The 13-Week Cash Flow Forecast for Project-Based Service Businesses

You have a client who pays late. Not sometimes. The same eleven days past due, every time, for two years. You have stopped worrying because the money always shows up.

Then one payment lands on the wrong side of payroll and an estimated tax installment. Both clear, but the balance gets close enough to zero that you feel it. The client pays eleven days later, the month closes with a profit, and the financials never show how little room that week had.

A 13-week cash flow forecast shows it before it happens. It projects the cash expected to enter and leave the account each week, with one week of verified actuals anchoring twelve weeks ahead. For a service business with project-based, lumpy revenue, it answers the question a monthly report cannot: Will the account clear payroll and taxes before the client payment arrives?

Why a monthly report misses the tight week

Your books may close two or three weeks after the month ends. By then, a cash squeeze from the middle of the month is old news. If the balance recovered before month-end, the financial statements may not isolate it at all.

That matters when the weeks inside the month do not look alike. Payroll, rent, and tax installments follow a calendar. Client receipts do not. A retainer may arrive within a week while a milestone invoice waits for approval from three people you have never met. That is how a business can post a healthy month and still spend one Friday wondering whether payroll will clear. Profit and cash flow are different numbers, and the distance between them often gets measured in weeks.

Thirteen weeks gives you time to act without asking the model to predict too much. The forecast uses the direct method, so each item goes in the week the money should move. A client payment belongs in the week the deposit should clear. Payroll belongs in the week the run debits. Nothing is backed into from net income.

“The direct method costs more setup time up front. It pays that back the first week it flags a collision the P&L never would have shown you.” — Derek Bungard, CPA, Senior Tax Manager

What goes in the forecast

Give each source of cash and each type of disbursement its own row. Keep a reliable retainer separate from a milestone payment waiting on someone else’s approval. Then copy these rows into a sheet with thirteen weekly columns.

Below those inputs, add six lines:

1. Opening cash
2. Total receipts
3. Total disbursements
4. Ending cash
5. Hard cash floor
6. Target reserve

Week 0 opening cash comes from the bank. After that, one week’s ending cash becomes the next week’s opening cash.

Ending cash=Opening cash+Receipts−Disbursements

The two guardrails answer different questions. The hard floor covers the next payroll run and estimated tax installment. Any week below it requires a decision. The target reserve is the larger buffer the business is working toward. The withdrawal test sizes it based in part on revenue concentration.

For project-based service businesses, milestone payments often carry the most timing risk. Payroll and rent rarely move by two weeks. Client receipts do.

Build the first forecast in five steps

1. Pull the source data

Start with today’s bank balance. That is the Week 0 anchor, even if the books are still closing last month. Pull recurring debits from the general ledger and use the trailing twelve-month P&L to separate fixed costs from those that move with revenue. Add open invoices, signed contracts, payroll dates, known one-time purchases, and the next estimated tax payment.

2. Give every item a cash date

Accrual accounting records revenue when earned and expenses when incurred, as defined in IRS Publication 538. That is right for the books and wrong for a cash forecast. Every item needs the date the money should move.

3. Place receipts using collection history

For each material client, count the days from invoice to deposit across the last six to twelve paid invoices. Start with the largest invoices and the clients whose stated terms differ most from their behavior. If the history is incomplete, match deposits back to invoices and use a conservative assumption for any remaining gaps.

Suppose a client’s last six invoices cleared in 41, 47, 44, 52, 43, and 45 days. Put the next one at about 45 days, not the 30 printed on the invoice.

4. Place disbursements when they debit

Use the same rule on the way out. If you pay a vendor at 25 days, place the payment at 25 days. Put wages on the payroll debit date and insurance in the week the charge clears.

5. Reconcile it every Monday

Compare the projected ending balance for the week that closed with the bank. Investigate any difference above a threshold set in advance. Replace the projection with the actual number, move anything that slipped, drop the oldest week, and add a new one at the far end.

A $900K Consulting Firm With Big Payment Gaps

Consider a consulting firm with $900,000 in annual revenue and $451,000 in net profit. Two clients carry the revenue.

Client A pays a $22,000 monthly retainer on the first and has never been late. Client B is billed through three $53,000 milestones each quarter on net-30 terms. Client B has historically paid in 45 days, so a milestone due in week 5 is expected to clear in week 7.

Cash leaves on a reliable schedule:

• $14,500 for payroll and contractors every other week
• $6,000 for rent, insurance, and software in the first week of each month
• A $33,825 estimated tax installment in week 6, based on the owner’s tax plan
• A $26,000 owner distribution each month

Week 0 cash is $78,000. The firm’s hard floor is $48,325, enough for one payroll run and the next estimated tax installment. Because two clients carry all the revenue, its target reserve is $112,250. The account begins above the hard floor but $34,250 short of the target reserve.

Here is what happens when the same milestone is placed two different ways.

If the forecast uses the invoice terms, week 6 closes at $86,675. No problem appears.

If it uses the client’s payment history, the same week closes at $33,675. Payroll, the owner distribution, and the tax installment leave before the $53,000 milestone arrives. Cash falls $14,650 below the hard floor.

The monthly P&L still shows a profit. It does not show that the account dropped to $33,675 before recovering to $72,175.

Now push Client B from 45 days to 60 because the milestone is waiting on an approval. If the owner leaves the distribution in week 6, no deposit arrives in week 7 and week 8 payroll leaves $19,175 in the account. The next payroll is covered, but the cash set aside for the next tax installment is gone.

The forecast creates a decision while there is still time to make one. Move the $26,000 distribution out of week 6 and the low rises from $33,675 to $59,675, back above the hard floor. If the milestone slips to 60 days, keep the distribution on hold until the receipt clears.

That solves the immediate week. It does not solve the $34,250 reserve gap the firm carried into the quarter. The forecast shows both problems without confusing them.

“The forecast doesn’t fix the tight week. It makes sure you’re not finding out about it on the day it happens.” — Derek Bungard, CPA, Senior Tax Manager

A planned hire belongs in the same grid. Add the salary and payroll taxes as disbursements, then look at the ending-cash line before committing. A hire can be profitable over a year and still create a cash problem before the related revenue arrives.

What to do when a week goes red

Five weeks of notice gives you options. Five days may leave you with only the owner distribution.

Pull a receipt forward

Add a deposit to the next milestone or split the invoice between acceptance and delivery so part of the cash clears sooner. For an invoice already outstanding, ask whether the client can approve or pay it before the scheduled date.

Push a movable disbursement

Move the software renewal or capital purchase. If a vendor gives you 30-day terms and you normally pay on day 10, use the terms you already have without paying late.

Confirm the line of credit

A line arranged in week 1 is financing. A line arranged in week 6 is a fire drill. Confirm the available amount, draw process, and bank requirements before the forecast says you need it.

Hold the owner distribution

In the example, moving the $26,000 distribution restores the hard floor without delaying payroll or the estimated tax installment. In the 60-day case, the distribution stays held until the milestone clears.

Payroll and estimated tax should not be the first levers. The underpayment penalty is figured separately for each payment period. If the week 6 installment remains unpaid until week 10, the charge is calculated on that underpayment for the days it remained unpaid, even if the annual tax is later paid in full.

Two mistakes that make the forecast unreliable

Building it once

A forecast is not a finished document. The client who has always paid at 45 days may pay at 60. A contractor invoice may arrive a week early. Every Monday, the forecast shows where last week’s assumptions were wrong. That variance is useful information, not something to smooth away.

Adding complexity before the habit exists

A lender-facing forecast may need covenant tests and variance commentary. A service business building this process for the first time does not need them.

Start with the rows above. Reconcile them every week for twelve weeks. Once the habit is stable, add detail only where it changes a decision.

The decision to make this week

Open a sheet and enter Monday’s bank balance in Week 0. Add every receipt expected over the next twelve weeks using the client’s payment history, not the invoice terms. Then add payroll, recurring debits, known one-time costs, owner distributions, and the next estimated tax installment.

Calculate ending cash and compare each week with the hard floor and target reserve. If a week falls short, decide now which lever you will use.

The setup is not the part that makes the forecast valuable. The upkeep is. Someone has to reconcile it, move the invoice that slipped, update the tax payment, and check the ending balance against the floor every week.

At Visor, the bank balance, current books, payment history, and tax plan stay connected inside one accounting operating system. The Monday reconciliation happens whether or not you remember it is Monday. You can start free with no credit card.

Built the forecast and found a week below your floor? Talk to our team and walk through which lever to pull first.

Frequently asked questions

What is a 13-week cash flow forecast?

A 13-week cash flow forecast is a rolling, week-by-week projection of cash receipts and disbursements. It uses the direct method, with one week of verified actuals anchoring twelve weeks projected from open invoices, signed work, payroll, recurring bills, tax payments, and other known cash movements.

Why use 13 weeks instead of a monthly budget?

A budget tells you whether the broader plan is holding. A 13-week forecast tells you whether the account can clear the next payroll run. The weekly view matters when receipts arrive unevenly and several large payments can collide before a client invoice clears.

Thirteen weeks is also close enough for the inputs to remain useful. Beyond that, more receipts depend on work that has not been signed or invoices that have not been issued.

How do you forecast cash flow when revenue is irregular?

Give each material client its own receipt row and place each invoice using that client’s payment history. A reliable retainer and an approval-dependent milestone carry different timing risk, so they should not be grouped together. Use a conservative assumption when history is incomplete, then replace it as actual deposits arrive.

How often should you update a 13-week cash flow forecast?

Update it once a week on the same day. Replace the week that ended with the actual bank balance, explain material differences from the forecast, move anything that slipped, and add one new week at the far end.

Want the forecast built and maintained for you? Talk to our team.