly forecast shows the reserve within reach in the next two quarters, a financing or covenant period applies, or billing is lumpy. Hold it to 100% accuracy on expense with revenue kept under what arrives, and judge it by whether every variance has an explanation and an owner.
| Metric | Value | What it means | Source |
|---|---|---|---|
| Forecast accuracy standard | 100% on expense; revenue under actual | weekly forecast versus what cleared the bankthe standard for a 13-week forecast kept evergreen by the controller or CFO | Golden Section operating viewGolden Section operating view |
| Collections outreach | 5 days before due, and on the due date | reminder cadence to the customer's billing contactpart of the accounts receivable best-practice process | Golden Section playbookAccounts Receivable Process play |
| Cash reporting cadence | daily cash report | cash balance, itemized AR and AP, and the day's credits and debitsemailed to the management team by the collections owner | Golden Section playbookAccounts Receivable Process play |
| Cash model refresh | at least every quarter | update of the cash model to current realityMistake 136; the founder stays responsible for its accuracy | Golden Section, publishedThe B2B Software Mistakes List |
A monthly forecast tells you whether the plan works. A 13-week forecast tells you whether you make payroll in week nine. They fail differently, because monthly models average away timing and software timing is uneven: annual invoices cluster at renewal, large customers pay late by habit, and a month with three biweekly payrolls arrives twice a year. Our growth capital guide puts it simply: a company with net-45 customers and net-15 payroll can be profitable on paper and out of cash in August.
The weekly view closes that gap only if receipts are forecast from behavior. Pull the receivables aging, renewal dates from the contract register, and payment history by customer. Then compare each week's forecast with the actual and write down why they differ. That variance log is what makes week ten more accurate than week two was.
The longer horizon needs its own discipline. Looking Glass, our venture intelligence platform, forecasts results automatically from actual performance, proprietary benchmarks from our portfolio and external benchmarks, which keeps a forecast of a year or more directionally accurate. A cash gap found ten weeks out is an ordinary collections project; found three weeks out, it gets decided under pressure.
A vertical software company at $3M in annual revenue bills most customers annually, heavily in the first quarter. The monthly forecast shows a comfortable second quarter. The 13-week view, built from the receivables aging, shows that two large renewals invoiced in March have historically paid about 50 days late, and that the month containing week 11 carries three biweekly payrolls plus an annual hosting renewal. Closing cash in week 11 dips below the company's reserve even though the month ends fine. The founder moves the collections calls forward and asks the hosting vendor for quarterly billing, and the gap never arrives. The numbers are invented.
A company with a large reserve, monthly card billing and no debt may not need a weekly view outside a financing or a downturn. Once cash approaches the reserve, or a lender's covenants apply, it stops being optional.
From the Golden Section mistakes list, each paired with the play that prevents it.
The 13-week view is the working tool for always knowing your cash position.
Aged receivables are the most common reason a weekly forecast misses, and the easiest to fix.
A forecast that is not rolled forward with actuals drifts from reality within weeks.
Weekly cash is the founder's responsibility even when someone else builds the sheet.
In the order we would run them. Each is on its own page, most with a free Excel template.
Supplies the aging, payment behavior and collections cadence the receipts forecast depends on.
Gives renewal and billing dates by customer.
Gives vendor renewals, notice terms and payment dates on the disbursement side.
Holds the monthly model the weekly view must reconcile to.
Executive plays Cash forecasting is an executive responsibility, and these plays cover the forecast, budget and reporting it connects to.
Expense should be 100% accurate, because payroll, vendor contracts and debt service are known in advance. Revenue should come in under what actually arrives, so the misses are upside. Track forecast against actual weekly, and treat any variance you cannot explain as the problem, even a small one.
The longer model tests the plan over a year or more and drives hiring, budget and capital decisions; a system like Looking Glass keeps it directionally accurate from actual performance and benchmarks. The 13-week forecast tests liquidity week by week from actual invoices and payment dates. Run both, and make the weekly view reconcile to the first quarter of the monthly one.
Lenders read cash discipline first, and our borrowers provide a monthly P&L, ARR bridge and cash flow statement. A company already running a weekly forecast has most of that work done.
Growth capital lending →Reviewed by Dougal Cameron, CEO & Co-Founder on 2026-09-23. Golden Section observations are labeled separately from external benchmarks and illustrative arithmetic.