A 13-week cash flow forecast is a spreadsheet with one column per week for the next 13 weeks and one row for each kind of money coming in or going out, ending in a running bank balance. It shows a cash gap weeks before it arrives, which a monthly profit figure cannot. Build it once in about two hours, then update it for 15 minutes every Friday.
Why monthly figures hide the problem
A monthly profit and loss tells you what you earned and spent over 30 days. For a business paid in uneven chunks, that is too coarse.
Say June shows $12,000 of revenue and $4,000 of costs. It looks healthy. But if the $12,000 arrived in one payment on June 28 and rent was due on June 1, you had a hole in your bank account for most of the month. The monthly number is correct and useless for what hurt.
Small businesses report this pain often. The Federal Reserve's 2024 Small Business Credit Survey, published in March 2025, found that 56% of firms cited paying operating expenses as a financial challenge and 51% cited uneven cash flow. The survey is a convenience sample, not a random one, and it covers employer firms, which are larger than one-person businesses, so read it as a sign of how common the issue is.
Thin reserves make gaps worse. A JPMorgan Chase Institute study, published in 2016 using 2015 data, found that the median small business held 27 "cash buffer days," meaning the days it could cover its outflows if inflows stopped, and that half held less than a month. It is old, but the shape has not changed for most small firms.
Payment speed adds to it. Xero's US small business data for the March 2026 quarter shows small businesses waited an average of 28.8 days to be paid, with invoices paid 9.0 days late on average. If you invoice on net 30, the gap between earning and having the money is the problem a weekly view makes visible.
Written by
The Delivvo team
Delivvo runs the client cycle for service businesses: inquiry, quote, contract, deposit and delivery. Payments go straight to your own account.
For 2026 the IRS kept seven brackets from 10% to 37% and raised the standard deduction to $16,100 for single filers. This post gives the thresholds, explains self-employment tax and the permanent 20% QBI deduction, and shows how to set tax money aside from each payment.
The first invoice feels bigger than it is. This post lists what every invoice needs, how to set the due date and payment terms, why a signed agreement should sit behind it, and how to turn the first one into a template you reuse.
Most late payments start with the invoice, not the client. This post goes through seven common mistakes, from a missing due date to no way to pay online, and shows what to change so the money arrives sooner.
8 min read
Your inquiry link is free.
Free plan. No card needed.
Blog
Guides for service businesses, from the first inquiry to the final payment. Written by the team behind Delivvo.
Thirteen weeks is one quarter. It is long enough to cover monthly bills, a quarterly tax payment and a slow-paying client, and short enough that each cell can be a number you actually know. Longer forecasts turn into guesses.
The rolling part matters more than the number. Each Friday you drop the week that just ended and add a new one at the far end, so you always look the same distance ahead. A forecast made once in January is decoration. One refreshed every Friday is a steering wheel.
Hands using a calculator next to a laptop showing a spreadsheet of weekly figures
The rows you need
The grid has one row for each cash line and one column for each of the 13 weeks. Choosing the rows is the real work.
Starting cash. The first cell is what is in your bank account on the morning you begin. If this is wrong, everything after it is wrong.
Cash in, by client and invoice. Enter each invoice in the week you expect to be paid, based on how that client actually pays, not the date you invoiced or the due date. A client on net 30 who always pays in 45 days goes in week 7.
Cash in, other. Refunds, one-off sales, anything dated and specific.
Fixed costs. Rent, insurance, software, phone and internet. They land on the same day each month, which makes them easy to forecast and easy to forget.
Variable costs. Subcontractors, travel, advertising, a coworking desk.
Tax. Give it its own row with a real number in the week a payment is due. In the US, the IRS lists estimated tax due dates of April 15, June 15, September 15 and January 15, which our guide to estimated quarterly taxes explains. Other countries have their own dates, so use yours.
Owner draw. The money you move to your personal account is real cash leaving the business. People leave this row out because it feels like double counting, and then the forecast lies to them.
Ending cash. Starting cash plus cash in minus cash out. It becomes next week's starting cash. This is the line you watch.
A small worked example
Here is a four-week slice for a consultant, with invented figures to show how the grid reads.
Week 1 starts with $6,000. A $2,500 invoice is expected, and rent of $1,800 and software of $200 go out. Ending cash: $6,500.
Week 2: nothing comes in. A $900 subcontractor payment goes out. Ending cash: $5,600.
Week 3: a $4,000 invoice was due this week, but the client usually pays two weeks late, so it sits in week 5. A US estimated tax payment of $2,400 goes out. Ending cash: $3,200.
Week 4: rent of $1,800 goes out again early in the month, plus $200 of software and a $1,000 owner draw. Ending cash: $200.
The monthly profit for those weeks looks fine, but the forecast shows cash of $200 at the end of week 4, with the $4,000 not due until week 5. That is a gap you can fix now by invoicing the next milestone early, asking for a deposit on a new project, or moving the owner draw. Found on the morning, it is a crisis.
The Friday check
The forecast is not the value. The weekly variance check is.
Every Friday, replace last week's forecast numbers with what actually happened, line by line. Compare them. Anything close is fine. Anything much earlier, later or smaller is a signal.
A sensible threshold is about 10%, or any single line over a fixed amount such as $500. When you hit it, ask one question: is this a one-off wobble or the start of a pattern? One late payment from a normally reliable client is a wobble. The same client paying late three months running is a pattern, and the pattern should change your terms.
The point of the check is to re-forecast the next 12 weeks while there is still time to act. A gap you see four weeks ahead can be fixed by sending an invoice sooner, delaying a discretionary cost or collecting a deposit. A gap you find the day it arrives cannot.
Income you control and income you do not
Treat these differently.
Income you control, such as a signed retainer or a deposit you have already invoiced, goes in the week you expect the money, at close to 100%.
Income you do not control, such as a proposal waiting for a signature, goes in only if you would bet the forecast on it. The cautious approach is to leave it out, or use a reduced figure, and treat a win as upside. An optimistic forecast hides the gap it exists to show.
What the forecast tells you to do
A useful forecast produces decisions. Three patterns come up.
A predictable dip. Ending cash falls below your minimum in week 7. Small moves now help: invoice a milestone a few days early, defer a renewal, or use a credit line for the dip and repay it when a larger payment lands.
Creeping late payments. Two clients drift from 30 days to 45. You see it because the cells where their money used to land are empty. Act before it reaches 60, using the steps in our post on handling late-paying clients, and think about whether a late fee is right for you, which we cover in late fees.
A thin pipeline. Weeks 10 to 13 are light because no new work is queued. That is a sales problem, not a finance one, and the forecast shows it eight weeks before the rent.
The forecast also tells you when not to act. One late payment with everything else on plan is variance. Reacting to every wobble wears you out, which is why the threshold exists.
For general financial housekeeping, the US Small Business Administration's guide to managing your finances covers bookkeeping and the financial statements that sit next to a cash forecast. A balance sheet shows where you are, and the 13-week view shows where you are heading.
How long it takes
A first build takes about two hours for a business with five to ten active clients: one hour to gather open invoices, recurring bills and the next tax date, and one to set up the grid. After that, the Friday update takes about 15 minutes.
The usual failure is building it once and never updating it. A stale forecast is worse than none, because it makes you confident about a future that has already changed. Pick a fixed slot and protect it like a client meeting.
How Delivvo makes the cash-in rows reliable
Delivvo does not build the forecast. You can do that in a spreadsheet. What it does is make the "cash in" rows real, because the dates and amounts come from invoices and schedules instead of memory.
Here is how it works:
When a client accepts your quote, the contract, deposit invoice and payment schedule are created together. A schedule splits the fee into up to 12 payments, each with its own due date, so each one is a dated line for your forecast.
On Pro, the client picks, signs and pays the deposit on one page, so a deposit often arrives before you start. That puts cash in week 1 instead of week 6.
Retainers use recurring invoices, so a predictable amount lands on a predictable date.
Invoices are numbered in order in your format, so your list of what is open has no gaps.
If an invoice goes unpaid, reminders run automatically at 7, 14 and 30 days and stop once it is paid, which shortens the slow tail that wrecks forecasts.
The client portal shows each client what they owe, so fewer payments are late because someone could not find the invoice.
Clients pay through your own gateway, and Delivvo takes 0%. Every invoice also carries a verification code. After a payment arrives, you replace the forecast number with the real one in your Friday update. Because the schedule and the invoice list live in one place, the Friday update gets quicker. You scan the open invoices, move any that slipped to a later week, and mark the ones that arrived. If you work with a team or subcontractors, their tasks and due dates sit in the team workspace, which helps you time your variable costs.
The Free plan has no card requirement, Solo is $12 a month and Pro is $29.
It is a weekly forecast of the cash coming into and leaving your business over the next 13 weeks, ending each week with a bank balance. You update it weekly so it always looks 13 weeks ahead.
Why 13 weeks and not 12 months?
Thirteen weeks is one quarter, long enough to cover monthly bills, a tax payment and a slow client, and short enough to forecast with real numbers. Longer forecasts become guesses.
Do I need accounting software to build one?
No. A spreadsheet with about eight rows and 13 columns is enough. Accounting software helps with the inputs, but the discipline is the weekly update.
How often should I update it?
Every week, ideally the same time each Friday. A forecast that is not updated gives false confidence.
Delivvo runs the whole client cycle for service businesses, from the first inquiry to the quote, contract, deposit, invoice and delivery, so your expected payments are dated, scheduled and chased automatically. Clients pay you directly through your own gateway, and Delivvo takes 0%. Start free
