ARX — Accelerant Holdings — Earnings Review Q3 FY25

Date: 2026-02-22 Quarter: Q3 FY25 (September 30, 2025) Task: First ARX earnings review — no prior Joe analysis

Opening Thought

Alright, ARX is new to me. I've seen the ticker floating around but this is my first real sit-down with the numbers. Insurance isn't exactly where my head usually goes — I tend to live in software and fintech — but when something is growing revenue 74% year-over-year and trading at 2.5x sales with a 66% drawdown from its highs, I owe it a serious look. Let me see what's actually here versus what might just be hope dressed up as a thesis.

The Results — What "Is" Here

Note on Q3 EBITDA: The 39.3% reported margin includes 39minnon − recurringinvestmentgains.TheunderlyingoperationalEBITDAwas 66m at ~24.7% margin. That's the real number.

Note on GAAP net income: -$1.37B net loss in Q3 due to a one-time, non-cash profits interest distribution — IPO-related equity comp to co-founders. This is accounting noise, not business performance.

What I Actually Like Here

Revenue growth is accelerating, not decelerating. 38.9% → 68.4% → 74.0% YoY over three quarters. That's not "oh things were slow but now they're picking up" — that's a business in the middle of acceleration. Most of the companies I follow that look like this either slow from here or hold the line. Very few continue to accelerate at this scale. At $267m quarterly revenue that's not some tiny base trick either.

The EWP QoQ number in Q3 is meaningful. Exchange Written Premium hit $1,043m, up 14.4% sequentially — the strongest QoQ growth in four quarters after a period of 4-5% bumps. When the underlying flow of business through your platform reaccelerates like that, it's telling you something real about demand.

NRR of 135% deserves to be said twice. 80% of their growth comes from existing members expanding their volume. These aren't customers barely staying — they're adding more business every year at a 35% clip net of any departures. That's the hallmark of a platform that delivers real value. I've seen sub-$3B companies in SaaS get bid up for far less compelling retention.

The loss ratio trend is solid. Gross loss ratio moved from 54.7% at its worst (Q2'24) to 50.1% in Q3. A portfolio that's 95% policies under $10,000 is genuinely less volatile. Management's claim that they're insulated from the industry-wide large-liability trends isn't just talk — the numbers show it.

The balance sheet is not a concern. $2.5B cash, $122m debt. This company has more cash than its market cap. That changes the risk profile. No solvency risk, no dilution imperative, no urgency that creates bad decisions.

The Big "Could Be" Tension

Here's where I slow down. The FY26 guidance is striking — and not in a good way.

• EWP $5.0B+ (28-33% growth, fine)
• Third-party DWP $2.1B (42% of volume from external insurers)
• Adj EBITDA $269M — 5.4% margin

They're guiding from ~25% underlying EBITDA margins today to 5.4% by next year. That's a massive, deliberate compression. The reason is strategic: they're transitioning from a model where Accelerant's own balance sheet bears the insurance risk (and earns the full premium economics) to a marketplace model where third-party insurers provide the capital (lower margins, but asset-light, scalable, diversified). It's the right long-term move.

But here's my "is" vs "could be" concern: the attractive platform economics are in the future, not now. Right now, the business is highly profitable on its own balance sheet. The transformation to a true exchange with 67% third-party mix (3-5 year target) will require years of margin suppression while they rebuild the earnings base on a more scalable but lower-margin model. Until I can see that model proving itself out at scale — what does a 42% third-party mix actually look like on the income statement — I'm dealing with a lot of "this is what it becomes" rather than "this is what it is."

That gap between today's performance (excellent) and tomorrow's promised architecture (unproven) is exactly where I've gotten burned before. The "one more quarter" trap is essentially this: the Q3 results look great, so you hold. Then Q4 shows margin compression. You hold because "that's the plan." Then Q1 FY26 shows the same. You tell yourself it's working. Eventually you own a structurally diluted business you overpaid for.

Management Credibility — Too Early to Know

This is only ARX's 7th quarter as a public company. That's not enough track record to know how they manage expectations. I can't tell you if this is a beat-and-raise team or a guide-and-miss team. The Q3 call sounds like the $3B pipeline disclosure was a new and positive addition, and the qualitative tone is disciplined. But I need more data points before I trust their guidance cadence.

The Glassdoor picture (3.3/5, 42% positive outlook, concerns about layoffs and leadership) is a yellow flag for a company of this scale in growth mode. Not a dealbreaker, but worth noting.

Beat-and-Raise Assessment

Incomplete — no explicit Q3 guidance was available from the Q2 call to benchmark against. What I do know:

• Q3 revenue was $267m, adjusted net income $79.8m, EWP $1,043m
• Market reaction and analyst commentary suggests in-line to slight beat
• The $3B pipeline disclosure and Q4 EWP guidance midpoint of $1,080m look conservative given Q3's acceleration

Q4 guidance at $59.5m EBITDA midpoint vs. Q3's $66m underlying feels conservative — likely deliberately so. If they hit $70m+ underlying in Q4, that's a beat worth noting.

Prior Beliefs / Updated Beliefs

No prior analysis. First look.

Going in: I knew the elevator pitch — specialty insurance platform, strong NRR, recent IPO, big drawdown.

Coming out: The business is genuinely interesting. The numbers are real — 74% revenue growth with 135% NRR and improving loss ratios is hard to fake. The platform model transition is strategically coherent. The valuation at 2.5x TTM revenue looks cheap given the growth rate.

But I have too much uncertainty on three fronts:

1. Does the third-party insurer onboarding model actually deliver the economics they expect at $5B EWP?
2. Is the management team capable of executing this complexity with discipline (need to see it)?
3. Can I hold through 18-24 months of guided EBITDA margin compression (5.4%) while waiting for the platform economics to emerge?

Sizing Recommendation

Tryout. Not a starter.

Here's how I'm thinking about it: the raw numbers in Q3 are genuinely impressive. The platform concept is differentiated. The valuation is low relative to growth. All of that says there's something here. But this is a 7-quarter-old public company asking me to be patient through deliberate margin compression while it transforms its business model. That's a lot of asks at once.

I'd want to own a small position — 2-3% — to keep me honest and paying attention. I'd watch:

• Q4 FY25: Does underlying EBITDA come in above the $59.5m guide?
• Q1 FY26: Is EWP tracking toward $5B? Does NRR hold at 130%+?
• Third-party insurer economics: What do the actual fee rates and take rates look like as this segment scales?
• Management commentary language: Do they stay disciplined, or do they start "managing" expectations around the platform transition?

If the first two quarters of FY26 show beat-and-raise behavior on EWP and the margin story is playing out as guided (not worse), this moves to a potential starter conversation.

If NRR starts breaking below 120%, if third-party insurer adds disappoint, or if management revises down the $5B EWP target — that's when you exit and move on. The "one more quarter" trap lives right here.

Summary

What I'm Watching

1. Q4 FY25 EBITDA: Beat or miss the $59.5m midpoint?
2. Q1 FY26: EWP quarterly run rate vs. $5B annualized trajectory
3. NRR: Does it stabilize at 130%+, or keep sliding?
4. Third-party mix: Does the 39-42% mix by year-end actually generate the fee economics described?
5. Any revision (upward or downward) to the FY26 $269M EBITDA target

First analysis of ARX. No position as of 2026-02-22. As usual, thanks for reading.