Most 13-Week Cash Forecasts Are Built Backwards

Every Monday, in finance teams at thousands of companies, someone opens a spreadsheet and rebuilds the 13-week cash forecast. It is the most-watched number in treasury and the most-rebuilt. And a surprising number of them are built backwards.

The backwards way is the intuitive one. Start with the cash in the bank. Add expected revenue, growing at some assumed rate. Subtract expected costs. Carry the balance forward thirteen weeks. It produces a clean chart and a confident-looking number.

It is also mostly a guess. You started from a balance and projected forward with assumptions, and assumptions are where forecasts go to die.

The credible way starts from the other end: from what you already know is going to happen.

You know who owes you money and roughly when, because it is sitting in your AR aging. You know what you owe and when, because it is sitting in your AP. You know your payroll dates and your debt service. None of that is a guess. It is in your books right now.

A 13-week forecast built up from those known obligations is not a projection. It is a schedule of things that are already largely committed. That is the difference between a forecast you can make a decision on and a chart you nod at.

Your 13-week forecast is actually three forecasts

Here is the part most people miss, and it is the single most useful thing to internalize about a 13-week.

The thirteen weeks are not equally reliable. They come in three zones.

Weeks 1 to 4 are near-certain. This is line-item territory: the invoices that will be collected, the bills that will be paid, payroll, rent, debt service. If your week 2 is wrong by more than a few percent, something is broken in your inputs, not your model.

Weeks 5 to 8 are moderate. You are into category-level estimates now. Collections by segment, spend by department, deals weighted by pipeline. Good, not exact.

Weeks 9 to 13 are directional. This is run-rate plus seasonality plus known large items. It tells you the shape of where you are heading. It does not tell you your balance on a specific Tuesday in week 11.

The cardinal sin of cash forecasting is treating week 11 like week 2. People make real decisions, hold a payment, delay a hire, off a number deep in the directional zone as if it were precise.

The number was never precise. It was a direction. Respect the zone the number lives in.

Why it drifts, and why you rebuild it every week

A 13-week forecast is a living thing because the inputs move. A big customer pays late. A deal slips. An invoice goes out a week behind. Each of those shifts cash across week lines, and the forecast you built last Monday is already stale by Thursday.

That is why finance teams rebuild it constantly, and why it eats hours a controller does not have. Most of that labor is not analysis. It is re-pulling the AR aging, re-keying the AP, reconciling against the bank, rebuilding the same structure with this week’s numbers.

It is maintenance, not insight.

The teams where this hurts most are the ones with the most places money can sit. If you run a lending book, the picture is spread across an operating account, a collection account that has not swept yet, a facility, and reserves you cannot touch without breaching a covenant. Assembling that by hand every week is not a modelling problem. It is a data problem wearing a modelling costume.

A forecast you don’t act on is just anxiety with a chart

Here is the thing nobody says out loud. The point of a 13-week forecast is not the forecast. It is the decision it triggers.

If you see a tight week 6 and do nothing, the forecast did nothing. It just gave you something to worry about on a nicer-looking chart.

The entire value is in what you do with it: pull a draw forward, time a large payment to a stronger week, move idle cash into something earning, accelerate a collection, delay a discretionary spend.

The best finance teams treat the 13-week as a control panel, not a report. Every tight week is a prompt to act while there is still time to act. Every loose week with a surplus is a prompt to put that cash to work instead of letting it sit.

The short version

Build it from your obligations, not your assumptions. The near term is a schedule, not a guess, so treat it that way.

Respect the three zones. Make sharp decisions on weeks 1 to 4, directional ones on weeks 9 to 13, and never confuse the two.

And act on it. A forecast that does not change a decision is a weekly ritual, not a tool.

The companies that stay liquid through a rough quarter are not the ones with the prettiest forecast. They are the ones who saw the tight week early and did something about it.