The customer wants net 60 instead of net 30, and they were already paying at 45. Or their accounts-payable department is asking whether you would take two percent off for payment in ten days. The work is the same, the price is the same, and somehow you are the one lending them money. "What do these payment terms actually cost me" is the question, and the answers you keep getting are a rule of thumb about what terms are standard in your trade, a factoring quote, or a shrug that this is the cost of doing business. None of those is your number. What the terms cost is set by four things you can read off documents already in your hands: the invoice, how long this customer really takes to pay, what your money costs, and what is left on the invoice after the direct cost of the work. This page runs that cost and turns it into a decision, and the free payment terms calculator runs the same arithmetic on your numbers with every field blank until you fill it.
This page is the payment-terms decision from the seller's side of the desk, in the order the tests have to run. First, the days the customer actually takes against the days on the paper, because the cost runs on behaviour, not on the contract number. Then the cost of the extension, per invoice and per year, at your own cost of money. Then the discount, if one is on the table, against the same cost of money. Then the test that decides it: whether the contribution on the invoice, less the cost of the wait, still clears the floor you set for the work to be worth doing. It ends in one of four answers -- accept the terms with the cost stated beside them, negotiate shorter with the exact net days named, price the terms into the quote with the exact percent named, or decline the contract -- and, when a discount is on the table, a separate answer on the discount.
The Real Problem: The Cost of Terms Is Set by How the Customer Pays, Not by the Number on the Invoice
Two mistakes sit under almost every payment-terms decision, and both are made before the cost is ever computed.
The owner does the cost math on the number printed on the invoice. Net 60 goes into the spreadsheet as 60. Then the money lands on day 75 because that is how this customer pays everybody, and the cost of net 60 from this customer was the cost of 75 days, not 60. Stated terms are not behaviour. The receivables aging for this customer, or the last three to five invoices against the bank statement, is the input that makes the cost true; the contract number is what you use only when you have nothing else, and the calculator says so on every figure when that is what it is running on.
The owner treats the terms as a cost of doing business and never asks what the same dollars would earn or cost on their own line. Carrying an invoice is lending. The loan is the invoice amount, the term is the days you wait, and the rate is what your money costs you: the rate on the line you would draw to cover the wait, or what the cash would earn if you had it. Once it is written that way it has a price, per invoice and per year, and a price can be compared with the price of the alternatives: shorter terms, a discount, a higher number on the quote, or walking away.
There is a quieter third one. The terms get judged by whether they are usual. Whether net 60 is common in your trade has nothing to do with whether this invoice, from this customer, at your cost of money, still leaves what the work has to leave you. Nothing on this page asks whether your terms are usual, and nothing on it holds a figure for what anyone else waits.
The question arrives in four ways, and each one changes which test leads. The customer is extending the terms: "They sent a new vendor agreement. Net 60 now, and they said it is company policy. Do I sign it?" That leads with the cost of the extension and the contribution test. The customer already pays late against the terms they have: "It says net 30. I have never once been paid inside 45 days by these people. What is that actually costing me, and is it worth the fight?" That leads with the gap between the paper and the behaviour and the capital tied up. Someone wants a discount for paying faster: "Their AP department asked if I would take two percent off for payment in ten days. Two percent sounds small. Is it?" That leads with the discount's effective annual rate against your own money, and with whether this customer has taken a discount and still paid late. The terms are on a quote you have not sent yet: "I am bidding a job for a customer I know pays slow. Should I put something in the number for that, and how much?" That leads with pricing the terms in. They are four doors into one diagnosis, not four products.
What Do Payment Terms Cost? The Six Questions That Decide It
Every input is something you read off a document already in your hands. Nothing is filled in for you: no field carries a default, a placeholder, a hint about what is usual, or a slider set to a midpoint, and a blank stays blank. Where the calculator cannot run a test without a number you have left out, it reports that test as not run and says which number would unlock it; it never falls back to a range. The right-hand column names who in your own finance setup can produce the number if you cannot. It is a fence, not a recommendation: no answer in that column is a provider, and no cell names a lender, a factor, a software vendor or a fractional executive.
| Question, in your words | Where the number comes from | Who can pull it | If it is blank |
|---|---|---|---|
| 1. How big is one invoice to this customer, the size you usually send, and how often do you bill them? | The last three invoices; the billing calendar (weekly, every two weeks, monthly, per milestone) | You, from the invoice file; the bookkeeper, from the sales ledger | Without the invoice amount the calculation does not run at all. Without the billing cadence, per-invoice results only: the capital tied up, the exposure if they stop paying, and every per-year figure are not tested. |
| 2. What are the terms on the invoice today, as written? | The invoice, the purchase order, the vendor agreement in hand ("due on receipt" is zero, entered by you) | You; the contract | The calculation stops: there is no comparison without today's terms. |
| 3. What is on the table: new terms, and/or a discount for paying faster? And has this customer ever taken a discount and still paid late? | The proposed agreement or the AP email; the discount as a percent and its window in days; your own experience with this customer for the last part | You; the contract; the aging report for the last part | No proposed terms: the cost runs on today's terms and the extension test is not run. No discount: the discount tests are not run. A discount entered as zero is a value, not a blank, and is scored as one. Late-payment history not stated: the abuse test is not run. |
| 4. How long does this customer actually take to pay, invoice date to money in the bank? | The receivables aging for this customer, or the last three to five invoices against the bank statement. Not the terms. | You, from the statements; the bookkeeper, from the aging | The calculator falls back to the terms as written, flags every figure as running on the contract number, and says so on the output, which is why the cash never arrives when the spreadsheet says it will. |
| 5. What does money cost you, and where does that number come from? | The rate on your line of credit if you would draw to cover the wait; otherwise what the cash would earn if you had it; otherwise your own figure. You name which. | You; the line agreement or the last bank statement | Cost of the terms in days and dollars of exposure only. No dollar cost of carrying is computed, the contribution test is not run, and no verdict is reached. The page never supplies a rate and never shows one beside this field. |
| 6. On one of these invoices, what is left after the direct cost of doing the work, and what does an invoice have to leave you for the work to be worth doing? | The bid worksheet or job cost for a recent invoice to this customer; the second number is your own floor | You; the bookkeeper for the direct cost if job costing exists. If it does not, that is the job-costing question, not this one. | Without the contribution, the deciding test is not run, no verdict is reached, and the one question is named. Without the floor, the floor is zero and the test only asks whether anything is left at all. |
| The bound: the largest price increase you believe this customer would accept, as a percent | Your own judgement of this customer; nothing on this page suggests a figure | You | The price-it-in lever is not decided. If it is the only lever that would cure, the page reaches no verdict and asks for this one number rather than assuming an increase you have not said you can get. |
What is deliberately not an input: your industry, your revenue band, your cash balance, your fixed outflow, a reserve floor, the customer's name. The cash trough of a large contract is a different decision, and the deposit question is another; this page routes to them rather than re-asking.
Net 30 Payment Terms vs How They Actually Pay: The Two Runs
The first thing the calculator computes is the slippage: the days this customer actually takes to pay, less the net days on the paper. If the terms say net 30 and the money lands on day 40, the slippage is ten days. Under proposed terms, the calculator assumes the customer slips by the same amount, and says so on every figure it computes that way. The alternative reading, that behaviour never changes whatever the terms say, would make every extension free, and it is not the reading this page takes. The headline days are the proposed terms, or today's terms if nothing is proposed, plus that slippage.
The carrying cost is then the invoice amount, times your cost of money, times the headline days, over 365; and per year, that figure times the invoices you send. Everything else on the page is a comparison against that cost: what today's terms and behaviour cost, what the extension adds, what the discount would buy, and what is left on the invoice after the wait. The difference between the run on the paper and the run on the behaviour is reported as its own number, because it is the number the owner has usually never seen.
The Seven Things the Calculator Checks, in a Fixed Order
Each check returns one of five states: absent (it ran on complete inputs and the condition did not hold), present (it ran and the condition held), undecidable (a required input was blank), not tested (an input was missing and the calculator says so rather than guessing), and a fifth state that no check on this page can enter, because no formula here depends on a benchmark.
| Check | In your words | What decides it | What it feeds |
|---|---|---|---|
| Stated terms versus behaviour | "It says net 30. They pay at 45." | Days-to-pay above the net days on the paper | The headline runs on behaviour; the gap is the sentence this page exists for |
| Terms-extension cost | "Net 60 instead of net 30: what is the extra?" | Proposed net days above today's; the extra days at your cost of money, per invoice and per year | The contribution test; the negotiate-shorter ask |
| Unearned-discount risk | "They took the two percent and still paid at 50." | Your own experience with this customer, yes or no | Dominates the discount arithmetic: the discount becomes pure cost and is not granted |
| Discount economics | "Two percent for ten days: is that expensive?" | The discount's effective annual rate over the days it actually buys, against your own cost of money | The discount verdict; the take-the-discount lever |
| Contribution erosion | "After waiting that long for the money, is there anything left in the job?" | Contribution on the invoice, less the carrying cost at the headline days, against the floor you set | The verdict. This is the check that decides the terms |
| Capital tied up | "I am permanently lending them how much?" | The invoice amount times the headline days over the billing cadence, against the same on today's terms | A fact beside the verdict, not a verdict |
| Exposure if they stop paying | "If they stopped paying tomorrow, how much would be out there?" | The invoices outstanding at the worst instant, counting a payment that lands on the day the relationship ends as still out | A fact beside the verdict; the negotiate-shorter and price-in asks |
The last row's counting rule matters more than it looks. A payment that lands on the day the customer stops paying is not money you have; it is money that was supposed to arrive. The calculator counts it as still out. On terms where the days outstanding divide evenly by the billing cadence, that is one more invoice than the loose count gives, and it is the number the owner would actually be chasing.
The Four Answers, in the Order They Are Tested
The contribution test decides the terms. If the contribution on the invoice, less the carrying cost at the headline days, clears your floor, the answer is accept the terms, with the cost stated beside it: a cleared test is not a free one, and the carrying cost, the capital tied up and the exposure sit next to the verdict whatever it is. If it does not clear, three asks are tested in turn, each at the minimum that restores the floor, cheapest first, because a shorter-terms ask changes nothing about the price the customer agreed to while a price increase does. Negotiate shorter: the longest net days at which the work still clears, named exactly. Take or grant the discount on the table, only if it is cheaper than your own money and this customer has not abused one. Price the terms into the quote: the smallest price increase that restores the floor, named exactly, and only if it is inside the bound you set for what this customer would accept. The first that cures names the verdict. If none does, decline the contract.
There are exactly two decline triggers, and they are independent. The first is that the terms consume the contribution and no owner-side ask restores it. The second is that the invoice misses your floor before a single day of waiting is counted: contribution below the floor at pay-on-receipt. That is not a terms problem, the page says so, and it routes to the pricing question rather than negotiating terms on a job that does not pay at net zero.
The levers are grouped by what the calculation computed, never ranked by what the customer might agree to: cures alone, with the minimum ask; cures only in combination; cannot cure, with the residual; not available, when the discount is already too expensive or this customer has abused one; and not decided, when the price-in bound is blank. You can select a combination and see its joint result; the verdict itself is reached on single asks.
Three things stop the page short of a verdict, and each is one question you can answer from your own records: the cost of money is blank, the contribution is blank, or the price-in lever is the only one left and you have not bounded it. The page names the one question and stops. It does not guess.
When a discount is on the table, a separate verdict rides alongside whichever answer the terms get: grant, grant conditionally (the calculator could not test whether this customer has abused a discount, and says so), do not grant, or undecidable without your cost of money. The terms verdict answers the terms; the discount verdict answers the discount.
Worked example: the $40,000 invoice moved from net 30 to net 90 by a customer who pays at 40
The figures below are a synthetic input set used to show the arithmetic, not defaults, not anyone's benchmark, and not a suggestion of what your numbers look like. A shop bills one customer $40,000 a month on net 30. The customer pays at 40 days, so the slippage is ten. The customer's new vendor agreement says net 90. The owner would draw on a line at 15 percent to cover the wait. A recent invoice leaves $2,200 after the direct cost of the work, and the owner's own floor is $1,200 an invoice. The largest price increase the owner believes this customer would accept is 2 percent.
| Check | Today's terms and behaviour | Proposed terms, same slippage | State |
|---|---|---|---|
| Days the money is actually out | 40 (net 30 plus 10 days of slippage) | 100 (net 90 plus 10) | Headline 100 days |
| Stated terms versus behaviour | 10 days late against the paper; $164.38 per invoice at 15 percent | Present | |
| Terms-extension cost | 60 more days; $986.30 per invoice; $12,000.00 per year at twelve invoices | Present | |
| Discount and unearned-discount risk | No discount on the table | Not tested | |
| Carrying cost at the headline days | $1,643.84 per invoice ($40,000 times 15 percent times 100 over 365) | -- | |
| Contribution after the wait | $2,200.00 less $1,643.84 = $556.16, against a floor of $1,200.00 | Present: the terms consume the job | |
| Capital tied up at all times | $40,000.00 | $133,333.33 | Present |
| Exposure if they stop paying | 2 invoices, $80,000.00 | 4 invoices, $160,000.00 | Present |
The terms fail the test, so the asks are run. Shorter terms cure it. The longest net days at which the floor still clears is net 50: at net 50 the carrying cost is $986.30 and the contribution after the wait is $1,213.70, above the floor; at net 51 it is $1,002.74 and $1,197.26, below it. Pricing it in would also cure it, at a 1.6786 percent increase ($671.43 an invoice), which is inside the 2 percent bound, but it is tested second and is not needed. There is no discount on the table. The verdict is negotiate shorter, at net 50 or better, with the cost of the extension, the capital tied up and the exposure stated beside it, and the ask is the owner's to make.
A second set shows the discount side. A $20,000 monthly invoice on net 60 from a customer who pays at 90, no new terms proposed, a 1 percent discount for payment within 10 days offered, this customer has never taken a discount and still paid late, a 12 percent line rate, $5,000 of contribution against a $2,000 floor. The discount buys 80 days (90 less the 10-day window); its effective annual rate is 4.6086 percent, below the 12 percent the owner's money costs, so the discount is cheaper than the wait and is worth $326.03 an invoice net of the $200.00 given up. The contribution after the 90-day wait is $4,408.22 against the $2,000 floor ($4,734.25 on the discount path), so the terms clear: accept the terms; grant the discount. And on today's terms the exposure if they stop paying is three invoices, $60,000.00, not two: 60 days outstanding divides evenly by a 30-day cadence, and the payment landing on the day they stop is counted as still out.
What to Do With Each Answer
| Answer | What it means | What to do next |
|---|---|---|
| Accept the terms | The cost is real and stated, and the work still clears what it must | Sign, with the carrying cost, the capital tied up and the exposure in front of you. Put the invoices on the 13-week cash flow forecast on the days the money actually lands, not the days the paper says |
| Negotiate shorter | The work does not clear at these terms; it clears at the named net days or shorter | Make that ask, and only that ask; the calculator has already told you the longest terms you can live with. If the customer will not move, the price-in and decline answers are the next two rows |
| Price the terms into the quote | Shorter terms will not get there; a named percent on the quote does, and you believe the customer would take it | Put the percent in the number before you send it. If the quote is already out, this is the renegotiation, not a surcharge after the fact |
| Decline the contract | Nothing you can ask for restores the contribution, or the invoice missed your floor before any waiting was counted | If it is the second trigger, the terms were never the problem: job costing and busy but not profitable are the pages, because the price or the cost of the work is what has to move |
| No verdict yet | One of your own numbers is missing: the cost of money, the contribution, or the price-in bound | Produce it and run again. If the contribution cannot be produced from your records, the making-money-on-each-job diagnostic is the page for the cost side and are my books a mess for the rest. A bookkeeper who runs an aging report, or a controller or fractional finance person who sets up job costing, is one way to get those two numbers produced; it is never the answer this page gives |
| Do not grant the discount | The discount costs more than your money, or this customer has taken one and still paid late | Say no to the discount on its own arithmetic, whatever the terms verdict is. The two answers are separate and so are the conversations |
Where the Numbers Come From, and Which Ones Are Yours
Every number that moves the verdict is yours: the invoice, the cadence, the terms, the proposal, the discount, the days this customer actually takes, your cost of money and its source, the contribution and your floor, and the bound on what this customer would accept. The only constant in the calculation is a 365-day year. There is no benchmark, no typical anything, and no rate supplied.
- No benchmark figure of any kind. Not a typical days-to-pay, not a typical net days by industry, not a typical discount, not what businesses like yours wait. Published days-to-pay panels exist and are not shown here: they are drawn from populations that are not yours, and the two-run gap on this page is built from your own numbers without them.
- No cost of money is supplied or suggested. The federal Prompt Payment interest rate and the Current Value of Funds Rate cited in the sources are the rates the federal government applies to its own bills. They appear there with their dates as the record that the mechanism this page applies is written down in law, and nowhere near the field where you enter your own rate.
- No interpretation of contract terms, late-fee clauses, lien rights or state prompt-payment statutes. Many states carry prompt-payment provisions for private construction contracts; they are per state, per contract type and conditioned, and this page names none. It asks what your document says and scores the answer.
- No prediction of whether the customer will pay, only your own observed behaviour with them.
- No accounting. If you cannot supply the days-to-pay or the contribution, the page reports which tests were not run and routes to the diagnostics that find out why.
- No factoring, no invoice financing, no collections service as a destination, and no fractional CFO. If you arrive holding a factor quote, the question underneath it is whether the receivable is any good, and that is the revenue-quality question, not this one.
- No pre-filled field, ever, and nothing any owner enters is shown to another.
When This Is Not the Right Starting Point
If the real question is whether to ask for a deposit or a split payment, that is a different decision, and the nearest place to start is the 13-week cash flow forecast on the dates the money actually moves. If this customer is not slow but stopped, 90 days past due and counting, the question is the quality of the receivable, and customer concentration and revenue quality with the revenue quality analyzer is the page. If the terms sit on one large contract with its own cost curve, progress billings and a cash trough, can I afford to deliver this contract runs that curve on the dates money moves and this page's terms cost is one input to it. If the invoice misses your floor before a day of waiting is counted, the terms were never the problem: job costing help for contractors and busy but not profitable are where the price and the cost of the work get fixed. If you are a contractor and the jobs look fine at bid and thin at the end, retainage and the final invoice are usually the leak, and why your jobs look profitable until the final invoice is the page for it. If the cash is short this month regardless of what the terms cost, the profitable-but-out-of-cash diagnostic sorts which drain is yours, and if the reason is that sales are outrunning the cash to fund them, growing too fast and out of cash names the mechanism. And if the question is how a discount you grant is treated on your return, that is your return preparer's, and the sources below cite where the treatment is written.
Frequently Asked Questions
What do net 30 or net 60 payment terms actually cost me?
The cost is your invoice amount, times your own cost of money, times the days you actually wait, over a 365-day year, per invoice; and that per invoice times the invoices you send in a year. The days you actually wait are not the number on the invoice: if the terms say net 30 and this customer pays at 45, the cost runs on 45. In the worked example on this page, a $40,000 invoice moved from net 30 to net 90 by a customer who already pays ten days late costs $1,643.84 to carry per invoice at a 15 percent line rate, against $164.38 on today's terms and behaviour. The calculator runs it on your invoice, your terms, your days-to-pay and your rate, and supplies none of them.
Should I accept net 60 payment terms?
Accept them if the contribution on the invoice, less the cost of carrying it for the days this customer will actually take, still clears the floor you set for the work to be worth doing. If it does not, three owner-side asks are tested in order at the minimum that restores the floor: shorter terms, the early-payment discount if one is on the table and it is cheaper than your own money, and a price increase inside the largest increase you believe this customer would accept. The first that cures is the answer. If none does, decline. The page never answers with a lender or a factor.
Is a 2 percent discount for paying in 10 days expensive?
It depends on two of your own numbers: how many days the discount actually buys, which is the days this customer really takes to pay less the discount window, and what your money costs. The discount's effective annual rate is the discount over one minus the discount, times 365, over the days bought. If that rate is above your own cost of money, the discount costs more than the wait and the verdict is do not grant it; if it is below, grant it. In the second worked example, a 1 percent discount for payment in 10 days from a customer who pays at 90 buys 80 days at an effective annual rate of 4.6086 percent against a 12 percent line rate, so it is granted. If this customer has taken a discount and still paid late, the discount is treated as pure cost and is not granted whatever the arithmetic says.
What if I do not know my cost of money?
Then the calculator reports the cost of the terms in days and dollars of exposure only: the days you wait, the capital tied up in this customer's invoices at all times, and what would be outstanding if they stopped paying. It does not compute a dollar cost of carrying, it does not run the contribution test, and it reaches no verdict; it names that one question and stops. The number is the rate on your line of credit if you would draw to cover the wait, what the cash would earn if you had it, or your own figure, and the page asks you to say which. It never supplies a rate and never shows one beside the field.
Does this tell me whether my terms are usual for my industry?
No. This page holds no typical days-to-pay, no typical net days by industry, no typical discount and no figure for what anyone else waits. Where you ask how your number compares, the answer is your number: the cost of the terms is set by how this customer pays and what your money costs, and neither of those is an industry figure. The one body of law that writes the mechanism down, the federal Prompt Payment rules, is cited in the sources as the record that the mechanism exists; its rates are the ones the federal government applies to its own bills and are not your number.
Sources
- 31 U.S.C. 3903, Prompt Payment Act, payment period: for federal agencies the required payment date is "30 days after a proper invoice for the amount due is received if a specific payment date is not established by contract." Read 2026-09-14 at the Office of the Law Revision Counsel; text current through laws in effect on September 12, 2026. Cited as the one body of law that fixes a default payment period, for federal agencies; it is context, not the population this page is written for, and no figure from it enters the calculation.
- 31 U.S.C. 3904, limitations on discount payments: an agency "may pay the discounted amount only if payment is made within the specified time," and the specified time "shall be determined from the date of the invoice." Read 2026-09-14; text current through laws in effect on September 12, 2026. Cited as the record that a discount window is counted from the invoice date, which is the convention this page adopts for the discount window you enter.
- 5 CFR 1315.7, Discounts (2025 annual edition): an agency "may take the discount if economically justified," and "the period for taking the discount is calculated from the date placed on the proper invoice by the vendor." Read 2026-09-14 at govinfo.gov. Cited as the provenance of the economic-justification test; the regulation refers its formula to the Treasury Financial Manual and states no number on the page.
- Bureau of the Fiscal Service, Prompt Payment interest rate: "The Prompt Payment interest rate for July 1 - December 31, 2026 is 4.75%." Page last updated June 30, 2026; read 2026-09-14. This is the interest penalty the federal government pays on its own late bills. It is not your cost of money and it is not used or displayed anywhere in the calculator.
- Bureau of the Fiscal Service, Current Value of Funds Rate: the rate for January 1 through December 31, 2026 is rendered on the page as "4:00%" (four percent), with an update note of January 2, 2026 and a page stamp of December 31, 2025; read 2026-09-14. It is the rate the federal government uses to evaluate cash discounts on its own payments. It is not your cost of money and it is not used or displayed anywhere in the calculator.
- Bureau of the Fiscal Service, Prompt Payment discount calculator, decision rule: "If the effective annual discount rate is larger than the current value of funds rate, accept the discount and pay early. If the effective annual discount rate is smaller than the current value of funds rate, reject the discount." Page last updated February 26, 2026; read 2026-09-14. This page applies the same comparison from the seller's chair with your own cost of money in place of the government's; no figure from the Treasury page enters it.
- Internal Revenue Service, Publication 538, Accounting Periods and Methods (revised January 2022; read 2026-09-14): "A cash discount is a reduction in the invoice or purchase price for paying within a prescribed time period. You can choose either to deduct cash discounts or include them in income, but you must treat them consistently from year to year." Cited for where the tax treatment of a discount is written; how it applies to your return is your return preparer's question.
- Every other number on this page is the worked example's own arithmetic or comes from your records. No days-to-pay, no net days, no discount, no rate and no contribution figure on this page is a benchmark, and the published days-to-pay panels the earlier research in this series tried to read were not retrievable on 2026-09-14 and are not shown by design.
Next Step
If the question in your head is what net 60, or the discount, or the customer who always pays at 45 is actually costing you, the first move is the arithmetic on your own four numbers, before the vendor agreement is signed and before the discount is granted. The payment terms calculator at GetAFractional runs it in a few minutes: the gap between the paper and how they pay; the cost of the extension per invoice and per year at your own cost of money; the discount's effective annual rate against that same money, and whether this customer has earned one; the contribution left after the wait against your own floor; the capital tied up and the exposure if they stop paying; the three asks at the minimum that cures, bounded by what you say this customer would accept; and the answer -- accept, negotiate shorter with the net days named, price it in with the percent named, or decline -- with every field blank until you fill it.
If the page stops on a question you cannot answer from your books, the making-money-on-each-job diagnostic sorts out whether the contribution is a tracking problem, and are my books a mess finds out why the aging report does not exist. If the cash is already gone while the invoices sit, the profitable-but-out-of-cash diagnostic is the retrospective version of this page.
This article is informational and does not constitute financial, legal, or tax advice. Consult a qualified professional for decisions specific to your situation.