Skip to main content

Guide

How to Build a 13-Week Cash Flow Forecast

A profitable business can still miss payroll. Profit is recognized when work is earned; cash moves when customers pay and suppliers insist. The gap between those two is where businesses get into trouble.

A 13-week forecast closes that gap. It is short enough to build from things you already know — open invoices, scheduled payroll, committed purchase orders — and long enough to see a problem while there is still time to act on it.

All resources

Structure

The weekly lines to include

Build it in columns of weeks, with each week starting from the prior week's closing cash. Keep receipts and payments separate — netting them hides the timing problem you are trying to see.

  • Opening cash: the reconciled balance across all operating accounts
  • Customer collections, timed from the AR aging and each customer's actual payment behavior
  • Other receipts: deposits, retainers, tax refunds, insurance proceeds, owner contributions
  • Payroll and payroll taxes, on the actual pay calendar rather than spread evenly
  • Supplier and subcontractor payments, timed from the AP aging and agreed terms
  • Rent, insurance, software and other fixed commitments on their due dates
  • Debt service, split between principal and interest
  • Sales, payroll and income tax payments on their filing dates
  • Capital purchases and any committed purchase orders not yet invoiced
  • Closing cash, and available headroom on any line of credit

If you would rather not run this yourself, cash flow forecasting for small businesses is the service that covers it.

Sourcing

Where each number comes from

  1. 01

    Start from reconciled cash

    Opening balance comes from the bank, not the general ledger, unless the ledger is fully reconciled. A forecast built on an unreconciled balance is wrong from week one.

  2. 02

    Time collections by customer, not by average

    Take the AR aging invoice by invoice and place each one in the week that customer usually pays. An average days-outstanding figure hides the two slow accounts that cause the problem.

  3. 03

    Take payroll from the calendar

    Use actual pay dates, and remember the months with an extra pay run. Add employer taxes and benefit remittances on their own dates.

  4. 04

    Take payables from terms, not habit

    Place each open bill on its due date first, then adjust for how you actually pay. Seeing the difference between the two is part of the value.

  5. 05

    Take debt service from the amortization schedule

    Principal and interest split, plus any covenant test dates that fall inside the window.

  6. 06

    Add committed but uninvoiced costs

    Purchase orders, subcontractor commitments and deposits already promised. These are the items that most often surprise a forecast.

A close has a date. Every step has an owner and a finish line, so statements arrive on a schedule instead of whenever the file happens to balance.
Month-end close timeline running from data cut-off on day one to delivered statements by day ten.

Discipline

Rolling it forward each week

The forecast is a habit, not a document. Once a week, replace the forecast week that just ended with what actually happened, add a new week 13 at the far end, and note the variance on any line that moved more than you expected.

The variance is the point. After a few cycles you learn which customers pay when they say they will, which cost lines you consistently underestimate, and how much cushion your business actually needs. A forecast that is never compared to actuals never gets more accurate.

Reading it

What the output tells you

  • The week cash is tightest, and how much headroom exists on that day
  • Whether the squeeze is a timing problem or a structural one that repeats every cycle
  • Which single collection or payment decision would resolve it
  • Whether a credit line needs to be arranged now, while the numbers still look healthy
  • What a delayed large payment or a lost month of receipts would do, if you model it as a second scenario
  • Whether growth is consuming cash faster than it produces it

Pitfalls

Common mistakes

  • Spreading monthly costs evenly across weeks, which hides every real due date
  • Forecasting revenue instead of collections
  • Leaving out sales and payroll tax payments
  • Recording loan payments as a single expense rather than splitting principal and interest
  • Building it once and never comparing it to what happened
  • Making it so detailed that nobody updates it weekly

FAQ

Questions we get asked about this

Next step

Want the forecast built and maintained for you?

We build the first 13-week model from your books and keep it rolling, so the conversation each week is about the decision rather than the spreadsheet.