You need to enable JavaScript to run this app.
By Travis Yule — CEO & Founder, Full Send Funding
A 13-week forecast has one output, the low point: a profitable 20-employee contractor bottoms at −$30,472 in week 9, and that gap sizes and dates the facility.
In one sentence: A 13-week cash flow forecast is not a rescue document but a quarter of cash movements on the days they clear, whose single output — the low point, its week, its length and its cause — sizes and dates any facility, and which a profitable, growing company needs at least as much as a troubled one.
A 13-week cash flow forecast is a week-by-week projection of the money that will enter and leave your operating account over the next quarter, built from the dates cash actually moves rather than the dates revenue is earned. It answers three questions no profit and loss statement can: how low the bank balance will go, in which week, and whether the business gets through that week on its own. For a 20-employee company the whole thing fits on one page, takes an afternoon to build the first time and an hour a week to keep alive, and it is the single document that turns "we might be tight in November" into "we are $30,472 short in the week of November 30, and here is why."
The usual framing gets the audience wrong. The 13-week forecast is taught as a restructuring tool — the document a turnaround advisor builds for a company in trouble, and the one a court or a lender asks to see when receipts and disbursements have to be reported week by week. That lineage is real, and it is why the tool is so well designed: it was refined by people who needed to know, to the week, whether a company could make payroll. But the design does not care whether the company is in trouble. A profitable, growing contractor with a slow-paying customer and a big equipment deposit has exactly the shape of problem the forecast was built to expose, and the worked example below is one of those, not a rescue.
The other thing the usual framing gets wrong is scale. A restructuring forecast has forty line items and a team maintaining it. A 20-employee company needs about a dozen lines, and most of them are known to the week already — payroll, rent, the truck note — so the work is in the two or three lines that are not: collections, big supplier payments, and anything that happens once. This article gives the full template as a table, builds every line, works a named company through thirteen weeks with real arithmetic, shows what moves the low point, and explains what we actually read when a business brings its forecast to us.
Three documents describe a business's money, and the forecast is the only one that is about the future and the bank account at the same time.
The profit and loss statement records revenue when it is earned and expense when it is incurred. It is the right way to measure whether the business made money, and the wrong way to know whether it can pay anyone on Friday, because an invoice issued in week 1 is revenue in week 1 and cash in week 8. A business can show its best quarter ever on the P&L and bounce a payroll tax deposit in the middle of it. In the Federal Reserve's Small Business Credit Survey, more than half of employer firms named paying operating expenses (56%) or uneven cash flow (51%) as a financial challenge in the 2024 survey year [1] — and those are not two separate problems. Uneven cash flow is what makes paying operating expenses hard for a business that, on its P&L, can afford them.
The bank statement is the record of what actually happened, day by day, produced by a third party. It is the primary document in working capital underwriting for exactly that reason. But it only looks backwards.
The 13-week cash flow forecast takes the bank statement's discipline — cash in, cash out, on the day it moves — and points it forward one quarter. Thirteen weeks is not arbitrary. It is long enough to contain a full receivables cycle for almost any business, a quarterly tax payment, an insurance renewal and the three-payroll month that biweekly payroll produces twice a year; and it is short enough that the numbers in it are estimates rather than guesses. Beyond thirteen weeks, the forecast stops being a cash tool and becomes a budget, which is a different instrument with different uses.
The SBA's own guidance to owners puts the cash flow projection at the centre of working capital planning, and says to build it from the historic cash flow statements the business already has [2]. That is the right instruction. The forecast is not a plan for what you would like to happen. It is a projection of what will happen if the pattern in your last three months of bank statements continues, adjusted for the specific things you know are coming.
Everything in the forecast is on a receipts and disbursements basis, and getting this one idea right is most of the skill.
A receipt is money that becomes available in the operating account. Not an invoice issued, not a job completed, not a card transaction at the point of sale — the day the deposit posts and clears. For a card receipt that is a day or two after the sale; for a customer check it is the day it clears, not the day it is mailed; for an ACH from a commercial customer it is the day their accounts payable run executes, which their terms describe and their behaviour predicts.
A disbursement is money that leaves the account. Not an invoice received, not an expense accrued — the day the check clears or the debit posts. A supplier invoice with net-30 terms is a disbursement in the week you pay it, and if you habitually pay on day 38, that is the week. Payroll is two disbursements, not one: net pay to employees on payday, and the tax deposit — the withholdings plus the employer's share of Social Security and Medicare tax, 6.2% and 1.45% of wages respectively [3] — a few days later, on the schedule the IRS assigns.
The consequences of that basis are what make the forecast useful:
The distance between the two bases is measurable. It is the cash conversion cycle — the days from paying for work to being paid for it, net of the credit suppliers extend — and the 13-week forecast is, in effect, that cycle laid out on a calendar with dollar amounts attached.
The template below is the whole forecast. Thirteen columns for the weeks, one row per line, and a beginning-to-ending cash chain that carries the balance forward. Every line names its basis — the document or rule it is forecast from — because a line without a stated basis is a guess, and the value of the forecast is that a reader can ask "where did this number come from" about any cell and get an answer.
The firm's working template, cut to the lines a business of roughly twenty employees needs. Split a line when two items move on different clocks; merge lines that fall on the same date. Lines are the same across industries; only the bases change.
Two rules about the lines.
Split a line when the timing differs, not when the category does. Collections from invoiced customers and cash paid at the point of service are both revenue on the P&L, and they belong on separate lines here because one arrives seven weeks after the work and the other arrives the same day. Rent and insurance are both fixed costs, and they can share a line because both fall on a known date. The test is whether the two things move on different clocks.
Anything that happens once gets its own cell and a note. An equipment deposit, an insurance renewal, an annual software bill, a bonus run, a tax payment. These are not hard to forecast — the dates and amounts are usually known months out — but they are easy to omit, because they are absent from the weekly rhythm the rest of the forecast is built on. Most of the low points we see in practice were caused by a known, one-time disbursement that nobody had put on a calendar.
Beginning cash for week 1 is the cleared balance in the operating account on Monday morning — cleared, not book, because checks you have written and not yet cleared are disbursements that will hit inside the window. Every later week's beginning cash is the previous week's ending cash. This chain is the whole mechanism: an error anywhere carries forward to every week after it, which is a feature, because it is how a small mistake in week 2 becomes a visible one by week 6.
Collections from receivables is the hardest line and the one the forecast exists for. Build it from the aged receivables report, one invoice at a time for anything large, placing each in the week its customer will actually pay — not the week its terms say. A customer on net 30 who has paid on day 45 for two years will pay on day 45. Anything more than 90 days old goes in with a value of zero until it arrives; carrying it as a receipt because it is owed is how a forecast lies. Then add the invoices you will issue inside the window, placed at your measured collection lag: if you invoice $60,000 a week and collect in seven weeks on average, week 8's collections include week 1's billing.
Government customers deserve a note, because their clock is fixed by rule rather than habit. Under the federal prompt payment regulation, payment on a proper invoice is due 30 days after it is received where the contract sets no other date, and an improper invoice must be returned within 7 days — after which the clock starts again when the corrected one arrives [4]. A federal receivable is therefore among the most forecastable lines a business has, and among the most damaging when a paperwork error restarts it. The mechanics are in the government contractor payment cycle.
Cash and card receipts are the point-of-sale line: service calls paid on completion, retail, deposits taken on order. Forecast from the trailing thirteen weeks' average, adjusted for anything you know — last year's November against this year's October, a price change, a closure. Where the business is seasonal, use last year's same weeks scaled to this year's run rate, not the trailing average, which will be wrong in exactly the weeks that matter.
Net payroll is the easiest line and the most important. Payday dates are known for the whole window. The amount is gross wages less withholdings; for a business with stable headcount it is the last payroll, adjusted for known changes. Biweekly payroll puts seven paydays into half of all 13-week windows and six into the other half — thirteen weeks is six and a half pay periods — and the seven-payday windows are the ones that hurt.
Payroll tax deposits are a separate line because they land on a different day. The deposit is everything withheld from employees plus the employer's share, and it is due on the IRS's monthly or semiweekly schedule. A semiweekly depositor with a Friday payday deposits by the following Wednesday, so the deposit for an odd-week payday lands in the even week after it. This is the disbursement a business is most tempted to slip when cash is tight, and the one it must never slip.
Materials, inventory and subcontractors is forecast from purchase orders and supplier terms. State the rule you use. The rule in the example below is that Northway buys about thirty days ahead of billing on net-30 terms, so its materials disbursement in any week is 55% of that week's billing — the two clocks cancel. Your rule will differ; the point is that it is written down and can be checked against the last quarter's bank activity.
Rent, insurance, leases and fixed costs fall on known dates in known amounts. Put them in the week they clear. Monthly items appear three times in a 13-week window, sometimes four; quarterly items appear once or not at all, and the ones that fall just outside the window are worth a note at the bottom so the next forecast does not meet them by surprise.
Debt service is every scheduled payment on every facility: truck notes, equipment loans, a term loan, an advance's daily or weekly remittance. If you carry a revenue-based position with a daily debit, it is 21 or so entries a month on the bank statement and one line here — the weekly total. The reason to list every facility is that total debt service against receipts is the ratio a funder reads first, and the forecast should show it plainly.
One-time items — deposits, renewals, tax payments, distributions, a bonus — each get their own cell. Then the arithmetic: total receipts, total disbursements, net cash flow for the week, and ending cash.
Northway Mechanical is a commercial HVAC service and installation contractor with 20 employees — 16 in the field, 4 in the office — and about $4 million of annual revenue. Its forecast window is the thirteen weeks beginning Monday, October 5. It bills installation and maintenance-contract work at $60,000 a week and collects in seven weeks; it takes $14,000 a week in service calls paid on completion, rising to $18,000 from week 7 as heating season starts; and its opening receivables of $420,000 are placed week by week from the aging report, with $20,000 over 90 days carried at zero. Two things are happening inside the window. A new maintenance contract and the install backlog take billing to $72,000 a week from week 5, and a rooftop-unit replacement at a federal facility requires a $48,000 equipment deposit in week 2 and a $22,000 balance on delivery in week 4, is invoiced at $95,000 at the end of week 6, and is paid on the 30-day federal clock in week 11.
Every input is in the source note under the table, and every cell is arithmetic on those inputs.
An illustrative company with stated inputs; every cell is arithmetic on them. Opening cash $48,000. Collections = opening receivables placed by week from the aging report ($55,000, $70,000, $75,000, $65,000, $55,000, $45,000 and $35,000 in weeks 1 to 7 — $400,000 of $420,000, with $20,000 over 90 days carried at zero) + billing collected seven weeks after invoice (billing is $60,000 a week in weeks 1 to 4 and $72,000 a week from week 5, so weeks 8 to 11 collect $60,000 and weeks 12 to 13 collect $72,000) + the $95,000 federal invoice issued at the end of week 6 and paid on the 30-day clock in week 11. Service receipts $14,000 a week in weeks 1 to 6 and $18,000 from week 7. Biweekly gross payroll $52,000 paid on the Friday of odd weeks: net pay is 78% of gross ($40,560, with 22% withheld) and the tax deposit — the 22% withheld plus the employer's 7.65% share — is $15,418, deposited the Wednesday after each payday, in even weeks. Materials and subcontractors are 55% of that week's billing ($33,000, then $39,600), plus the federal job's equipment deposit of $48,000 in week 2 and $22,000 balance in week 4. Rent and fixed costs $23,500 on the first business day of November (week 5) and December (week 9); the insurance renewal $14,000 in week 7. Debt service: a truck note of $2,400 due on the 15th (weeks 2, 7, 11) and an equipment note of $3,100 due on the 20th (weeks 3, 7, 12). Ending cash = beginning cash + total receipts − total disbursements, carried into the next week.
The quarter is a good one. Receipts total $1,089,000 against $1,012,328 of disbursements, so the business ends the window with $76,672 more cash than it started with, and it never had a losing month on its P&L. It is also below zero in the operating account for four consecutive weeks, from week 7 to week 10, with a low point of −$30,472 in week 9.
Weeks 5, 7 and 9 are payroll weeks that each also carry a monthly or one-time item while collections are at their thinnest; the federal receivable in week 11 refills the account.
The net cash flow column of the 13-week table: total receipts less total disbursements in each week, from the inputs stated under that table.
Read the weekly net line and the cause is visible. Weeks 5, 7 and 9 are all payroll weeks, and each also carries a monthly item — rent and fixed costs in weeks 5 and 9, the insurance renewal and two loan payments in week 7 — while collections through weeks 5 to 7 are the thinnest of the window, because the opening receivables have mostly been collected and the new billing is still seven weeks from landing. The growth ramp in week 5 makes it worse: materials for the higher billing are paid immediately, and the receipts for it do not arrive until week 12. The federal job is the same shape in miniature — $70,000 out in weeks 2 and 4, $95,000 in during week 11.
None of that is a problem with the business. It is what growth and a slow-paying customer look like on a calendar, and every one of those dollars comes back by week 11. The forecast's job is to say, nine weeks in advance, that the business needs about $30,000 it will not have for about a month.
The low point is the lowest ending cash balance in the window and the week it occurs. It is the only number in the forecast that a reader who is not going to read the forecast needs, and it carries four pieces of information.
How much. The low point is the size of the gap. It is not the amount of funding needed, for two reasons. First, no business should run its operating account to zero, so the requirement is measured against a minimum operating balance — the cash the business decides never to go below, enough for a tax deposit, a repair or one late customer. Northway sets its floor at $25,000. Second, the forecast is an estimate, and the requirement should be sized against the forecast's error, which is the subject of the next section. Northway's requirement is its floor less its low point: $25,000 − (−$30,472) = $55,472.
When. The week of the low point sets the deadline. Money that arrives after week 9 does not help week 9, and money that arrives in week 9 is late, because the disbursements in that week clear through the week. Whatever solves the gap has to be in the account before the first negative week, which for Northway is week 7 — so the decision has to be made by week 5 or 6. That is the real reason to build the forecast now rather than in week 6: it converts a Friday emergency into a decision with a month's notice.
For how long. The number of weeks below zero is the term of the problem. Northway's is four weeks, which says the gap is a timing gap — the money exists and arrives later than the expense — and that a facility repaid from week 11 onward is the right shape. A forecast whose ending cash falls and stays down is describing something else, and a facility does not fix it; that case is at the end of this article.
Why. The line that causes the low point is the line that fixes it cheapest. Northway's low point is made of three things — a growth ramp, a federal receivable and an insurance renewal — and each has a different remedy. The ramp could be financed; the federal invoice could be issued a week earlier by finishing the paperwork the day the units are commissioned; the renewal could have been paid in monthly instalments. The forecast does not choose. It makes the choice visible.
A forecast is only as good as its worst assumption, and the assumption that matters is the one the low point is most sensitive to. The way to find it is to change one input at a time and watch the low point move. Five variations on Northway's forecast:
Each row re-runs the 13-week table with exactly the one change described and reads off the lowest ending cash, its week, the count of weeks whose ending cash is below zero, and week 13's ending cash. All other inputs are as stated under the 13-week table.
Three findings, in order of how often we see them in practice.
Slow collections are the sensitivity that matters. If every opening receivable arrives 10% short inside the window — the customers pay, a little later than the aging report predicted — the low point more than doubles to −$70,472, the business is below zero for six weeks, and the window ends $40,000 poorer. Nothing about the business changed except the week its money arrived. That is why the collections line is built one invoice at a time, and why the forecast is re-run every week against what actually came in.
Growth causes the hole. Hold billing at $60,000 instead of stepping up to $72,000 in week 5 and the low point shrinks to −$7,994, for a single week. The business would end the quarter with $35,400 more cash by growing less — because the materials for the higher billing are paid in the week the work is done and the receipts for it arrive seven weeks later. Every added dollar of weekly revenue costs 55 cents of materials now and returns a dollar seven weeks later, and a business that does not fund that gap is choosing to stay smaller. Growth consumes cash before it produces it, and the forecast is where the amount becomes visible.
A slipped federal invoice does not move the low point; it lengthens it. If the federal facility returns Northway's invoice as improper and the corrected one is paid in week 13, the low point stays at −$30,472 in week 9 — the money was never in that week anyway — but the business is now below zero for five weeks instead of four. Small errors in the near weeks change the depth of the low point; errors in the far weeks change its length.
The fifth row is the control. Had Northway opened the window with $100,000 instead of $48,000, there would have been no low point at all — the same forecast bottoms at +$21,528 — which is the plainest statement of what a cash reserve is for. The business did not have it, because the reserve was spent on last quarter's growth, which is the same finding as the third row.
We do not require a 13-week forecast. Most of what we fund is decided from three months of bank statements and an application, and a business under $10,000 a month in revenue or under three months old has nothing to forecast from yet. But a business that brings one is read differently, and it is worth knowing how.
The most common reason employer firms seek financing is the one Northway has: in the Federal Reserve’s 2026 report on employer firms, 56% of those that sought financing did so to meet operating expenses [5], and a forecast is the difference between a request that names the week and one that names a round number.
The first thing read is the low point against the request. A forecast showing a −$30,472 low point in week 9 and a request for $60,000 is a coherent file: the requirement is stated, the floor is stated, and the number asked for is that sum, rounded up. A forecast with the same low point and a request for $200,000 raises a question the forecast has to answer — what is the other $140,000 for — and if the answer is not in the forecast, the forecast is not the basis of the request. That is not disqualifying. It is just a different conversation, about a different use of money, and it is better to have it openly.
The second thing read is the recovery. Ending cash at week 13 is $124,672, up from $48,000. That says the gap is timing, the money exists, and a facility drawn in week 6 can be repaid from week 11. In underwriting terms it answers the first of the two questions — can the business afford the payment — with a date. The federal receivable is especially legible: it is a fixed clock, on a customer that pays, and the forecast shows the exact week.
The third thing read is the debt service line. Northway's is $5,500 a month on two notes — $1,269 a week against $83,769 a week of receipts, or 1.5% — and a forecast that shows every existing facility plainly is a forecast a reader trusts. A forecast that omits one is found out by the bank statements, and the omission costs more than the position would have. Existing positions are the most common reason for a decline, and the forecast is the natural place to disclose them.
The fourth thing read is whether the forecast matches the statements. A forecast projecting $69,000 of receipts in week 1 is checked against what the last thirteen weeks of deposits actually averaged. If the statements show $55,000 a week and the forecast shows $85,000, the forecast is a plan, not a projection, and it is read as one.
What the forecast changes about the structure of an offer is the shape. Northway's requirement is $55,472, needed by week 6 and repayable from week 11. That is a six-week problem, and the right instrument for it is a line of credit drawn once and repaid when the federal receivable lands, or a short working capital facility on a term matched to the gap — not a three-year term loan for a six-week gap, and not a daily remittance against a business paid in weekly and monthly lumps. The figure below shows the forecast with a $60,000 draw in week 6 and $30,000 repaid in each of weeks 11 and 12, before the cost of the money.
The forecast alone is below zero from week 7 to week 10 with a low point of −$30,472 in week 9; a $60,000 draw in week 6 keeps the balance above the $25,000 floor and is repaid from the week-11 federal receivable, before the cost of the money.
Series 1 is the ending cash column of the 13-week table. Series 2 adds a $60,000 receipt in week 6 and a $30,000 disbursement in each of weeks 11 and 12, with every other input unchanged and the facility's cost excluded (shown separately in the prose: $60,000 × 15% × 6 ÷ 52 ≈ $1,038 at a 15% annualized cost).
The cost sits on top of that line and is a separate calculation. Six weeks on $60,000 at, for example, a 15% annualized cost — inside our published cost of capital of 4.5% to 45%, where the position depends on the file and the term — is $60,000 × 15% × 6 ÷ 52, about $1,038. Against a $95,000 federal receivable and a quarter that ends $76,672 up, it is a rounding error; against a business whose forecast does not recover, no price is cheap. The comparison between instruments, and the arithmetic of a factor against a rate, is worked in APR, factor rate and the math nobody shows you. Our own sizing rule — 80% to 150% of monthly revenue — is what the calculator applies, and Northway's forecast is the reason the answer should be closer to the bottom of that range than the top: it needs $60,000, not $300,000.
A 13-week forecast built once is a snapshot that is wrong by week 3. The instrument is the weekly re-forecast, and it takes an hour.
Over a quarter this produces something more valuable than any single forecast: a measured record of how good the assumptions are. A business whose collections variance runs −$5,000 a week for eight weeks has learned that its collection lag is eight weeks, not seven, and its next forecast is built on that. That measured error is also the number to size the requirement against: a business whose forecasts run $10,000 optimistic should carry that $10,000 in its floor.
Two disciplines hold it together. The forecast is built from the bank, never from the accounting system, because the accounting system records what should happen and the bank records what did. And it is maintained by whoever pays the bills, because the person deciding what clears on Thursday is the person who knows.
The 13-week forecast answers a timing question, and it is the wrong instrument when the question is not about timing.
A structural loss. If the ending cash line falls across the window and does not recover — each week's net is negative in the ordinary run of business, with no one-time item and no receivable coming — the forecast is describing a structural loss: the business spends more than it makes in a normal month. The forecast is still worth building, because it shows the date the account runs dry, but it is a measurement of the problem, not a plan for it, and borrowing against it adds a payment to a business that cannot make the ones it has.
A business with no receivables and no lumps. A café paid at the till, buying stock weekly from a distributor on seven-day terms, with weekly payroll and monthly rent, has a cash conversion cycle near zero and a bank balance that tracks its P&L to within a few days. Its forecast will be flat, its low point will be the rent week, and the instrument that matters more is the daily balance and a reserve equal to the largest fixed payment. Build a 13-week forecast once to prove that, then stop.
A gap longer than the window. Retainage held for a year, a twelve-month receivable on a government program, a seasonal business whose peak funds its trough — these are real gaps with real remedies, but they are longer than thirteen weeks, and a forecast that cannot see the recovery cannot tell you the term. The 13-week forecast will show a falling line and stop; the right tool is a monthly forecast covering the full cycle, and for the seasonal case the trough is the number to size against, not the week.
A forecast built to justify a number. If the collections line was placed at the terms rather than the behaviour, the 90-day receivables were carried at face value, and the one-time items were left off, the forecast will produce whatever low point the person building it wanted. We read those against the statements, and the gap is visible. A forecast that says the business needs $55,472 in week 6 is useful; one that says it needs $200,000 for reasons the lines do not show is a loan application wearing a spreadsheet.
A decision that has to be made this week. A business that is going to miss payroll on Friday needs a different first move, not a forecast. Build the forecast the week after — it is exactly the tool that stops the next one — but it is a planning instrument, and it is at its most valuable with a month of notice, which is why it is built before it is needed.
A 13-week cash flow forecast is a quarter of cash movements, by week, on the day they clear: receipts by the week the money lands, disbursements by the week it leaves, a beginning-to-ending chain that carries the balance forward, and one cell for every one-time item. Its output is the low point — how much, in which week, for how long, and which line caused it — and the funding requirement is the business's minimum operating balance less that low point, needed by the week before the first negative one. Northway Mechanical, a 20-employee contractor having a good quarter, is below zero for four weeks with a low point of −$30,472 in week 9, needs $55,472 by week 6, and repays it in weeks 11 and 12 from a $95,000 federal receivable; its low point is most sensitive to slow collections, is caused by growth, and is lengthened rather than deepened by a late invoice. A funder reads the forecast for the low point against the request, the recovery, the debt service line, and whether it matches the bank statements — and a forecast that passes those four reads shapes an offer to the gap it shows. It is the wrong tool for a structural loss, a business with no timing gap, a gap longer than a quarter, or a decision due this week.
If you want to check your own low point against what the business would qualify for, the calculator applies the same sizing rule we do — or call 518-312-0382 with the forecast open, and we will read it with you.
Northway Mechanical is a constructed company, not a client. Every figure in its 13-week table is arithmetic on the inputs stated in the table's source note — opening cash, the placement of opening receivables by week, weekly billing and a seven-week collection lag, service receipts, a $52,000 biweekly gross payroll split into net pay and a tax deposit, materials at 55% of billing, the federal job's equipment payments and 30-day receipt, fixed costs, an insurance renewal and two loan schedules — and the net cash flow figure, the funded-scenario figure and the sensitivity table are re-runs of the same model with one input changed. The site's arithmetic audit recomputes every published value from those inputs before publication; any input can be replaced with a reader's own. The 15% annualized cost used to price the six-week draw is an example inside the firm's published range, not a quote.
Three external figures are quoted and none is a computation of ours. The shares of employer firms naming operating expenses and uneven cash flow as challenges, and the share that sought financing to meet operating expenses, are from the Federal Reserve Banks' 2025 and 2026 Reports on Employer Firms — nationwide convenience samples of firms with fewer than 500 employees, offered as prevalence, not cause. The 30-day and 7-day federal payment clocks are stated from the prompt payment regulation cited, and the employer payroll tax shares from the IRS employer's tax guide. The description of the forecast's origin in restructuring practice, the IRS deposit-schedule timing used for the example, and the four-step order in which a forecast is read are stated from the firm's own practice and from general professional usage, and no survey of either is claimed.
A week-by-week projection of the cash entering and leaving the operating account over the next thirteen weeks, built from the dates money actually clears rather than when revenue is earned or expense incurred. Its output is the low point — the lowest projected balance, its week, how long the balance stays down and which line caused it — and it is re-forecast every week against actuals.
Thirteen weeks is one quarter: long enough to contain a full receivables cycle for almost any business, a quarterly tax payment, an insurance renewal and the three-payday month that biweekly payroll produces, and short enough that the numbers are estimates rather than guesses. Beyond that horizon the document becomes a budget, which is a different tool.
The P&L records revenue when earned and expense when incurred; the forecast records cash when it clears. An invoice issued in week 1 is revenue in week 1 and cash in week 8. Non-cash items such as depreciation vanish from the forecast, while loan principal, equipment deposits, tax payments and owner distributions — absent from the P&L — appear, and those are usually the lines that cause the low point.
From the aged receivables report, one invoice at a time for anything large, placed in the week the customer actually pays rather than the week its terms say. Carry anything over 90 days at zero until it arrives. Then add invoices you will issue inside the window at your measured collection lag — if you collect in seven weeks, week 8 includes week 1's billing.
The minimum operating balance you decide never to go below, less the low point, plus the measured error of your own forecasts. A −$30,472 low point against a $25,000 floor is a $55,472 requirement. It has to be in the account before the first negative week, so the decision date is a week or two earlier than that.
We do not; most working capital decisions are made from three months of bank statements and an application. A business that brings one is read for the low point against the amount requested, the recovery by week 13, the debt service line and whether the receipts match the statements — and a forecast that passes those reads shapes the offer to the gap it shows rather than to a round number.
When the ending cash line falls and does not recover, which describes a structural loss rather than a timing gap; when the business has no receivables and no lumpy payments, so its balance tracks its P&L; when the gap is longer than a quarter, such as retainage or a seasonal trough; and when the decision is due this week, which needs a different first move.
Travis Yule founded Full Send Funding in 2021 and leads it from Middle Grove, New York. He writes about working capital from the underwriting side of the table — what the numbers actually have to say before a business gets funded.