Marqeta (MQ) 2026财年第三季度业绩电话会：TPV突破1260亿美元，EBITDA增长28%

核心要点

• Marqeta公布2026财年第三季度总处理规模（TPV）达到1260亿美元，同比增长29%。年化TPV首次突破5000亿美元。
• 净营收增长8%至1.76亿美元，毛利润增长7%至1.22亿美元。来自Block的净营收占比为41%，低于去年同期的44%。
• 调整后EBITDA增长28%至3900万美元。调整后EBITDA利润率创下历史新高，占净营收的22%，占毛利润的32%。
• 按GAAP计算的净利润为1.15亿美元，其中包括1.07亿美元的非现金税收收益。稀释后每股收益（EPS）为1.09美元，其中1.01美元来自税收资产减值准备金的释放。
• 管理层维持全年毛利润增长11%–12%的指引，将净营收增长指引下调1个百分点至11%–12%，并将调整后EBITDA增长指引上调至30%低段至中段区间。
• Cash App的新发卡速度如预期放缓。Marqeta预计到12月底新发卡量将微乎其微甚至降至零，不过庞大的存量卡用户仍保留在该平台上。

关键财务数据

Marqeta在该季度以16.25美元的平均价格回购了70万股股票。截至季度末，公司在1.5亿美元的回购授权额度中还剩余1.4亿美元。

业务与运营表现

非Block TPV的增长率继续保持在Block TPV增速的两倍以上。扣除Block后，金融服务业务的增长明显快于Marqeta的整体增长率。

在灵活的网络凭证使用和客户地域扩张的支撑下，包含“先买后付”（BNPL）在内的借贷业务同比增长近40%。费用管理规模增长在50%高段，且连续第五个季度加速增长。即时配送业务实现高单位数增长，反映出其作为Marqeta最成熟应用场景的地位。

Marqeta拓展了与Block的合作关系，在运通（American Express）网络上启用了Square信用卡。Square现在可以通过Marqeta的平台跨不同银行和网络运营借记卡和信用卡项目。此次资产组合迁移是Marqeta在24个月内的第四次迁移，信用卡和借记卡项目各占一半，管理层表示更多已签约的迁移正在推进中。

其他客户动态包括利用Marqeta的代币化和消费控制功能推出的儿童版谷歌钱包（Google Wallet）服务，以及Satispay消费者借记卡项目的上线。Satispay在意大利全境拥有超过650万用户和45万家合作门店。

产品投资方面，通过与Coinbase、BVNK和zerohash的合作，推出了稳定币卡支持能力。Marqeta还通过Mastercard Open Finance和Plaid增加了供应商支付直通式处理（STP）以及开放银行功能，用于账户关联、余额查询、身份验证、客户开户与资金划转。

管理层业绩指引

对于2026财年第四季度，管理层预计：

• 毛利润增长5%–7%。
• 净营收增长3%–5%。
• 调整后运营费用环比增长，同比基本持平。
• 调整后EBITDA增长在20%中段区间。
• 按GAAP计算的净利润在数百万美元的中高段。

第四季度净营收预期较低，部分反映出有一家客户直接与发卡银行签订了合同。相关的银行和网络费用将不再经过Marqeta的损益表，从而等额减少净营收和营收成本，但不会影响毛利润或处理规模。

对于2026财年全年，管理层预计：

• 毛利润增长11%–12%，与此前指引持平。
• 净营收增长11%–12%，下调1个百分点。
• 调整后EBITDA增长在30%低段至中段区间，高于此前预期的30%低段区间。
• GAAP净利润为1.35亿至1.40亿美元，其中包括1.07亿美元的非现金税收收益。

公司表示尚未看到支出或消费者行为发生显著变化，并假设2026年剩余时间的支出模式将保持一致。

风险与关注焦点

Cash App发卡渠道的多样化仍是一项主要不利因素。Marqeta预计其在新发卡中的份额将继续下降，到12月底新发卡量将微乎其微甚至降至零。管理层表示，这种动态将在2027年造成更大的拖累，不过传统Cash App卡的大部分存量用户仍留在Marqeta平台上。

第三季度毛利润增速较第二季度放缓约10个百分点。管理层将增速放缓归因于客户续约、与TransactPay收购案的对比基数、去年同期强劲的借贷业务增长以及预期中的Cash App发卡多样化。

净营收可能会受到客户与发卡银行及卡组织构建合同架构的方式的影响。管理层强调，这些穿透式变化可能会在不改变毛利经济效益的情况下，对报告的营收带来波动。

对于代理支付（agentic payments），管理层指出欺诈责任、持卡人身份识别和未确定的市场规则是制约因素。Marqeta预计发卡机构将发挥更大作用，因为它们掌握着客户身份、授权和消费场景数据。

分析师问答环节亮点

管理层表示，Square信用卡的迁移展现了Marqeta通过单一平台支持多种产品、银行和网络的能力。其迁移引擎在处理复杂数据映射和加密技术的同时，致力于避免对持卡人造成干扰。

在谈及2027年增长框架时，管理层指出，金融科技繁荣时期谈判的两项合同续约以及Block规模较大的价格阶梯式下调将不会再次发生。然而，Cash App的发卡过渡仍将是较大的拖累因素。正式的2027年指引预计将于2月公布。

在费用方面，管理层表示既定的投资计划仍在按计划推进。2026年前三个季度中，每季度的调整后运营费用均稳定在8300万至8400万美元。公司预计2027年费用将会增加，但表示其规模和工程效率应能保证业务量和产品能力的扩展，而无需按比例增加人员投资。

业绩电话会议完整文字记录

完整财报电话会议逐字稿

管理层陈述

Operator

Greetings. Welcome to the Marqeta, Inc. Third Quarter 2026 Earnings Call. [Operator Instructions] Please note, this conference is being recorded.

I will now turn the conference over to Jessica Miller, Head of Corporate Communications. Thank you, Jessica. You may begin.

Jessica Miller

Thank you, operator. Hello, everyone, and welcome to Marqeta's Third Quarter 2026 Earnings Call.

Hosting today's call are Mike Milotich, Marqeta's CEO; and Patti Kangwankij, Marqeta's CFO.

Before we begin, I would like to remind everyone that today's call may contain forward-looking statements. These forward-looking statements are subject to numerous risks and uncertainties, including those set forth in our filings with the SEC, which are available on our Investor Relations website, including our annual report on Form 10-K and our subsequent periodic filings with the SEC.

Actual results may differ materially from any forward-looking statements we make today. These forward-looking statements speak only as of the time of this call, and the company does not assume any obligation or intent to update them, except as required by law. In addition, today's call includes non-GAAP financial measures. These measures should be considered as a supplement to and not a substitute for GAAP financial measures.

Reconciliations to the most directly comparable GAAP measures can be found in today's earnings press release and earnings release supplement materials, which are available on our Investor Relations website.

With that, I'll turn the call over to Mike.

Mike Milotich

Thank you, Jessica, and thank you for joining us for Marqeta's third quarter 2026 earnings call. I'll start with a few highlights of our third quarter results, then I'll cover how customers are using our platform across a range of products, use cases and geographies, including executing another program migration onto Marqeta.

I'll also highlight the new capabilities we're adding to our product suite, which are enhancing our end-to-end platform and easing the burden on enterprise customers utilizing Marqeta. Finally, I'll turn the call over to Patti, who will walk through our financial performance in the quarter and our outlook for the remainder of 2026.

Our third quarter results reflect strong fundamental trends. TPV was $126 billion and grew 29%, marking the first quarter where annualized TPV exceeded $500 billion. Gross profit growth was on the high end of our expectations at 7%, but below recent levels driven by the discrete items we discussed last quarter. Adjusted EBITDA grew 28%, resulting in a 32% margin based on gross profit, continuing to show the benefits of our scale and efficiency.

GAAP net income of $115 million included $107 million noncash tax benefit. These third quarter results demonstrate the continued progress of the durable, differentiated business we are establishing. With many proof points enabling payments innovation for a diverse set of customers operating at scale, we believe we are still only scratching the surface of our addressable market. We continue to successfully serve and expand with our long-standing fintech partners while broadening our customer base with embedded finance enterprises.

In conjunction, we are also enhancing the platform that leading companies trust to build and scale their payment programs. The range and configurability of our offering, spanning debit and credit, consumer and commercial sets us apart from other issuer processors and is resonating in the market.

Let me share a couple of notable expansions with customers where we are deepening the partnership with new product offerings, adding value across their portfolio and supporting their growth objectives over the long term. First, our relationship with Block remains strong as we continue to expand our work across Cash App, Square and Afterpay with new programs and services.

The latest example is our recent announcement that Marqeta now enables the Square Credit Card on the American Express Network, extending a relationship that already includes Square's debit program. Originally launched in 2023, the Square Credit Card gives Square sellers access to a credit line that grows with their business volume. By consolidating debit and credit onto Marqeta's end-to-end platform, Square now has a single foundation to build and iterate on card products for sellers.

Our portfolio migration capabilities allow Square to seamlessly migrate existing accounts and card credentials onto Marqeta without any disruption to the cardholder experience. This highlights the major strength of our platform.

Marqeta's migration engine and white-glove approach help reduce the complexity of moving a program, so customers can transition with confidence and keep their focus on their end users. This mark -- this also marks our first program live with American Express, enabling our customers to access benefits of their network and increasing the breadth of our platform. With the support of all major card networks, Marqeta's platform offers increased flexibility and differentiation for customers across credit and debit card programs.

With Cash App, we continue to onboard new users, both to our long-standing program and the newer flexible credential offering. However, the diversification of new issuance did progress through the quarter in line with our expectations. This remains a new issuance dynamic only with the large existing card base, including millions of highly engaged users continuing as is.

Looking beyond Block, the caliber of business we are winning from expanding with existing customers to signing significant new logos continues to be a strong signal of our competitive position. We are now providing a new offering in Google Wallet for kids, an expansion of our long-standing relationship with Google. Marqeta's tokenization and spend control functionality enables Google to provide a safer digital-first way for kids to receive and spend allowances without needing a traditional bank account.

Parents are able to set daily spend moments, monitor activity and lock and unlock the balance for their kids and teens at the very start of their financial lives. We have experience with similar family and teen programs with other leading consumer brands, including Cash App and OnePay. Marqeta's platform enables our customers to set granular controls for spending based on transaction elements such as amount, merchant type or location and offers parents full visibility into spending patterns to help create healthy financial habits from an early age.

We also continue to launch new customers in Europe. This quarter, we enabled Satispay, the Italian fintech that makes money management and financial services simple and accessible, to launch its consumer debit card program, allowing users to pay with Satispay anywhere in the world in-store and online. With over 6.5 million users and 450,000 affiliated stores across Italy, Satispay selected Marqeta for its proven scale and trusted infrastructure to help power its expansion into new financial offerings for its broad user base.

Turning now to our product suite. We continue to innovate and enhance our platform to support the full range of our customers' needs. We offer an integrated product continuum that enables our customers to meet consumers and businesses wherever they are in their financial journey, whether that's prepaid, stand-alone debit, secured credit, transaction-based lending integrated with a card credential, net 30 charge card or revolving credit.

Our customers increasingly want to serve several of those needs through a single comprehensive partner, and that is where Marqeta stands apart. I'll start with stablecoin, where interest in digital asset-backed payments is building, and we are enabling stablecoin-backed card solutions through a strong ecosystem of infrastructure partners.

Following our strategic partnerships with BVNK and zerohash, we recently announced a similar collaboration with Coinbase, linking stablecoins directly to existing card rails and making it possible to use that funding source for purchases anywhere a card is accepted. Coinbase's infrastructure and global reach strengthen network and reinforce our position at the intersection of crypto and fiat payments. The collaboration builds on our existing relationship with Coinbase, where we already issue crypto-backed cards. We're also investing in commercial payment capabilities to give users the right payment method for their needs without the cost and complexity of stitching together multiple providers.

Whether it's a virtual card, a real-time bank transfer or push-to-card payment, multiple rails are available on the Marqeta platform with a single view of each payment from start to finish. Most recently, we are integrating straight-through processing at the network level, which strengthens our supplier payment capabilities by removing manual payment steps that can stall virtual card acceptance.

Suppliers enroll once and every subsequent card payment can be processed automatically directly with the supplier's payment processor. No PIN entry, virtual terminal or manual intervention is required. Straight-through processing has been around for some time, but not as a service facilitated directly by the networks, and that gives us much broader acquirer and supplier reach than before, which our customers can leverage without additional work.

Another example of additional integrations being incorporated into the Marqeta platform is functionality for account linking, balance checking and identity verification via our open banking partners, Mastercard Open Finance and Plaid. Our customers can offer fast, seamless onboarding and funding for consumer and commercial programs so the users can start transacting right away with authentication and identity verification happening seamlessly in the background.

Importantly, we deliver these capabilities in a choice of best-in-class providers through a single integration and contract with Marqeta, meaning no separate vendor to manage while leveraging our scale and expertise. Across all these product enhancements, the goal is the same, give our customers one platform that enables them to meet all the payment needs of consumers and businesses while also providing program management and value-added services that reduce our customers' burden and give the customer the tools to provide an elite end user experience.

To wrap up, our business continues to have strong momentum on several fronts, the customers we serve, the geographies where we serve them and the platform capabilities that power their innovation and end user engagement. The enhancements we're making to our end-to-end platform bring choice, scale and expertise to enterprises that run on Marqeta, strengthening our position to meet the growing demand for modern card issuing with a comprehensive solution.

We are in the very early stages of a wave of modernization in the issuing ecosystem. Successful fintechs leveraging modern platforms, many of whom using Marqeta are pushing incumbents to evaluate whether their current capabilities enable them to compete effectively. This early trend has embedded finance enterprises and financial institutions engaging with Marqeta more and more to explore their options, both for de novo offerings and existing programs requiring migration. The market can see the quality of the programs we are onboarding, demonstrating how the scale and flexibility of our platform is helping customers build and grow with confidence.

I'll now turn the call over to Patti to discuss our Q3 financial results and the expectations for 2026.

Patti Kangwankij

Thank you, Mike, and good afternoon, everyone. Our third quarter results reflect consistent execution and the operating leverage inherent to our business, which allows us to drive adjusted EBITDA margin expansion and increasing profitability over time. Net revenue grew 8% and gross profit grew 7% on a year-over-year basis, both at the high end of our expectations driven by TPV growth of 29%. Adjusted operating expenses were down 1% year-over-year, a bit lower than we expected.

We achieved this while continuing to invest in the priorities we've outlined and those investments remain on track. The combination of gross profit and adjusted operating expense outperformance resulted in adjusted EBITDA growth of 28% year-over-year, which was above our guide.

GAAP net income of $115 million included $107 million noncash tax benefit related to the release of a significant portion of the valuation allowance on our deferred tax assets. Q3 TPV was $126 billion, growing 29% year-over-year, marking our first quarter with annualized TPV of more than $500 billion.

Non-Block TPV continued to grow more than twice as fast as Block TPV. Growth within our financial services use case continues to run a little slower than the overall company. But excluding Block, financial services growth is meaningfully faster than the overall company. Cash App new issuance slowed through the quarter as expected, and we continue to expect little to no new issuance by the end of December.

Lending, including Buy Now, Pay Later, grew nearly 40% year-over-year, still very strong against a tough year-over-year comparison. Growth in this use case continues to be driven by expanding flexible network credential usage and our customers' ongoing geographic expansion on our platform. Expense management growth accelerated for the fifth consecutive quarter with volume growth rate in the high 50s year-over-year. This robust growth reflects our fastest-growing customers continuing to take share by acquiring new end users made possible by their utilization of our uniquely configurable capabilities. On-demand delivery grew in the high single digits year-over-year, below the company's overall growth rate as this is our most mature use case.

Turning to the P&L. Q3 net revenue was $176 million, growing 8% year-over-year, at the high end of our expectations. Block net revenue concentration was 41% in Q3, down from 44% a year ago and consistent with Q2 as Block programs continue growing well on our platform. Q3 gross profit was $122 million, growing 7% year-over-year at the high end of our expectations. As we expected, the renewal we discussed last quarter was signed in Q3.

That renewal, together with lapping the TransactPay acquisition, lapping very strong lending, including Buy Now, Pay Later growth and the expected diversification of Cash App new issuance accounts for the roughly 10-point deceleration from Q2 that we outlined last quarter. Q3 adjusted operating expenses were $83 million, down 1% year-over-year, slightly better than our outlook of nearly flat.

Adjusted operating expenses have consistently been $83 million to $84 million in each of the first 3 quarters this year, following the step-up in the second half of 2025 and reflect the benefits of scale. While we remain focused on efficient execution, we are continuing to invest at the pace we feel positions us well for future growth while maintaining the flexibility to pursue additional attractive investment opportunities to drive profitable growth. Q3 adjusted EBITDA was $39 million, up 28% year-over-year, ahead of our 20% to 25% outlook. This represented a margin of 22% based on net revenue and 32% based on gross profit, both representing new highs for Marqeta.

Q3 GAAP net income was $115 million, including a discrete noncash tax benefit of $107 million from the release of a significant portion of the valuation allowance on our U.S. federal and certain state deferred tax assets. This is a milestone. The release reflects our judgment that these assets will be realized based on our cumulative U.S. income and the consistency of our profitability over the last few quarters and our confidence that it will continue.

Profitability has been running well ahead of our expectations this year, and we have raised our adjusted EBITDA and net income outlook at every earnings call since the start of the year. As a reminder, this is a GAAP accounting item with no cash impact. We still hold considerable deferred tax assets and our cash tax outlook is unchanged. GAAP diluted EPS was $1.09 in Q3, including a benefit of $1.01 per share from the release of our tax valuation allowance.

EPS reflected a weighted average diluted share count of 106 million shares, down 6% year-over-year adjusted for the 1-for-4 reverse stock split that became effective on June 30. Our share repurchase activity remained ongoing. In Q3, we repurchased 700,000 shares at an average price of $16.25. We continue to believe the current valuation does not fairly represent the company's value or the market opportunity ahead of us.

As of September 30, we had $140 million remaining under the $150 million share repurchase authorization approved by the Board in August. We ended the quarter with $698 million in cash and short-term investments.

Now let me turn to our outlook. For the fourth quarter of 2026, we expect gross profit to grow between 5% and 7%, consistent with the third quarter. We now expect net revenue to grow between 3% and 5%, lower than we shared last quarter. One factor is how our customers choose to structure their programs. Choice across banks and networks is a top priority for our customers and an important source of competitive advantage for us. We structure our customer, bank and network contracts to accommodate these choices.

Our top 10 customers by gross profit work across 17 banks and 11 primary and secondary networks globally, which our platform enables seamlessly. Many of our large customers evaluate our pricing after bank and network fees. So whether those fees run through us or are paid directly generally doesn't change what we earn from them. In some cases, our scale with bank partners also benefits our customers.

In this case, one customer is choosing to contract directly with another issuing bank, which is a bank we already support. We continue to process their volume and the related bank and network fees no longer run through our P&L. Therefore, for this customer, net revenue and cost of revenue both come down by the same amount with no impact on gross profit.

This is another example of why we emphasize gross profit as the best measure of our performance. We expect Q4 adjusted operating expenses to increase sequentially and be nearly flat on a year-over-year basis as we continue to efficiently manage our investment initiatives against the second half 2025 base that stepped up meaningfully last year.

Adjusted EBITDA growth is expected to be in the mid-20s. We expect Q4 GAAP net income in the mid- to high single-digit millions. For the full year, we are maintaining our gross profit growth expectations of 11% to 12%. We are reducing our full year net revenue growth expectations by 1 point to a range of 11% to 12%, reflecting our Q4 net revenue outlook. We are not currently seeing a notable shift in spend or consumer behavior and are assuming consistent spending patterns for the remainder of the year.

Given the Q3 outperformance on the bottom line, we are raising our full year adjusted EBITDA percentage growth rate expectations from the low 30s to the low to mid-30s. We expect full year GAAP net income in the $135 million to $140 million range, including the $107 million discrete noncash tax benefit. We are continuing to strengthen our end-to-end platform by enhancing our product suite, including stablecoin-backed card solutions, supplier payment features and seamless onboarding and funding capabilities. We believe this continues to support and advance customer growth and expansion.

We see that in the programs we are onboarding, which reflect demand across both new and existing customers. Our continuum of products across geographies and use cases, along with our expertise and scale positions us to capture an evolving set of opportunities that we believe will drive long-term value for customers and shareholders.

In conclusion, we delivered strong third quarter results with gross profit growth at the high end of expectations and adjusted EBITDA and profitability above our expected ranges. Adjusted EBITDA grew 28% and adjusted EBITDA margin expanded to 32% of gross profit, a new record for Marqeta.

Looking ahead, our sizable market opportunity, strong competitive position and scalable financial model give us confidence that we can sustain healthy top and bottom line growth.

I will now turn it back over to the operator for questions.

Operator

[Operator Instructions] Our first question is from Sanjay Sakhrani with KBW.

分析师问答

Sanjay Sakhrani

Congrats on the card issuing -- credit card issuing partnership with Square. I guess, Mike, when we think about the potential from here for that business given it validates it in some ways, are you seeing more interest to pick up? Maybe you can just talk about how we should think about this win.

Mike Milotich

Yes. Thanks for your question, Sanjay. Yes, what's really interesting about this win that, again, we're seeing more and more demand for is the optionality that our platform provides allows our customers to do lots of different things and use our platform as the sort of one place that they see everything that keeps it simple for them. So if you think about with the Square American Express credit program coming over to us, and we already have their debit program, so that means they have a debit and credit program on different banks with different networks.

But the kind of one glue that brings it together is the Marqeta platform that allows that to be very, very easy for them. And so that's really where a lot of customers -- existing customers and prospects are coming to us for is they want the ability to offer a variety of products, and they might want to use different players in the ecosystem to do that. And they may even want to utilize different services. And they want like -- but they don't want to have lots of integrations and stitch things together across lots of different partners. And so a lot of our strategy is to build those components into our platform and make it simple for customers, and that's what Square is doing.

As I mentioned in the call, they originally -- or in my remarks, they originally launched this program about 3 years ago, and at the time, we didn't support American Express. And so it really wasn't an option for us to be able to support it. But I think now that it's come on to the Marqeta platform, which is very capable and something they're very familiar with, I think they believe it can be a very successful program that now -- that will drive a lot more interest because of the capabilities that they will be able to offer. So we're pretty excited about the potential of bringing the credit program now together with the debit program for Square.

Sanjay Sakhrani

Perfect. And I guess just a follow-up question, and I don't want you to give guidance necessarily for next year, but just thinking about the growth algorithm for next year, you have some renewals. Maybe you could just talk about how you think the confidence in the underlying growth trajectory for next year is panning out. You have some easier comparisons, you have some momentum here. Maybe you could just talk a little bit about how you're thinking about it.

Patti Kangwankij

Yes. Why don't I kick off with that? Thanks for the question. So we're -- yes, as you mentioned, we're not going to be giving expectations for 2027 today. We'll do that in February. But let me give you a sense of how we're thinking about it. First, we've discussed a few times this year, there are very -- a few specific things for 2026 that won't repeat again. The first being the 2 renewals of the deals negotiated during the fintech boom, one of which will have a -- one which had been in place for the full year and the second, which will affect the second half.

And then the Block reaching its larger price step down as Cash App scale to kind of this next level of growth, those won't repeat. And then more broadly, we continue to be very confident about the underlying business. As we mentioned, our fast-growing customers continue to perform well and grow on us and in many cases, expand with us. And the new deals that we're signing are bigger and more -- with more established companies. And Europe and value-added services are continuing to scale very nicely for us.

But at the same time, Cash App's diversification of new business is still phasing in, and we expect little to no new issuance by the end of December. So that -- and that will be a bigger drag next year. So there are still a number of factors that are working through, and that will affect timing and the size of what shows up in the P&L. So we'll share more in February.

Operator

Our next question is from Connor Allen with JPMorgan.

Connor Allen

I also wanted to ask about the Square Credit Card win, but maybe just more broadly, it feels like Marqeta has been scaling the ability to migrate portfolios for a few years now. And I was curious if you could talk about that a little bit, whether your ability to migrate portfolios more effectively has changed the win rate on some of these migrations versus de novo programs. Have you seen any of that shift in the last few years?

Mike Milotich

Sure. Thanks, Connor. I'll take that question. So agree that, as I mentioned in my remarks, there's a wave of modernization that's underway, and it's in the very early stages. More than 90% of the card volume still runs on legacy platforms, which makes it difficult for issuers to compete with fintechs and more and more embedded finance companies as they grow larger and get more aggressive utilizing more modern infrastructure like ours.

And so our migrations reflect that sort of wave of modernization arriving. So programs are looking to come off the older platforms and they see the benefit of what we can provide. And we do that without disrupting the cardholder experience. So the Square migration onto our platform is the fourth migration we've done in the last 24 months, 2 in credit, 2 in debit. And we have more signed that are underway as we speak. And so this continues to be a major strength and differentiator for Marqeta.

In the past, doing a migration was pretty scary and relatively unheard of. I would say, as we do more and more of these and customers are happy and they feel it went very smooth, we're seeing more and more people receptive to it. We can talk them through kind of the migration, almost product. We've created a product to facilitate this. And that combined with our white-glove approach and all our expertise about how to do this, because the 2 most complex parts of this, Connor, are related to kind of the data mapping and all the cryptography shifting that over. And the more of these we do, obviously, the better and better we get, which we think should lower the barriers for other companies looking to take on a migration to get to a modern processor.

Connor Allen

Perfect. And I did want to ask separately, if you don't mind, just about this revenue revision. If you wouldn't mind just talking a little bit more about it. It sounds like on a gross profit level, the economics aren't shifting with this customer, but does their engagement with -- directly with the issuer, the network change the way that you're engaging with them? Just maybe a little bit more on this revenue revision dynamic would be helpful for us, I think.

Patti Kangwankij

Yes. So why don't I start? So yes, for Q4, we expect our gross profit to grow 5% to 7%, and we expected slightly slower net revenue growth for 3% to 5%. And the important thing here is that our gross profit growth didn't change relative to kind of the net revenue that we're talking about.

One of the reasons that we cite is how our customers set up their programs. And one of the stats that we talked about was of our top 10 gross profit customers by gross profit, they work across 17 banks and 11 primary and secondary networks globally. When we double-click on the U.S., these top 10 customers utilize 10 banks. 7 of them use more than 1 U.S. bank and then 4 use at least 3 banks. And this flexibility is a big reason why our customers choose us and keep scaling with us. They can work across various networks and banks, and Square is a good example of that here as well.

And they can do that all seamlessly on one platform. And then we also have -- given our scale, we have sometimes better economics with banks, and we pass that on with -- to our customers. And our largest customers, they know these bank and network costs in detail, so we think about -- they think about this as all-in pricing, and so they could see where we add value and the pricing very clearly.

But this is the exact dynamic we see in Q4 with this one customer. They are working with us in the same way. We still have the same volume, but they came with us and are -- asked us to contract with their issuing bank directly, and it was easy for us to move over. And the only thing that changes now is the cost of bank and networks comes out of our revenue as well as our cost of revenue, and this didn't impact gross profit at all. And this is probably another good example of why we use gross profit as the key measure of performance.

Mike Milotich

Yes, if I was just going to add, I would say, from a business perspective, this is the kind of flexibility exactly we want to provide customers. And many -- as Patti just gave some stats, many of our customers take advantage of this. It's just sort of unfortunate that then it can be disruptive to the revenue line on our P&L, which again is why we emphasize gross profit so much because this is a great example of why that is, where the value we're adding and what we're making from this customer remains unchanged. It's just creating some noise in our revenue.

Connor Allen

Yes, the pass-through is just changing.

Operator

Our next question is from Holt Bashinsky with UBS.

Holt Bashinsky

I was hoping you could help us better understand Marqeta's role across the different Cash App card programs. So when we think about the traditional Cash App card post-purchase BNPL and prepurchase BNPL, how is Marqeta's role and the underlying economics differ across the programs, if at all?

Mike Milotich

Sure. Maybe I'll start with that one. So I would say, Holt, we -- there's -- I would classify it 2 ways. There's a component where there's the traditional Cash App product, which is what's been in the market for many, many years. And then recently, in the past year or so, they introduced a flexible credential product that brings the Afterpay capability directly into Cash App.

Both of those, we support the bulk of the traditional business, obviously, the overwhelming majority, high 90% of that business still remains on our platform. And we still probably have the majority also of the flexible credential business at this time, but the diversification of the new issuance is -- our share of new issuance has been going down for the last couple of months.

I think there's a third use case you're talking where they might be providing some lending that happens outside of the card experience, and that is not something that affects our business. That's something that's done with a direct lending agreement between Block and the end user and is done outside of the card.

I think the flexible credential was introduced so that they could make it a little bit easier to do installment-based lending and integrate that into the cardholder experience. But we are involved in both of those use cases.

Operator

[Operator Instructions] Our next question is from Michael Infante with Morgan Stanley Investment Management.

Michael Infante

I just wanted to ask, Mike, on your perspective just in terms of some of the industry commentary around single-use virtual cards and whether or not they may be perhaps too restrictive for agentic transactions. Like where do you generally see the limitations? And how is Marqeta positioning to support other credentials and payment mechanisms as that market develops?

Mike Milotich

Yes, that's obviously been a hot topic over the last few weeks. I think the, I don't know, discussion about single-use virtual cards in this case, the challenge is that it's being disconnected from the ultimate cardholder. So what's being lost with the single-use virtual card is who's actually making the purchase, and that's being masked in some of the early experimentation that's going on with agentic payments right now, which is what we think is what some people do not like.

Our view is still that virtual cards could -- are still -- could be a useful part of agentic commerce, but done from the issuer so that, that link is not lost. So you still might want to create a separate credential that then is configured very specifically and limits your risk, but it's done by the issuer itself, which then doesn't disconnect everyone from who's actually making the purchase and allowing them to manage fraud versus some of the early solutions that are in the market now are breaking that into 2 steps and therefore, the ultimate consumer is being lost to everyone else in the ecosystem in terms of who is actually making that purchase and who authorized that agent.

So our view is that the issuer for agentic to be very successful that the issuer is going to have to play a larger role because they hold the context needed to authorize agents' actions. So we expect that issuers will play a critical role alongside networks, wallets, merchants, everyone.

And it's really that the issuer has 2 advantages, and they're both related to fraud, which is sort of what is holding back agentic and has everyone nervous, right? And the first is that issuers already know their customers.

They know their spending and behavior patterns, which makes them best positioned to use the KYC or KYB that they've already done to ensure that the agent is authorized by the actual account holder. And so that's one aspect of it. The second is that the issuer already has the cardholders' trust, so -- which is very important because the cardholder is also worried about potential fraud if you look at all the surveys and everything done. And so that trust is important. So what we're focused on with agentic is building with intention and getting it right.

And there is a lot of things being tried in the marketplace. But with the rule still unsettled, we -- particularly as opposed to fraud and who owns that risk, we think it's good to be deliberate and then pick the right path. But we do think like a platform like ours has the building blocks that are needed to do agentic successfully. So we have deep experience with persistent virtual cards with tight limits, real-time authorization, spend controls, disputes, chargebacks, tokenization, kind of risk decisioning, all the component pieces that you would need.

We just think ultimately, what the solution will be is probably going to have a larger role for issuers than maybe what's being done right now on a limited basis in the market. The last thing I'll just say while we're talking about agentic that I just wanted to add is that something that's talked about less, I think, in the market, but we believe there's a real opportunity in commercial payments that is not discussed as much kind of in the public domain. And a lot of that has to do with consumer spending patterns are more random and a little more inconsistent versus in commercial AP and supplier payments, the counterparties are consistent, the flows are pretty predictable, and the compliance is very well understood.

So -- and there are already instances where businesses delegate spend of kind of software and user-defined controls. So we expect that things like procurement, bill pay and other expense flows are good candidates for agentic commerce. And so that's also something that we think in the coming quarters and months potentially, you're going to hear more about that, that maybe you don't hear as much about today.

Michael Infante

That's really good context. And then maybe just for a quick follow-up on OpEx framework as we sort of think about next year, understanding you guys aren't explicitly guiding to it today. But if I just sort of look at the year-to-date OpEx numbers in terms of trends through the third quarter, right, I think by my math, you're sub down 2% in terms of year-to-date OpEx, roughly, you're guiding to 4Q being basically flat. So as you just sort of think about the investments that you need to make in the business or perhaps not to drive the level of gross profit growth that you ultimately want to deliver, like where do you guys think you're positioned? Where do you want to lean in? And how do you sort of think about the durability of the recent trend in OpEx as we -- as you sort of think about setting models for next year?

Patti Kangwankij

Yes. Why don't I start on that one? So I'll just emphasize that our planned investments are really on track, and we're managing them very efficiently. Although the growth rate had declined to flattish versus last year, they've been very consistent through the year at $83 million to $84 million, which was a step up after the second half of last year during the CEO transition.

And that -- so -- and that has continued to -- and that will continue to grow, I think, in Q4 on a dollar basis as we scale. And so I do think we've talked a lot about OpEx efficiency over -- in general, as payment platforms scale nicely, and we're at a scale right now of $500 billion, where we won't need to invest as much for the volume growth, and it won't grow proportionately. And so we do expect to continue to get EBITDA margin expansion from this as well over time.

But right now, I feel very good about kind of where we're spending and where we're investing. And I think we also have the flexibility of our balance sheet to continue to lean in as we see fit.

Mike Milotich

Yes. I mean the only thing I'd add, I think we kind of happen to have some easy comps here because of the weird pattern of our expenses last year. So the -- as Patti mentioned, the growth rate looks odd, but the dollars tell a different story. And so I think I encourage you to look at those. There's definitely investments we want to make. So we will be growing our expenses next year, at least that's our plan. We'll talk about more how much, obviously, in February.

The only other thing I'd just point out is that our investment capacity is also pretty significant. So we do have a large engineering team that continues to get more efficient. And so that means we can still make a lot of investment and deliver a lot of new functionality without necessarily needing to make a lot of incremental investments in headcount. So we'll continue to do what we think is right for the business. But right now, what you're just seeing is a little bit of noise that's caused from last year.

Operator

This now concludes our question-and-answer session. Ladies and gentlemen, thank you for your participation. This does conclude today's teleconference. Please disconnect your lines, and have a wonderful day.