First Fully Consolidated Quarter Clears Every Bar; Compass Alliance Widens the Distribution Moat — Maintaining Outperform.

Rating Action

We initiated coverage of Rocket Companies at Hold at Q1 2025 on a clean operating story but unresolved integration and rate-cycle questions: the Redfin and Mr. Cooper transactions had been announced or were closing into a quarter where Rocket was still a stand-alone direct-to-consumer mortgage originator, the synergy targets were management aspiration rather than executed reality, and the gain-on-sale margin trajectory was still tied to a mortgage market where the 30-year was hovering around 6.5%–7%. We upgraded to Outperform at Q2 2025 when the first-three-weeks Redfin signal cleared the most binding execution gate (200K prequal clicks, 23% lead conversion, 7K agent referrals, "exceeding expectations"), the $4B Mr. Cooper bond was 3x oversubscribed at investment-grade quality, and the operating beat (adj rev $1.34B above guide, +13% rate-lock volume vs. flat MBA industry view, GOS held at 280 bps, $80M independent cost-action layer separate from M&A synergies) demonstrated the standalone Rocket franchise was inflecting independently of the integration story. We maintained Outperform at Q3 2025 through the actual close of Mr. Cooper on October 1, with day-1-through-day-30 integration evidence (40K servicing leads into the Rocket pipeline, first Mr. Cooper client closed in three days, 400 LOs onboarded), Redfin attach climbing to 40% from 27% in four months, and the September AI/refi case study showing 10% conversion lift — every operating thesis point underwritten at the upgrade had confirmed or improved.

The Q4 2025 print and the FY 2026 framework reset clear the bar on each open question simultaneously, and they do so by a wider margin than we had calibrated. Integration execution: $140M of Redfin synergies captured in under six months, Mr. Cooper synergies ahead of plan five months in, combined synergy realization pulled forward to end of 2026 from end of 2027 — a full year of operating leverage advanced. The 600,000-loan single-day migration onto a unified servicing platform is the kind of operational milestone that, if it had failed, would have headlined the call; it didn't. Recapture and gain-on-sale: 320 bps in Q4 (highest Q4 print since 2021), more than 50% of refinance volume from the servicing book (versus 30% in Q4 2020), and explicit confirmation that the Mr. Cooper book recapture rates are already lifting in the first quarter post-close. Top-line momentum: $2.44B of adjusted revenue, $140M above the high end of the guide, 24% adjusted EBITDA margin (versus 20% in Q3), $200B annualized run rate with half the headcount of the last comparable cycle. FY26 framework: Q1 2026 guide of $2.6–$2.8B that, after backing out the $150M warehouse-interest reclassification, implies underlying revenue continuing to step higher from the Q4 base, in a market that management and forecasters expect to grow at the high end up to +25%. Distribution: the Compass alliance widens the agent surface from ~2,000 W-2 Redfin agents to 340,000 agents with referral economics and a preferred-pricing mortgage bundle.

We initiated at Hold at Q1 2025, upgraded to Outperform at Q2 2025, and maintained Outperform at Q3 2025. Today we are maintaining Outperform. The asymmetry has continued to shift in our favor: a business that needed to demonstrate two transformational acquisitions could be integrated without operating-leverage drag has now demonstrated they can be integrated with operating leverage acceleration, in a rate environment that is rotating from headwind to tailwind, with a distribution overlay (Compass) announced concurrent with the print. Downgrade catalysts: a stall in the rate compression that flips the refinance pipeline back to a multi-quarter wait (a 30-year stuck above 6.5% through H1 2026 would re-test the thesis), a Compass-alliance economic structure that proves more dilutive than the referral-fee framing suggests, capital-allocation discipline slipping into a transformative-scale acquisition that diverts the synergy flowthrough, or any operational hiccup in the Mr. Cooper system migration that would surface in Q1 or Q2 2026. Reinforce-Outperform catalysts: a clean Q1 2026 print confirming the underlying revenue step-up, the first material month of Compass-bundled mortgage volume (likely H2 2026), or an explicit capital-return signal once integration is fully completed.

Results vs. Consensus

Q4 2025 is the first quarter where the Rocket P&L is consolidated with both Redfin and Mr. Cooper for the full period — meaning the Q4 print is the first clean comparable on the new platform. Management led with the metrics that matter to the bull case (adjusted revenue versus guide, gain-on-sale margin, recapture mix from the servicing book, market share), and the Q4 numbers cleared each one by a wider margin than the Q3 framework had set up.

The headline framing of the print is that this was the first quarter where investors could see what the consolidated business looks like in a steady state — and the steady state is materially better than the pre-deal stand-alone Rocket. The $2.44B adjusted revenue print is the first time Rocket has crossed $2B in a quarter since the 2020–2021 cycle. The 24% adjusted EBITDA margin is the first time the business has printed a 20%+ adjusted EBITDA margin in any quarter post-2021. The 5.5% Q4 market share is the first time the business has been above 5% on a quarterly basis since the post-IPO peak. None of these data points individually re-rate the stock; together, with a clean FY26 setup, they do.

Segment / Channel Performance

Direct-to-Consumer (Origination — the legacy core)

The DTC purchase business was up double digits Q/Q in Q4 — an unusual move into a seasonally low quarter, and a clean read on the integrated-ecosystem thesis. The Redfin preferred-pricing bundle was the proximate driver: bundle volume up 40% Q/Q, and now scaling materially against a Compass-overlay set to launch in 2026. Closed-end second mortgages (CES) had a record month in December with $1B in originations — the first time the CES product has crossed $1B in a single month, and the second consecutive quarter of YoY-doubling on the line. Jumbo loans grew nearly 70% YoY on a deliberate product expansion to address higher-price-point Redfin (and now Compass) customers. The 320 bps gain-on-sale margin reflected the rate-cycle tailwind (rates compressed into the low 6s in late Q4 and tested a 5-handle in early 2026) but also a more favorable origination mix as the recapture engine pulled higher-quality refi business through the funnel.

Servicing Portfolio (the load-bearing wall)

Year-end UPB $2.1T generating ~$5B in recurring annual cash revenue. The structural disclosure that matters: over $300B of the in-house portfolio carries a note rate above 6% — meaning, with rates compressing toward 6% and below, that $300B is the addressable refinance pool sitting inside the captive book, before Rocket has to compete for a single externally sourced lead. Q4 closed-loan volume from the servicing portfolio hit an all-time high; more than 50% of refinance closings came from the servicing book, versus 30% in Q4 2020. Brian Brown was explicit that recapture rates on the Mr. Cooper portfolio specifically lifted post-close, ahead of the deal-model assumptions. Combined refinance recapture is running at ~3x the industry average.

Pro / Mortgage Broker Channel

Pro is now described as the second-most-purchase-skewed channel after DTC purchase. Q4 production was up sequentially, and into Q1, Pro is "doing really well" with a heavier mix shift toward purchase. The blended gain-on-sale margin will move modestly lower in Q1 simply because Pro — which carries lower per-loan economics than DTC — is taking a larger share of the production mix. Per-channel margins are holding; the blended print compresses a few basis points on mix. This is healthy, not concerning, and management framed it that way.

Correspondent (Mr. Cooper inheritance)

The headline disclosure of "$36B ex-correspondent" raised the question of whether Rocket was deprioritizing the channel. Brown was clear: correspondent stays. The framing is that correspondent functions as an MSR-acquisition tool whose lifetime-value math is governed by Rocket's industry-leading recapture rate — which means the same dollar of correspondent flow is worth more to Rocket than to peers. Correspondent will not be the volume driver; bulk MSR acquisition and co-issue are similar capital-allocation levers, and all three are sized against a returns-based framework rather than a volume target.

Redfin (Search & Brokerage)

Redfin contributed 50M monthly active users to the consolidated MAU pool of 62M across Rocket and Redfin. Redfin's ~2,000 W-2 agents continue to operate as the highest-conversion arm of the agent footprint. Mortgage attach rates on Redfin transactions are on track to surpass 50% — a number that, in a Redfin stand-alone world, would have been a mid-teens-percent comp. The Compass alliance now adds 340,000 third-party agents on a referral basis with a preferred-pricing-bundle integration on the mortgage side.

Mr. Cooper (Servicing & Subservicing)

The integration milestone was the migration of 600,000 loans in a single day onto Rocket's unified servicing platform — a transfer Brown described as the largest of its kind, executed without service disruption. Mr. Cooper loan officers were transitioned to Rocket's proprietary LOS system and AI tooling within the first month after close; recapture was lifting from the first month, and accelerated through Q4 as the propensity models were overlaid on the Mr. Cooper book.

Key Topics with Management Commentary

1. The integration is ahead of plan, and the synergy timeline has been pulled forward by a full year.

This was the single most material disclosure on the call relative to the Q3 setup. Brown:

"Our original target was to recognize those expense synergies by the end of 2027. And I'm sitting here today, and I think we can do that by the end of 2026. So that is really good news."

And on the Redfin synergy line specifically:

"We've already captured $140 million in Redfin expense synergies in under 6 months, and we're on pace to exceed our initial goal."

Pulling a synergy realization timeline forward by 12 months is unusual. The flowthrough into FY26 operating leverage is mechanical: a year's worth of duplicative cost runs out a year earlier than the deal model assumed.

2. The Mr. Cooper recapture lift is happening earlier and is larger than the deal model implied.

Brown, on what is in some ways the central economic question of the entire Mr. Cooper transaction:

"We are farther along on the recapture goals. And frankly speaking, I thought we would be at this time, and it's happening at a perfect time because rates are cooperating."

And on the operational mechanics that drove the lift:

"We actually transferred 600,000 loans over. I think that might be the largest transfer of loans ever, and it went off without a hitch."

The combination of a single-day migration onto the unified servicing platform and the centralized propensity-model overlay produced a near-instantaneous lift in conversion. This is the bull-case mechanical assumption of the Mr. Cooper deal arriving on schedule — or, more accurately, ahead of schedule.

3. The macro setup — a refinance pool newly in-the-money on a scale not seen in years.

Brown:

"In January, the population of homeowners benefiting from a rate and term refinance surged to $4.8 million or over $1 trillion in unpaid principal balance. That's a 4-year high. We have reached a tipping point. For the first time in this cycle, the cohort of mortgages with rates at or above 6% now exceeds those below 3%."

This is the macro disclosure that converts the FY26 framework from a hopeful directional call into a quantified setup. $1T of UPB newly able to refinance into savings, of which Rocket's captive book accounts for $300B, against a recapture rate roughly 3x the industry average. The math implied by those three numbers is the single largest reason the rating is moving.

4. Compass — the distribution moat that addresses the supply side of affordability.

Krishna framed the partnership in terms of the structural barrier the company is now willing to take on:

"Today, the way we view the world is that affordability is constrained on many fronts, but two of the big ones, one is supply. Like we just don't have enough supply and inventory is the key to doing that. And then the second is on the other side, demand, the transaction process is too expensive and too fragmented."

The mechanics of the alliance: Redfin becomes the exclusive search portal for Compass private and "coming soon" listings (supply-side advantage); Compass becomes Redfin's largest brokerage referral partner (demand-side distribution); Rocket Mortgage becomes Compass' digital-mortgage partner with a preferred-pricing bundle (transaction-economics advantage). Krishna on the consumer-facing economics:

"Preferred pricing, which includes up to 1% off the first year of the mortgage or up to $6,000 off of closing costs."

Brown clarified the leads-economics structure as a traditional referral model, with the Redfin partner-network playbook scaled across the Compass agent base. The 1M-leads-over-three-years number Krishna cited is a floor disclosure, not a ceiling.

5. AI and operating leverage — doubled Q1 2022 volume with half the headcount.

Krishna offered the cleanest formulation of the operating-leverage disclosure we have heard from the company so far:

"We just delivered nearly $50 billion in loan volume. That is an annualized run rate of $200 billion, effectively double our full year 2024 volume. To put this scale in further perspective, the last time we originated this level of volume was the first quarter of 2022, but here is the critical difference. Today, we delivered that same volume with half the headcount."

And on the framing of the AI moat versus the broader market sentiment:

"We are not a pure software company. We're not an AI company. We're not even really a mortgage company anymore. We're in the business of homeownership and a home is physical. It's a heavy asset."

The implication is that the operating-leverage disclosure (volume parity at half the headcount) is structural rather than cyclical, and would translate into incremental margin expansion in any market — meaning the 24% Q4 adjusted EBITDA margin print is a floor, not a ceiling, in a market that grows.

Guidance: FY 2026 Setup

The FY26 framework is the most consequential part of this report. Management did not give a hard FY26 number for revenue, EBITDA, or EPS — consistent with their prior practice of one-quarter-out guidance — but did give a quantified Q1 guide and an explicit alignment with industry forecasts on market growth.

Three points on the guide that are easy to miss:

• The $150M warehouse-interest reclassification is non-economic. It moves a contra-revenue item to a direct-expense line to standardize reporting between Rocket and Mr. Cooper. Backing it out, the Q1 underlying revenue range is ~$2.45–$2.65B, which compares against the $2.44B Q4 print — meaning the underlying revenue trend is up, not flat.
• Per-channel gain-on-sale margin is holding. Brown was explicit that GoS margins on a per-channel basis are holding into Q1; the blended margin compresses modestly only because Pro (lower-margin, purchase-skewed) is taking a larger share of mix. This is a healthier composition, not a deteriorating one.
• The FY26 framework is asymmetric to the upside on rates. The Q1 guide does not assume a 5-handle on the 30-year for the full quarter; it assumes the rates that prevailed in late Q4 and into early Q1. If rates compress further into the back half of Q1, the guide could prove conservative.

Analyst Q&A

Six questions on the call. The themes that drew the most depth: Compass economics and mortgage-product fit, FY26 expectations / market share / operating leverage, expense bridge and seasonality, regulatory and bank-reentry risk, market share trajectory and correspondent strategy, and the AI / technology operating-leverage story.

Ryan McKeveny (Zelman) opened with two questions on the Compass alliance: the structure of the lead-flow economics (referral-fee model? exclusive to Compass?) and whether Rocket's mortgage product set was deep enough for Compass' higher-ASP customer base. Krishna framed the partnership architecture (inventory + lead flow + mortgage integration) and Brown answered the economics directly — the leads structure is a traditional referral model, the 1M-leads-over-three-years figure is a floor, and on the jumbo question, jumbo production was up ~70% YoY in Q4, with the preferred-pricing bundle already live and operational on day one of the alliance announcement.

Ryan Nash (Goldman Sachs) asked the FY26 framework question directly: market size, share gains, operating leverage, EBITDA. Krishna framed the macro (high end of forecasts at +25% market growth, rates with a 5-handle, inventory creeping up) and Brown walked through the Q4-to-Q1 bridge in granular form: production up Q4-to-Q1, per-channel gain-on-sale holding, blended GoS modestly compressing on mix, and Q4 explicitly framed as "a good jumping-off point" for the year. Brown's specific phrase — "we expect the first quarter to be very similar, if not better" — is the cleanest forward-looking color we have heard from the company.

Jeff Adelson (Morgan Stanley) pressed on the Q1 expense bridge and operating-leverage trajectory. Brown gave the cleanest expense walk we have seen on a Rocket call: ~$2.6B at the midpoint, of which ~$50M is one-time acquisition costs, ~$110M is intangibles amortization, ~$80–$90M is SBC (lower than Q4 because of acquisition-accelerated awards), and the underlying expense run rate is ~$2.2B. Adelson's follow-up on bank-reentry risk drew a more political answer from Brown — banks haven't seen mortgage as a productive channel, the unit economics haven't penciled out, and the path back is "long" — but the substantive point was that Rocket has invested in this stack for two decades and reentry is not a near-term concern.

Mihir Bhatia (Bank of America) asked for explicit FY26 market share targets and pressed on the correspondent-channel framing. Krishna reaffirmed the through-2027 share targets without quantifying the 2026 step (which is consistent with prior practice) and emphasized that progress is non-linear: in fast-growing markets share takes itself, in compressing markets discipline is what wins. Brown was clear that correspondent stays as a channel, sized as an MSR-acquisition tool against returns-based capital allocation. The implication: Rocket is not retreating from correspondent; it is sizing it on lifetime-value math rather than volume targets.

Mark DeVries (Deutsche Bank) asked for quantitative AI / operating-leverage disclosures. Krishna offered a longer narrative on the AI-as-accelerant framing rather than incremental new numbers, but reinforced the structural disclosure on volume parity at half the headcount and the pre-approval-conversion math (2.5x higher conversion on AI-qualified leads). The substantive new disclosure was the $1B+ in incremental monthly volume from automated communication tools (800K chats, 1.8M texts, 2M outbound calls, 5M document processes per month).

Don Fandetti (Wells Fargo) closed with a focused question on the Mr. Cooper recapture lift. Brown's answer is the one that, if extracted from the call, would have been the headline: the recapture engine on the Mr. Cooper book is running ahead of where Brown expected at this stage of the integration, the 600,000-loan migration was the trigger, and the rate setup ("you could end up in a 5 handle") is amplifying the lift.

What They’re NOT Saying

• No FY26 hard number on revenue, EBITDA, or EPS. Consistent with prior practice, but worth flagging. The Q1 guide is $2.6–$2.8B; from there, investors are extrapolating against management's "double-digit market growth" framing and the synergy pull-forward. We read the absence of a hard FY26 number as conservatism rather than caution — if management had wanted to signal a deceleration, this was the call to do it on, and they did not.
• No explicit FY26 gain-on-sale margin range. Brown gave per-channel and blended directional color (per-channel hold, blended modestly lower on mix) but did not commit to a quarterly or annual range. Given that 320 bps was the highest Q4 GoS print since 2021, and FY25 was 283 bps, the absence of a forward range can be read either way; we read it as management not wanting to anchor to a number that might prove conservative if rates compress further.
• No explicit FY26 capital-return signal. Year-end available cash $2.8B, total liquidity $10.1B. No discussion of buyback authorization or dividend on the call. We read this as a signal that capital is being directed at integration completion, the Compass-alliance launch, and balance-sheet flexibility through the rate-cycle inflection — not as a signal that capital return has been deprioritized. The buyback signal, if it comes, is most likely to come once integration is fully realized end of 2026.
• No specific Rocket Money revenue or subscriber number. Rocket Money was mentioned only in the context of January marketing seasonality — "Rocket Money's January marketing campaign that drove record subscriber growth" — without a quantified subscriber count or a revenue contribution. Given the asset's strategic role in the consumer-engagement flywheel, the absence of disclosure is a watch item; we expect it to surface in 2026 as the unit becomes a more material contributor.
• No explicit servicing-book growth target for FY26. The book is $2.1T at year-end; the in-the-money cohort is $300B+; the natural recapture mechanic will run loans off the book even as new originations come on. Whether the book holds at $2.1T, grows, or modestly compresses through a refi-heavy 2026 is a question management did not directly address.
• No discussion of the regulatory environment beyond the bank-reentry question. Given the political backdrop (CFPB posture, GSE guarantee fees, FHFA leadership), a no-comment on the broader regulatory framework is conspicuous. The framing was neutral-to-positive on bank-reentry risk and silent on everything else.

Market Reaction

The print and the Compass alliance landed concurrently on a single news day — an unusual disclosure choreography that telegraphs management's confidence in both events being received positively. The framing in the press release led with the Compass announcement and the Brown promotion to President, with the Q4 numbers as supporting material; the call inverted that order, leading with Q4 and FY framework before pivoting to Compass. Either reading, the news combination is a narrative reinforcement: integration completing ahead of schedule, a new distribution overlay announced same-day, and an internal promotion that signals leadership continuity into the FY26 framework.

Pre-print, the buy-side debate had crystallized around three questions: (1) would the first fully consolidated quarter validate the Mr. Cooper recapture deal-model, (2) would the synergy timeline hold to end-of-2027 or compress, and (3) what was the implied FY26 revenue base off the Q4 jumping-off point. Each of those questions cleared positively, simultaneously. We expect the print to be received as the cleanest validation moment of the post-Redfin / post-Mr. Cooper Rocket since the deals were announced, and the rerating risk-reward to shift toward upside on a per-quarter basis through H1 2026 as each successive print either confirms or extends the FY26 framework.

The most likely sources of post-print debate: (a) how much of the Q4 GoS strength was rate-cycle versus structural, (b) the dilutive vs. accretive math of the Compass referral economics at scale, and (c) whether the "double-digit market growth" framing for FY26 will hold if the 30-year backs up above 6.5% in H1.

Street Perspective

The Street had been split on Rocket through 2025 along a familiar fault line: the fundamentalist construction (integration risk + rate-cycle exposure dominate the operating story; wait for the consolidated print) and the structuralist construction (the integrated ecosystem is a category-of-one platform that will compound through any rate cycle; do not wait). The Q4 print is the first quarter where the structuralist construction has empirical backing on a quarterly basis — consolidated revenue, GoS margin, recapture mix from servicing, market share, and synergy timeline all moved in the structuralist direction at the same time.

The bull case being made on the Street had been anchored to the recapture-engine math: a $2.1T servicing book with $300B+ in-the-money UPB, recapture at 3x industry, a refinance market newly opening up — mechanically implies a multi-quarter origination tailwind that flows through to GoS-rich revenue. The Q4 print delivers the first full-quarter empirical test of that math, and the test resolves in favor of the bull case. The bear case had been anchored to integration coordination tax and rate-cycle path dependency; the synergy pull-forward and the rate compression both move against the bear-case anchor.

The post-print Street question that we expect to dominate is FY26 market-share trajectory. Management is reaffirming the through-2027 share targets but declined to quantify the 2026 step, which leaves room for both constructive and skeptical extrapolations. We are constructive: the Q4 5.5% share is up 170 bps YoY against a market that itself was up modestly, meaning Rocket took share in a market that was growing — the cleanest definition of structural share gain.

Model Implications — FY 2026 Setup

The FY26 model setup, as we read the framework off this call:

• Revenue trajectory: Q1 2026 guide midpoint $2.7B (which, after backing out the $150M reclassification, implies underlying Q1 of ~$2.55B versus Q4 of $2.44B, a ~5% Q/Q step-up). Across FY26, against a market growing in the high single digits to low double digits and a captive book with $300B+ of in-the-money UPB feeding a recapture engine running at 3x industry, the underlying revenue trajectory should compound through the year, with rate-cycle-amplified upside through H1 if rates hold a 5-handle.
• Gain-on-sale margin: 320 bps in Q4 is the highest Q4 print in four years; FY25 average was 283 bps. We model a FY26 GoS in the range of 285–315 bps, with the upper end requiring the rate environment that prevailed in late Q4 / early Q1 to persist. Per-channel margins are holding; the blended print compresses only on a mix shift toward Pro purchase volume.
• Adjusted EBITDA margin: Q4 24% versus FY25 19% (which itself is up from 18% in 2024). The synergy pull-forward to end of 2026 means a year of duplicative-cost runout advances by 12 months, which mechanically amplifies operating leverage in FY26. We model FY26 adjusted EBITDA margin in the 21–24% range, with the upper end requiring the rate-cycle tailwind to persist through H1 and the synergy realization to track to plan.
• Servicing book and recapture: $2.1T UPB at year-end is the load-bearing wall; we model the book holding flat-to-modestly-up through FY26 as new origination retention offsets natural runoff. Recapture mix from the servicing book at 50%+ of refi closings is the new normal, not a one-quarter event.
• Capital deployment: $2.8B available cash, $10.1B total liquidity. We do not model an explicit FY26 buyback authorization but assume capital is reserved for integration completion, the Compass alliance ramp, and selective bolt-on optionality. A buyback signal — if it comes — is most likely H2 2026 once synergy realization is in the rear-view.
• Compass alliance contribution: Modest material contribution to FY26, ramping more meaningfully into 2027. The 1M-lead figure is a three-year commitment; the mortgage-attach economics will scale with the bundle adoption rate at Compass agents, which is a 2026 ramp.
• Watch items for downside: a 30-year backing up above 6.5% through H1 (refinance pool compresses); any operational hiccup in the unified-servicing-platform migration; a Compass economic structure that proves more dilutive on the GoS line than the referral framing suggests; the Brown promotion-to-President structure introducing decision-coordination friction with Krishna.

Thesis Scorecard — State of the Thesis at End of FY 2025

Net read of the scorecard: every operating pillar has moved in the direction of the structural bull case over the course of FY25, and the FY26 macro setup is the most constructive backdrop the thesis has had to underwrite against since the 2020–2021 cycle. The single watch item is whether rate compression holds; even if it does not, the recapture engine inside the captive book provides a multi-quarter origination tailwind that is materially insulated from the spot rate path.

Bottom Line

The Q4 2025 print is the cleanest operational quarter Rocket has delivered since the post-IPO peak, and the first quarter where the consolidated entity (Rocket + Redfin + Mr. Cooper) has shown what its steady-state operating economics look like. The same call delivered (a) a $2.44B revenue print $140M above the high end of the guide, (b) the highest Q4 gain-on-sale margin in four years, (c) a 24% adjusted EBITDA margin (versus 20% one quarter ago), (d) a synergy realization timeline pulled forward by a full year, (e) the largest single-day loan migration in industry history executed cleanly, (f) a 340,000-agent distribution overlay via the Compass alliance announced concurrent with the print, and (g) an internal promotion that telegraphs leadership continuity into the FY26 framework. The Q2 upgrade thesis has continued to compound through Q3 and now Q4 — this is the print where the consolidated economics are visibly better than the standalone case the upgrade had to underwrite. Maintaining Outperform.