Prepared remarks
Hello, and welcome to the Klaviyo Q2 2026 Earnings Call. Operator Instructions: To ask a question, please press star then one on your telephone keypad. Also, as a reminder, this conference is being recorded. If you have any objections, please disconnect at this time. With that, I would like to turn the call over to Brad Sills, Vice President, Investor Relations. Brad, you may begin.
Welcome, everybody. We appreciate you joining us. Joining me today are Klaviyo Co-Founder and Co-CEO, Andrew Bialecki; Co-CEO, Chano Fernandez; and CFO, Amanda Whalen. Andrew, Chano and Amanda will first share their views on the quarter, and then we'll open up the line for your questions. Our earnings press release, investor presentation, SEC filings and a replay of today's call can be found on our IR website at investors.klaviyo.com. As a reminder, our commentary today will include non-GAAP measures. Reconciliations to the most directly comparable GAAP measures can be found in today's earnings press release and our supplemental materials, which can be found on our Investor Relations website. Additionally, some of our comments today contain forward-looking statements that are subject to risks, uncertainties and assumptions, which could change. Should any of these risks materialize or should our assumptions prove to be incorrect, actual company results could differ materially from these forward-looking statements. A description of these risk factors, uncertainties and assumptions and other factors that could affect our financial results are included in our filings with the SEC. We do not undertake any responsibility to update these forward-looking statements, except as required by law. Andrew, that concludes my introductions. We're ready to begin.
Thanks, Brad, and welcome, everyone. Let's start with the headlines. Klaviyo has scaled to nearly $1.5 billion in annualized revenue run rate, growing quarterly revenue 26% year-over-year. We signed our largest deal ever last quarter, an eight-figure multiproduct contract with one of the fastest-growing brands in e-commerce. More than 205,000 brands now rely on Klaviyo. We offer our customers an autonomous B2C CRM. We vertically integrated the data infrastructure, experience infrastructure and the AI and agents to decide which experiences to deliver each consumer to drive better engagement, revenue and growth. Our customers are proving that the right infrastructure with all of your customer context and the right agents can deliver stunning consumer experiences that stand out, turn visitors into customers and keep them coming back. We're continuing to bet on being the source of truth for a business' consumer context and embedding that real time into agents and end consumer experiences.
Using the scale and volume of our consumer data points, we tune and guide AI models and agents, building all of this faster and with more tailored guidance to each of our customers through our internal use of large language models and agents. Every business will have an agent that decides, delivers and autonomously optimizes the experiences their customers have. That's the promise of an autonomous B2C CRM. I'd like to take a few minutes to share some updates on both the infrastructure and agentic layers of our products. Our data infrastructure now stores more than 9 billion consumer profiles and ingests and indexes more than 0.25 trillion data points every quarter. We know context is the key for high-performing agents and experiences. We spent the past few months improving the scalability and burst capacity of Klaviyo's infrastructure to handle data and messaging workloads that require personalizing and powering up to 100 million marketing messages and experiences in less than 20 minutes.
This helps our largest enterprise customers with massive audiences as well as AI agents that want to ask more questions and do greater personalization. Having clean data and context is important, too. We've taken that a step further by using machine learning and AI to train models to predict consumer behavior, and this powers features like personalized send times, channel preference, audience optimization and personalized product and content recommendations. Many of these features are included as part of our data and analytics products, which includes marketing analytics, and those products are collectively growing ARR more than 100% year-over-year. Our customers are finding there is real demonstrable ROI and growth from using AI to improve the marketing and experiences they deliver to consumers. Our marketing platform and infrastructure continue to deliver. While our messaging volume continues to scale into the hundreds of billions of messages per quarter, our compliance and deliverability infrastructure are continuing to improve.
Delivery and engagement rates improved year-over-year, while unsubscribe, spam and bounce rates decreased. We're also seeing the flexibility and scalability of our marketing platform being leveraged by enterprises and advanced users, both human and AI. Multichannel campaigns have increased nearly 50% quarter-over-quarter and marketing flows have seen a 48% increase in the number of messages and decision-point actions added to those automations. Our B2C CRM platform sets the table for agents, ones our customers have built outside of Klaviyo and those we built for our customers. In June, we launched our Composer agent to all of our customers, and the response has been very exciting. As a reminder, Composer is our purpose-built agent harness that's trained to excel at analyzing your customers and past marketing and taking those learnings to create and optimize marketing campaigns and automations.
In the first month since launch, Composer already has over 95,000 users. Very encouragingly, nearly one quarter have turned into recurring weekly Composer users. We've given all of our customers 10,000 credits to experiment and use with Composer, and the feedback has been overwhelming. We're shipping updates multiple times a day to improve Composer's performance, breadth and usability. As recently as last week, credit consumption grew 30% week-over-week. Usage is broad from entrepreneurs to enterprise, including 27% of our mid-market and enterprise customers being Composer users and exhibiting strong repeat usage. To share two examples, a fashion brand used Composer to design and send two marketing campaigns about shoes that were back in stock. They used Composer to select the audiences and consumers those campaigns would go to, asked Composer to calculate the best time to send those campaigns, had Composer run in a loop to review, audit and optimize their campaigns, ran a test send and then scheduled both campaigns, all from our agentic interface.
Another enterprise U.S. retailer used Composer to audit the performance and logic of marketing flows and made adjustments that increased performance by tens of thousands of dollars, all in a matter of hours. This is real agent ROI. I want to comment here on some of the reasons Composer is gaining traction. First, we've built the ability to understand the structure of data, the ontology and semantics into Composer to improve our agents' reasoning abilities. Second, Composer understands the style of a brand, what we call their taste, because it has the direct context of marketing and business decisions they've made in the past. Third, Composer is available where and how users want to work. And fourth, it has access to aggregated knowledge we've curated about what makes marketing and consumer experiences convert. Focusing on these areas, we believe there is a long runway in front of us to improve agent performance and grow usage.
As an example, in just the past few weeks post launch, we improved the percentage of generated campaigns users ended up using as part of their marketing to 46%, up from 35% just a few weeks ago as a result of improvements to taste, alignment and validation. We have a long list of these types of gains in front of us. Finally, Customer Agent, which gives every business an always-on agent for their customers covering both support and sales conversations, has seen adoption grow 40% quarter-over-quarter and weekly resolution volumes grow nearly 80% since early June. Boston Proper launched their Customer Agent and within two weeks had generated three times ROI in incremental revenue by being embedded on their website while also resolving autonomously more than half of their support tickets. Earlier this year, we built out an AI agent which we've embedded into Composer to help brands automate the setup and ongoing optimization of their Customer Agent.
We believe Customer Agents are going to be ubiquitous, but many businesses aren't versed in how to configure or improve an agent and how to measure its performance and quality. For businesses of all sizes, from entrepreneurs and SMBs to enterprises, we're leading in building agents to train the agents that understand your business and can automatically configure, experiment and optimize agents you deploy to customers that not only handle support cases, but also delight customers and drive revenue and growth. All of this has resulted in our customers generating nearly $50 billion in attributed revenue, or Klaviyo Attributed Value (KAV), in the first six months of the year. And we've done this while building agents for ourselves, allowing us to increase ARR per Klaviyo by 28% year-over-year. We're finding ways to increase our capacity. And with that, our ambition is increasing as well. The opportunity to give every business an agent to help them grow their business and delight customers is clear.
As we announced earlier today, I'm excited to build on our results so far with the acquisition of the Agency team. I've known the founder of Agency, Elias Torres, for over a decade, and he's been on the forefront of AI agents for customer experiences and is a technologist and a builder I deeply respect. He'll be joining Klaviyo as our Chief Product Officer, and together we'll build on the momentum of Composer, Customer Agent and the entire B2C CRM. Before I hand it over to Chano, I'm excited to welcome Erica Smith as our next CFO. I'd also like to say a big thank you to Amanda for her leadership over the last four-plus years, growing and scaling Klaviyo, championing our customers and furthering our mission of empowering businesses and creators to own their destiny. And with that, over to you, Chano.
Thanks, Andrew. I'm about two quarters into this role, and the go-to-market changes we've made are showing up in the numbers. We strengthened sales leadership and brought more rigor to our process. In Q2, we saw gains in sales efficiency, both in the Americas and globally. B2C CRM is a $160 billion-plus market opportunity, and we're capturing more of it every quarter, moving upmarket, expanding internationally, cross-selling and pushing beyond retail represent a great long-term growth opportunity for us. Agents open up an entirely new layer on top of that. Let me walk through where growth is coming from and what's next. Enterprise momentum is where these changes are showing up first. Many of the enterprise conversations we are having start with the same problem: a business looking to replace a legacy system because it can no longer keep up. In Q2, our larger customers, those with $50,000-plus ARR, grew 36% year-over-year to 4,477.
This group now represents approximately 40% of total ARR. Our $1 million-plus ARR customers continue to outpace growth of the overall customer base. This quarter, we were thrilled to welcome Warner Music Group, one of our most significant new contracts to date. They have over 1,400 artists and millions of fans, and every artist needs to reach their audience in their own voice. Our platform makes that possible by consolidating everything a business knows about its customers in one place so that every relevant team can act on it in real time. Their Head of E-commerce and Merchandising recently shared that Klaviyo's tools are second to none and will be integral to supercharge the e-commerce business they are building. We also won Claire's, replacing two legacy vendors in a competitive bid across e-mail, text and analytics. That's our enterprise motion working as designed. We're building champions inside the brand, educating executives on the value of consolidation and clearing procurement and CIO approval.
We're also selling deeper into the existing customer base with a strong expansion led by text messaging. We're also winning down market as can be seen with strong net adds, which were up 29% year-over-year. We're seeing more multiyear commitments and more products being sold across our base as our 205,000-plus customers increasingly use Klaviyo as a full, complete platform, which brings me to our second growth lever, multiproduct adoption. In Q2, nearly 20% of ARR came from customers using three or more products. This is important because our multiproduct customers retain more than six points better on gross retention than single-product customers. Our largest ever omnichannel deal closed this quarter, an eight-figure two-year agreement with one of the fastest-growing brands in e-commerce and a top seller on TikTok Shop. They're running their entire program on Klaviyo—e-mail, text and marketing analytics—because of the value of unifying one data-powered platform.
We're also extending Klaviyo's relevance beyond traditional e-commerce. This quarter brought our first NFL franchise, the San Francisco 49ers, plus hospitality names like SuiteHost. Every business that wants a direct personalized relationship with its customers is a business we can serve, and that keeps widening the market in front of us. That brings me to our third growth lever, international. Revenue outside the Americas was up 35% year-over-year. With the go-to-market and product investments we've made to internationalize Klaviyo, the runway ahead remains significant. EMEA revenue outside the U.K. was up 41% year-over-year, and we moved into Q3 with our strongest large deal pipeline in EMEA yet. K:London drew more than 1,000 attendees and our Paris and Berlin events were at capacity. We are backing that momentum with a new France office and an EU data center coming in the second half of the year.
In EMEA, The Body Shop expanded from the U.K. into e-mail, WhatsApp and marketing analytics across Germany, Switzerland, Belgium, Austria and the Netherlands. In APAC, we closed one of our largest new logo deals ever in the region with Country Road Group. I was in the region and met with their team, so I heard firsthand what choosing Klaviyo meant for them. It was a competitive multistage RFP win that consolidated five brands' fragmented stack onto one platform, sourced through our first global AWS Marketplace deal. The focus is to get the same big deal motion and country-by-country expansion moving at the pace EMEA is moving at today. In the Americas, we're focusing resources where we see the biggest opportunities—enterprise, a stronger cross-sell motion and new verticals. We're also investing in product specialist roles so our teams can go deeper with customers as they adopt more of the platform.
Before I close, echoing Andrew's note, Amanda, thank you for everything you built here. We're grateful you will stay on an advisory role through November, and we're thrilled to welcome Erica Smith as our next CFO on September 1. She brings deep software experience at the scale we are heading into, and the handoff is going to be seamless. With that, I'll turn it over to Amanda.
Thanks, Chano and Andrew. The growth we're building for the long term is broad-based as we expand across enterprise, international and multiproduct and make advancements with our agentic solutions. Q2 was a strong quarter with revenue of $370.6 million, up 26% year-over-year and ahead of our guidance, driven by notable strength in text messaging, WhatsApp and growth in marketing analytics. It was a strong quarter for new business, driven by sizable enterprise deals and cross-sell momentum across the base. This is proof that customers of all sizes are increasingly valuing Klaviyo as their B2C CRM platform. As Chano mentioned, in the enterprise we're winning new customers from larger legacy vendors. The pipeline in this segment is growing, and we have the team and the platform to serve this market at scale. The value that customers realize from our platform continues to expand. Our attributed revenue or KAV per message continues to grow, driven by increasing personalization powered by automation.
Our customers continue to shift their messaging volumes towards automated flows, which generate 10 times more revenue per message compared to static campaigns. Because our usage-based model is directly aligned with customer success, as customers realize more value, they grow and expand, and we grow alongside them. Net revenue retention was 109% in Q2, reflecting strengthening gross retention, increasing text messaging expansion and our strongest quarter of cross-sell since our IPO. These were offset by the lapping of last year's profile enforcement, which will continue to impact NRR through Q1. It's important to remember that NRR is a trailing 12-month metric and the leading indicators of our business are strong. Customers are staying with us, expanding with us and buying more products from us. Turning to the P&L. Non-GAAP gross margin was 73.4%, down three points year-over-year, driven by strong growth in text messaging, along with higher carrier fees, which we chose to fully absorb in prior quarters.
In Q3, we updated our mobile pricing to pass the higher carrier fees on to customers going forward, along with other changes to reduce friction in our mobile pricing mechanics. The new pricing will start to flow through gradually as we cycle through renewals. Due to timing and various puts and takes, this will be neutral to 2026 revenue and gross margin, but an overall positive impact for our customers and for the business. While growth in the text business impacts gross margin, it's an important channel for Klaviyo. Text has strong unit economics, driven by its lower cost of acquisition and higher rates of expansion. It also positions us as our customers' omnichannel platform of choice, driving higher retention and enabling a long runway to cross-sell other products with higher margins over time. Non-GAAP operating expenses were approximately 60% of revenue, down roughly three points year-over-year.
Relative to last year, we saw leverage in sales and marketing and G&A, while R&D as a percentage of revenue increased slightly from investments we're making behind product innovation to drive our next phase of growth. The results of these investments can be seen in our increased pace of product launches in the quarter, including the important advances with Composer and Customer Agent. Non-GAAP operating income was $50.9 million in Q2, representing a 13.7% non-GAAP operating margin. We continue to drive efficient growth at scale with another quarter of operating at the Rule of 40. Turning to the balance sheet. We generated $83 million of cash and ended the quarter with a total cash balance of $833 million. We used approximately $240 million in cash to repurchase shares during the quarter, leaving us with $160 million in capacity under the $500 million buyback authorization we announced in March.
We've remained active in the market and expect to continue repurchasing our stock under this program as we believe our current valuation represents an attractive opportunity. Our strong cash position provides us with the flexibility to also continue investing in growth and pursuing M&A that accelerates our roadmap, as Andrew noted earlier, with our acquisition of Agency. Turning to our outlook. Based on the Q2 outperformance and the broad momentum we're seeing, we are raising our full year 2026 revenue guidance by $12 million at the midpoint, higher than our beat this quarter. We now project 2026 revenue between $1.526 billion and $1.534 billion, representing 24% year-over-year growth. We are revising our full year 2026 non-GAAP operating income guidance down to a range of $212 million to $218 million, with non-GAAP operating margin of approximately 14%. This amounts to a $10 million reduction from the midpoint of our prior outlook and includes $10 million to $12 million in costs associated with our Agency acquisition as well as continued investment behind product innovation.
For Q3, we expect revenue in the range of $377 million to $381 million, representing growth of approximately 21.5% to 22.5%. We expect non-GAAP operating income of $40 million to $43 million or a non-GAAP operating margin of 10.5% to 11%. We expect non-GAAP gross margin to be down slightly in Q3 versus Q2, followed by a greater-than-normal seasonal step down in Q4 as we continue to see text growth outpace total company. To close, this was a strong quarter as customers increasingly lean into Klaviyo as a platform solution. Our business model is built for the long term with compounding growth across enterprise, international and multiproduct adoption. With agents, we're at the very start of our next S-curve. The B2C CRM market is large and the opportunity ahead for Klaviyo is significant. On a personal note, this is my last earnings call as CFO of Klaviyo. I still remember my first whiteboarding session with Andrew.
I left that conversation convinced of the power of the model, the size of the opportunity and the strength of the team. Everything since has reinforced it. It's been a privilege to build this business alongside you, Andrew and Chano and this entire team. I'm grateful to all of you on the line as well. I leave with confidence in the business and confidence knowing it is in strong hands with Erica as she steps in. Thank you. And with that, we'll open the call for questions.
Questions and answers
Operator Instructions: To ask a question, please press star then one. Our first question comes from Elizabeth Porter from Morgan Stanley.
Amanda, I just want to say it's been great to work with you and wishing you the best of luck in the next chapter. My first question: I wanted to dig in a bit on Composer. It was really great to hear about some of the adoption metrics. Historically, we think about the number of campaigns a brand could run really being constrained by available marketing headcount and production capacity. So as Composer removes some of that bottleneck, what have you seen from the early adopters in terms of campaign creation velocity and how that's translating to potentially higher customer contract volumes and incremental platform usage?
Thanks, Elizabeth. We're obviously a month out from launch, and the results have been really encouraging. Let me frame Composer for folks. Composer helps in three main areas: research that comes before actually building a marketing campaign or automation; generating the campaign or marketing flow; and helping verify and optimize ongoing performance. We've focused a lot on the first two categories: helping customers do the research, figure out which customers might be interested in which products or services and where opportunities exist, and then help with creative generation. We're seeing many weekly active users come in, build prompts with Composer or bring briefs from other sources, generate content and then use historical data in Klaviyo to allow Composer to pick the right audiences and send times. Regarding incremental usage: customers are often rate-limited by ideas and execution capacity. We're seeing folks using Composer to do more marketing across channels, including more text messages, because Composer is creating those campaigns. Our core pricing axis is the profile, and what's happening is Composer is driving more engagement per profile, more Klaviyo attributed value and more attributed revenue. We're watching this closely. We expect many customers will spend on Composer credits because it produces incremental revenue and sales.
Our next question comes from Arjun Bhatia from William Blair & Co.
I'm curious, two-parter, on SMS. It seems like that's seeing a lot of solid traction. Has there been an inflection there or is it more a factor of Klaviyo moving further upmarket and seeing enterprise success? And then a follow-up for Amanda on the gross margin front: what is the pass-through of the carrier fees and how much of that is weighing on gross margin this quarter? Any quantification to help with modeling would be helpful.
Thanks for the question. On text, that's the right read: we've seen really strong growth in text and WhatsApp as channels. Customers are telling us they want to put these messaging channels in one platform so they can use the same personalization and coordinate across channels. We're seeing a dramatic increase—about a 50% quarter-over-quarter rise—in multichannel campaigns. Many of our larger enterprise customers are consolidating their channels onto Klaviyo. They might start with e-mail, move to text messaging and WhatsApp, then use our marketing analytics product and Composer to stitch it all together. That's a common consolidation pattern.
On gross margin, there were three primary drivers this quarter: carrier fee impacts, infrastructure investments and the increasing presence of text in our business. We changed our approach on carrier fees in Q3 to pass through the higher carrier fee increases to our customers, and over the balance of the year we expect this to be neutral to revenue and gross margin due to the timing of renewals. On infrastructure, we're making investments to drive product innovation and growth, and historically we've made these investments and then seen them deliver leverage as we scale. Lastly, text is making up a growing portion of our business. Text has strong stand-alone unit economics with lower customer acquisition cost because it is primarily cross-sold and high expansion rates as customers expand usage. Strategically, text positions us as the omnichannel partner of choice, driving higher retention and enabling cross-sell of higher-margin products over time. Bottom line: gross margin was driven by a combination of strategic choices. Carrier fees are now being passed through, and the growth in text provides long-term benefits for the business.
Our next question comes from Tyler Radke from Citi.
Amanda, again, echoing the pleasure in working with you. Can you dive in a little more on the operating margin side of the equation? I understand dynamics with mix and SMS, but strategically, how are you thinking about the right framework for investing in growth? Anything we should read into that in terms of margin expansion beyond 2026?
Thanks, Tyler. On operating margin, we're making deliberate choices this year to invest behind the acquisition of Agency and behind product innovation and capabilities that will drive growth over the long term. In the back half of the year, you'll see investments show up in infrastructure and R&D. Those investments are driving strong early results—Customer Agent adoption up 40% quarter-over-quarter, over one quarter of our mid-market and enterprise customers using Composer and strong enterprise traction like Warner Music Group, Claire's and the 49ers. The Agency acquisition will impact the P&L by about $10 million to $12 million this year, and that's included in our outlook. If you exclude Agency, we're still driving over one percentage point of leverage year-over-year. This is a business that can continue to grow, make investments in innovation and drive leverage over time.
Our next question comes from Samad Samana from Jefferies.
Maybe let's dig into the Agency acquisition, especially given Elias will be the Chief Product Officer going forward and Andrew is focusing more on product as part of the co-CEO model. Help us understand what Agency brings more specifically and the decision to add a Chief Product Officer with Andrew's focus as well. There's a lot going on with AI, so it's great to bolster the bench, but help us think about the division of labor.
Great question. Elias is someone I've known for over 10 years; we've built and scaled companies together and work well together. We share a belief that every business will have an agent that helps them grow and run their business, making great decisions and personalizing experiences at scale. The acquisition accelerates how quickly we can build on Composer, Customer Agent and other agent initiatives. Regarding division of labor, Elias and I both come from engineering backgrounds and work closely with customers. As we do more in enterprise, I'm spending time with customers to understand large-scale, unique needs. Elias will help us scale product development across the agent surface area, helping us to move faster across Composer, Customer Agent and related products.
Our next question comes from Raimo Lenschow from Barclays.
Bringing Chano in as well, if you think about the broader setup—Customer Agent and Composer—how does that help in your conversations with enterprise customers? You mentioned legacy replacements. Talk a bit about how that broader vision helps you engage with larger customers.
Thanks, Raimo. We're seeing two key trends in enterprise: consolidation and an innovation play, which come together in our 'better together' story. Enterprise companies are looking for innovative solutions on a single platform and they increasingly value having unified customer data for context and better outcomes from agents. That consolidated data, our architecture, and our infrastructure are significant advantages. Take Claire's—we replaced two legacy platforms—and Warner Music required even broader consolidation. These wins show enterprise customers want a strategic vision for AI driving Klaviyo Attributed Value and revenue. Composer and Customer Agent are two strategic components of that vision. In many cases, it is a rip-and-replace effort in enterprise, and customers want someone who can deliver a better total cost of ownership and superior outcomes. Our improvements in scale—such as supporting 100 million messages in less than 20 minutes—are important for reliability and volume, and as customers grow, they have confidence we can support them.
One thing we're seeing: Composer adoption is already faster with larger enterprise customers. They use Composer not only to create net-new marketing but to manage complex automations and personalization across larger data sets. Common patterns include using Composer to review and audit existing marketing, to ramp new hires, and to make changes with more confidence because Composer validates the work. While Composer is only 30 days out, enterprise usage is already meaningful because it helps them manage complexity and do more.
To give a concrete example: I was with Country Road and asked their CIO why we won. He said we won because of our AI vision and openness, our interoperability with forward players like Anthropic and OpenAI, and the robustness of our infrastructure. Those were the reasons they chose Klaviyo.
Our next question comes from Matthew VanVliet from Cantor Fitzgerald.
When you look at the rate of adoption you're seeing on the various agents and Composer, how should we think about monetization over the next year or two? How much uplift are you getting at existing customers as they use the platform more? For customers who have landed with those products, how is average ARR trending versus similar cohorts without agents fully deployed?
On monetization: Composer is a credit-based model. When we released Composer, we gave everyone 10,000 credits for a 90-day trial to experiment. Klaviyo has always been focused on helping customers grow, and we measure that via Klaviyo Attributed Value. If customers use more Composer intelligence, they get better results—either through generating net-new marketing or improving existing marketing—and we can demonstrate ROI. We've already seen customers start to pay for Composer even during the trial period, and we expect that to grow. For Customer Agent, the pricing is outcome-based and resolution-based: customers pay when our AI resolves a conversation autonomously. If the conversation proxies back to another contact center or help desk, customers don't pay for that. Growth in Customer Agent is driven by pipeline in enterprise and by cross-use cases where marketing and service inform each other.
We've also built an agent to train other agents to automate setup and ongoing optimization, making it easier to run proofs of concept and increase resolution rates. We've seen notable improvements in resolution after deploying that capability—for one customer, resolution jumped from 52% to 79% in seven days without extra engineering effort. We want everyone to have a customer agent, including smaller businesses that need help answering questions. Because we now have an agent that helps build agents, even nontechnical customers can deploy agents, which is increasing adoption. So both Composer and Customer Agent are early but promising monetization drivers; Composer is only 30 days out and Customer Agent is seeing strong adoption.
To add color on Customer Agent with enterprise customers: we engaged our top enterprise accounts to prove our agent delivers better results than their current solutions. We run proofs of concept and are now in the high single digits on conversion for those conversations. Those customers are very large and handle millions of conversations yearly, so if they convert, the impact could be substantial. The solution is ready and we believe we can drive significant value.
On uplift: it's meaningful. There's a wide variety depending on adoption, since pricing can be outcome-based and depends on frequency and reliance. But for customers who have adopted these agent solutions, we are definitely seeing a meaningful contribution to average ARR.
Our next question comes from Terrell Tillman from Truist.
AB, Chano, and welcome aboard, Brad and Erica, and Amanda good luck. My question is on the enterprise business. You mentioned an eight-figure two-year deal this quarter and prior mid-seven-figure transactions; those are sizable. You also noted newer sales leadership. What about sales capacity—do you have the right people in the field, or are you still building the sales team? And last, on enterprise product-market fit: is anything missing, or do you feel you have everything at this point?
Great questions. On sales leadership: we've brought in a new CRO and a new Head of Americas who have worked with me before and know enterprise business well. They have helped attract great sales leadership and talent. We had large deals this quarter and strong pipeline in EMEA. Enterprise is a journey: we're moving from a transactional motion to a more consultative, multi-stakeholder sales motion that requires more senior engagement and business-case selling. That takes time. Early signals are strong—our large deals are outpacing regular business growth—but enterprise growth will be a process. Regarding sales capacity, we have the capacity we need and we prefer to focus on productivity rather than simply adding headcount. Productivity per head is higher than a year ago. We want the sales team to qualify deals properly and focus on high-probability opportunities. When it's time to scale, we'll invest further in headcount. We're also expanding agency partnerships and our ecosystem. We're investing in infrastructure such as an EU data center and stronger security and compliance, which matter for large customers. So the pieces are in place, performance is improving, and while it's a journey, we're excited about the long-term opportunity.
Our next question comes from David (DJ) Hynes from Canaccord Genuity.
Amanda, on net revenue retention: improving gross retention, SMS strength and multiproduct customers would imply NRR could inflect higher. Questions: a) How much is profile enforcement overhanging that metric? b) Does the intra-quarter picture tell us something different than the trailing 12-month metric? c) Is it reasonable to think NRR should go up from here?
Thanks, David. On NRR: we're seeing strength in the metrics that drive long-term health—improving gross retention, text expansion and cross-sell. This quarter, those benefits were offset by lapping last year's profile enforcement. As you look forward, the impact of profile enforcement will gradually phase out since NRR is a trailing 12-month metric and the effect winds out by Q1 next year. Over the longer term, the business indicators are positive: customers are staying, expanding and buying more products from us. Those are the drivers that will push NRR higher over time.
Our next question comes from James Wood from TD Cowen.
International continues to be a bright spot. Over the last couple of years you've been expanding languages, local support and sales teams. What are the next phases on the international front for both product and go-to-market to sustain this momentum?
International is a big opportunity—our share outside North America is still small relative to the market. Revenue outside the Americas was up 35% year-over-year. On product: we've invested to help businesses operate across markets, including managing product catalogs and differing strategies across regions, and we improved our Klaviyo organizations features to support multi-brand and multi-market setups. We've expanded data center footprint into Europe and enhanced capabilities for multi-market operations. On go-to-market: we've expanded events, held K:London, Paris and Berlin, and added sales headcount in France and Germany. WhatsApp has been particularly impactful internationally and even opened doors in markets like Brazil. Our aim is to grow the ex-North America portion of revenue substantially and we're on track.
To add: the international opportunity is in the early innings. We've made progress in markets like Germany, Spain and Italy, and Australia is strong in APAC. We're making leadership changes to accelerate growth in Southeast Asia, Singapore and potentially Japan. While current growth is good, the long-term opportunity internationally is much larger than what we see today.
Our final question comes from Brett Huff from Stephens Inc.
Andrew and Chano—welcome Erica and good luck Amanda. I'm asking an AI question from a different angle as the market continues to debate disruption, especially within SMBs and mid-sized customers. You noted Composer adoption is strong in enterprise, but can you share any data points, anecdotal or systematic, about how smaller customers are choosing AI from Klaviyo versus other vendors, such as general LLMs?
We've been watching sign-ups and trials from entrepreneurs and small businesses. Those using Composer tend to be more successful with Klaviyo in the first few weeks and are more likely to convert to paid customers. For many SMBs, they don't always know what marketing or support automation they need; Composer helps define that and make it executable. Composer has advantages over general-purpose models because we organize a business' data—orders, transactions, product catalogs—and understand the brand's past marketing and preferences. We recently upgraded the system behind the scenes to include Klaviyo proprietary knowledge about what works in marketing, which customers query for best practices on timing, audience selection and more. We believe SMBs will prefer domain- or task-specific AI embedded in a product they already use, and Composer is built to be that. We've also integrated Composer into Klaviyo's UI to make it accessible for customers who prefer drag-and-drop interfaces, and Composer integrates with other agent systems. We're already seeing entrepreneurs and SMBs adopt Composer and realize value, even if we still have work to do in educating the market on what's possible with AI.
That was our final question. I will now hand back to management for closing remarks.
All right. Thanks, everybody, for joining us today. Amanda, thank you for helping us continue to grow Klaviyo. And as we'd like to say, we're one percent done. Look forward to seeing everybody next quarter.
This concludes today's call. Thank you for joining us. You may now disconnect.