Beat the Guide, Shrank the Business: Etsy's Take-Rate Quarter, a Negotiable Margin Floor, and a New CEO

Results vs. Consensus

Q3 2025 Scorecard

Year-Over-Year Comparisons

Quarter-Over-Quarter Comparisons

Etsy's quarters are not seasonally comparable in the way a semiconductor company's are: Q4 carries the holiday, and Q3 is the setup quarter. Sequential dollar comparisons are therefore of limited use here. What is sequentially meaningful is the direction of the growth rates, which is the metric management chose to steer the narrative around this quarter.

The Sequential Trend That Actually Matters

Management's central claim this quarter is not about dollars, it is about second derivatives: that the year-over-year decline in the core marketplace is narrowing, and that Depop is accelerating. Both claims check out.

The ex-Reverb consolidated line crossing from −2.6% to +0.9% is the single fact management most wants investors to hold onto, and it is a real fact. It is also, at +0.9%, a fact about a company whose largest asset is still contracting and whose second asset is 11% of GMS. A consolidated marketplace that grows 0.9% because a $292M business grew 39% while a $2.43B business shrank 2.4% is arithmetically in growth and structurally still in decline.

Quality of Beat

Revenue: The $22.7M revenue beat is roughly 3.5% and is genuinely well-sourced: GMS came in above the guidance ceiling and take rate came in 40bp above the guided 24.5%, so both the volume and the monetization line beat their own bars. The problem is the composition. Services revenue, which is overwhelmingly Etsy Ads plus Depop's boosted-listings product, grew 12.7% year-over-year. Marketplace revenue, which is the transaction and listing fees that scale with units, fell 1.7%. Strip out the Reverb divestiture and revenue grew 6.1%; the delta between that and the 12.7% services line tells you where the growth lives. This is an advertising business growing inside a marketplace that is not.

Margins: Gross margin fell 68bp to 71.3% as cost of revenue grew 4.9% against 2.4% revenue growth. Operating margin fell 91bp to 12.2%. Marketing rose to 30.7% of revenue from 29.7%, driven by Depop brand spend, and product development rose to 16.7% from 16.2% on higher engineering headcount. The only line that gave leverage was general and administrative, down 7.8% in dollars and 129bp as a percentage of revenue. A quarter in which the sole source of operating leverage is the overhead line is not a quarter with operating leverage.

EPS: This is where the print is most misleading, and it deserves to be spelled out. Diluted EPS grew 40.0%, from $0.45 to $0.63. Income from operations declined 4.7%. The bridge between those two facts contains no operations at all. Other income swung $22.4M in Etsy's favor, driven by a $0.8M foreign exchange gain this year against a $16.8M foreign exchange loss last year, plus higher interest income on the $1.6B cash and investments balance. That lifted pre-tax income 24.8% on declining operating income. The effective tax rate then fell from 22.3% to 18.0%, adding roughly another 5 points of net income growth. Finally, the buyback shrank the diluted share count 4.4%. Sum those three: currency comparison, tax rate, share count. None of them recurs by right, and none of them is the marketplace.

Segment Performance

Marketplace Mix — Q3 2025

The reported 6.5% consolidated decline is an artifact of the Reverb divestiture and should be read as such: $213.7M of prior-year GMS simply is not in the base any more. The honest comparison is the ex-Reverb +0.9%. But the mix table also makes the structural point plainly. Etsy marketplace is 89.3% of consolidated GMS. For consolidated growth to be durably positive on the strength of Depop alone, Depop would have to hold a near-40% growth rate indefinitely against a core that keeps shrinking, and the arithmetic of that gets harder every quarter as Depop's base grows and Etsy's shrinks.

Etsy Marketplace — The Decline Is Narrowing, the Customer Base Is Not

Etsy marketplace GMS of $2,432.6M was down 2.4% year-over-year and down 3.2% on an FX-neutral basis, meaning currency contributed roughly 80bp of help. Sequentially the year-over-year comparison improved 300bp from Q2's −5.4%, the second consecutive quarter of narrowing decline. Management attributes that improvement to the four strategic priorities the incoming CEO defined, and there is supporting evidence: app GMS outperformed non-app GMS by 13 percentage points, app home-screen GMS grew 20% year-over-year, and the app's share of total GMS rose to 46% from 42.8% a year ago.

"While we're not satisfied with any decline in Etsy GMS, we are encouraged that year-to-year GMS comparisons continued the momentum established earlier this year and improved by another 300 basis points from Q2 to Q3." — Lanny Baker, CFO

The customer-count picture is less encouraging and management spent less time on it. Etsy active buyers fell 5.0% year-over-year to 86.6M and 0.8% sequentially. Active sellers fell 10.9% year-over-year to 5.5M, though they grew 1.7% sequentially and both US and international seller counts returned to sequential growth. GMS per active buyer was $121 and has been flat all year. So the improvement in the GMS growth rate is not coming from more buyers or from each buyer spending more on a trailing-twelve-month basis. Management's own explanation, offered on a month-by-month rather than trailing basis, is higher average item values and improving purchase frequency since April 2025. That is plausible and it is also unverifiable from the disclosed data.

Assessment: Two consecutive quarters of narrowing decline is a genuine trend and deserves credit, but the composition matters. A marketplace whose GMS decline is narrowing while its buyer count is still falling 5% is improving its monetization of a shrinking audience, which is a different and more fragile achievement than reversing the audience decline. The signpost for next quarter is unambiguous: Etsy marketplace GMS growth crossing zero, with the active buyer count stabilizing sequentially. One without the other is not the inflection.

Depop — Real Momentum, Newly Expensive

Depop GMS grew 39.4% to $292.1M, an acceleration of roughly 400bp from Q2's 35.3%. US GMS grew 59%. Active buyers grew 38.8% to 6.6M and active sellers grew 40.8% to 3.0M, so the growth is broad-based across both sides of the marketplace rather than a per-user monetization effect. Management now describes the business as running above a $1B annualized GMS rate.

The cost of that momentum arrived in the same breath. Depop's segment margin declined sequentially as brand marketing accelerated, and the Q4 guide embeds a further "significant sequential increase." The incoming CEO ran Depop for two years before returning to Etsy, so the decision to press here carries her personal conviction and her personal track record.

"This investment in Depop is discretionary and opportunistic arising from our excitement about the scale and growth of the apparel resale market as well as Depop's own very encouraging momentum as the business is now at an annualized run rate of $1 billion plus." — Lanny Baker, CFO

Assessment: Depop is the best thing in this print and the most expensive. A 39% grower with 40% user growth on both sides at a $1B GMS run rate is a genuinely valuable asset, and the CEO said as much in the bluntest terms available to him: he floated the idea that Depop could be "the Venmo to our PayPal." The bear reading is that Depop is 10.7% of GMS and is now consuming enough marketing dollars to move the consolidated margin by roughly 300bp against Street expectations. Both readings are true. The question the next two quarters answer is whether the brand spend buys durable awareness or rents temporary GMS.

Key Operating Metrics

Assessment: The KPI table splits cleanly in two. Everything measuring the size of the Etsy audience is negative. Everything measuring the intensity of engagement within that audience is positive: app share, app downloads, reactivations, seller counts sequentially, habitual buyers. That is the profile of a business successfully defending its best cohort while the long tail leaves. It can support flat-to-modest GMS for a long time. It does not, by itself, produce growth.

Key Topics & Management Commentary

Overall Management Tone: Measured and unusually candid for a quarter that beat, with the confidence concentrated in the strategy rather than in the results. Management repeatedly volunteered its own dissatisfaction with the growth rate rather than waiting to be pressed on it, and the phrase "far from satisfied" appeared without prompting. The one place the posture turned assertive was the Depop marketing decision, which was defended as a deliberate choice rather than presented as a cost. The succession announcement was handled without defensiveness, and the incoming CEO spoke about a strategy she had already been running for a year rather than one she was unveiling.

1. The CEO Transition: A Handoff, Not a Rescue

The largest news of the morning was not in the numbers. After eight and a half years, Josh Silverman is stepping down as CEO effective January 1, 2026, moving to Executive Chair. Kruti Patel Goyal, currently President and Chief Growth Officer and previously CEO of Depop, takes over. She has been at Etsy for more than fifteen years across two stints, spent roughly two years in London running Depop, and returned to Etsy last year specifically to own growth.

"After 8.5 years as Etsy's CEO, I've decided that this is the right time for me to hand over the leadership baton to the next generation." — Josh Silverman, CEO

The structural detail that matters more than the announcement itself: the strategy the incoming CEO will execute is the strategy she already authored. The four priorities described on this call, showing up where shoppers discover, matching them with the right items, deepening loyalty, and amplifying human connection, are hers. She has been running day-to-day operations for months. The 300bp sequential improvement in Etsy GMS growth is being attributed to initiatives she defined. This is materially different from an outside hire arriving with a blank slate, and it substantially reduces the usual risk that a transition produces a strategic reset and a lost year.

"If I was the right person to lead Etsy's last chapter, Kruti is the right person to lead our next." — Josh Silverman, CEO

The counterweight is timing and interpretation. Silverman ran the 2017 turnaround and the pandemic scaling, and he is leaving with GMS below its peak and the buyer count falling. Whatever the internal logic, the market reads a CEO departure announced simultaneously with a soft guide as an admission, and it read this one that way.

Assessment: Constructive on the merits, poorly timed on the optics. Continuity of strategy is the important variable and it is intact; the incoming CEO's Depop record (three flat years turned into two accelerating ones) is the most relevant credential available. But she now owns a 2026 she has not yet framed publicly, and the market will not extend the benefit of the doubt indefinitely. The first genuinely informative event of her tenure is the 2026 outlook, which was conspicuously not given on this call.

2. Take Rate: The Engine, and Its Ceiling

Consolidated take rate reached 24.9%, up 220bp year-over-year and 90bp sequentially, and 40bp ahead of the guided 24.5%. Management attributes the expansion to two things: the Reverb divestiture, which removed a structurally lower-take-rate business from the mix, and growth in services and ads revenue at both Etsy and Depop. Services revenue grew 12.7% year-over-year on Etsy Ads model enhancements that optimize seller budget pacing and improve ad relevance, plus Depop's expanding boosted-listings product.

This is the mechanism by which Etsy produced revenue growth and a revenue beat in a quarter when volume fell. It works, it is well executed, and it has a limit. Take rate is the share of a seller's sale that Etsy keeps. Etsy raised it 220bp in a year during which its seller count fell 10.9%. Those two facts are not necessarily causal, and management offered a partially reassuring counterpoint: seller satisfaction scores are up more than 10 percentage points year-over-year, and new sellers making a first sale grew at strong double-digit rates. But the direction of travel is a marketplace taking a larger cut from a smaller seller base, and there is no version of that which runs indefinitely.

Assessment: Take-rate expansion is the highest-quality earnings lever Etsy has and the one with the shortest runway. Q4 take rate is guided to step down to 24.5% on seasonality, which is normal, but it also means the 24.9% is not yet a new floor. The number to watch in 2026 is whether take rate can hold in the mid-24s while seller counts stabilize. If take rate keeps rising while sellers keep falling, that is monetization borrowing from supply.

3. The Q4 Margin Guide and What It Reveals

Q4 adjusted EBITDA margin was guided to approximately 24%, against a Street carrying roughly 27%. This, more than the CEO news, is what a portfolio manager should sit with.

"Consolidated adjusted EBITDA margin will be approximately 24% reflecting stable, strong profitability for the Etsy marketplace paired with a significant sequential increase in brand marketing investment at Depop, which will compress margin performance." — Lanny Baker, CFO

Run the arithmetic. At the GMS guidance midpoint of $3.575B and a 24.5% take rate, implied Q4 revenue is approximately $876M. At a 24% adjusted EBITDA margin, implied Q4 adjusted EBITDA is approximately $210M. Q4 2024 delivered $250.6M of adjusted EBITDA at a 29.4% margin on $852.2M of revenue. On a reported basis that is roughly 3% revenue growth and roughly a 16% decline in adjusted EBITDA, with margin down about 540bp year-over-year in the company's single most profitable quarter. Both readings carry the same basis break, and it pushes them in opposite directions: the prior-year column includes Reverb and the guide does not, so like-for-like revenue growth is mid-single-digit rather than 3%, while Reverb's removal also lifts the consolidated take rate it was dragging down.

No row in that table is apples-to-apples: the prior-year column includes Reverb and the guide does not, which is why management frames the GMS midpoint as further sequential improvement in the ex-Reverb growth rate while a reported-basis read shows a low-single-digit decline. Both framings are defensible. The same break runs through the revenue and margin rows, and Etsy has not disclosed Reverb's own margin contribution, so the 540bp is the reported-basis deterioration rather than a clean like-for-like measure. What is not in question is the direction, because management guided it: a materially lower Q4 margin.

Management was explicit that the spend is a choice and is reversible, which is the right disclosure and also the point.

"Now if things look a lot better than we thought, we might step it up. If it doesn't look like it's going as we expect, we can be very maneuverable on our spending there." — Lanny Baker, CFO

Assessment: The investment logic is sound. Depop's unit economics are proven, the resale market is large, and awareness is the binding constraint. But the market had been valuing Etsy partly as a high-margin cash machine with a fixed profitability floor, and this guide reprices that floor as a variable under management control. Forecasting a company whose margin is a strategic dial is harder than forecasting one whose margin is a constraint, and multiples compress for exactly that reason. Investors did not lose 300bp of Q4 margin here; they lost the assumption that 29% was the base case.

4. Agentic Commerce: First Mover, Undisclosed Economics

Etsy became the first live partner for OpenAI's instant checkout feature in September, letting users complete purchases inside ChatGPT. Etsy provides a dedicated product feed covering the substantial majority of its listings; transactions run through Etsy Payments and surface as guest checkout, with Etsy branding and the human seller's information attached to the confirmation.

"Agentic visits represent a small slice of e-commerce traffic today but they're growing quickly. An early analysis suggests that these buyers come to Etsy with higher purchase intent than those from traditional search." — Kruti Patel Goyal, President and Chief Growth Officer

The strategic case management makes is that Etsy is structurally advantaged in agentic commerce rather than threatened by it, because agents commoditize commodities. Where a product is identical across ten retailers, an agent reduces the decision to price and delivery speed. Etsy's inventory is largely unique, so an agent presenting a small curated set has a reason to include it.

The economics were disclosed only in structure, not in magnitude. Etsy pays OpenAI a success-based transaction fee per purchase, described as comparable to what it would pay an affiliate or influencer, and is not currently passing that cost through to sellers.

"We pay a commission, a success-based transaction fee to OpenAI for each of those purchases. And it's not unlike what we would pay to an influencer, for example, or an affiliate." — Josh Silverman, CEO

Assessment: Correct posture, unquantified risk. Being first to the table is genuinely valuable when the presentation format, the data contract, and the branding conventions of agentic commerce are all still being written, and Etsy's unique inventory is a real structural edge. But the rate is undisclosed, Etsy is currently absorbing it rather than passing it through, and any material shift of demand from the Etsy app to a third-party agent moves transactions out of the surface where Etsy sells its highest-margin product, which is seller advertising. The advertising pool is the same services revenue growing 12.7% and carrying the take rate. Agentic commerce is a growth channel and a margin-mix risk at the same time, and only the first half of that has been quantified.

5. The App as the Real Growth Program

The most concrete operational progress in the quarter is in the app, and the disclosed metrics are unusually specific.

"In the third quarter, on average, Etsy's app users visited roughly 5x more often than our non-app users. They viewed 3x more pages per visit and were 1.5x more likely to convert with a purchase." — Lanny Baker, CFO

App GMS outperformed non-app GMS by 13 percentage points. App home-screen GMS grew 20% year-over-year. The app's share of total GMS rose to 46% from just under 45% a quarter ago and 42.8% a year ago. Downloads grew 9%, faster among new buyers. Behind it sits a redesigned home screen, simplified navigation, and a rebuilt discovery feed powered by new machine-learning models in two areas management calls buyer understanding and item understanding.

A 320bp year-over-year gain in app share of GMS is worth roughly $87M of GMS shifted onto a surface that converts 1.5 times better and visits five times more often. That mix shift is also what feeds the owned-marketing channels, email and push notifications, which the CFO described as becoming high-return channels that drive attributed GMS without external ad spend. There is a coherent flywheel here: app engagement produces data, data improves targeting, targeting improves owned-channel returns, owned channels reduce paid-marketing dependence.

Assessment: This is the most credible internal growth program Etsy has, and it is the mechanism most likely to be responsible for the 300bp GMS improvement. It is also, at 46% of GMS, roughly at the halfway point of its own runway. The relevant question for 2026 is whether app share can keep climbing 300bp a year and whether the conversion advantage survives as the marginal app user becomes less committed than the current cohort.

6. De Minimis Expiry and the Tariff Backdrop

The US de minimis exemption expired at the end of August, removing duty-free treatment for low-value imports. For a marketplace where a meaningful share of US demand is served by non-US sellers, this is a direct headwind.

"The expiration of the de minimis exemption at the end of August weighed on performance immediately thereafter, and our business stabilized as we moved through the quarter, resulting in only a modest headwind to quarterly results." — Lanny Baker, CFO

Management's explanation for the resilience is inventory substitution: over 60 million active US listings, a very high replacement rate on disrupted items, and proactive outreach helping non-US sellers find viable shipping alternatives. Sequential improvement in the US domestic trade route offset pressure on the US import route. The word "modest" is doing real work here and was never converted into a number, even when asked directly.

Assessment: The substitution argument is structurally sound and is a genuine advantage of a long-tail marketplace over a curated one: when a listing becomes uneconomic, another seller's listing fills the slot, and GMS is preserved even if the individual seller is not. The risk this obscures is compositional. If import listings are being replaced by domestic listings at different price points and different take-rate mixes, the GMS number can hold while the underlying seller base churns. Nothing disclosed lets an outsider check that, and the absence of a quantified impact after a direct question is the kind of omission worth carrying forward.

7. Loyalty and the GMS-per-Buyer Problem

GMS per active buyer is $121 on a trailing-twelve-month basis and has been flat all year. Management's stated answer is the Etsy Insider loyalty program, now entering a second beta iteration restructured around frequent buyers, offering shipping discounts and 5% back in Etsy credit on every purchase.

The CFO framed the opportunity in terms that are honest about the size of the gap.

"I would start out by thinking about Etsy generating $120 in GMS per buyer on a 12-month time frame. And our opportunity to increase purchase frequency through rewards and through incentives and through retention benefits, that's a huge lever for us. We feel like that number — we know we've got plenty of customers who are spending 5 and 10x that amount with us in a given year." — Lanny Baker, CFO

The first beta taught management that rewards drove both adoption and a meaningful uplift in purchase frequency. The second beta reorients toward the most frequent buyers specifically. What has not been disclosed is membership, cost per member, or incremental GMS per member, so the program's economics are entirely unauditable from outside.

Assessment: $121 of annual spend per buyer is a genuinely low number for a marketplace of Etsy's category breadth, and closing even part of the gap to the heaviest cohort is a larger lever than anything else on this list. But 5% back plus shipping discounts is a real margin cost per transaction, and a loyalty program that succeeds at driving frequency among buyers who would have purchased anyway is a margin transfer, not growth. Management said it is "seeking to deliver sustainable economics for Etsy," which is the right ambition and not yet a result. Watch for a disclosed member count.

8. Capital Allocation: The Buyback Is Doing the Heavy Lifting

Etsy repurchased 2.1 million shares for approximately $120M in the quarter, generated more than $200M of free cash flow, and finished with $1.6B of cash and investments against approximately $3.0B of convertible debt. The cumulative effect of the program is substantial.

"you can see in the chart on the right that we've reduced our share count by 17% since December of 2023 through our stock buyback program, delivering significant value to our shareholders" — Lanny Baker, CFO

Free cash flow was $635M on a trailing-twelve-month basis. Against a fully diluted market capitalization of roughly $8.15B at the post-print price, that is a 7.8% free-cash-flow yield; against the basic share count it is 9.8%. Asked about M&A, management pointed back at the buyback and at the two brands it already owns rather than signaling appetite for a third.

Assessment: This is the floor under the stock and it is a real one. A 17% reduction in share count in under two years means that a business with flat GMS still delivers mid-single-digit growth in per-share cash flow, which is why EPS can grow 40% in a quarter when operating income falls. The caution is that a buyback is an accelerant, not a thesis: it multiplies whatever the underlying trajectory is. Retiring shares of a business whose GMS compounds negatively converts a slow decline into a slower one, not into growth. The $3.0B convertible balance also means the fully diluted count is materially above the basic count, which mutes the per-share benefit relative to the headline retirement figure.

9. Seller Health: The Metric That Turned

Active sellers fell 10.9% year-over-year to 5.5M, the worst-looking number in the release. Underneath it, three things improved. Sellers grew 1.7% sequentially, with both US and international counts returning to sequential growth. The number of new sellers making a sale grew at strong double-digit rates year-over-year. And the percentage of total sellers with a sale in the last twelve months rose year-over-year, rebounding from a trough in Q1 2024.

Management paired that with an operational claim: seller satisfaction scores are up more than 10 percentage points from a year ago, attributed to new AI tools that generate listing titles and draft buyer messages, plus improved access to live support.

Assessment: The year-over-year seller decline is largely a trailing-twelve-month artifact of a purge that has already happened; the sequential turn is the live signal and it turned positive. More important than the raw count is the rising share of sellers who actually transact, which is the quality measure. A smaller, more active seller base is a better marketplace than a larger, dormant one. This is the one place in the release where the trailing metric looks worse than the business.

10. The Consumer: Strength Concentrated at the Top

Management sliced GMS by estimated household income and reported a favorable year-over-year inflection across all income levels, with the strongest performance among the highest-income buyers. Category strength was in Vintage Home & Living, jewelry above $100, and wedding and engagement rings. Comparisons improved sequentially across most of the top six categories.

Those are high-ticket, discretionary, gift-adjacent categories, and the pattern is consistent with the broader bifurcated-consumer picture. It also means Etsy's improvement is disproportionately exposed to the cohort most sensitive to asset prices rather than to wage growth.

Assessment: Constructive for the near term and a concentration risk for the medium term. High-income strength is why the quarter beat and why the Q4 guide can be framed on a stable-consumer assumption. It also means the Q4 range is wider in practice than it looks: management explicitly said the upper end assumes a healthier consumer and the lower end a more challenging one, and the swing factor is a cohort whose spending correlates with markets that are themselves near highs.

Guidance & Outlook

Management's framing of the Q4 GMS range is that the midpoint represents further sequential improvement in the apples-to-apples growth rate, both consolidated and for the Etsy marketplace specifically. The upper end assumes a healthier consumer and more traction from the internal initiatives; the lower end assumes a more challenging consumer environment. The whole range is built on an explicit assumption that consumer health stays roughly where it has been year to date.

Implied sequential ramp: going from $2,724.7M of Q3 GMS to a $3,575M Q4 midpoint requires 31.2% sequential growth. For context, Etsy's Q4 is structurally its holiday quarter, so a ramp of that order is normal rather than aggressive. The judgment call embedded in the guide is not the seasonal step-up, it is the assumption that the year-over-year comparison keeps improving through the highest-volume weeks of the year, which is when a marketplace has the least control over its own outcome.

Street at: the revenue and GMS ranges were broadly in line with what the market was carrying. The gap was entirely on profitability, with the ~24% adjusted EBITDA margin guide landing roughly 300bp below a Street at ~27%. That single line accounts for most of the reaction.

Guidance style: the Q3 guide given in July (GMS $2.60–2.70B, take rate ~24.5%, adjusted EBITDA margin ~25%) was cleared on all three metrics, which is one data point suggesting a conservative posture. One quarter is not a pattern, and we have no basis yet for assuming a habitual sandbag. What is notable is that management chose to disclose the Depop spend as a margin guide-down rather than absorbing it quietly, which is the more credible of the two available approaches and cost them 12.8% in a session.

What was not guided: no 2026 framework of any kind was offered. No preliminary GMS growth expectation, no margin range, no indication of how long the Depop brand investment runs beyond "the first half of next year." For a company changing CEOs in nine weeks, the absence of even a directional 2026 marker is the single largest information gap in the release.

Analyst Q&A Highlights

Whether the OpenAI Fee Is an Offsite Ads Extension or a New Cost Line

The opening question went straight to the economics of the agentic-commerce partnership rather than its strategic merit, asking whether the fee Etsy pays OpenAI is best understood as an extension of the offsite advertising program that sellers help fund, or as a cost Etsy absorbs on its own income statement. The answer settled the accounting question and left the magnitude open.

Q: "And maybe secondly, can you maybe help us think through sort of the performance fee since Etsy's covering this fee, is it fair to think about this sort of an extension of your offsite ads? Or are there any sort of transaction cost on your end that you will not be incurring sort of to compensate for this additional fee?"
— Maria Ripps, Canaccord Genuity

A: "We pay a commission, a success-based transaction fee to OpenAI for each of those purchases. And it's not unlike what we would pay to an influencer, for example, or an affiliate. So it's a CPA-based transaction fee. At this time, we are not passing that through to our sellers. So it's not part of our offsite ads program, and we'll consider as time goes on, how to think about that."
— Josh Silverman, CEO

Assessment: The important phrase is "at this time." Etsy is currently eating a cost that its offsite ads program was purpose-built to share with sellers, and management explicitly reserved the right to revisit that. While agentic volume is immaterial the choice is costless; if it scales, Etsy faces a decision between absorbing a growing per-transaction fee or passing it to a seller base whose count is already down 10.9%. Neither branch is free, and the fee rate remains undisclosed, so the exposure cannot be sized from outside.

How the Holiday Guide Was Constructed

A recurring line of questioning probed whether the sequential improvement in core GMS carries into the fourth quarter given easier comparisons, and how management approached the holiday given last year's compressed calendar. The response was notable for how much of the range it attributed to consumer uncertainty rather than to internal execution.

Q: "So great to hear about the sequential improvement at core Etsy GMS. Are you seeing further improvement here in 4Q, just given the momentum of the initiatives and the easy compares that the business is lapping?"
— Anna Andreeva, Piper Sandler

A: "As we said, the consumer outlook remains uncertain. There's probably a higher degree of uncertainty around the consumer going into this holiday season than there has been for prior seasons, for some prior seasons anyway. But our assumption is that, if the consumer health stays about where it has been year-to-date into the fourth quarter."
— Lanny Baker, CFO

Assessment: Management drew the range around a macro variable rather than around the initiatives, which is the honest construction and also the one that offers the least forward visibility. The bull and bear ends of the GMS guide are separated by $150M, and management explicitly assigned that spread to consumer health plus initiative traction rather than to anything it controls. Investors should treat the Q4 print as a read on the US discretionary consumer at least as much as a read on Etsy's turnaround.

Whether Industry-Leading Profitability Should Be Spent to Buy Growth

The most strategically loaded question of the call was directed at the incoming CEO and asked, in effect, what her margin philosophy will be: whether Etsy's peer-leading profitability represents stored optionality that should be deployed to reignite growth, or a structural advantage worth preserving. The answer came with a worked example attached rather than a principle.

Q: "And then, Kruti, maybe it's early for this question, but Etsy's profitability is some of the highest among the peer set. Just strategically, do you think there's an opportunity to invest more just to reignite growth even faster? Or do you think it's important to preserve profitability?"
— Anna Andreeva, Piper Sandler

A: "it is early to comment on the specifics there. But what I will say is that we have seen the profitability of our marketplace is the real strength of our marketplace. And we've always been really disciplined and thoughtful about our investments. And so I would point to Depop as really a great example of where we've been willing to make deeper investments where we think there is a lot of opportunity."
— Kruti Patel Goyal, President and Chief Growth Officer

Assessment: This is the closest thing to a 2026 framework offered on the call, and it points toward more spending, not less. Naming Depop as the template for how deeper investment gets made, in the same quarter that Depop investment is guided to cut roughly 300bp off the consolidated margin, is a signal about disposition rather than a commitment about magnitude. The answer preserves flexibility, which is appropriate nine weeks before taking the job, and it should also lower any investor's confidence that 29% margins are the resting state.

Whether Agentic Commerce Disintermediates the Buyer Relationship

A pointed line of questioning moved past the near-term traffic opportunity to the structural concern: that routing demand through third-party agents erodes direct buyer relationships and, with them, the on-site advertising profit pool that marketplaces depend on. The response leaned on brand pull-through and on the slow pace at which consumer habits actually change.

Q: "But investors are starting to worry longer term about what this means from a retention of buyer perspective and what it means for direct traffic and core profit pools for businesses like yours with on-site ads. So when you evaluated this partnership, how did you kind of think about the risk/reward longer term of something like this?"
— Nikhil Devnani, Bernstein

A: "It's also been my experience from decades of doing this, that customer behavior changes much more slowly than people think. We will have an app, I believe, for many years to come, and people are going to shop agenticly. And both are going to be good and both are going to be complementary."
— Josh Silverman, CEO

Assessment: The answer addressed the traffic question and stepped around the profit-pool question, which was the substance of what was asked. The claim that "customer behavior changes much more slowly than people think" is a defensible read on adoption pace and says nothing about what happens to seller advertising revenue if and when agentic share does become material. On-site ads are inside the services line growing 12.7% and carrying the take-rate expansion, so this is not a peripheral concern. The honest position is that nobody knows yet; the answer given implied more confidence than the disclosure supports.

Duration and Reversibility of the Depop Brand Investment

The Q4 margin guide drew a direct question on whether the Depop marketing step-up is a one-quarter holiday phenomenon or a structural change to the cost base. The answer removed the one-quarter interpretation immediately and put a soft floor under the commitment.

Q: "And then, Lanny, can you maybe unpack a little bit more the nature of the investments planned for Depop in Q4, which will compress margins near term? Is this a 1 quarter type of phenomenon around the holidays? Or is this more structural going forward?"
— Youssef Squali, Truist Securities

A: "And so it's a brand campaign. It's not going to be concluded in 3 months. It will carry over into the first half of next year. It will take some time, frankly, for us to be able to really assess how we're performing, and we're committed to staying with that brand investment through the time that it will take to be sure that we're really expanding awareness of that product."
— Lanny Baker, CFO

Assessment: This is the answer that should reset 2026 models, and it is why the reaction was as large as it was. The spend is committed through at least H1 2026, and management said explicitly that it will take time before performance can even be assessed, which means there is no near-term checkpoint at which the investment can be judged and withdrawn. Brand campaigns are the hardest marketing dollars to attribute; committing to one through a period in which you concede you cannot yet measure it is a governance choice as much as a marketing one. It is defensible for an asset management genuinely believes can become multi-billion-dollar, and it removes roughly two quarters of margin visibility.

Sizing the Loyalty Opportunity Against a Flat GMS-per-Buyer

Questioning on the loyalty program pushed past the mechanics to the sizing: what the learnings from the first beta were, and how large the GMS opportunity attached to loyalty could be over time. The response reframed the whole question around a single number.

Q: "What have been your key learnings so far as you've rolled out more loyalty on the platform? How do those learnings inform the innovation or iteration of loyalty you're announcing today that looks like it's going into beta this week? And how should we think about the GMS opportunity attached to loyalty over the longer term?"
— Eric Sheridan, Goldman Sachs

A: "What we've learned from the initial version is that the rewards that we offered both drove adoption of the loyalty program and drove a meaningful uplift in frequency of engagement and purchasing. And so that was a really important learning from this first version. What we're changing in this next version is we're really reorienting it to really focus on our more frequent buyers."
— Kruti Patel Goyal, President and Chief Growth Officer

Assessment: The learning disclosed is that rewards drive behavior, which is the least surprising possible finding and does not address whether the behavior is incremental or subsidized. Reorienting version two toward the most frequent buyers is a deliberate narrowing: it maximizes retention of the highest-value cohort and minimizes the number of marginal buyers whose purchases get discounted. That is the right design if the goal is defending the core, and it is the wrong design if the goal is lifting a $121 average. Both goals were invoked on this call. Membership counts and cost per member remain undisclosed.

Quantifying the De Minimis and Exogenous Impact

A closing question asked management to size the exogenous factors in the quarter, specifically the de minimis expiry, to let analysts separate underlying progress from external drag. The answer described the shape of the impact without ever attaching a number to it.

Q: "And then, Lanny, a quick follow-up. Can you just maybe help us size a bit the exogenous factors, the de minimis impact to the extent you saw it over the course of Q3?"
— Robert Coolbrith, Evercore ISI

A: "And from a trade and tariff perspective, there — we've seen the resilience in Etsy marketplace of stronger growth in U.S. domestic with a little bit slower growth in the international imports in the United States, right around a couple of weeks of the end of the de minimis exemption, we saw a little bit of pause in the marketplace, but that came back and was a very modest headwind for the total quarter."
— Lanny Baker, CFO

Assessment: Asked to size it, management characterized it. The phrase "a very modest headwind" is the third distinct qualitative framing of the same impact on this call and there is still no basis point attached. The analytically important consequence is that the 300bp sequential improvement in Etsy GMS growth cannot be decomposed: some unknown portion of it is initiative traction and some unknown portion is the trade-route mix normalizing after an August disruption. Without the split, the durability of the improvement is unverifiable, which is exactly the uncertainty a Hold is designed to price.

What They're NOT Saying

1. Any 2026 framework at all. A CEO transition takes effect January 1 and the call passed without a single directional marker for next year: no GMS growth expectation, no margin range, no revenue framing. The Depop brand spend was said to run into H1 2026, which is the only forward statement of substance, and it is a cost commitment rather than an outlook. For a company asking investors to underwrite a new leader, giving them nothing to underwrite against is a choice.
2. The OpenAI commission rate. Described as CPA-based and "not unlike what we would pay to an influencer, for example, or an affiliate," which spans a wide range. Etsy is absorbing it rather than passing it to sellers, so it lands directly in Etsy's own cost of revenue or marketing line. At current volumes it is immaterial; the point of disclosure is to let investors model what happens when it is not.
3. The dollar size of the Depop brand campaign. The margin impact is guided (roughly 300bp below Street at the consolidated level), but the spend itself is not quantified, nor is the awareness baseline it is meant to move, nor the metric by which success will be declared. Management said it will take time before performance can be assessed, which means there is no disclosed checkpoint.
4. Quantification of the de minimis impact. Asked directly, management answered with "a very modest headwind." Without a number, the 300bp GMS improvement cannot be split between execution and trade-route normalization.
5. Etsy Insider membership, cost, or incremental GMS. Version two of the loyalty beta launches with a 5% credit and shipping discounts on every purchase, which is a real per-transaction margin cost, and no membership count, retention rate, or incrementality figure has been disclosed at any point in the program's life.
6. What the 10.9% seller decline did to listing supply. Management pivoted immediately to the sequential rebound and to seller satisfaction scores, both of which are legitimate. Neither addresses whether total active listings, the actual inventory a buyer sees, grew or shrank. Total listing count was not disclosed.
7. Habitual buyers in context. The 6.0M trailing-twelve-month habitual buyer figure was disclosed without a year-over-year comparison, in a release where every other customer metric carried one. Habitual buyers are the cohort that matters most to the loyalty strategy, which makes the missing comparison conspicuous.
8. Whether take rate has a stated ceiling. Take rate rose 220bp year-over-year while sellers fell 10.9%, and no one on either side of the call connected those two facts or asked management where the monetization ends.

Market Reaction

• Pre-print setup: ETSY closed at $74.78 on October 28, up 41.4% year to date against the S&P 500's 17.2%, up 57.5% over the trailing twelve months, and up 0.6% over the trailing thirty days. The 52-week closing range entering the print was $40.80 to $75.56, so the stock was sitting within 1% of its 52-week closing high.
• Reaction session (October 29, the print and call day): the shares gapped down to open at $70.26, a 6.0% gap, traded a $65.20 to $72.73 range, and closed at $65.21, down 12.8% or $9.57. The close was one cent off the low of the day.
• Volume: 13.2 million shares against a 5.5 million 30-day average, 2.4 times normal.
• Market context: the S&P 500 was unchanged on the session, so the entire move was idiosyncratic.

The shape of the session is more informative than its size. The stock gapped 6% lower on the release and the leadership announcement, then lost another 7% over the course of the trading day and closed within a penny of the low. That is not a first-reaction overshoot that stabilizes as the details are digested; it is a market that liked the print less the longer it looked. The obvious candidate for the second leg is the call itself, where the Depop investment was confirmed to run into the first half of 2026 and the incoming CEO declined to frame 2026.

The pre-print setup explains the magnitude. A stock up 41% year to date and sitting a percent below its 52-week closing high has no cushion: the positioning coming in assumed the turnaround was working, and the quarter beat every guided metric while conceding that operating income and adjusted EBITDA both fell year-over-year and that margins would fall further. Ownership that has been paid for improvement is unusually sensitive to a guide that defers it.

What did not happen is also worth noting. This was not a valuation de-rating in search of a floor. At $65.21 the shares still sit 23% above where they started the year and 60% above the 52-week closing low. The market moved Etsy from pricing an inflection to pricing an option on one, which is a smaller move than it feels like on a 12.8% day.

Street Perspective

Debate: Is the Narrowing GMS Decline an Inflection or a Comparison?

Bull view: Two consecutive quarters of improving year-over-year GMS comparisons, 300bp in the most recent one, with the ex-Reverb consolidated line crossing back into positive growth. The improvement traces to identifiable product work with disclosed metrics behind it: app share of GMS up 320bp year-over-year, app home GMS up 20%, a rebuilt discovery feed driving double-digit engagement gains. The Q4 guide midpoint implies the trend continues. This is what the early innings of a turnaround look like.

Bear view: The comparisons are getting easier and the trade-route disruption of August is normalizing, and management has quantified neither effect, so none of the 300bp can be attributed with confidence. Meanwhile the underlying customer base is still eroding: buyers down 5.0%, GMS per buyer flat at $121 all year. A GMS decline that narrows while both of its inputs deteriorate is a mix effect, not an inflection.

Our take: The bear has the better of it on evidence, the bull on direction. The improvement is real and is more likely than not partly attributable to the app and discovery work, which are genuine and measurable. But the refusal to size the de minimis impact after a direct question makes the decomposition impossible, and the buyer count is the metric that ultimately governs. We need to see the Etsy marketplace GMS growth rate cross zero and the active buyer count stop falling sequentially in the same quarter. Q3 delivered neither.

Debate: Is a ~24% Margin the New Floor or a One-Off Investment Window?

Bull view: The spend is discretionary, opportunistic, targeted at a $1B-run-rate asset growing 39% with 40% user growth on both sides, and explicitly reversible; management said it can be "very maneuverable." Etsy marketplace profitability itself was described as stable and strong, just shy of a 30% segment margin. Consolidated margin is being temporarily diluted by an investment in the fastest-growing asset in the portfolio, which is the correct use of a high-margin core.

Bear view: Brand campaigns are the least attributable marketing dollars there are, management conceded it will take time before performance can be assessed, and the spend is already committed through H1 2026. The core marketplace is not growing, so this is not surplus profitability being redeployed, it is a shrinking base funding a growth experiment. Once a margin has been shown to be negotiable it never fully recovers its status as a floor, and the multiple adjusts accordingly.

Our take: Both sides are describing the same fact with different time horizons, and the resolution is empirical. The strategic logic is sound and Depop is the right asset to press on. But the market's repricing was not irrational: the difference between a business with a 29% margin floor and a business whose margin is a management dial is a real difference in the multiple that business deserves, independent of whether this particular dial-turn pays off. We treat roughly 24% as the operating assumption through H1 2026 and would need to see Depop's US GMS growth hold near 59% into the second quarter of the campaign to underwrite it as money well spent.

Debate: Does Agentic Commerce Help Etsy More Than It Threatens It?

Bull view: Etsy's inventory is unique rather than commodity, so an agent presenting a small curated set has a structural reason to include it, and Etsy gets brand exposure in purchase occasions where it was previously never considered. Early data shows agent-sourced traffic converting better than organic search traffic. Being the first live partner puts Etsy in the room while the presentation format, data contracts, and branding conventions are still being written, which is worth more than the near-term volume.

Bear view: The volume is currently trivial, the fee is undisclosed and currently absorbed by Etsy rather than shared with sellers, and the medium-term risk is to the highest-margin revenue line in the business. Etsy Ads is sold against on-site attention. Demand that clears inside a third-party agent does not generate an on-site ad impression. Services revenue growing 12.7% is what is carrying the whole revenue line right now, and that is precisely what agentic commerce erodes at scale.

Our take: The bull case is right about the next two years and the bear case is right about the risk that matters. Nothing about agentic commerce threatens Etsy's 2026, and the option value of being first is genuine. But the question asked on this call was specifically about the on-site advertising profit pool, and the answer was about adoption pace, which is not the same thing. We score this as a genuine long-term uncertainty that management has not yet engaged with substantively, and we would upgrade our view of it the moment Etsy discloses either the fee structure or an explicit view on how ad monetization survives agent-mediated demand.

Model Update & Valuation Framework

This is an initiation, so there is no prior model to revise. The table below states the assumptions we are initiating with and the basis for each.

Valuation framework: at the $65.21 post-print close, Etsy carries a fully diluted market capitalization of approximately $8.15B on the 125.0M diluted share count, or approximately $6.48B on the 99.3M basic count. The gap between the two is the convertible structure: roughly $3.0B of convertible debt sits against $1.6B of cash and investments, and the if-converted treatment pulls those shares into the diluted count. Against $635M of trailing-twelve-month free cash flow, that is a 7.8% free cash flow yield fully diluted and 9.8% basic. For a business with 71% gross margins, negative working capital characteristics, and a demonstrated willingness to return essentially all of its cash flow, both of those are cheap in absolute terms.

Why cheap is not enough here: a free cash flow yield is only a floor if the cash flow is stable. Etsy's is currently produced by a marketplace whose volume is shrinking, monetized at a take rate that has risen 220bp in a year against a seller base down 10.9%, with a margin that management has just demonstrated it will spend down when it sees an opportunity. Each of those three is individually defensible and together they mean the denominator of the yield is less anchored than the headline suggests. The bear case does not require a catastrophe; it requires the GMS decline to resteepen while take rate plateaus, which produces $520M of free cash flow and, at a multiple appropriate to a declining marketplace, a $45–52 stock.

Risk/reward: base case $70–78 implies +7% to +20% against the $65.21 close. Bull case $95–105 implies +46% to +61%. Bear case $45–52 implies −20% to −31%. The base-to-bear ratio is roughly 0.5 to 1, which is the arithmetic definition of a Hold. The bull case is genuinely attractive and requires two specific things to happen that have not yet happened: Etsy marketplace GMS growth crossing zero, and the margin recovering after the Depop campaign. Neither is unreasonable. Neither is visible in this print.

Thesis Scorecard: Establishing Coverage

This is our first published view on Etsy, so there is no standing thesis to grade. The table below establishes the pillars we will carry forward and score every quarter from here.

Overall: thesis established. Four bull pillars, three confirmed on the print and one awaiting an outcome. Five bear pillars, two confirmed and three neutral-but-live. The balance is genuinely even, which is why the rating is Hold rather than a directional call in either direction. The company generates a great deal of cash from a business that is getting smaller, and it has just told investors it intends to spend some of that cash on making a different business bigger.

Action: hold. Do not chase the 12.8% decline. The valuation floor is real and the strategic direction is coherent, but the two facts that would convert this to Outperform, a positive Etsy marketplace GMS growth rate and a stabilizing buyer count, are both absent, and the incoming CEO has not yet published the 2026 plan against which she should be judged.

Bottom Line: Initiating at Hold

Rating decision: we initiate coverage of Etsy at Hold. The quarter beat every metric management guides to and simultaneously showed operating income down 4.7% and adjusted EBITDA down 6.4% year-over-year. Those two statements are both true and the second one is the business. Net income grew 31.6% and earnings per share grew 40.0% on a currency comparison, a lower tax rate, and a smaller share count, with no operational contribution at all.

What makes this a Hold rather than an Underperform is the cash. A 7.8% fully diluted free cash flow yield, a 71% gross margin, a buyback that has retired 17% of the shares in under two years, and $1.6B of cash and investments constitute a genuine floor. What makes it a Hold rather than an Outperform is that every mechanism producing the current earnings has a visible limit: take rate cannot rise forever against a falling seller count, the tax rate and currency comparisons do not repeat by right, and the buyback multiplies the underlying trajectory rather than changing it.

The most important thing in this print is not the CEO change. Internal succession with strategic continuity is close to the best version of a leadership transition available, and the incoming CEO's Depop record is directly relevant to the problem she has been handed. The most important thing is the Q4 margin guide, because it converts Etsy from a company with a profitability constraint into a company with a profitability policy. Investors were valuing the former. The 12.8% decline is the difference.

What would change our mind:

• Upgrade to Outperform: Etsy marketplace GMS growth crossing zero with active buyers flat or growing sequentially in the same quarter; or Depop sustaining 35%+ GMS growth with segment margin recovering as the brand campaign annualizes; or a 2026 framework from the incoming CEO that pairs a credible GMS growth path with a defined margin floor.
• Downgrade to Underperform: Etsy marketplace GMS decline resteepening beyond 4% with take rate flat or falling; or the Depop brand investment extending past H1 2026 without disclosed evidence it is working; or a Q4 GMS print below the low end of the $3.50B guide, which would signal the consumer assumption underpinning the whole outlook is wrong.

Signposts for Q4 2025 earnings (February 2026):