Pinterest reported second quarter revenue of $1.180 billion on August 4, 2026, an 18 percent increase that cleared the top of its own guidance range. The quarter also exposed a widening split between the company's two largest advertising regions. Revenue growth in the United States and Canada accelerated five points to 18 percent, while European revenue growth fell to 12 percent on a reported basis and 7 percent stripped of currency effects, after Asia-based cross-border retailers pulled back under regulatory pressure that Pinterest expects to persist through the third quarter.

A beat on revenue, a loss on the bottom line

The headline figures were comfortably ahead of what Pinterest told investors to expect three months earlier. Second quarter revenue of $1,179.7 million exceeded the $1.133 billion to $1.153 billion range the company guided to in May. Adjusted EBITDA of $311.3 million landed above the $256 million to $276 million guided range by a wide margin, and the company raised its full-year Adjusted EBITDA margin outlook to approximately 30 percent from approximately 29 percent.

Underneath that beat sits a GAAP net loss of $46.7 million, against net income of $38.8 million in the same quarter a year earlier. The swing is almost entirely explained by share-based compensation, which rose to $324.5 million from $227.2 million, and by $14.3 million of restructuring charges that did not exist in the year-ago period. Loss from operations widened to $55.2 million from $4.3 million. Interest income, which contributed $28.0 million in the second quarter of 2025, fell to $7.3 million as the company spent down its cash balance on buybacks.

On a non-GAAP basis the picture inverts. Non-GAAP net income reached $249.5 million, up 9 percent, and non-GAAP earnings per share came in at $0.43 against $0.33 a year earlier, a 30 percent increase. The gap between the 9 percent income growth and the 30 percent per-share growth is share count: diluted weighted-average shares used for the non-GAAP calculation fell to 577.6 million from 689.8 million.

Cash generation held up. Net cash provided by operating activities reached $292.9 million, a 41 percent increase, and free cash flow was $269.9 million, up 37 percent. On a trailing twelve-month basis Pinterest generated $1,280.4 million of free cash flow against $1,365.4 million of Adjusted EBITDA, a conversion rate of 94 percent.

"Our Q2 results reflect the scale and strength of our platform: more than $1.1 billion in revenue, growing 18%, and 640 million monthly active users, our 11th consecutive quarter of double digit user growth," said Bill Ready, chief executive officer, according to the earnings release.

Where the growth came from, and where it stopped

The regional breakdown is the most consequential part of the release for media buyers, because it shows Pinterest's monetization engine performing very differently on either side of the Atlantic.

Revenue apportioned by the estimated geographic location of users came in at $879.9 million for the United States and Canada, up 18 percent. That figure represents a five-point acceleration from the 13 percent growth the region posted in the first quarter. Europe delivered $212.7 million, up 12 percent on a reported basis and 7 percent in constant currency. Rest of World contributed $87.0 million, up 38 percent reported and 32 percent in constant currency.

The European deceleration is stark against the prior quarter. In the first quarter of 2026 the region grew 27 percent. Three months later that rate had more than halved. Chief financial officer Julia Donnelly attributed the change to three separate forces on the earnings call: a lapping effect, a structural go-to-market disruption, and a mid-quarter shock.

The cross-border squeeze

The mid-quarter shock is the item with the clearest read-across to other platforms. According to Donnelly, Pinterest was lapping a significant influx of cross-border spend into Europe from the prior year that followed changes in the United States tariff environment. Advertisers that had been diverted into European auctions in 2025 were no longer providing the same base.

Then a second effect landed inside the quarter itself. "We faced incremental pressure mid-quarter from Asia-based cross-border retailers impacted by regulatory actions, particularly in Europe, and that pressure is continuing in the third quarter," Donnelly said, according to the earnings call transcript.

Pinterest did not name the retailers or the specific regulatory actions. The timing lines up with a documented shift in European commerce advertising. The European Union abolished its 150 euro customs duty exemption on July 1, 2026, replacing it with a flat 3 euro charge applied per item rather than per parcel. Auction data covering roughly 500 European advertisers showed Temu's presence in Google Shopping halving from March onward while Shein approached a full exit, with the retreat intensifying through late May and June. Because customs enforcement is triggered by a parcel's arrival date rather than its order date, reducing new order volume through advertising was one of the fastest levers available to limit exposure before the deadline.

Pinterest is not a Google Shopping competitor in the direct sense, but it sells against the same budgets. The company's European monetization has leaned on retail, and Donnelly identified retail as the driver of European growth in the quarter. When a category of high-volume, low-margin cross-border sellers withdraws from paid acquisition across a continent, the platforms carrying their lower-funnel spend register it inside a single quarter.

The third factor is self-inflicted and deliberate. Pinterest flagged on its previous earnings call that international growth would moderate as it restructured its go-to-market organization. A new head of international sales is now in place, focused on market prioritization, coverage, and deeper relationships with advertisers and agencies in the region. According to Donnelly, some level of disruption is expected to continue in the third quarter, and the third quarter represents the most difficult year-over-year comparison of 2026 for the European region specifically.

Rest of World laps its resellers

The Rest of World number looks healthier at 38 percent, but it too decelerated. The region grew 59 percent in the first quarter. Donnelly identified the cause as lapping the ramp of resellers in the region during 2025, a base effect rather than a demand problem.

The region now carries 377 million monthly active users, up 15 percent, against $87.0 million of quarterly revenue. Average revenue per user in Rest of World is $0.23, up 21 percent year over year. For comparison, the United States and Canada generated $8.30 per user, up 14 percent, and Europe $1.35, up 4 percent. Global ARPU was $1.86, up 7 percent.

That spread is the entire international monetization argument in one line. Europe holds 157 million users, 51 million more than the United States and Canada, and produces roughly a quarter of the revenue. Ready described the situation on the call as being a long way from having fully monetized the commercial intent already present on the platform.

A technical note on two different Europes

Pinterest publishes European revenue twice, using two different methodologies, and the figures do not match. In the management discussion, revenue is apportioned by the estimated geographic location of users when they perform a revenue-generating activity, producing $212.7 million for Europe. In the notes to the financial statements, revenue is disaggregated by customers' billing addresses, producing $223.4 million. The same divergence runs the other way for the United States and Canada: $879.9 million by user location against $842.6 million by billing address.

The gap reflects advertisers headquartered in one region buying audiences in another, which is precisely the cross-border dynamic under discussion. United States revenue alone, by billing address, was $800.5 million. No individual country other than the United States exceeded 10 percent of total revenue. Pinterest also notes that its Europe segment includes Russia and Turkey.

Impressions grew, pricing barely moved

For advertisers, the auction mechanics matter more than the revenue total. Pinterest served 16 percent more advertisements year over year in the second quarter, while the price of advertisements increased 1 percent.

That 16 percent impression growth is a deceleration from prior quarters. Across the first six months of 2026, impressions rose 20 percent while pricing fell 2 percent. Donnelly attributed the impression slowdown partly to lapping the ramp in ad impressions from previously under-monetized international markets, a supply effect rather than a demand one.

The pricing figure carries a compositional wrinkle. According to Donnelly, the 1 percent increase was driven in part by stronger demand in the United States and Canada, and in part by a higher relative mix of impressions from that region, which carry higher average pricing overall. Strip out the mix shift and the underlying price movement in the auction is close to flat.

For a buyer, a platform growing supply at 16 percent while clearing at roughly flat prices is a platform where incremental reach is available without proportional cost inflation. That is a different environment from Meta, which reported second quarter ad impressions up 14 percent and average price per ad up 12 percent, and from Snap, where pricing outpaced revenue.

What accelerated in the United States

The five-point sequential acceleration in the United States and Canada is the result Pinterest wants investors to focus on, because it is the proof case for a playbook the company now intends to export.

Donnelly broke the acceleration into three components. First, ad platform improvements led to pockets of large-retailer spend accelerating, with certain retailers increasing budgets in response to return-on-ad-spend enhancements and AI-driven bidding optimizations. She also cited Asia-based cross-border retailer spend flowing into the region, the mirror image of the European pressure. Second, growth outside the largest retailers continued across mid-market, managed small and medium business accounts, and emerging verticals. Third, the region benefited from three timing items.

Those timing items are worth isolating, because they do not repeat. Prime Day moved from the third quarter of 2025 into the second quarter of 2026, running June 23 to June 26, which Pinterest quantified as an approximately 0.5-point benefit to second quarter growth and an equivalent headwind to the third. FIFA World Cup spending contributed a nearly 1-point benefit that will not repeat. Foreign exchange provided a 1-point tailwind in the second quarter and is expected to turn into a modest headwind. Together those three factors account for a 3-point sequential swing.

Across verticals, Donnelly identified retail as the leading contributor globally, alongside smaller but faster-growing categories including financial services, travel, and health.

The Performance+ adoption curve

Pinterest Performance+, the automated campaign product launched globally in October 2024, remains the company's primary lever for broadening its advertiser base. As of the first quarter, approximately 30 percent of lower-funnel revenue ran through Performance+ campaigns, with adopters growing lower-funnel spend at more than twice the rate of non-adopters. Pinterest did not update that adoption percentage for the second quarter.

What it did provide was a new performance figure and a new product. According to Ready, updates to Pinterest's ad-delivery models drove a 28 percent improvement in return on ad spend during testing for small and medium business advertisers using Performance+ with ROAS bidding.

Smart Assembly opens automation to advertisers without catalogs

The structural limitation of automated creative optimization has been that it presumes a product catalog. Advertisers promoting brands, services, or content rather than stock-keeping units have been unable to access the creative layer of Performance+ in full.

Smart Assembly, launched during the second quarter, addresses that gap. Advertisers upload multiple images and Pinterest automatically builds and serves the best-performing combination for each impression. In early alpha testing the feature delivered a 6 percent improvement in click-through rate on average, according to Ready.

The figure is modest in isolation. Its significance lies in eligibility rather than magnitude: it extends automated creative testing to a category of advertisers that previously had no route into it. Pinterest listed Smart Assembly as a beta capability alongside A/B testing and title and call-to-action optimization on its product roadmap for the first half of 2026, with a simplified campaign-creation flow targeted for early next year.

Ready described the destination as nearly every lower-funnel campaign starting in an AI-powered best-practice setup, with added controls for more complex buyers who need them. He also set expectations on pace, noting that other platforms deploying automation suites saw adoption play out over several years, and that even the earliest movers are still working through their adoption cycles.

The measurement integration pilot stays small

The item most likely to shape enterprise buying on Pinterest over the next two years is also the least visible. The company is piloting direct integrations between large advertisers' in-house measurement systems and its own AI bidding stack, allowing the bidding engine to optimize toward outcomes those advertisers define internally rather than toward Pinterest's default conversion signals.

According to Ready, the initial group is seeing strong performance, and that performance partially contributed to the second quarter beat. The pilot is expanding to a small number of additional advertisers over the course of the third quarter.

The mechanics matter. By connecting to an advertiser's measurement source of truth, Pinterest can align bidding to metrics such as customer lifetime value or incremental return on ad spend rather than last-click conversions. Ready framed this as part of an industry-wide shift beyond last-click attribution, one that Pinterest expects to reach a broader set of advertisers through third-party measurement platforms as those capabilities become easier to deploy.

He was explicit that the most customized version of this will stay confined to a relatively small number of advertisers with the data infrastructure and technical resources to support it. The learnings, he said, are informing broader bidding enhancements that can scale further.

That framing sets a realistic expectation. Advertisers without in-house measurement stacks will receive the second-order benefits through platform-wide model improvements rather than direct integration. The dependency on third-party measurement is consistent with other recent developments in the ecosystem, including AudienceProject's direct integration with Pinterest in the United Kingdom and Germany in June.

Pinterest also disclosed changes to its shopping ads delivery systems covering which products to show, when to show them, and how to allocate budgets against the highest-value opportunities, including improved product selection for advertisers with large catalogs and greater variety in the products users see. Business Assistant, the conversational collaborator for advertisers first shown in June, remains in beta.

Pinterest Assistant moves to the foreground

On the consumer side, the substantive change in the quarter is distribution. Pinterest Assistant, the conversational layer that launched in beta during the fourth quarter of 2025, became available to the vast majority of United States users as of the end of July 2026.

That is a substantial expansion from the controlled rollout Ready described three months earlier, when he emphasized deliberate pacing and pointed to competitors that had to pare back more aggressive launches. The product is now woven throughout the existing app with contextual entry points and prompt suggestions rather than sitting in a separate surface, and Pinterest has begun introducing memory and conversational history.

Ready framed the monetization implication directly on the call: because the assistant sits inside the existing experience rather than beside it, the advertising inventory already present in that experience applies. Pinterest is not treating this as a build-now-monetize-later product.

The scale figures Pinterest cited to justify the approach are unusually specific. The platform records more than 80 billion monthly searches, of which the vast majority are visual and over half are commercial. Users have created more than 16 billion boards. And over 96 percent of text-based searches on the platform are unbranded, meaning the searcher has commercial intent but has not yet selected a product or brand.

That last figure is the core of Pinterest's pitch to advertisers, and it distinguishes the platform from retail search environments where the query already names the destination. Pinterest also reported that 100 percent of its users are logged in and that 85 percent arrive directly through the mobile app rather than through search engines or referrals.

Generation Z now represents over half the user base and remains the largest and fastest-growing cohort. Ready cited an Adobe study, conducted independently, finding that 39 percent of Generation Z consumers think of Pinterest as a first place to search, and that more than 70 percent see the platform as more personalized.

The open-source cost argument

The most quantified technical claim in the entire earnings package concerns model economics. According to Ready, Pinterest achieves cost per transaction at less than 8 percent of the cost of comparable closed proprietary models when running open models post-trained on its own data.

He extended the argument to quality as well as cost, stating that Pinterest's latest multimodal models outperformed closed models by 34 percentage points on the relevancy of shopping recommendations, a narrow benchmark he was careful to bound as applying to that specific task rather than to general model capability.

"At this point, any CEO who is not taking advantage of open-source models is almost certainly wasting a lot of shareholder money," Ready said on the call.

Pinterest describes its approach as model-agnostic, combining in-house compact models built for specific use cases, open-source models post-trained in its own cloud environment, and closed third-party models in limited cases. A centralized model-routing layer directs production traffic based on quality, cost, latency, and reliability, reserving higher-cost models for complex work.

The company completed its transition to GPU serving more than two years ago, which Ready identified as the precondition for the personalization gains that followed.

Donnelly extended the routing infrastructure argument to internal operations. Pinterest is expanding model routing to internal use cases across engineering, sales, and other functions in the second half, with costs appearing in operating expense rather than cost of revenue. The tracked metrics are adoption, developer throughput, cycle time, incident rates, and service uptime. In July, weekly pull requests per engineer increased 45 percent versus the prior year while incident rates and service uptime remained relatively consistent, according to Donnelly.

Connected TV: the integration is a 2027 event

Pinterest closed its acquisition of tvScientific on February 17, 2026, and the second quarter was the first full quarter of contribution. The 10-Q discloses terms that were not public when the deal was announced in December 2025.

Total purchase consideration was $465.1 million, primarily in cash. Pinterest also issued replacement share-based awards with a grant date fair value of $24.1 million, of which $2.5 million was attributed to pre-combination services and $21.6 million will be recognized as compensation expense over the remaining service periods. The preliminary allocation assigns $59.0 million to developed technology with a five-year useful life, $25.0 million to customer relationships over eight years, and $375.1 million to goodwill, which is not deductible for tax purposes.

Goodwill on the balance sheet rose from $100.2 million at the end of 2025 to $475.3 million at June 30, 2026, almost entirely from this transaction. Intangible assets, net climbed from $6.1 million to $83.0 million.

The commercial timeline, however, extends further out than the accounting one. Ready stated that Pinterest expects to fully integrate tvScientific capabilities directly into Performance+ in 2027, at which point the company intends to position itself as a combined search, social, and connected TV performance solution. Until then, tvScientific operates as a distinct capability with what Ready described as strong early advertiser reception.

For media buyers evaluating Pinterest as a CTV channel, that means a second-quarter contribution to cost of revenue arrives well ahead of a unified buying workflow.

The cost side: GPUs, compensation, and a shrinking headcount

Cost of revenue on a GAAP basis reached $257.4 million, up 27 percent, and 22 percent of revenue against 20 percent a year earlier. On a non-GAAP basis it was $245 million, up 25 percent and up 6 percent sequentially. Donnelly attributed the increase to the full-quarter impact of tvScientific and investment in additional GPU capacity.

Research and development reached $451.0 million on a GAAP basis, up 25 percent, driven primarily by a $66.6 million increase in share-based compensation and a 15 percent increase in personnel expenses. Sales and marketing reached $374.3 million, up 20 percent, driven by an $18.3 million increase in marketing expense tied to the brand campaign, a 14 percent increase in personnel expenses, and a $13.4 million increase in share-based compensation. General and administrative rose 9 percent to $137.9 million.

Non-GAAP operating expense totalled $629 million, up 13 percent.

Headcount is moving in the opposite direction from expense. Reported headcount was up 3 percent year over year in the second quarter, a figure that includes employees who joined through the tvScientific acquisition. Excluding that acquisition, headcount was down year over year and down 6 percent since the end of 2025, according to Donnelly.

That reduction traces to the global restructuring plan Pinterest initiated in January 2026, which cut approximately 780 roles and which the 10-Q describes as a workforce reduction of less than 15 percent alongside office space reductions. The stated purposes were reallocating resources to AI-focused roles, prioritizing AI-powered products, and accelerating the transformation of the sales and go-to-market approach.

Restructuring charges totalled $14.3 million in the quarter, comprising $8.0 million of severance and other personnel costs, $4.8 million of share-based compensation, and $1.6 million of office space reductions. Six-month charges reached $61.4 million. Pinterest expects total charges of up to $69.6 million through the end of the third quarter, implying roughly $8.2 million of remaining expense.

Share-based compensation peaked, by design

Share-based compensation of $324.5 million represented 28 percent of revenue in the quarter. Donnelly stated that the second quarter should be the highest quarter for share-based compensation expense in 2026, and that both the dollar amount and the year-over-year growth rate are expected to step down from the third quarter onward.

The explanation given was the annual equity grant cycle combined with a lower stock price at the time of grant, which required issuing more shares to remain competitive on compensation, plus a timing effect in which a greater portion of annual grant expense was recognized in the quarter. Unrecognized share-based compensation stood at $1,443.8 million at June 30, 2026, expected to be recognized over a weighted-average period of 2.1 years.

Pinterest recognized income tax benefits of $65.1 million on share-based compensation expense during the quarter. Donnelly also stated the company expects to be profitable on a full-year GAAP net income basis, as in recent years.

Buybacks, notes, and a capped call

Pinterest has spent the first half of 2026 restructuring its capital base as aggressively as its organization.

Under the November 2024 programme, the company repurchased and retired 27,664,663 shares for $472.9 million at an average of $17.09 per share during the six months. In March 2026 the board authorised a new $3,500.0 million programme and cancelled the earlier one, under which $499.9 million had remained available. Under the March 2026 programme, open market purchases retired 28,950,481 shares for $551.0 million at $19.03, and an accelerated share repurchase agreement retired 54,796,613 shares at $18.25 for $1,000.0 million. Total repurchases across the six months come to approximately 111 million shares for just over $2 billion, at a stated average of $18.17. As of June 30, 2026, $1,948.0 million remained available under the March 2026 programme. The company recorded $14.9 million of excise tax on these repurchases under the Inflation Reduction Act.

One figure does not reconcile cleanly across the source documents. Donnelly stated on the call that Pinterest allocated $58 million toward share repurchases in the second quarter, while the condensed consolidated statement of cash flows records $78.6 million of repurchases of Class A common stock for the three months ended June 30, 2026. Pinterest did not reconcile the two figures, and differences in settlement timing or excise tax treatment could account for the gap.

The buyback was funded in part by $1,000.0 million of 1.75 percent convertible senior notes issued in March 2026 and maturing March 1, 2031. The initial conversion rate is 44.0063 shares per $1,000 principal, equivalent to approximately $22.72 per share. In June 2026 Pinterest entered privately negotiated capped call transactions at a cost of $99.2 million covering the 44,006,300 underlying shares, with a strike matching the conversion price and an initial cap of $30.59 per share. The effect is to eliminate expected dilution at maturity for stock prices up to roughly $30, according to the illustrative table in the earnings presentation.

Net dilution for the quarter was down 12 percent year over year, which Donnelly connected directly to the 30 percent growth in non-GAAP earnings per share. Cash, cash equivalents and marketable securities stood at $1,274.9 million at quarter end, down from $2,467.2 million at the end of 2025.

The company's infrastructure commitments have grown in parallel. Under a private pricing addendum with Amazon Web Services signed in May 2026, Pinterest must purchase at least $4,000.0 million of cloud services through May 2031, with any shortfall payable as the difference. The remaining contractual commitment was $3,927.9 million at June 30, 2026, implying roughly $72 million consumed in the first weeks of the agreement.

A labelling error in the earnings presentation

Analysts working directly from the slide deck should note a transposition in the appendix. On the GAAP to non-GAAP reconciliation slide, the first table is headed "Quarter ended March 31, 2026" and the second "Quarter ended June 30, 2026", but the figures are reversed. The first table's cost of revenue of $257.4 million, research and development of $451.0 million, and restructuring of $14.3 million match the second quarter figures in both the press release and the 10-Q. The second table's restructuring charge of $47.1 million matches the first quarter. The tables are correct; the headings are swapped.

Guidance and the second-half setup

For the third quarter, Pinterest guided revenue to $1,190 million to $1,210 million, representing 13 percent to 15 percent growth. Adjusted EBITDA is guided to $335 million to $355 million.

Donnelly made the arithmetic explicit: the high end of the third quarter range is consistent with what the company just delivered once adjusted for the 3-point sequential impact from foreign exchange, Prime Day, and World Cup spending. Read that way, the guidance implies stable underlying performance rather than deterioration, with the reported deceleration from 18 percent to 13 or 15 percent explained largely by comparison effects and the European drag.

Non-GAAP cost of revenue is expected to be roughly flat versus the second quarter, which Donnelly attributed to accelerated recognition of certain contractual benefits in the third quarter associated with the recently executed multiyear infrastructure agreement. That is a one-time timing benefit inside an otherwise rising cost line: Pinterest continues to expect modest headwinds from cost of revenue as a percentage of revenue in the second half from GPU capacity and the tvScientific inclusion.

The raised full-year margin outlook of approximately 30 percent sits within the 30 to 34 percent long-term range Donnelly reiterated from a prior investor day. She declined to provide specifics for 2027.

What this means for the marketing community

Pinterest's second quarter lands in the middle of an unusually well-documented earnings sequence for digital advertising. Alphabet reported second quarter advertising revenue up 14 percent to $81.6 billion, with Google Search up 17 percent to $63.3 billion. Meta grew advertising 27 percent to $59.36 billion. Amazon advertising services reached $19.8 billion, a 26 percent increase. Reddit advertising revenue rose 64 percent to $762 million. Microsoft search advertising growth slowed to 10 percent.

Pinterest's 18 percent sits in the middle of that distribution, below the large-platform leaders and well above the search laggards. Three practical implications follow for buyers and publishers.

The first concerns auction economics. A platform adding 16 percent more impressions at roughly flat pricing is expanding available reach faster than it is repricing it. The mix effect flatters the headline pricing number, but the direction is clear: Pinterest is not currently in a position to extract meaningful price increases, and the European softness reduces auction pressure further. For advertisers with European retail budgets, the withdrawal of high-volume cross-border sellers from paid acquisition changes the competitive set they bid against, and Pinterest has signalled that the effect persists into the third quarter.

The second concerns automation adoption. Pinterest is following the pattern established by larger platforms: automated campaign types become the default path, manual controls are added back for sophisticated buyers rather than preserved by default, and the adoption curve runs for years rather than quarters. The company held its Performance+ share of lower-funnel revenue at the previously disclosed 30 percent rather than updating it, and Ready explicitly compared the trajectory to platforms still working through their own adoption cycles. Advertisers evaluating whether to move budget into Performance+ now have a disclosed small-business ROAS improvement of 28 percent in testing and a 6 percent click-through improvement from Smart Assembly to weigh, both from Pinterest's own testing rather than independent verification.

The third concerns measurement. The bidding integration pilot represents a structural change in how large advertisers can direct platform optimization, and it remains confined to a small group expanding by a small increment in the third quarter. The broader route runs through third-party measurement systems. Pinterest's willingness to optimize against externally defined outcomes is a concession that platform-defined conversion signals are not sufficient for its largest customers, and it is a concession several platforms are now making simultaneously.

Underneath all three sits the same unresolved question the company has carried for several quarters. Clicks to advertisers are up more than fivefold over three years, according to Ready, without a corresponding fivefold increase in revenue. Pinterest reached 640 million monthly active users, its twelfth consecutive quarter of record users, and monetizes them at $1.86 each. Whether the go-to-market changes that worked in the United States and Canada transfer to a European market simultaneously absorbing a cross-border retail shock is the question the third quarter will begin to answer.

Timeline

Summary

Who: Pinterest, Inc. (NYSE: PINS), the San Francisco-based visual search and discovery platform, with chief executive officer Bill Ready and chief financial officer Julia Donnelly presenting results. The disclosures affect advertisers buying lower-funnel and consideration inventory on the platform, European retail buyers in particular, and the mid-market and small business advertisers targeted by the Performance+ automation suite.

What: Second quarter 2026 results showing revenue of $1,179.7 million, up 18 percent reported and 17 percent in constant currency; 640 million global monthly active users, up 11 percent; a GAAP net loss of $46.7 million against net income of $38.8 million a year earlier; Adjusted EBITDA of $311.3 million at a 26 percent margin; and free cash flow of $269.9 million. Regionally, United States and Canada revenue grew 18 percent to $879.9 million while Europe grew 12 percent reported and 7 percent in constant currency to $212.7 million, and Rest of World grew 38 percent to $87.0 million. Ad impressions rose 16 percent and ad pricing rose 1 percent. Product disclosures included Smart Assembly creative automation delivering a 6 percent click-through improvement in alpha testing, a 28 percent return-on-ad-spend improvement in testing for small business advertisers using Performance+ with ROAS bidding, and the expansion of Pinterest Assistant to the vast majority of United States users.

When: The results cover the three months ended June 30, 2026, and were released after market close on August 4, 2026, with the earnings call held the same day at 1:30 p.m. Pacific time. Third quarter guidance covers July through September 2026.

Where: San Francisco, California, with revenue reported across three regions: the United States and Canada, Europe including Russia and Turkey, and Rest of World. The regulatory pressure cited as a mid-quarter headwind was concentrated in Europe.

Why: Pinterest is attempting to close the gap between its user engagement and its revenue, a gap the company quantifies through a fivefold increase in clicks to advertisers over three years that has not produced a proportional revenue increase. The second quarter tested that thesis in opposite directions: a sales and go-to-market transformation combined with AI bidding improvements produced a five-point acceleration in the United States and Canada, while the same transformation, difficult prior-year comparisons, and a withdrawal of Asia-based cross-border retail spend cut European growth by more than half. Third quarter guidance of $1,190 million to $1,210 million, representing 13 to 15 percent growth, reflects a 3-point sequential drag from foreign exchange, the Prime Day calendar shift, and non-repeating World Cup spending, alongside continued European disruption.