Goosehead Filed the Restatement Before It Read the Revenue
The 8-K landed before the call did
Maddie Middleton's disclaimer was not the usual webcast housekeeping. Before a single revenue number was read aloud, she told listeners that Goosehead's “previously issued annual and quarterly financial statements for and as of the years and quarterly periods in 2023, 2024 and 2025 and our quarterly financial statements for the first half of 2026 should no longer be relied upon” — Maddie Middleton, Unknown · 2026-11-09. That is the whole story of this quarter, and everything else — a 17% revenue print, a record producer count, an AI roadmap — orbits it.
The mechanics are almost quaintly benign. Agent compensation costs that had been expensed as incurred must now be deferred over a benefit period under ASC 606, pushing expense onto the balance sheet and lifting operating income by an estimated 0.5%–5.5% of total employee comp for FY25 and 1H26. Separately, franchise technology cost-recovery fees get grossed up one-for-one into revenue and G&A — roughly $11–16M for 2025 and $8–13M across the first three quarters of 2026. CEO Mark Miller framed the direction of travel plainly: “These adjustments have no impact on our cash flows or business operations, but they result in an increase in reported revenue and earnings in the affected periods.” — Mark Miller, CEO · 2026-11-09 A restatement that improves the P&L is an unusual thing to have to apologize for.
What follows is the real cost. John Martin closed the financial section with a disclosure blackout:
As the restatement work remains ongoing, we intend to focus the balance of today's discussion, including Q&A, on key business drivers and operating trends, which we can speak to with confidence. We'll provide a more fulsome discussion of expenses and profitability once the financial statements are finalized.
And when Ryan Tunis of Cantor pressed the obvious question — ASC 606 has existed since 2018, so why do restated periods begin in 2023? — Mark Jones Jr. answered, in full: “Yes. I would just refer you to the 8-K on the disclosures related to this topic.” — Mark Jones Jr., President and COO · 2026-11-09 financial statements is now the single highest-momentum keyword in this company's vocabulary, and compensation expense and G&A expense entered the top-30 alongside it, while accelerated core revenue growth rate and margin language slid out of the frame.
The tape was already unforgiving. Goosehead has given back 17% over 90 trading days and sits 36.8% below its August high of $73.00, itself 74% under the 2021 peak of $177.30. Price to revenue has compressed to 4.3x from 39.2x at the 2020 peak. A restatement landing on a stock in a 37% drawdown is a different animal than one landing on a high.
The market healed, then it got competitive
Here is the quiet conceptual flip of the quarter. Last period, product environment was this company's number-one keyword — carriers returning, capacity restored, a hard market ending. That same healing is now a revenue headwind, and the vocabulary has inverted into pricing pressure. Jones was blunt about the geography: “really where you're seeing the impact is pricing in the Texas book for the corporate side of the business. That's been probably a larger decline in pricing than what we would have expected even 3 months ago.” — Mark Jones Jr., President and COO · 2026-11-09 Auto is down mid-single digits, home low-single digits, and the arithmetic is clean: policies in force grew 14% against premium growth of 11%, implying roughly a 3-point average price decline — versus the pricing tailwind the same math produced a year ago.
Core revenue grew 13%, a deceleration from the second quarter, on what Martin called “a bit tougher comparisons in the third quarter than in the second quarter” — John Martin, Unknown · 2026-11-09 plus more pricing erosion than planned. That matters because in April management had explicitly promised the back half would inflect: “We are still expecting acceleration in the second half of the year as the improvements in client retention begin to outpace the offset of the pricing impacts on year-over-year premium changes.” — Mark Jones Jr., President and COO · 2026-04-22 Eleven weeks before year-end, on a call where they declined to discuss profitability, they retracted the timing:
It's hard for me to put a stake in the ground and tell you exactly when pricing is going to level off. I think I tried to do that in '23 and in '24 and was ultimately not correct on those topics because of the general extremeness of the hard market.
The offsets are real, if increasingly familiar. Client retention reached 86%; contingent commissions were guided up to 80–110bps of total written premium from a prior 70–100; the enterprise team posted its eighth consecutive quarter above 50% new-business growth and now writes 23% of new business commissions; top-50 franchise same-store sales rose 29%; producers per franchise hit a record 2.6. And the corporate agent launched franchises thesis — 19 launches this year — is genuinely working, at 6% of producers producing 15% of new business.
A margin story on hold, on a suddenly levered balance sheet
Strip out the accounting and Goosehead still cannot tell you what it will earn this year. Jones defended the record — “margins have only effectively gone in really one direction for the last 6 years” — Mark Jones Jr., President and COO · 2026-11-09 — and the first half broadly supports him, though not in the direction he implied: operating margin ran 20.4% across 1H26 versus 21.8% a year earlier, on total revenue of $113M in 2Q26.
The balance sheet is the under-discussed line. Effective net cash swung from +$82M in 2Q25 to −$296M in 2Q26, and the call disclosed $33M of cash against $322M of debt. Over the same four quarters, repurchases ran $58M, $23M, $50M and $4M. In February, Jones had framed the buyback case on conservatism: “we've been really conservative with our balance sheet for our entire history. So we have a lot of optionality... We have a revolving credit facility of $75 million.” — Mark Jones Jr. · 2026-02-17 That optionality has now been spent, at precisely the moment the company cannot publish audited expenses.
The MCP server and the $250 bot
The one place management sounded genuinely forward-leaning was distribution architecture. Goosehead is building a Model Context Protocol server — Model Context Protocol debuts in its top-30 keywords — to let third-party personal AI agents transact against its carrier network, live by year-end, with generative-engine optimization to make Goosehead legible to models. This is a personal AI agents theme the market is debating too, not a Goosehead invention — Lemonade and Airbnb both leaned on agent language this reporting season, and it sits among the market's higher-momentum global keywords this quarter. Goosehead is riding the wave, not creating it.
Its defense is refreshingly concrete. Randy Binner's bot test: two agents, identical information, quoting carrier direct-to-consumer auto sites, returned quotes “one being $1,000, one being $1,250 with all of the exact same information” — Mark Jones Jr., President and COO · 2026-11-09 — and “the bots were unable to get homeowners quotes” — Mark Jones Jr., President and COO · 2026-11-09. Add errors-and-omissions cover and the "liability transfer" argument, and you have a real moat claim. It is also a notably different posture from February, when the answer to disintermediation was purely defensive: “there's only from what we know of 1 business that can actually today bind policies end-to-end without human intervention, and that's us.” — Mark Jones Jr. · 2026-02-17 Today the pitch is plumbing, not protection — be the intermediary whichever model the client sends.
Which is why independent agent ranks second in this quarter's keywords. When the financials are un-auditable, you argue structure: agents place north of 70% of a $560B personal lines market, carriers keep investing in the channel, and “the vast majority of them, somewhere throughout the process, want to kick out and talk to a human being” — Mark Jones Jr., President and COO · 2026-11-09. The human capital and producer base arguments are the ones Goosehead can still prove.
Mark Miller retires at year-end handing Mark Jones a company with 2,840 producers, a record franchise productivity line, a soft-market revenue base, a pending restatement, and $322M of debt. The growth engine looks intact. The reporting around it does not yet.