Key takeaways
- Profit and cash diverge structurally: receivables, inventory, debt principal, taxes, owner draws.
- Growth consumes cash. The faster you grow, the wider the gap gets.
- A cash flow statement records history. A forecast changes decisions.
- A rolling 13‑week forecast, updated weekly, is the tool most owners lack.
- Its value is lead time, not precision. A gap seen seven weeks out still has options.
Almost every owner who has had a genuine cash scare describes it the same way. The profit and loss statement looked fine. Revenue was up. And then payroll was three days out and the money was not there.
That is not bad management. It is what happens when you run a business on a report never designed to tell you about cash.
Why profit and cash are different numbers
Your profit and loss statement is built on accrual accounting. Revenue is recognized when earned and expenses when incurred, regardless of when money moves. That is the right way to measure performance and a poor way to predict a bank balance.
Receivables. You invoice in March and collect in June. March looks profitable; the bank does not change until June. If your terms are net 30 and clients treat that as a suggestion, you are financing them.
Inventory. Cash leaves when you buy. The expense appears only when the item sells. A full warehouse is an outflow your P&L has not acknowledged.
Debt principal. Interest is an expense; principal is not. A loan payment can take thousands out each month while only a fraction shows on the P&L.
Owner distributions and taxes. Neither is a business expense. Distributions cover personal tax on business profit, and the quarterly estimates leave the account without appearing on the profit statement.
Capital purchases. A vehicle bought outright is cash today and depreciation spread over years, a small monthly cost for an amount that already left.
Underneath all of it sits the effect owners find hardest to believe: growth consumes cash. To deliver more work you hire and buy materials, and each of those outflows lands before the matching collection.
The most dangerous month is a great month
Landing a large project means hiring and buying ahead of a payment that arrives sixty or ninety days later. The month you win the work is often the month your cash position gets worst. Model it before you sign.
A statement is history. A forecast is a decision tool.
The cash flow statement reconciles profit to the actual change in cash for a closed period. It is useful for diagnosis. What it cannot do is change anything, because by the time it exists the period is over.
A forecast points the other way. It projects what your balance will be on each future date, built from what you expect to collect and what you must pay. Its purpose is to alter a decision you have not made yet.
The standards differ too. A statement has to be exact; a forecast has to be timely and approximately right. Insist on precision and you end up with something too laborious to update.
Owners often ask me to walk them through last month’s cash flow statement. The more useful conversation is the next thirteen weeks, because that is the only part they can still do something about. Ali Kafoo, CPA
The rolling 13‑week forecast
Thirteen weeks is one quarter, and the choice is not arbitrary. It is long enough to contain several payroll cycles, a full collection cycle, and a quarterly tax payment, and short enough that your estimates stay credible.
Rolling is the other half. Each week you drop the week that passed and add one at the far end, so thirteen weeks are always ahead of you. A forecast built once a quarter is thirteen weeks long in January and one week long in March.
What goes in as inflows
Expected collections, listed by customer and invoice, dated when you actually expect payment, not the invoice date, and not the due date unless that client pays on time. Add deposits, financing draws, and refunds.
What goes out
Payroll and payroll taxes on their real pay dates. Rent. Recurring software and insurance. Vendor payments by when you intend to pay them. Debt service including principal. Estimated taxes. Capital purchases. Owner distributions.
Every line is dated by when money moves, never by when revenue was earned. Two rows finish it: net movement for the week, and the running balance. That last row is the reason for all of this.
The weekly update is the whole discipline
Pick a day and keep it. Twenty to thirty minutes once the format is set: update collections against what actually landed, add new invoices, move payment dates that slipped, add the new week at the end.
Compare forecast to actual, every time
Keep last week’s projection beside what really happened. Within a month or two you will see your own biases: collections arrive a week later than you assume, one client always pays at the outer edge of terms, you forget the second payroll in a three‑payroll month.
That is what makes a forecast accurate, correction applied repeatedly to a simple model. One prerequisite is non‑negotiable: a balance drawn from an unreconciled account is a guess dressed up as a number. If bookkeeping is behind, fix that first, the point of our compliance work.
Modeling the decisions you actually face
A forecast earns its keep the first time you use it to answer a real question rather than to monitor a number.
- A hire. Model the fully loaded cost, including the ramp before that person generates anything. The question is not whether you can afford the salary, but what the running balance looks like in week nine.
- A lease. A multi‑year fixed obligation against a revenue line that is not fixed. Model it in the downside case, because that is the case in which you still owe rent.
- A price change. The revenue arithmetic is easy. The cash question is the lag before collections reflect it.
- A large purchase. Compare cash against financing by the trough of the running balance under each, not the total cost. The cheaper option that leaves you three weeks from empty is not better.
Keep three versions: a base case, a downside where revenue falls short and collections slow, and an upside. The downside is the one worth the time, because it tells you how much room you have before a decision becomes irreversible.
The longer horizon is a different instrument
The 13‑week forecast is a control tool, and it works because it is built from specific invoices and payments you can name.
An annual model runs monthly and is driven by assumptions, growth rate, gross margin, headcount plan, capital spending, rather than individual transactions. Nobody can name next October’s invoices, and pretending otherwise produces a worse forecast, not a longer one.
Its purpose is capacity: whether the hiring plan is fundable, whether debt service works at the projected margin, when outside capital would be needed. Both models are standard in our CFO and growth advisory, what a fractional CFO costs covers the economics.
Leading indicators worth watching
A few metrics warn you about cash before the forecast does, because they move first.
Track these four, consistently
Days sales outstanding, average time from invoice to payment; a rising trend is the earliest reliable warning. Collection rate, the share of what you invoiced that you actually collected. Pipeline coverage, signed and likely work against next quarter’s costs. Runway, weeks the current balance lasts at your current burn.
Four tracked every month beat twelve tracked occasionally. The value sits in the trend line, which needs consistent measurement more than a perfect definition.
Where forecasts go wrong
- Building on stale books. If the last reconciliation was six weeks ago, your opening balance is fiction and everything downstream inherits the error.
- Over‑modeling. An elaborate model nobody updates is worse than one sheet maintained every Monday.
- Never comparing to actual. Without it there is no correction loop, and the forecast stays as accurate as your first guess.
- Forecasting revenue instead of collections. The most frequent error by a wide margin. Revenue is a promise. Collections are money.
- Leaving out owner taxes and distributions. Consistently the largest missing outflow, and the one that causes the actual emergency.
What to do when the forecast shows a gap
Eventually the running balance dips below your comfort line in week seven. That is the forecast working: you have weeks to respond rather than days, and the levers are worth pulling in order.
- Accelerate collections. Work the aging report by phone, not by email. Offer a modest early‑payment discount where margin supports it. Require deposits on new work.
- Reschedule outflows deliberately. Ask vendors for extended terms before the invoice is due. Negotiated is a different thing from missed.
- Draw on a line of credit. The time to set one up is when the forecast is healthy.
- Pause owner distributions. Uncomfortable, fast, within your control.
- Cut discretionary spending. Last rather than first, usually the smallest lever.
None of these work under a three‑day deadline. All of them work under a seven‑week one. That gap is the return on twenty minutes a week.
A note on the numbers in this article
The thirteen‑week horizon, the collection assumptions, and the example downside case are illustrative conventions rather than rules. The right horizon depends on your collection cycle and your obligations. Build the first version on your own history, then correct it weekly.
A cash forecast is not a sophisticated instrument. It is a list of what is coming in, a list of what is going out, both dated honestly, and the discipline to update it weekly. What it buys is the ability to decide early and with information. If you want help building one on your actual numbers, schedule a consultation.
Frequently asked questions
Because profit and cash measure different things. Your profit and loss statement recognizes revenue when it is earned, while the money arrives when clients pay. Several large cash movements never appear on it at all: loan principal, inventory, owner distributions, estimated taxes, and equipment bought outright. Growth widens the gap.
Thirteen weeks, updated weekly, for operational decisions. That horizon covers a full quarter including payroll cycles, a collection cycle, and a quarterly tax payment, while staying close enough that your estimates hold up. Pair it with an annual model built on monthly assumptions for hiring, capital spending, and financing decisions.
A spreadsheet and current books. One sheet with weeks as columns, expected collections and payments as rows, and a running balance at the bottom is enough for most businesses. Forecasting software saves time once the habit exists, but it will not rescue a forecast built on unreconciled accounts.
