Sales You Can't Collect Aren't Sales

Under Section 280E, federal tax is owed on revenue when it's booked, not when it's collected. A brand that ships $500,000 on net-30 terms owes tax on that revenue this quarter, even if the cash shows up 300 days later, or never. Operators are paying real taxes on phantom income while the money sits in their customers' bank accounts.

That's why the standard framing of extended terms undersells the problem. A $100,000 order on terms isn't a sale waiting to settle. It's a loan, and a uniquely bad one: no underwriting, no collateral, zero interest, and, because cannabis remains federally illegal, often no practical legal recourse across state lines. Operators are extending credit on terms no lender in America would accept, to counterparties no bank will touch, in an industry where the debt can't even be resold. The stakes are compounded by margins that leave no room for error: only 24.4% of U.S. cannabis operators are profitable on an after-tax basis, compared to 47% of U.S. employer firms, and 280E compresses net margins to 5-12% before a single invoice goes unpaid.

The Metric Nobody's Tracking

Most cannabis brands report sales the way retail reports foot traffic: a top-line win, independent of whether the cash ever showed up. Ask a sales team for its monthly number and you'll get gross bookings. Ask the same team for its collections-to-sales ratio (cash actually collected, divided by sales booked in the same period) and most can't answer. Nobody's calculating it.

That ratio, not gross sales, is the number that predicts whether growth is real. For a steady-state brand it should sit near 1. One caveat before anyone panics: a brand growing fast will run below 1 by construction, because this month's collections trail last month's smaller sales. A sub-1 ratio isn't automatically a red flag. The signal is the ratio falling while sales are flat, or the gap widening faster than the growth rate explains. A brand booking $500,000 a month with flat sales and collecting $350,000 is running a 0.7 ratio with no growth alibi, and that $150,000 gap isn't "aging AR" parked in a spreadsheet waiting to be worked. Once carrying costs are priced in, it was never a real sale to begin with. It just hadn't been marked down yet.

This is also where the complaints piling up on LinkedIn actually come from: retailers stretching payment to 90 or 120 days "if it gets paid at all," sales reps extending terms to hit a number with no read on whether the buyer can pay, brands matching competitors' looser terms to protect shelf space and then watching those terms stretch even further the next quarter. All of it traces back to the same root cause: nobody is tracking what fraction of booked sales actually land as cash, so nobody catches the problem until an invoice is already 90 days old.

The Shelf-Space Trap Is Real. The Answer Is Pricing, Not Refusal

The obvious objection to "tighten your terms" is that terms are how shelf space gets won. Refuse net-30 and a competitor offers net-60, and now your product is off the menu. That objection is correct, and it's why generic advice to "just extend less credit" fails in cannabis wholesale.

The way out is to stop treating terms as a binary. Terms are a pricing decision. If a retailer's payment history says they pay in 90 days, the cost of carrying them for 90 days belongs in their price. At the double-digit effective cost of capital most cannabis operators face, carrying a $100,000 order for a quarter costs $2,500 to $4,000. A brand that knows this prices the slow payer accordingly, or offers an early-payment discount that beats what a factor would charge. A brand that doesn't is giving away free credit and calling it a sales strategy. You don't have to refuse terms to stop being the cheapest lender in the industry. You have to stop quoting net-30 prices to customers who behave like net-120.

Why Booked Revenue Lies

The decay is measurable. A receivable 30 days past due is worth roughly 80 cents on the dollar in expected recovery. At 180 days it's worth 20 to 40 cents. Past two years, under 10. And recovering anything at that age typically means handing 25-40% of whatever comes back to a third party. Every month an invoice ages, the sale it represents gets marked down, whether or not the books say so.

280E hits the same receivable twice. The tax on the booked revenue was owed up front, and if the invoice eventually becomes a bad-debt write-off, the deduction rarely offsets cleanly what was already paid. By the time a net-30 invoice becomes a net-120 invoice, the margin it was priced at no longer exists. The sale has moved from profitable to breakeven to a net loss, but the income statement still shows it as revenue recognized on day one. Booked revenue measures intent to get paid. It doesn't measure whether anyone actually did.

What Sales-Aligned Collections Actually Looks Like

Operators who manage this well share three practices.

First, every account gets classified by actual payment history instead of gut feel or the sales rep's relationship with the buyer. The classification that matters most is means versus motive: an account that can't pay right now and an account that won't pay are different problems requiring opposite responses. The retailer who is genuinely short on cash but communicates gets a payment plan and a preserved relationship, because they usually come back. The retailer who has the money and ghosts gets escalated immediately, because every month of politeness costs recovery odds. Misdiagnosing one as the other is expensive in both directions: you burn a good account with premature aggression, or you spend six months being patient with someone who was never going to pay.

Second, sales compensation gets tied to collected cash instead of booked orders. Done naively, this backfires: reps stop selling to slow-but-reliable accounts that are perfectly profitable once priced correctly, and start sandbagging orders near quarter-end. The version that works uses holdbacks rather than clawbacks (the bonus vests when the cash lands, instead of being revoked on default), and pairs the comp change with the account classification above, so a rep selling to a finance-approved slow payer isn't punished for finance's pricing decision. The point isn't to make reps afraid of terms. It's to make the cost of terms show up in the same place the commission does.

Third, the tracking has to happen in real time, not once a quarter. A spreadsheet updated monthly can't catch a ratio sliding from 1.0 to 0.7 while it's still fixable, and it can't show aging buckets migrating (current invoices sliding to 30, 30 sliding to 60) which is the earliest signal an account is drifting toward trouble. A system that surfaces aging, contact history, and payment classification per account, updated continuously, turns the ratio from a year-end surprise into a number someone checks every week.

The Real Fix

Three things worth doing this quarter:

1\. Calculate the collections-to-sales ratio monthly, adjusted for growth. Two consecutive months below 1 with flat sales means it's time to revisit terms and underwriting, before it turns into a bigger collections push.

2\. Reprice your slowest payers. Pull the ten accounts with the worst payment history and check whether their pricing reflects their real payment behavior. If a net-120 customer is paying net-30 prices, the carrying cost is coming out of your margin.

3\. Tie sales incentive compensation to collected cash, using holdbacks paired with account classification, so reps and finance are pricing the same risk instead of fighting over it.

Revenue realized beats revenue recognized. Until the cash lands, a sale is just an open question, with the product, the payroll, and the tax bill on the other side of it already spent.