The 13-week cash flow forecast for marketing agencies

Agencies rarely fail because of weak revenue. They fail on timing: the invoice goes out in March, the client pays in May, and payroll clears on the 1st. A profit and loss statement will never show you that, because it works on accruals, not on money in the bank. The tool that shows it is a rolling 13-week cash forecast.

Why 13 weeks

Thirteen weeks is one quarter at weekly resolution. It is long enough to reveal the effect of a retainer ending in June, and short enough that the numbers are still credible. Annual forecasts break down in agency life because the pipeline changes every week, while the week is the real operating unit: payroll, software subscriptions, taxes and card bills all land on fixed dates inside a week.

Rolling means that every Monday you drop the week that just closed and add a new one at the end. The horizon stays at 13 weeks and never shortens.

The four lines you actually need

Most agencies build a 40-line model and abandon it in month two. Start with four blocks:

• Opening balance: what is in the account on Monday. One number, read off the bank statement, never estimated.
• Contracted inflows: payments from signed work, dated by invoice due date, not by delivery date.
• Probable inflows: open proposals, weighted by close probability. A $9,000 proposal at 50% enters as $4,500, in the week the payment would land.
• Outflows: payroll and owner draws, contractors, taxes, paid media fronted by the agency, software, rent.

The closing balance of one week is the opening balance of the next. That is the only formula that matters.

The mistake that breaks the model: fronting paid media

Performance agencies routinely put client ad spend on the company card and get reimbursed later. On paper it nets to zero. In cash it does not: the card is due in 10 days and the reimbursement arrives in 30. An agency managing $150,000 of monthly ad spend is financing its clients with its own working capital, for free.

In the forecast, that spend needs two separate lines with two different dates: the outflow on the card due date and the inflow on the reimbursement date. Once you see the gap drawn week by week, the conversation about prepaid media budgets stops feeling awkward and starts feeling obvious.

Forecast the client you have, not the contract you signed

Do not date inflows by contract terms. Date them by observed behaviour. If a client consistently pays 12 days late, enter the payment 12 days late. An uncomfortable forecast that is accurate beats an optimistic one every time.

Keep a simple column with each client's average days-late over the last six months. Beyond improving the forecast, that number becomes your argument for changing billing dates, requesting deposits, or enforcing late fees.

Reading the output

The forecast has exactly one job: show the first week where the closing balance goes negative. If that week is fewer than four weeks out, your levers are short term, in this order: collect overdue invoices, factor receivables, delay non-critical spend, negotiate supplier terms.

If the critical week sits between week five and week thirteen, the problem is structural and the levers change: revisit pricing, move new contracts to 50% upfront, cut underused software, or grow the share of recurring revenue.

If no week goes negative, you have slack. Then the question becomes: what is the minimum cash buffer this agency wants to hold? A practical benchmark for service businesses is three months of fixed costs. Below that, losing one large client becomes an emergency rather than a setback.

Cadence beats sophistication

Every Monday, 20 minutes, same person. Cash forecasts decay fast: a model last updated three weeks ago is worse than none, because it produces wrong decisions that look data-driven. If nobody has that slot blocked in their calendar, the model will not survive month two.

When the spreadsheet stops working

A spreadsheet holds up to roughly 15 active contracts. After that the bottleneck is not the maths, it is data entry: someone has to remember every new proposal, every price change, every contract that ended. That is where the forecast silently drifts from reality.

Eluvie connects proposal, contract and payment in one place, so the forecast updates when sales does its job rather than when finance remembers. If you want a picture of your current situation first, the free diagnostic takes a few minutes.