The 13-week cash-flow model your board will actually read

Profitable companies run out of cash. It is such a well-worn line that boards stop hearing it, right up until a receivable slips a month and a statutory payment falls in the same week. The 13-week cash flow exists for precisely that window: long enough to see trouble coming, short enough to be honest.

Why thirteen weeks

A quarter is the natural rhythm of Indian compliance — advance tax, GST cycles, TDS, PF and ESI, and for most businesses at least one large seasonal swing. Thirteen weeks captures a full cycle without inviting the fiction that sets in beyond about ninety days.

It is deliberately not a forecast of profit. It is a forecast of the bank balance, which is a different and far less flattering number.

Build it on receipts, not revenue

The discipline that makes the model work is refusing to let accrual thinking in:

• Collections, not sales. Take the actual ageing, apply the collection pattern you observed last quarter rather than your terms, and be specific about the three or four large customers who drive the number.
• Payments by week, not by month. Statutory dues, payroll, rent and loan servicing land on known dates. Put them where they fall.
• One row per certainty level. Committed, likely, hoped-for. Boards make better decisions when the model shows them which is which.

The rolling discipline

Every week, replace forecast with actual for the week just ended, add a new week 13, and — this is the part everyone skips — record the variance and its reason in one line. After a quarter you have something more valuable than the model: an evidenced view of how wrong your forecasting is, and in which direction it is consistently wrong.

In our experience most owner-managed businesses discover they are optimistic on collections by a predictable margin. Once that is known it can be priced in.

Where it earns its keep

Three situations, repeatedly: negotiating a working-capital limit, where a bank that sees a disciplined weekly model asks fewer questions; deciding whether a large order can be funded without borrowing; and the week a major customer slips, when the difference between knowing on Monday and knowing at month end is the difference between a phone call and a default.

Keep it to one page

Thirteen columns, perhaps twenty rows, opening and closing bank balance in bold, and the lowest projected balance highlighted with its date. If it needs a second page, it has stopped being a decision tool and become a report.

A last word on ownership. The model belongs to whoever can pick up the phone to a customer about an overdue invoice, not to an analyst producing a file. The value is not the spreadsheet; it is the weekly half hour in which somebody with authority looks at week nine, sees the balance dip below comfort, and still has time to act on it.