Twenty for Twenty, and a Q3 Guide Five Points Above the Street: Shopify Takes Back Half the De-Rating in a Day

Results vs. Consensus

Q2 2026 Scorecard

GAAP net income of $1,502M is again distorted, this time favourably: equity investments contributed a $1,063M after-tax gain, the mirror image of the $941M charge that made last quarter's GAAP line a loss. On the company's own basis, net income excluding equity investments was $439M against $338M, up 29.9%. We continue to ignore the GAAP bottom line entirely for this company and read operating income and free cash flow instead.

Delivery Against Shopify's Own Guidance

Year-Over-Year Comparisons

One line in that table is going the wrong way and it is the same line that went the wrong way last quarter. Transaction and loan losses grew 76.3% against revenue growth of 33.7%, reaching 3.9% of revenue against 3.0%. Everything else compounds.

Quarter-Over-Quarter Comparisons

Quality of Beat

Revenue. A roughly 4% beat, and management decomposed it into three sources rather than leaving it to inference: broad-based GMV outperformance, higher-than-expected payments penetration, and strength in other merchant solutions driven primarily by partner revenue shares and financial services. That last item is worth marking, because it is the first time financial services has been named as a contributor to a revenue beat, and it is the same business whose credit losses are the one deteriorating line in the report.

Margins. The gross margin story remains resolved, with a wrinkle. Merchant solutions gross margin rose to 38.4% from 37.9%, the second consecutive year-over-year expansion, and management explained it in a way that reframes the whole two-year compression debate: payments mix pressure was more than offset by growth in higher-margin revenue streams, and international payments carry structurally better margins than domestic because of higher debit prevalence and lower interchange. Subscription solutions gross margin was 79.7% against 81.6%, down 187 basis points year over year but flat with Q1, and this is where the vast majority of Sidekick's AI costs land. The blended 91 basis point decline is therefore now a subscription-margin story rather than a merchant-margin story, which is a complete inversion of where we started coverage.

Operating leverage and cash. Operating expense at 34.0% of revenue is 367 basis points better year over year, with sales and marketing under 14% of revenue (a roughly 160 basis point improvement) even while the company funded additional merchant acquisition, and research and development, and general and administrative both improving as a share of revenue despite research and development absorbing the majority of internal AI spend. Free cash flow of $654M at an 18.3% margin exceeded the guide; the CFO stated the underlying expansion excluding the accounting change was roughly 150 basis points, driven by operating margin flow-through and offset by about a point of higher taxes.

Segment Performance

Revenue and Gross Profit by Segment

Merchant Solutions: Penetration 68%, and a Margin Explanation That Changes the Model

Merchant solutions grew 37.4% to $2,781M, contributing $757M of the $903M of total incremental revenue, or 84% of the quarter's growth. Payments penetration reached 68% of global GMV, three points higher year over year, and Shopify Payments launched in the UAE, bringing the total to 40 countries. European payments penetration rose more than 350 basis points year over year even though many of those markets launched only last year. Shop Pay GMV grew 53% and surpassed $400B of lifetime accelerated GMV in June. Cumulative gross payment volume through Shopify Payments crossed $1 trillion during the quarter.

The margin explanation is the analytically important part.

"Margin pressure from increased payments volume was more than offset by growth in higher-margin revenue streams. Also, to my earlier comments regarding local payment methods, our payments gross margins are generally higher in international markets due to the greater prevalence of debit transactions and lower interchange fees." — Jeff Hoffmeister, Chief Financial Officer

Assessment: This retires the last version of our original bear case. We initiated coverage arguing that every incremental point of payments penetration would be dilutive to gross margin by construction, and that the fastest-growing region monetized worst. The second half of that was wrong: international payments carry higher gross margins than domestic because debit prevalence and interchange economics differ. So the international runway, which is where essentially all remaining penetration upside sits, is margin-accretive rather than dilutive. Merchant solutions revenue as a percentage of GMV rose to 2.41% from 2.30%, so monetization per dollar of merchant volume continues to climb.

Subscription Solutions: Growth Accelerates Again, Sidekick's Cost Shows Up

Subscription solutions grew 22.3% to $802M, a fourth consecutive quarter of acceleration from 14.6% at initiation. The largest contributor was monthly subscriptions for standard plans on what management called a strong quarter of merchant net adds, with Plus subscriptions and variable platform fees contributing roughly equivalent dollar growth. MRR grew 19.5% to $221M with Plus at 34% of MRR and also growing 19%.

The margin line is where the AI investment becomes visible.

"As a reminder, the vast majority of AI costs related to merchant use of Sidekick appear in Subscription Solutions gross profit. We were able to hold gross margins at a relatively consistent level quarter-over-quarter while Sidekick usage scaled, which reflects some cost efficiencies and support as well as our ability to continue providing merchants unique AI solutions like Sidekick, while diligently managing cost." — Jeff Hoffmeister, Chief Financial Officer

Assessment: Two facts sit uneasily together and both are true. Subscription gross margin held flat sequentially while Sidekick's daily active merchants grew 3.6x and daily sessions grew 4.8x, which is genuinely impressive cost management. And it is down 187 basis points year over year, which is a real transfer of value from the income statement to the product. We opened this as a new risk last quarter and it is behaving exactly as named: an open-ended claim on the highest-margin segment, actively managed, not sized. The offsetting fact is that subscription revenue growth has accelerated for four straight quarters, so the product appears to be paying for itself in adoption.

Growth by Lens

Assessment: The retention figures are the most valuable disclosure Shopify has made in our four quarters of coverage, and they arrived without a question prompting them. A 92% five-year retention rate at $1M of annual GMV, rising to 97% at $10M, converts the qualitative "once they come to Shopify, they stay" claim into a number that can be modelled. Combined with the cohort statistic, that the 2015 cohort now does five times its original quarterly volume, the business is properly described as an annuity with an embedded growth rate rather than as a subscription base with churn. The eMarketer figure is the other one worth holding: capturing roughly half of incremental United States ecommerce dollars while holding 14% of the installed market is the definition of share gain, and it is sourced to a third party rather than to management.

Key Operating KPIs

Key Topics & Management Commentary

Overall Management Tone: Assured and unusually structured. The prepared remarks were built around a single organizing principle, that Shopify builds what most merchants need most of the time and opens the platform for everything else, and every product update was placed against it rather than listed. The disclosure quality also stepped up: retention rates, cohort compounding multiples and a third-party market-share statistic all appeared unprompted. The one topic handled with visibly less rigour was credit, where the loss line rose for a second consecutive quarter and was described as normalized.

1. The Q3 Guide, and Why It Landed So Far Above the Street

Q3 revenue is guided to a low-thirties growth rate. The Street was at 26.3%. On a Q3 2025 base of $2,844M the guide implies roughly $3.70B to $3.78B against a consensus near $3.59B. Management named the drivers as the same ones that worked in Q2 and, critically, removed the currency ambiguity that made the previous quarter's guide so easy to misread.

"We expect Q3 revenue growth in the low 30s year-over-year. The expected sources of growth are consistent with the drivers that we saw in Q2, broad-based across geographies, merchant sizes and channels. We do not expect any significant FX impact." — Jeff Hoffmeister, Chief Financial Officer

Three months ago the identical guidance line pointed the other way and was clouded by an assumption of half a point of currency tailwind against more than two points in the prior quarter. The stock fell 15.6% on a guide the Street itself considered in line on revenue. This quarter, with no currency asterisk to strip out, the same disclosure mechanism moved the stock 17% the other way.

Assessment: A guide four to seven points above consensus, on a clean currency basis, from a management team that has beaten the top of its guided range four times out of four, is the single most valuable piece of information in this release. It also rewrites the forward estimate curve: a Street modelling 26.3% for Q3 was almost certainly modelling something similar for 2027, and that number has to move. Most of the 17% move is a revision event, not a multiple event.

2. Agentic Commerce Refutes the Cannibalization Case

For four quarters the agentic bear argument has been that AI-originated orders substitute for search-originated orders, redistributing demand rather than growing it. Management addressed that directly with the one statistic that settles it.

"This is not just AI taking share of search. In fact, search remains one of our largest sources of buyer traffic to our merchants, and it's still growing. Traditional search sessions are up 1.3x over the past 2 years, holding roughly 1/3 of all storefront sessions. That is AI as a complement to search rather than a substitute for it." — Harley Finkelstein, President

The supporting data is the strongest yet: AI-driven traffic and AI-driven orders both tripled year over year; new buyer orders from AI channels arrive at nearly twice the rate of other channels; conversion from AI search runs roughly 80% higher than traditional organic search; catalog-powered AI searches convert at twice the rate of searches using scraped data; and half of AI-referred sessions land directly on a product description page, 2.5x the rate of traditional search.

The distributional finding is the one with the most strategic weight: 75% of AI-attributed orders in the quarter came from outside Shopify's top 100 categories. Management's explanation is mechanical rather than promotional. Search engines rank by popularity against a handful of keywords; agents make multiple calls into a structured catalog and match against the buyer's full set of constraints at once.

Assessment: This is the most complete answer to the agentic question in four quarters and it moves the argument from positioning to mechanism. If AI-originated demand disproportionately finds specialized long-tail products, and the long tail is the majority of Shopify's merchant base, then agentic commerce is structurally biased toward Shopify's customers rather than toward the large retailers who dominate keyword search. Traditional search still growing at the same time removes the substitution objection. The remaining honest caveat is unchanged and management repeated it: the volume is small relative to $116B of GMV, and there is still no absolute dollar figure, in the sixth consecutive quarter.

3. Credit Rises Again, and the Disclosure Gap Reaches Four Quarters

The one line moving the wrong way, for the second consecutive quarter. Transaction and loan losses were $141M, or 3.9% of revenue, against 3.0% a year ago and 3.7% last quarter. The full-quarter provision for transaction and loan losses in the cash flow statement was $94M against $44M, up 113.6% on revenue growth of 33.7%.

"Transaction loans and losses came in at 3.9% of revenue. It will scale with volumes in our payments, capital, and credit products. Capital was a larger driver this quarter, while loss rates and payments and credit are both at normalized levels." — Jeff Hoffmeister, Chief Financial Officer

There is genuine mitigation in the balance sheet. Loans and merchant cash advances reached $2,184M, but growth decelerated sharply to 4.1% sequentially from 17.5% in Q1. Originations of $1,584M exceeded repayments of $1,417M by $167M, roughly half the $306M of net cash absorption in Q1. Management characterized loss rates in payments and credit as normalized, with Capital the larger driver this quarter.

Assessment: The trajectory improved and the disclosure did not. Four quarters into coverage there is still no loss rate by product, no vintage performance, no reserve coverage, no duration and no concentration data, while the loss line has risen from 3.0% to 3.9% of revenue year over year and the provision has more than doubled. Management's framing shifts the driver each quarter, from payments experimentation to credit to Capital, which is a rotating explanation for a persistently rising number. We are keeping this at materializing rather than escalating further, because the book's growth decelerated and the absolute exposure of $2,184M is roughly 1.2% of enterprise value. It remains the only unmodellable item in an otherwise legible business and the single largest constraint on our conviction.

4. Operating Leverage: Fourth Consecutive Quarter, Now With an AI Cost Philosophy

Operating expense reached 34.0% of revenue against 37.7%, a nearly four-point improvement, with sales and marketing, research and development, and general and administrative each improving as a percentage of revenue. Sales and marketing was under 14% of revenue, roughly 160 basis points better, even while funding additional merchant acquisition. Research and development absorbs the majority of internal AI spend and still delivered leverage.

"We've moved from a place of just reflexive use of AI to a place of AI leverage. Our AI philosophy is straightforward. Maximum leverage paired with thoughtful cost management. We use the best model for the job, frontier intelligence where it matters, less expensive models where it doesn't." — Jeff Hoffmeister, Chief Financial Officer

Management described routing every model call through an internal proxy for visibility, control and security, and moving repeatable tasks to cheaper models, including distilled models where a frontier model teaches a smaller one a specific use case, producing something faster, cheaper and sometimes better at the narrow task.

Assessment: This is the answer that was missing last quarter, when two analysts asked how AI investment translates to margin and neither received one. A model-routing architecture with an internal proxy and distillation pipeline is a cost-control framework, not a slogan, and it is corroborated by subscription gross margin holding flat sequentially while Sidekick usage grew 3.6x. The operating expense pattern is now four quarters long and remarkably consistent: roughly 400 basis points of improvement per year with every category contributing. Our standing caution stands, that this decelerates, and Q3 is guided to 33% to 34%, implying roughly 330 basis points, slightly less than delivered.

5. Free Cash Flow Margin Finally Expands, and the Accounting Change Is Sized

Free cash flow of $654M produced an 18.3% margin against 15.7%, a 250 basis point expansion. Management sized the accounting component precisely rather than leaving it to inference.

"This represents roughly 1.5 points of year-over-year margin expansion after excluding the benefit of the accounting change in merchant cash advances that we mentioned last quarter. This free cash flow margin expansion was primarily the flow-through of our operating margin expansion, partially offset by about 1 point of increased taxes compared to last year." — Jeff Hoffmeister, Chief Financial Officer

The cash flow statement footnote corroborates: the presentation change, moving merchant cash advance flows from operating to investing, produced a net cash use of $37M now shown in investing, which is roughly one point of margin on $3,583M of revenue.

Assessment: We flagged in February and again in May that free cash flow margin was being guided flat-to-lower while operating margin expanded sharply, and named taxes and the accounting change as the reconciling items. Both resolved as described. Roughly 150 basis points of real expansion, driven by operating margin flow-through and offset by about a point of higher taxes, is the honest underlying number and it is the first genuine free cash flow margin expansion of our coverage. Management proactively quantified the non-operational half, which is the correct handling and the second consecutive quarter of it.

6. Catalog Becomes the Named Strategic Asset

Management elevated the product catalog from a feature to the company's central strategic claim, with a specific and unusual assertion of permanence.

"Catalog will be one of Shopify's most important assets for years to come. And here's why. We're seeing that AI searches powered by Catalog converted twice the rate of those using scraped data… Catalog is the discovery engine for the future and Shopify built it and owns it." — Harley Finkelstein, President

The catalog now holds over one billion products built over nearly two years of investment, is integrated with Shop sign-in so agents can recognize returning buyers and personalize on purchase history, and is being extended with what management called taste-driven attributes: whether an item is formal enough for a wedding, whether a fabric is breathable, wrinkle tendency, whether a shoe suits an endurance run. Every builder can now access the catalog API and the commerce protocol across millions of merchants using the same infrastructure as the major AI partners.

Assessment: The 2x conversion advantage over scraped data is the moat expressed as a number, and the identity integration is the part competitors cannot easily copy: a catalog is replicable, a catalog joined to a purchase history and a payment credential is not. The taste-attribute work is the more interesting long-term play, because it moves the matching problem from specification to preference, which is where the long-tail advantage compounds. The commercial question remains open. Catalog is currently a demand-generation asset that monetizes only through the resulting GMV, exactly as management has said four quarters running.

7. Sidekick's Usage Curve Steepens and Its Job Changes With Merchant Age

Daily active merchants using Sidekick grew 3.6x year over year and daily sessions grew 4.8x, so usage per merchant is also rising. The assistant handled nearly 34 million conversations in the quarter and was used to build more than 36,000 custom apps, up from 12,000 in Q1. It can now reach third-party applications, with Klaviyo named, without the merchant leaving the interface.

The most analytically useful disclosure was how the product's role changes over a merchant's life.

"In a merchant's first 30 days, roughly half of their conversations with Sidekick are about store setup, design and theme configuration. Now for merchants 5 years in, that drops to about 8%, while analytics and reporting claims passed 40% as they use Sidekick as their intelligence layer to interrogate their own data and make better decisions. Same product, different job." — Harley Finkelstein, President

Management also gave a measurable outcome: Sidekick's onboarding guidance drove an 8% increase in merchants reaching five orders within 15 days.

Assessment: The 8% figure is the first hard efficacy number for Sidekick and it matters more than the usage counts, because early order momentum is the strongest predictor of merchant survival. The lifecycle finding is the retention argument: a merchant five years in who uses the assistant as the interface to their own analytics has embedded Shopify in their decision-making, not just their checkout. Sidekick is still a cost line with no revenue line, and asked directly whether premium AI capabilities could become a direct pricing opportunity, management said it reserves the right to wake up smarter but that monetization today runs through merchant success. We continue to model it as a retention investment.

8. Enterprise: Balmain Migrates in Weeks, and the Cross-Sell Motion Shows Its Work

Q2 signings included Guess, Fred Segal and Avon, with e.l.f. Cosmetics, Claire's, Burton and Suitsupply going live. On the offline side Shopify welcomed Holt Renfrew, powered multiple Canada Goose locations, and expanded its relationship with furniture retailer Arhaus into offline, business-to-business and payments. Management disclosed in real time that Balmain had migrated in a matter of weeks.

"We're seeing a lot of success right now in the sort of cross-sell motion where they come to us for one particular channel, online store, for example, that's the on-ramp in. And then over time, they really migrate towards this unified commerce platform with Shopify." — Harley Finkelstein, President

Asked whether there is an upper bound on serviceable merchant size, management said the constraint is not GMV, citing merchants doing billions with very small teams, and that where a full migration does not fit, the accelerated checkout is offered as a component to start the relationship. Management also flagged that fewer than 1% of global retail sales run through Shopify and characterized international as the company's most underpenetrated opportunity.

Assessment: The Arhaus example is the one to hold onto because it is the cross-sell thesis with a name attached: online first, then offline, then business-to-business, then payments, one back end. That is the mechanism by which enterprise wins compound into revenue rather than sitting as logos, and it is why merchant solutions revenue as a percentage of GMV keeps rising even as the average merchant gets larger. A weeks-long migration for a European luxury house is also a direct answer to the implementation-risk objection that legacy platforms have historically used against Shopify upmarket.

9. The Buyback Nearly Completes, and the Question Becomes What Comes Next

Shopify repurchased $1,420M of stock in Q2, on top of $491M in Q1, bringing cumulative repurchases to $1,911M against a $2B authorization. The company has now spent roughly 96% of the program in two quarters, buying most heavily in the quarter when the shares traded between roughly $105 and $145.

Assessment: In February we described a $2B authorization with no minimums as potentially a sentiment gesture and said we would watch execution. It was executed, aggressively, into weakness, which is the best version of this behaviour. Combined with the $1,043M cash settlement of the convertible notes, the company has committed roughly $3B of actual cash against dilution in three quarters. The forward question is now unavoidable and unaddressed: with the authorization nearly exhausted, roughly $2.9B of annual free cash flow in prospect and $4.9B of cash on hand, does the board renew, enlarge, or redirect toward acquisitions? No comment was offered and no question was asked.

10. The Open Ecosystem Model Gets Named as the Agentic Strategy

Management organized the entire call around a single operating principle and then argued it is uniquely suited to the current moment.

"Our open ecosystem model is uniquely well suited to this agenetic era. We have always focused on most merchant needs while making the platform extensible for everything else. We build the primitives, we open them up, and we let the best developers and the best companies in the world build on top of us." — Harley Finkelstein, President

The evidence offered: OpenAI, Google, Meta and Microsoft have all partnered to open commerce front doors; dozens of retailers and platforms have adopted the commerce protocol; every Shopify merchant is protocol-ready by default with checkout logic and fulfilment rules preserved; and the AI Toolkit now ships connectors to Claude, ChatGPT, Perplexity, Manus, Replit and Vercel, plus integrations with vibe-coding platforms. In May the company shipped an agentic section in the merchant admin providing cross-channel attribution for agentic selling.

Assessment: The attribution tooling is the underrated item. A merchant who cannot measure agentic performance cannot optimize for it, and Shopify is currently the only place that measurement exists across channels. That is how a distribution advantage becomes a data advantage. The strategic risk we have carried since initiation is unchanged in principle: an open, payment-agnostic protocol is by design easier for others to adopt and extend. Four quarters of evidence now suggest that openness has bought Shopify the center of the ecosystem rather than commoditized it.

11. A Quiet Change in the IR Chair

The call was hosted by a new Director of Investor Relations, Shane Kleinstein, replacing Carrie Gillard who had hosted every prior call in our coverage. No announcement accompanied the change.

Assessment: Immaterial to the investment case and worth noting only for continuity. Disclosure quality improved rather than declined this quarter, with retention rates, cohort multiples and a third-party market-share statistic all appearing unprompted, so the transition has not cost the market anything.

Guidance & Outlook

Implied Q3 revenue. On a Q3 2025 base of $2,844M, a low-thirties growth rate implies roughly $3.70B to $3.78B. Given four quarters of exceeding the top of the guided range, we model $3.80B, or roughly 33.6% growth. The absence of any currency caveat is what makes this guide readable at face value, unlike the Q2 guide three months ago.

Implied Q3 operating income. Revenue of $3.74B at the guide midpoint, gross profit dollars up a mid-to-high twenties rate on Q3 2025's $1,391M (roughly $1.77B), and operating expense at 33.5% of revenue (roughly $1.25B) implies operating income near $520M against $343M, up roughly 52%, at a 13.9% margin against 12.1%. That is roughly 184 basis points of operating margin expansion, against 276 delivered in Q2 and 345 in Q1. The rate of improvement is decelerating, which is the honest counterweight to an otherwise excellent guide.

The one soft spot in the guide. Gross profit dollars guided to a mid-to-high twenties rate against revenue in the low thirties implies roughly 150 basis points of year-over-year gross margin compression, wider than the 91 basis points delivered in Q2 and the 76 in Q1. Two things temper it. Shopify has guided gross profit conservatively and beaten it in each of the last three quarters, by roughly four points in Q1 and six in Q2. And Q3 2025's 48.9% was itself a seasonally strong comparison. We model roughly 100 basis points of actual compression rather than the guided 150.

Guidance style, with four quarters of evidence. Twenty guided items, twenty cleared. Revenue growth above the top of the range four times out of four; operating expense below the low end four times out of four; stock-based compensation below the guided figure four times out of four. We treat every Shopify guide as carrying two to three points of embedded conservatism, and we would treat any quarter in which the company merely meets its guide as a genuine negative signal.

What is still absent. No annual framework, for the fourth consecutive quarter. This is the longest-running unmet item on our list. The cost of it was demonstrated in May, when a single quarter's guided growth rate took 15.6% out of the stock in a session, and again in August, when a guide four to seven points above consensus added 17%. A company whose forward disclosure produces double-digit moves in both directions has a communication problem that an annual framework would substantially solve.

Analyst Q&A Highlights

Eight questions were taken. For the first time in our coverage, one of them addressed the free cash flow margin trajectory directly. None addressed the credit book, the loan portfolio or the rising loss line, in the fourth consecutive quarter.

Whether Shopify Will Harvest More of the Value It Creates With AI

The opening question asked the monetization question plainly: given the value the AI products deliver to merchants, does the company intend to capture more of it, or keep the doors open as wide as possible. The answer was the fourth consecutive quarterly restatement of the same model, now with a distributional argument attached.

Q: "How are you thinking about potentially harvesting some of that value back to Shopify? Or is the intent still to kind of open the doors as wide as possible to bring in as many merchants as you can?"
— Ken Wong, Oppenheimer

A: "The agentic transactions carry the exact economics as an online store transaction. There's no new fees. There's no separate pricing. But more agenetic GMV, it means more Shopify revenue, and that's the model and it's been working really well for almost 2 decades… 75% of AI-attributed purchases in Q2 were from outside the top 100 categories."
— Harley Finkelstein, President

Assessment: Consistent to the point of being a policy rather than an answer, and consistency is worth something: anyone who modelled an agentic take-rate premium has now been told four separate times not to. The genuinely new content is the second half. A finding that three quarters of AI-attributed orders come from outside the top 100 categories means the channel favours exactly the merchant profile Shopify over-indexes to, which is a better argument for the platform than a pricing change would have been.

Compressing the Enterprise Sales Cycle and Lifting the GMV Ceiling

A question asked about the ability to onboard enterprises faster and what would move the target merchant cohort from $100M of GMV toward $200M. The answer arrived with a same-morning proof point.

Q: "Can you just talk about the ability to onboard new enterprises quicker and truncate that selling cycle? And I guess what's next on the product road map here where you can lift and shift that GMV target from $100 million-plus to north of $200 million over time?"
— Bryan Smilek, JPMorgan

A: "I woke up this morning to an e-mail from Balmain, one of the most iconic, legendary retailers in Europe, luxury retailer in Europe, saying that they migrated to Shopify in a matter of weeks… What these larger brands are looking for is they want a future-proofed commerce partner. They don't want to have to think about ever migrating again."
— Harley Finkelstein, President

Assessment: Implementation speed is the enterprise objection that legacy platforms have leaned on hardest, and a weeks-long migration for a European luxury house is a concrete rebuttal rather than a claim. The "final migration" framing is the actual sales proposition and it is a strong one in a market where replatforming has historically been a multi-year, multi-million-dollar project undertaken every few years.

Whether There Is an Upper Bound on Serviceable Merchant Size

A question probed the ceiling on addressable market: is there a GMV level or a business model, such as marketplaces, that simply does not fit. The answer separated size from complexity and produced the most useful framing of the addressable market in our coverage.

Q: "For your total addressable market, what is serviceable? And we'd love to know, is there an upper bound on a GMV per merchant? Or is there maybe a business model, like marketplaces that you don't think are a right fit for Shopify over the long term?"
— Michael Morton, MoffettNathanson

A: "In terms of upper bound, I don't think it's a GMV issue. We have merchants that are doing billions of dollars with a very, very small team… I think we've captured less than 1% of global retail sales. And in the core geographies where we operate, the opportunity is still massive, but I still think we are, I mean, my team hears me gripe about this all the time, I think we are underpenetrated internationally."
— Harley Finkelstein, President

Assessment: The honest content is the admission that some business models, marketplaces among them, are not full-migration candidates, and that the answer there is to lead with the accelerated checkout as a component. That is a narrower addressable market than the maximalist framing usually offered, and stating it improves the credibility of everything else. The sub-1% global retail penetration figure is the right denominator for a long-duration story and the international underpenetration point is corroborated by international GMV growing 37% against North America's 28%.

Whether the Catalog Is Now a Merchant-Acquisition Driver

A multi-part question asked how often the catalog is a primary driver of demand generation and new merchant acquisition, and how the current wave of entrepreneurship compares to the pandemic-era surge. The answer supplied a conversion statistic not in the prepared remarks.

Q: "On Shopify Catalog, I mean, how often is this actually now being a primary driver of demand gen and actually getting the new merchant?… This feels like we're having another moment here. How does this compare to that moment in terms of this kind of builder economy around AI?"
— Terrell Tillman, Truist Securities

A: "When you zoom it even further, conversion from AI search runs nearly 80% higher than traditional organic search as well… I meet these very large retailers and the executive teams there literally every single week. Every one of them is being pushed to figure out what their agentic strategy is. By coming to Shopify, we take their agentic strategy off their plate."
— Harley Finkelstein, President

Assessment: The 80% conversion premium of AI search over organic search is a bigger number than the 2x catalog-versus-scraped comparison and it was volunteered in Q&A rather than in prepared remarks, which is where the useful disclosures often sit. The second half describes the actual enterprise sales motion in 2026: the agentic strategy is now a board-level requirement at large retailers, and outsourcing it is a purchase decision. That is a genuinely new demand driver relative to the "replatform for total cost of ownership" argument of prior years.

Whether Sidekick Becomes a Direct Pricing Opportunity

A question asked for the merchant-outcome evidence behind Sidekick's usage growth, on retention, GMV and product attach, and then asked directly whether premium AI capabilities could become a direct pricing lever rather than an indirect one.

Q: "Could we see how retention is happening? How GMV growth is happening? How product attach is happening where merchants using Sidekick as well as the traditional Shopify offerings? And over time, do you see that indirect monetization as the main driver? Or could those premium AI capabilities become a more direct pricing opportunity for you?"
— Adam Wood, Morgan Stanley

A: "We reserve the right to wake up smarter every single day. But in terms of monetization now, the business model is predicated on merchants doing well. The better merchants do, the better Shopify does and Sidekick helps with that."
— Harley Finkelstein, President

Assessment: "We reserve the right to wake up smarter" is the closest management has come to leaving the door open on direct AI monetization, and it is deliberately non-committal. The retention and attach evidence the question asked for was not provided in the form requested; what came instead was the lifecycle data and the 8% onboarding improvement. Given that Sidekick's costs are explicitly landing in subscription gross margin and pulling it down 187 basis points year over year, a future decision to price it separately would be the single largest swing factor in this segment's economics. Worth tracking as an option, not modelling as a plan.

Is Roughly 20% Free Cash Flow Margin the New Normal

The most direct financial question of the call, and the first time in our coverage that an analyst pressed on the free cash flow margin trajectory. Management declined to set a new baseline and pointed at the guide instead.

Q: "I wanted to ask about free cash flow margins. Is roughly 20% the new normal? And how should we think about the levers around that?"
— Thomas Ingham, CIBC

A: "The gross margin is something where we've talked about in terms of any pressure we see on gross margin, we'll more than offset that with OpEx success… I don't think we can say, can we have a new normal? I would focus you on the guidance we gave for Q3. And I would tell you, we continue to drive more and more leverage through the system."
— Jeff Hoffmeister, Chief Financial Officer

Assessment: The refusal to declare a new normal is appropriate discipline and the sentence before it is the whole financial model in one line: gross margin pressure gets more than offset by operating expense. That has been true in each of the four quarters we have covered, and it is the mechanism by which a business with a falling gross margin produces expanding operating and free cash flow margins. Declining to anchor a level while guiding high-teens to low-twenties for Q3 is, functionally, anchoring a level.

What Is Working in the Headless Strategy

A question asked what is working in headless commerce given the third-party tooling merchants are building on, and what the second-order implications are as Shopify becomes embedded in that tooling. The answer described the architecture as a set of on-ramps rather than a single product.

Q: "Could you talk to us a little bit about what you think is working well with the headless strategy? And the second derivative implications for your model longer term as you become more entrenched in some of this third-party tooling."
— Gabriela Borges, Goldman Sachs

A: "We want merchants to build however they want. The key though is underneath all of it sits this commerce retail operating system, which is Shopify. That's where your inventory is, that's where your transactions are, your customer data, your analytics are."
— Harley Finkelstein, President

Assessment: The consistent through-line across four quarters is that Shopify is indifferent to the presentation layer and insistent on the system of record. Headless, agentic surfaces, third-party builders and the merchant's own storefront are all treated as interchangeable front ends over the same back end. That is the correct strategic posture in a period when interface layers are changing faster than infrastructure, and it is why the intermediation bear case has weakened rather than strengthened over our coverage.

Which Agentic Surfaces Are Showing Traction, and Where Shop Fits

The closing question asked which specific surfaces are producing an uptick in transactions, and how the consumer app fits the agentic picture. The answer restated the aggregate data and then gave an unusually concrete illustration.

Q: "I'm curious if there's any surfaces that are worth calling out where you're seeing an uptick in transactions or usage initially? And as a sort of maybe a follow-up to that, just how do you view the Shop app sort of playing a role in this agentic commerce area?"
— Arjun Bhatia

A: "We were looking for a screen-less phone for our daughter, for our 10-year-old, and I just started putting in my own prompt to look for different products. Five years ago, I would've done that in search, and I probably would have received some big box retailer selling a bunch of different random products. Instead, I was directed to the Tin Can Phone, which is this amazing product, a Shopify merchant."
— Harley Finkelstein, President

Assessment: The anecdote is doing real analytical work rather than decorating the answer, because it is the mechanism behind the 75%-outside-top-100-categories statistic made concrete. Keyword search rewards the merchant with the biggest budget for the broadest term; constraint-based agent search rewards the merchant whose product actually matches. The question about which specific surfaces are converting was not answered, and the absence of any surface-level breakdown after six quarters of agentic commentary is now conspicuous.

What They're NOT Saying

1. Anything quantitative about the credit book. Fourth consecutive quarter. Losses have risen from 3.0% to 3.9% of revenue year over year and the provision has more than doubled, while the explanation has rotated from payments experimentation to credit to Capital. No loss rate by product, no vintage curve, no reserve coverage, no duration, no concentration. Nobody asked.
2. Any absolute figure for AI-originated commerce. Sixth consecutive quarter. Traffic and orders both tripled, conversion is 80% above organic search, 75% of orders come from outside the top 100 categories, and the base remains undisclosed and described only as "small relative to $116 billion."
3. Any annual framework. Fourth consecutive quarter, and the item whose absence has now demonstrably cost the stock a 15.6% down day and produced a 17% up day on the same disclosure mechanism.
4. What happens after the buyback. $1,911M of a $2B authorization spent in two quarters, roughly $4.9B of cash on hand, and no comment on renewal, expansion or redirection.
5. Sidekick's cost, in dollars. Confirmed as the location of the vast majority of merchant AI costs, confirmed as the reason subscription gross margin sits 187 basis points below last year, and still unsized. The cost-management architecture was described; the cost was not.
6. Which agentic surfaces are actually converting. Asked directly in the final question. The answer gave aggregate multiples and an anecdote. Six quarters in, there is no breakdown by partner.
7. Sidekick or Campaigns revenue, or any pricing intent. "We reserve the right to wake up smarter" is the entirety of the forward disclosure on direct AI monetization.
8. Merchant count. Fourth consecutive quarter. The retention disclosures this quarter are genuinely better than a count, but they are not a substitute for one.
9. The economics of the partner revenue shares. Named for the first time as a driver of the revenue beat, alongside financial services, with no sizing of either.
10. Why the IR chair changed. A new Director of Investor Relations hosted the call with no announcement. Immaterial, but unremarked.

Market Reaction

• Pre-print setup: Closed at $123.30 on August 4, still down 23.4% year to date against an S&P 500 up 13.0%, and down 2.9% over the trailing twelve months. It had recovered 14.6% from the $107.63 close on the day of the Q1 print but sat 31.1% below the $179.01 top of its 52-week closing range.
• Reaction session (August 5, before-open reporter): Gapped up 21.7% to open at $150.00, traded a range of $142.52 to $153.88, and closed at $144.24, up 17.0% or $20.94. Contemporaneous coverage recorded shares up 26% in premarket, so roughly nine points of the initial move came off during the session.
• Volume: 41.5 million shares against an 8.7 million thirty-day average, 4.8 times normal.
• Market context: The S&P 500 fell 0.2% on the session, so the stock outperformed the index by roughly 17 points. The close leaves it down 10.4% year to date and 19.4% below its 52-week closing high.

The same disclosure mechanism, in reverse. Three months ago Shopify beat on revenue, beat on adjusted operating income, and lost 15.6% because its forward revenue guide implied deceleration that was mostly currency and comparison arithmetic. This quarter it beat by a similar margin and gained 17.0% because its forward revenue guide implied acceleration, cleanly, with management explicitly stating no significant currency impact. The prints were of similar quality. The difference was the guide and how legible it was.

This was a revision event, not a re-rating. A Street modelling 26.3% for Q3 was given a low-thirties guide from a company that has beaten the top of its own range four times out of four. That does not move one quarter's estimate; it moves the whole forward curve, because the market's implied 2027 growth rate was built on the same assumption the Q3 guide just invalidated. Enterprise value moved from roughly 9.1x our forward revenue estimate to roughly 11.8x on the day, and roughly half of that is the estimate rising rather than the multiple.

The fade matters for the anchor price. Premarket +26%, open +21.7%, close +17.0%. Large gap-ups routinely give back part of the move as overnight buyers take profit, and nothing in the session suggested a change in the message. But the correct anchor for valuation is the $144.24 close, not the premarket print, and we use it throughout.

Our position in it. We upgraded at $118.71 in February, watched the position go 9.3% underwater in May, and maintained. The stock is now 21.5% above the upgrade price and 34.0% above the Q1 close. Three quarters of coverage have produced one badly-timed upgrade and two correct maintains, and the reason the maintains were correct is that the business never stopped compounding while the price did whatever it was going to do. That said, roughly a third of the upside we identified in February has now been realized, which is the honest reason our conviction on the next twelve months is lower than it was on the last twelve.

Street Perspective

Debate: Is a Low-Thirties Growth Rate Sustainable Into 2027?

Bull view: Constant-currency GMV growth has sat in a 29% to 30% band for five consecutive quarters while the base grew by more than a third and the comparisons hardened, which is the definition of durability. Q3 is guided to low-thirties revenue with no currency help. Subscription revenue growth has accelerated four quarters running to 22.3%. Payments penetration is 68% with 40 countries live and Europe's newest markets still immature and, it now turns out, structurally higher margin. Merchants captured roughly half of incremental United States ecommerce dollars since the start of 2025.

Bear view: Every growth number here is being compared against a base that was itself inflated, and the guided gross profit growth of mid-to-high twenties implies 150 basis points of margin compression, wider than either of the last two quarters. Comparisons get materially harder from Q4 onward, when the company laps 30.6% and then 34.3%. And a business guiding to low-thirties while its highest-margin segment grows at 22% is still mix-shifting toward its lowest-margin revenue.

Our take: With the bulls into Q3 and genuinely uncertain beyond. The five-quarter constant-currency band is the strongest evidence anyone has produced on either side, because it holds the comparison and the currency constant and still shows no deceleration. The bear's point about the Q4 and Q1 comparisons is correct and is exactly where we expect the reported growth rate to step down toward the high twenties. Our model has FY2027 revenue growing 27%, which assumes deceleration, and the base case still clears our hurdle.

Debate: Is Agentic Commerce Additive or Substitutional?

Bull view: The substitution question has been answered with data. Traditional search sessions are up 1.3x over two years and still hold roughly a third of storefront sessions, while AI traffic and orders both tripled. New buyer orders from AI channels arrive at twice the rate of other channels, and half of AI-referred sessions land directly on a product page, 2.5x the traditional search rate. Three quarters of AI-attributed orders come from outside the top 100 categories, which is structurally favourable to Shopify's merchant base.

Bear view: All of that is true on a base management still refuses to disclose and describes as small relative to $116B of GMV. A tripling of something immaterial is immaterial. And the economics are explicitly identical to a storefront order, so even full success delivers no incremental take rate, only whatever volume the channel eventually adds.

Our take: Both correct, and the bear's framing is the one that should govern position sizing. Agentic commerce is confirmed as a genuine, additive, structurally Shopify-favourable channel that currently contributes an undisclosed and probably small amount of volume at unchanged economics. That is a real long-duration option and it is not a 2026 or 2027 earnings driver. Our valuation work assumes no agentic contribution beyond what is already in the GMV run rate, which is why we are comfortable being wrong about it in either direction.

Debate: Does the Credit Book Eventually Bite?

Bull view: Book growth decelerated sharply to 4.1% sequentially from 17.5%, net cash absorption halved to $167M, management describes payments and credit loss rates as normalized, and at $2,184M the exposure is roughly 1.2% of enterprise value. Lending against a merchant's own payment flows is the best-collateralized small-business credit that exists, and Capital is a decade-old product.

Bear view: Losses have risen from 3.0% to 3.9% of revenue in a year and the provision has more than doubled against revenue growth of 34%. The stated driver has changed three times in three quarters. Four quarters of coverage have produced zero credit metrics, and the company is simultaneously pursuing money transfer licenses to expand further into regulated financial services. Financial services was also named this quarter as a contributor to the revenue beat, meaning it is now large enough to move the top line and still small enough to escape disclosure.

Our take: With the bear on disclosure and the bull on magnitude, which is the same place we have been for three quarters. We are not forecasting a credit event. We are objecting to being unable to forecast one either way. The rotating explanation is the part that bothers us most: a number that rises for three consecutive quarters with a different named cause each time is a number the company may not fully understand. This single item is why our conviction is 7 rather than 9, and disclosing loss rates and reserve coverage would move it more than any other action management could take.

Debate: Has the Stock Already Priced the Good News?

Bull view: Even after a 17% day the stock is down 10.4% year to date and 19.4% below its 52-week closing high, at roughly 9.3x our FY2027 revenue estimate. The Street's forward numbers have to rise after a guide four to seven points above consensus. Operating margin is expanding, free cash flow margin finally expanded on an underlying basis, and the company has bought back $1.9B of stock in two quarters.

Bear view: Enterprise value went from 9.1x to 11.8x forward revenue in a single session. A stock that moves 15% down and 17% up on consecutive guides is being traded on quarterly sentiment, and the next guide is the one that laps a 34% comparison. Buying after a 26% premarket gap has historically been a poor entry in this name, twice in the last four quarters.

Our take: The bear's last point is the one we take most seriously and it is why our conviction is lower today than in May. But the framework has not changed: our base case is built on revenue growth and a multiple below where the stock traded three quarters ago, not on multiple recovery. At $144.24 the base case offers roughly 23% against roughly 19% downside, which is a materially narrower spread than the 37% against 11% we published in May. That is still an Outperform. It is a meaningfully less comfortable one, and we would move to Hold in the high $170s absent further upward estimate revision.

Model & Valuation Framework

Where the stock trades. At the August 5 close of $144.24 on roughly 1.29 billion diluted shares (reduced by $1.4B of Q2 repurchases), market capitalization is approximately $186B. Cash and marketable securities were $4,947M at quarter-end with no debt, giving an enterprise value near $181B. That is roughly 11.8x our FY2026 revenue estimate and roughly 9.3x our FY2027 estimate. On free cash flow it is roughly 63x FY2026 and roughly 46x FY2027. The comparable FY-forward figures across our coverage: 14.3x at initiation, 10.0x at the upgrade, 9.1x last quarter, 11.8x today.

On the roll-forward. Our prior targets of $187, $147 and $96 were anchored on FY2026 revenue at 16.0x, 12.5x and 8.0x. A twelve-month horizon from August 2026 lands in August 2027, by which point the market prices FY2027 and FY2028, so we roll the anchor to FY2027E revenue of $19.5B and step the multiples down accordingly to 14.5x, 11.5x and 7.5x. That is a lower multiple on a larger number, which is the conservative version of a roll-forward. The base case moves from $147 to $178, and essentially all of that increase is the estimate, not the multiple.

The asymmetry has narrowed, and that is the honest headline. Across our coverage the base-case spread has run 8% against 30% at initiation, 24% against 19% at the upgrade, 37% against 11% last quarter, and 23% against 19% today. The 17% session consumed roughly a third of the upside we published in May. This is still an Outperform on the arithmetic, and it is the least comfortable one we have published on this name. We would move to Hold in the high $170s absent further upward revision to the forward estimates.

Thesis Scorecard Post-Earnings

Scored against the standing thesis carried since our November initiation. Same pillars; only the status tags move.

Scoring Our Own Q2 Signposts

Scorecard summary: Eight bullish (two strongly), one half-met, one neutral, three bearish. The three bearish items are the three we have flagged every quarter: the credit loss line, the absent agentic denominator, and the absent annual framework. Every test of the operating business passed again.

Overall: Thesis intact and strengthened on the operating side. All four bull pillars on track, three of them strengthened. Bear 1, 4 and 5 remain contained or resolved. Bear 2 stays materializing with genuine mitigation in the balance sheet and none in the disclosure. Bear 3, valuation, is re-emerging after a 17% session.

Action: Hold the position, do not add aggressively here. The business case is stronger than at any point in our coverage; the price no longer offers the same margin for error.

Bottom Line: Maintaining Outperform, With Less Room

Rating decision. We maintain Outperform and raise the base case twelve-month price target to $178 from $147, against the August 5 close of $144.24. The increase is a roll-forward, not a re-rating: we move the valuation anchor from FY2026 revenue to FY2027 revenue of $19.5B and step the multiple down from 12.5x to 11.5x. Essentially all of the target increase comes from the estimate, and the estimate rose because the company guided Q3 revenue to a low-thirties growth rate against a Street at 26.3%.

Four quarters, twenty guided items, twenty cleared. That is the fact that has governed this coverage. Revenue growth has exceeded the top of the guided range four times out of four, operating expense has landed below the low end four times out of four, and stock-based compensation has come in below the guided figure four times out of four. Over the same period constant-currency GMV growth has sat in a 29% to 30% band while the base grew by more than a third. There is no reasonable reading of this record in which the forward guides should be taken at face value rather than as floors.

The bear case we initiated on is now fully inverted. We began coverage in November arguing that blended gross margin would drift structurally lower and that merchant solutions gross margin was the number that would settle it. That line has now expanded year over year for two consecutive quarters, and management explained why in a way that reverses one of our own assumptions: international payments carry higher gross margins than domestic because of debit prevalence and interchange, so the remaining penetration runway is margin-accretive rather than dilutive. The residual gross margin decline has migrated entirely to subscription solutions and is the cost of Sidekick, which is a different argument with a different set of tradeoffs.

Agentic commerce answered its hardest question. The substitution objection, that AI-originated orders simply cannibalize search-originated ones, was refuted with the only evidence that could refute it: traditional search sessions up 1.3x over two years, still holding roughly a third of storefront sessions, while AI traffic and orders both tripled. Add the distributional finding, that 75% of AI-attributed orders come from outside the top 100 categories, and the channel is not merely additive but structurally biased toward the long-tail merchant base Shopify over-indexes to. It still has no disclosed denominator, and the economics are explicitly unchanged, so we continue to assign it no value in our numbers.

What has not improved in four quarters is credit. Transaction and loan losses have risen from 3.0% to 3.9% of revenue, the provision has more than doubled against revenue growth of 34%, and the named driver has rotated three times in three quarters, from payments experimentation to credit to Capital. There is genuine mitigation: the book grew only 4.1% sequentially against 17.5% in Q1, net cash absorption halved, and management describes payments and credit loss rates as normalized. There is no mitigation at all on disclosure. Four quarters, zero loss rates, zero vintage data, zero reserve coverage, and not one analyst question. This is why our conviction is 7 and not 9, and it is the single action management could take that would move our rating framework most.

Why the conviction is lower even though the business is better. The base-case spread across our coverage has run 8% against 30% at initiation, 24% against 19% at the upgrade, 37% against 11% last quarter, and 23% against 19% today. A 17% session consumed roughly a third of the upside we published in May. We are not going to pretend that a stock 21.5% above our upgrade price offers the same opportunity it did at $118.71. It offers a smaller one, in a better business, and that still clears the bar for Outperform.

What would move us to Hold:

• A share price in the high $170s absent further upward revision to forward estimates, which would close the base-case gap entirely.
• Constant-currency GMV growth breaking below the 29% to 30% band it has held for five quarters.
• Operating margin expanding less than 100bp year over year in any quarter, which would break the leverage pillar the whole thesis rests on.
• Transaction and loan losses above 4.5% of revenue, or the loan book resuming double-digit sequential growth with continued silence on credit metrics.

What would move us to Underperform:

• A credit event in the lending book, defined as a reserve build exceeding a quarter's operating income, or a disclosed deterioration in vintage performance.
• Revenue growth guided below 22% on a constant-currency basis for any quarter.
• Operating margin contracting year over year, which has not happened in any quarter of our coverage.
• Merchant solutions gross margin resuming a decline of more than 100bp year over year, which would reopen the bear case we just retired.

Signposts for Q3 2026 earnings (expected November 2026):