GeneDx Chose Paid Volume Over Volume — and the Tape Is Making It Pay
A deliberate documentation bottleneck lifted genome collections 400bps and turned adjusted net income positive, but it cost a third volume guide cut, a new PAMA disclosure, and half the market cap from the September high
WGS · Earnings Call · 2026-11-09
The trade the Street didn't want
GeneDx spent the third quarter proving a point that nobody on the call wanted proven: that a diagnostics company can choose which volume to run. Revenue came in at $123.1M with $110.4M of exome/genome, up $10M sequentially. Adjusted gross margin expanded to 72% from 70% in Q2 and 69% in Q1, and adjusted net income swung to $8.6M from an $8.2M loss in the first quarter. The blended average reimbursement rate rose roughly 5% sequentially to $3,436, and outpatient genome collection rates went from 32% to 36%. By almost any operating measure, this was the best quarter the company has printed. The stock disagreed. As of the day after the call the shares sat 48% below the September 17 peak of $104.95, down 16% over 90 trading days, with the tape's most recent leg a six-week slide. The reason is in the fine print of the volume line: management cut full-year exome/genome volume growth to "at least 28%" from the 30% it had reaffirmed in August, and pushed revenue to the low end of a $475-490M range.Documentation as strategy
The mechanism is unusually concrete. GeneDx moved genome workflows to collect payer-required documentation upfront before the lab starts testing. The consequence shows up in two numbers placed side by side: accessions grew 28% year over year while reported result volume grew 25%. Mark Gardner, the new President, was blunt about why the industry hadn't done this before — that the lab services business has habitually run tests "in as frictionless a way as possible, to grow volume without regard for whether a test will be paid for or not." He then quantified the denial pool he inherited: “we found that 62% of them were administrative and procedural in nature” — Mark Gardner, President · 2026-11-09, with the remaining 38% split between subjective medical-necessity disputes and patients with no coverage at all. This is where the quarter gets interesting, because it contradicts the company's own guidance cadence. On the August call, CFO Kevin Feeley told investors to expect “the collection rate to remain roughly flat with Q3 with meaningful improvements beginning in Q4” — Kevin Feeley, Chief Financial Officer · 2026-08-03. Instead, the collection improvement arrived a quarter early — genome up 400bps, coverage-policy-matched volume up from 67% to 82% — and the volume cost arrived with it. Gardner conceded the workflow changes were “implemented really at the beginning of this quarter” — Mark Gardner, President · 2026-11-09, which is exactly the answer Guggenheim's Subbu Nambi wasn't satisfied with when she asked why the softness only surfaced now. The sharpest challenge came from Craig-Hallum's Bill Bonello, and it deserves to be read in full because it is the bear case in one breath:Gardner's answer was a partial yes. The 38% is real, he said, and the company isn't giving up on it — but "there is a certain percentage of folks that we have been running the tests on where no matter how well we administratively do it, that we were not actually going to get reimbursed for it." Feeley added that substantially all of the 3-point accession-to-volume gap should eventually pull through. That is the whole bull case and the whole bear case in tension: either the gap is a timing delay in a queue, or it is a permanent culling of bad revenue that will never come back. What got sacrificed to get there is the mix. Genome fell to 26% of outpatient insurance-based volume from 32% in Q2 and roughly 40% earlier in the year — a mix management decision, in Feeley's phrasing, because exome carries a higher contracted rate and higher collection rate today even though genome now collects better than it did. And genome cost per test is down 43% year-over-year, with genome COGS targeted for another "few hundred dollars" of reduction over coming quarters, mostly reagents and dry-lab AI. The company is deliberately shrinking the product it says is the future in order to fund the economics of that future.have you sort of discovered or do you believe that there is a portion of the orders that you were getting that no matter how well you filled out the claim or followed up on denials, they just were not reimbursable? In other words, that the breadth of ordering had sort of expanded beyond the coverage policy?
PAMA: the disclosure that wasn't there before
Buried near the end of the prepared remarks was genuinely new information. Feeley framed the payer base — insurance based revenue is about 85% of the total — and then quantified PAMA for the first time: a little more than half of contracts are tied to a percentage of the Clinical Lab Fee Schedule, and “with approximately 45% of our revenue subject to PAMA, we expect a mid-single-digit headwind to total 2027 revenue” — Kevin Feeley, Chief Financial Officer · 2026-11-09. That is a structural price cut on nearly half the business, disclosed two months after a call where the emphasis was entirely on the upside from collections. Management's offset argument — that RCM work more than covers it — is plausible, but it is now a company asking investors to accept a visible 2027 headwind in exchange for an invisible 2027 tailwind. Global context confirms the general direction: payer reimbursement pressure has been a persistent market-wide theme, but GeneDx is the rare name quantifying a statutory fee-schedule cut rather than a negotiated one.The cash register doesn't care about adjusted net income
The quarter's loudest silence was liquidity. Adjusted net income of $8.6M is a real milestone, but the balance sheet tells a different rhythm. Through the June quarter, effective net cash fell from $120M at year-end to $72M and then $34M, with free cash flow less stock comp running at -$45M in the June quarter after -$48M in March. Some of that is capital expenditure — capex jumped to $10M in the June quarter — but much of it is the collection problem itself showing up as balance sheet: receivables reached 73% of revenue, up 26 points year over year. A delayed collection is a working-capital loan to a payer. Notably, the Blackstone financing that featured on the prior call — and ranked as a theme in Q2 — went unmentioned this quarter, as did Epic Aura, which has decayed from a top-of-mind integration story to a passing reference. When a company stops narrating its funding and its flagship IT partnership in the same quarter it starts celebrating adjusted profitability, that's worth flagging.What's actually left standing
Strip out the noise and GeneDx still owns something hard to replicate. GeneDx Infinity — the variant database — is the moat argument, and the biopharma/data business grew for a second consecutive quarter with Mirum and Beren added as sponsored-testing partners. Long-read sequencing is entering the portfolio via an early-access program, a long read capability that matters for the difficult-to-sequence regions where panels still win. GenPeds, launched in February, is already paying for itself. The walk-back inside the call was telling though: Gardner's prepared text said general pediatrics would drive more unit growth in 2027 than "all other segments"; in Q&A he clarified it would be the largest single segment, not larger than all others combined. At 4.5x price-to-revenue, down from 8.9x at the 2025 peak, the market has already priced a credibility discount. The third volume cut in a year explains why. But the operating line — 72% adjusted gross margin, positive adjusted net income, a 400bps collection gain delivered a quarter early — argues the discount is being applied to the wrong number. The next two quarters have to resolve which one is right: the volume line, or the cash line.That was Feeley in May. He is now asking for the same patience a second time — but this time with a leverage metric that is finally moving the right way, and a cash balance that is not.We understand we need to rebuild credibility, and the way we expect to do that is by setting a guide we believe we can execute on and deliver against.