Net 30, net 60, 2/10 net 30: payment terms explained

Payment terms are the sentence on your invoice that decides when you get paid, and most small businesses inherit theirs rather than choose them. Someone typed "net 30" into the first invoice template years ago, and it has governed cash flow ever since. This is a plain-English tour of what the standard terms mean, what they cost, and how to pick deliberately.

The vocabulary, decoded

• Net 30 : the full (net) amount is due 30 days from the invoice date. The workhorse of B2B billing. Net 15, net 45, net 60 and net 90 work identically with different clocks.
• Due on receipt : payment is expected as soon as the invoice arrives. In practice it means "as fast as your AP process allows", which for many companies is still a weekly payment run.
• 2/10 net 30 : an early-payment discount. The customer may take 2% off if they pay within 10 days; otherwise the full amount is due in 30. The pattern generalizes: 1/10 net 30 means 1% off for paying in 10 days.
• EOM and MFI : end of month and month following invoice. "Net 30 EOM" starts the 30-day clock at the end of the invoice month, so a June 3 invoice and a June 28 invoice are both due July 30. Common where customers batch monthly payment runs.
• CIA / CBD : cash in advance / cash before delivery. Payment before work ships. The standard answer to a customer with a bad payment history.
• Installments and milestones : the amount is split across dates or deliverables, common in projects. Each installment behaves like its own little invoice with its own due date.

What a month of float actually costs you

Terms are an interest-free loan from you to your customer, so price them like one. Carrying $50,000 of receivables for an extra 30 days, at a time when your own credit line costs 12% per year, costs you roughly $500 a month in financing, before counting the hours spent chasing or the risk that an older invoice becomes an uncollectible one. That does not make net 60 wrong; large customers often simply will not onboard a vendor on shorter terms. It makes net 60 a priced concession, something you give to win business you want, not a default you never examined.

Choosing your default: net 15, 30, 45 or 60

• Net 15 suits ongoing service relationships (retainers, agencies, bookkeeping) where invoicing is monthly and trust is established. Half the float of net 30, and reasonable customers rarely blink.
• Net 30 is the safe B2B default: long enough for any competent AP process, short enough that a late invoice is unambiguously late. If you have no reason to deviate, this is the answer.
• Net 45 and net 60 are concessions to size. Enterprises and public bodies often mandate them in procurement. Take them consciously, and consider pricing the float into the rate.
• Due on receipt works for consumer-ish transactions and one-off small jobs with a payment link attached. For invoiced B2B it mostly produces the same 2-3 week reality as net 15 with less clarity about when "late" begins.

One term to avoid: no term at all. An invoice without an explicit due date has no agreed moment at which it is overdue, which neuters every downstream remedy, from polite reminders to late fees to statutory interest (some state defaults, like Texas's 6%, key off the due date). Whatever you choose, print it and the resulting date on every invoice.

Early-payment discounts: do the math before offering 2/10 net 30

2/10 net 30 sounds like a small courtesy. Run it as a financing rate: the customer pays 20 days early and keeps 2%. There are about eighteen 20-day periods in a year, so you are paying an annualized rate of roughly 36% for the acceleration. If your alternative financing costs 12%, the discount is expensive money.

• When it makes sense: cash is genuinely tight and expensive, you have a few large, reliable customers who will actually take the discount, and the certainty is worth the fee.
• When it does not: margins are thin (2% off revenue can be a fifth of a 10% margin), or your real problem is customers ignoring due dates, which a discount does not fix. Slow payers do not become fast payers for 2%; organized payers just get cheaper.
• Watch for the worst case: customers who take the discount and still pay on day 30. Decide in advance whether you will claw it back, and say so in the terms.

The clauses that prevent late payment before it happens

A complete terms block is more than a net number. The additions that earn their keep:

1. A late-fee clause. "Balances unpaid 10 days past the due date accrue 1.5% per month, simple interest." Agreed in advance it is enforceable in most states, subject to caps; check yours in the state late fee library.
2. A deposit for new customers. 30-50% up front is the single most effective de-risking tool for project work. The customer with money already in the job pays the balance.
3. A stop-work trigger. "Work pauses when any invoice is more than 30 days past due." Stated up front, it converts an awkward personal standoff into a documented procedure.
4. Who pays collection costs. A sentence assigning reasonable collection and legal costs to the delinquent party changes the calculus for a customer deciding whether to stall.
5. The clock's start. Invoice date and delivery date can differ by days; say which one starts the term and keep your invoicing prompt either way.

Terms only work if the follow-through is real

Here is the uncomfortable part: none of the vocabulary matters if nothing happens on day 31. Customers learn a vendor's real terms within two or three invoices, and the real terms are defined by the follow-up, not the template. Net 30 with a reminder on day 33 and a firm notice on day 44 is net 30. Net 30 with silence until day 75 is net 75 with extra steps.

So pick your terms deliberately, then make the follow-through mechanical: a heads-up before due, a friendly nudge just after, and a firm escalation ladder on a schedule that runs whether or not you remember. That combination, clear terms plus reliable chasing, is what most "payment terms" problems turn out to be missing. Watch your DSO against the terms you chose to see whether it is working.