The Accounts Receivable x Revenue Cohort

This is FYE 2025. The next fiscal year-end is close enough that those 10-Ks will replace this window.

The concept

Revenue is a claim the company has decided it earned. Accounts receivable is the part of that claim that has not yet turned into cash. One line without the other is an incomplete demand print.

Revenue growing while the receivable book grows faster is not the same event as customers pulling product. It can be looser terms, a stuffed channel, or booking work that has not been collected. A receivable book shrinking with no revenue growth is just a collection sweep. The useful read is the crossing:

sales accelerating, and receivables lagging or falling. Customers are paying at least as fast as the company is selling. The growth is pulled by cash-backed demand. That is the archetype: growth that collects.

The two legs discipline each other. A name cannot sit here on a revenue explosion unless collections kept pace, and cannot sit here on a shrinking receivable book unless the sales line actually moved. High conviction in the ranking comes from that mutual check, not from either number alone.

How it was built

The window is FYE 2025, versus the prior year. It holds until the FY2026 10-Ks print. Every name in the set that booked revenue and carried a receivable in FY2024 is in. Pre-revenue names, and names that do not disclose receivables, are out. That is why a launch texture dominates the ten: those are the names where both lines exist and the growth is loudest.

Two measures, same two years, taken from the Forms 10-K:

1. Demand. How fast revenue grew, FY2024 to FY2025.
2. Collection gap. Revenue growth minus receivable growth, in percentage points. Positive means the receivable book grew slower than sales, or shrank.

A name is ranked on both together. The revenue-growth leg breaks a tie. The ten below are the top of that list. The rule was fixed before the names were seen. Changing the rule after seeing the output is fitting. This pull did not.

Ten names. Not a recommendation. This ranking dies when the FY2026 10-Ks print.

The ranking

#	Ticker
1	JOBY
2	NUVB
3	EOSE
4	ONDS
5	MDGL
6	PHAT
7	GERN
8	DCTH
9	SOUN
10	NVAX

What the ranking is saying

1. JOBY. Revenue arrived and receivables fell. The percentage is a base artifact off a $0.1M FY2024 line, so the rate is not the read. Carry the texture: sales showed up, the receivable book shrank. Least stable row in the ten.

2. NUVB. Revenue up about seven-fold. Receivables barely moved (+19%). Widest clean gap after Joby among names with a real base.

3. EOSE. Revenue from $16M to $114M. Receivables grew, but much slower than sales. Hardware ramp that collected.

4. ONDS. Another hardware ramp: revenue up about six-fold. Receivables grew fast too, and still lagged the booking. The growth collected, just with more working-capital use than the names above it.

5. MDGL. The scale row. $180M to $958M. A launch that is no longer a rounding error. Receivables rose, and still at well under half the pace of sales.

6. PHAT. Revenue more than tripled. Receivables about doubled. Drug-launch texture, collections slower than the booking but not running away.

7. GERN. Revenue more than doubled. Receivables almost flat. Sales stepped up and the receivable book stayed still — collections did the work.

8. DCTH. Same shape as GERN, smaller dollars. Revenue more than doubled, receivables barely up. Another drug launch that collected.

9. SOUN. Revenue about doubled, receivables up less than half that. The first name in the ten that reads as a ramp rather than an ignition.

10. NVAX. The large-base name. $682M to $1.12B, and receivables actually fell. Sales rose and the receivable book shrank. That is the cleanest collection print in the ten once you ignore the tiny-base rows.

Usage notes

These are straight reads, not gates. Nothing below knocks a name out of the ten. The notes tell you how to hold the ranking. They are the part that matters once the list exists.

The dominant texture is the commercial launch. MDGL, PHAT, GERN, DCTH, and NVAX are drug launches. EOSE, ONDS, and SOUN are hardware and AI ramps. Revenue is exploding off a small base while collections keep pace. That is clean-demand growth at its loudest. It is a different animal from a mature compounder growing 30% with falling days sales outstanding. The dollar facts under the ranking make the distinction readable. Do not read #1 through #10 as a quality ladder of franchises. Read them as the loudest collection prints at FYE 2025.

JOBY is the least stable read in the ten. FY2024 revenue of $0.1M is noise-scale, so the year-over-year rate and the gap are base artifacts. It ranks first because the rule says it does. The read to carry forward is only “revenue arriving, receivables falling” — texture, not magnitude. Do not treat the percentage as a fact about the business.

Small-base loudness is the pattern, not a flaw in one row. A raw rank on growth and collection gap will keep handing the top to launch ignitions. That is what the rule produces, and it is a true print of the loudest collected growth. It is also why a mature name growing cleanly at 30% does not appear here. A scale-banded version — rank inside revenue-size bands — would let that quieter cleanness surface next to the ignitions. That refinement was not applied. The pull ran as pre-registered. Changing the rank rule after seeing the names is fitting.

Carry a name by reading two things, in order. First the two legs: did sales move, and did the receivable book lag or fall? Then the dollar base. If the base is noise, keep the texture and throw away the rate. If the base is real, the gap is the collection print. NVAX is the large-base version of the same test. JOBY is the noise-base version. MDGL sits in between — a launch that is no longer a rounding error.

This is not a cash test and not a buy list. Receivables keeping pace is not the same as cash from operations turning positive. A launch can collect and still be early, thin, or one-product. The ranking does not grade the business. It grades the demand print at FYE 2025. When the FY2026 10-Ks print, every row is replaced.

The clock

FY2025 is still the last completed fiscal year for these names. The FY2026 year-end will replace every row. This piece is FYE 2025, written before that handoff.

Sources

1. Joby Aviation, Inc.; Nuvation Bio Inc.; Eos Energy Enterprises, Inc.; Ondas Holdings Inc.; Madrigal Pharmaceuticals, Inc.; Phathom Pharmaceuticals, Inc.; Geron Corporation; Delcath Systems, Inc.; SoundHound AI, Inc.; Novavax, Inc. — Forms 10-K for fiscal years ended 2024 and 2025 (revenue and accounts receivable).