SoundHound AI, Inc. (SOUN) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to SoundHound AI Q2 2026 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your first speaker today, Scott Smith, Investor Relations.

Good afternoon, and thank you for joining our second quarter 2026 conference call. With me today is our CEO and co-founder Keyvan Mohajer; and our CFO and co-founder, James Hom. We will begin with some short remarks before moving to Q&A. We'd also like to remind everyone that we will be making forward-looking statements on this call. Actual results could differ materially from those suggested by our forward-looking statements. Please refer to our filings with the SEC for a detailed discussion of the risks and uncertainties that could affect our business and for a discussion of the statements that qualify as forward-looking statements. In addition, we may discuss certain non-GAAP measures. Please refer to today's press release for more detailed financial results and further details on the definitions, limitations and uses of those measures and reconciliations from GAAP to non-GAAP. Also note that the forward-looking statements on this call are based on information available to us as of today's date. We undertake no obligation to update any forward-looking statements, except as required by law. Finally, this call is being audio webcast in its entirety on our Investor Relations website. An audio replay will be available following today's call. With that, I'd like to turn the call over to our CEO, Keyvan Mohajer.

Thank you, Scott, and thank you to everyone for joining the call today. I'm delighted to report an outstanding second quarter. Coming off an already strong Q1, the team executed flawlessly in Q2. This was not only a record Q2, but our largest all-time quarter with $62 million in revenue. Revenue was up 45% year-over-year, 40% sequentially and 10x higher than Q2 4 years ago, which was our first quarter as a public company. We also improved all key profitability metrics. Our gross margin improved, our adjusted EBITDA improved and our EPS improved both year-over-year and sequentially. Our results were beyond our own expectations, and a significant part of that is attributed to OASYS, SoundHound's self-learning agentic AI platform that we launched in May this year. With OASYS, we are winning in demos, we are winning in RFPs, we are winning in pilots, and we are winning as we scale with our customers in production. OASYS demos are absolutely delighting new and existing customers, and it shows in our renewal rates and our exceptional win rates. Because of our standout technology, we are now moving quickly with our prospects from demo to contract signing in a matter of months. And we're also converting pilots to large implementations at a record pace. We recently signed an 8-figure commitment in less than 90 days from the initial demo to contract signature. This is all while enterprise market for conversational and agentic AI grows exponentially with every quarter that passes. In a recent report, Gartner projected that agentic AI software spending will hit nearly $1 trillion by 2030 with a compound annual growth rate of over 60% between 2025 and 2030 as enterprises scale adoption. We have everything we need to seize a sizable portion of this. We continue to invest in growth and convert on the massive pipeline we built to start the year. And even with those key investments, we have maintained a disciplined approach to spending, achieving a 33% improvement in adjusted EBITDA this quarter. We saw success across all key metrics and every part of our business. SoundHound's enterprise AI business is fundamentally changing the market, emerging as a clear leader in what was once a fragmented landscape. SoundHound's excellence in conversational AI for enterprise also recently received heavyweight validation with SoundHound being named as a leader by analyst firm Gartner in their Magic Quadrant for conversational AI platforms. We expect these high-profile recognition to drive even greater momentum with enterprise buyers who rely on Gartner to validate market leaders with both a strong vision and proven ability to execute. Looking ahead, we are poised to increase our scale again with the acquisition of LivePerson, which is expected to close by the end of this year. When it does, we will accelerate our leadership position and extend our footprint in conversational AI with 25 of the Fortune 100 brands and a significant increase in the number of enterprise brands using our platform across a range of verticals. Our track record shows that our repeatable M&A formula working, we've been able to realize high potential turnaround opportunities with our technical innovation and financial discipline. We look forward to repeating our process with LivePerson. Our success in Q2 was driven in large part by our new OASYS platform, which we demoed extensively with existing customers and prospects before launching publicly in May. Early reactions confirm what we already knew. This is a category-leading enterprise AI platform that addresses the critical pain points that large multinational businesses face when trying to deliver a quality AI-based customer experience consistently across billions of interactions. One large health care company leader told us they evaluated other options and found OASYS was the most intuitive and easiest to navigate. OASYS has a unique feature where AI builds AI, and they were able to create agents using this feature in just a few minutes, which took much longer in other platforms. Agentic AI is an important step change for our industry. Previously, technology stacks were vertically integrated. Meaning that, for example, phone-based customer service was attached to telephony channels or product-based conversations were limited to the product platform itself. OASYS is built for a new era in which the AI conversational layer is no longer attached to a specific modality or channel. It lives independently with cross-channel orchestrated agents that are built for any and every touch point, including physical environments. This means that AI agents built on OASYS can power enterprises and businesses of all sizes in vehicle voice assistance, restaurant and retail voice AI transactions and complex customer tasks across our multiple verticals. SoundHound is a company that puts its customers first, supported by products that deliver tangible value. That's why we are proud that our biggest Q2 deals are with customers that chose OASYS. And as we continue to onboard and upgrade customers, introducing them to features that enable AI to build and continually optimize agentic workflows on their behalf. We are expanding our capabilities to support our rapid growth across key verticals. One area we are seeing very strong traction is in health care with one top 20 provider quadrupling their spend with us in Q2. We also added Champion Payer Solutions and managed health care services organization based in California and either expanded or renewed with five additional health care organizations. And we continue to strengthen our presence in banking and financial services, renewing or expanding contracts with 2 of the top 7 global banking institutions as well as one of the largest global insurers. At the same time, we are seeing strong demand from regional banks and credit unions who look to leverage our platform to scale customer service efficiently and extend operating hours. We recently partnered with a top-tier regional credit union and help them introduce automated AI agents without compromising their famously high standard of customer service or sacrificing containment rates. And in Q2, we announced that Qualitas, one of our insurance customers in Latin America now uses SoundHound AI agents to handle over 100,000 calls per month, an increase of 150% over 4 years, with our automated agents interacting with customers over issues like card assistance requests and broken glass claims, handling the complexities from the initial call to the claim resolution. These are very human moments of need and the message is clear, end customers are not resistant to automation. They're only resistant to bad service. This momentum in highly regulated industries like health care and financial services underscores our enterprise readiness. And with built in guardrails and rigorous agent evaluation, SoundHound platform provides the security and traceability needed for mission-critical AI workflows. Because of this trusted foundation and our growing credentials, we are seeing our vertical specific strength build across a number of other critical industries, including telecommunications and energy and utilities. A notable component of SoundHound's ongoing success is the excellent traction our solutions are getting in Asia. In each of the last 5 consecutive quarters, we have signed major deals in the region and we are seeing particular success with the automotive sector in China, a market that has been notoriously difficult for Silicon Valley companies to compete in. Time and time again, large Chinese OEMs are choosing SoundHound because of our mature flexible proprietary stack and due to the pure-play nature of our offering. The Chinese automotive industry is booming and where the winners are, we will be there. In Q2, we signed a 7-figure deal with a large infotainment software company to provide agent AI solutions, and we also had an IoT win with an established automotive maintenance and diagnostic company that will use our AI agents to boost operational productivity on site. Both customers will integrate with our OASYS platform which expands their possibilities for taking any built-for-purpose agents and deploying it across multiple channels. Elsewhere in the region, we also signed a 6-figure unit expansion with Indian 2-wheeler company Ultraviolette. And we entered an agreement with another large OEM customer to roll out real-time generative AI feature to vehicles in India. The first LLM enabled voice technology, the brand has rolled out in the country. Our ability to outpace our competitors in high-impact Asian markets is not accidental. These companies consistently partner with us based on the performance of our agentic technology, our broad vertical expertise and our growing experience within the region. And the same story holds for automotive globally. In Q2, we expanded with a number of our major OEM customers, including Stellantis in Europe, which saw overall higher unit adoption and added live generative AI capabilities. And Hyundai, one of our prominent automotive customers in the U.S., Latin America and India, who is adding generative AI enhancements which allow end users to retrieve LLM power search results and ask multilayered questions on the go. We continue to innovate and see opportunity and growth across restaurants and retail. Our agentic AI solution for SMBs, which we call Smart Answering more than doubled year-over-year in Q2, largely driven by excellent platform performance and increasingly sophisticated capabilities that allow for outbound calls to help businesses with retention, new registration and win-back campaigns. We also signed 3 major U.S. restaurant chains in the quarter. In Q2, restaurant renewal rates reached 100% for key accounts while client location count grew with major brands, including Jersey Mike's, Five Guys, IHOP and a prominent pizza brand with thousands of locations that are now live with more than 75% of their total locations. Last earnings, I talked about one well-known QSR customer that had conducted an analysis and reported that drive-through locations deploying SoundHound AI solutions are generating greater revenue than comparable locations that don't currently use it. I'm delighted to say that those metrics continue to improve across locations, which is fueling the path to broader expansion. Our restaurant AI solutions get better and better, and customers and businesses alike have become even more confident of their capabilities and consistency. This is proving fertile territory for cross-selling, using our land and expand strategy with our OASYS platform, which allows us to also provide IT help desk and guest relation services to our existing restaurant base. By positioning SoundHound as a one-stop shop for AI we aim to become central to the AI strategies of our top restaurant clients, thereby growing our total addressable market. Another part of our business that is experiencing rapid expansion is Voice Commerce. In Q2, we saw strategic advancements engaging with some of the biggest global manufacturers across automotive and IoT, all of whom we're excited about the prospect of this new monetization channel and what it can do for their business and the end user experience. I'm delighted to reveal that our first commercial projects will launch soon. In Q3, we will begin to pilot direct in-vehicle transactions, with a world-renowned automotive brand with a recently signed contract. We are integrating these vehicle native voice commerce capabilities seamlessly into the car navigation system, via the OASYS agentic AI platform with a broader public launch to follow. And later this year, we will launch with a global electronic manufacturer who will initially utilize SoundHound technology to enable agentic transactions, including retail and restaurant purchases right from the television screen in homes across the U.S. without the need for a phone, tablet or even an app. With Voice Commerce, we are observing and accelerating interest within the automotive industry to implement agentic AI for a broader range of use cases as well as monetizable transactional opportunities with OEMs adjusting their business models to accommodate it. It's amazing what we are doing. We are transforming the way OEMs can improve the user experience and monetize their products while delighting their customers. Our leadership position in this space is unprecedented, and we are leading the charge to a shift in the way OEMs are looking at the next-generation capabilities. The breadth of our tech suite and the depth of our stack have made us hugely appealing to enterprise clients looking for consistency across all their channels and operations. OASYS delivers on this with a single pane of glass view, alongside capabilities like AI builds AI, which slashes build times from months to minutes, and save vital resources. And we always say innovation is in our DNA. So those businesses already adopting OASYS can expect features to be added and existing capabilities to grow even more powerful. Last quarter, I gave an update on our investment in our R&D. OASYS will be powered by Polaris, SoundHound's own speech foundation model that consistently outperforms its competitors. We are also making a calculated investment to augment Polaris with our own specialized LLMs and speech synthesis, built using our own rich interaction data. We expect the results of these investments will be better quality and lower cost for our customers and more control, independence, higher profit margin and differentiation for SoundHound. I'm proud to say that we have made incredible progress with thanks to the strength of our engineering teams, which also bring together some of the best minds from the pioneer companies we've acquired over the last few years. Our SMB customers are now entirely on our own stack using our models, independent of frontier models. And we are about to go live with one of our largest healthcare customers after seeing significant improvements by making the switch. Once again, this is not just about cost savings, but also improvement in quality, such as accuracy and latency. As we lean deeper into our specific verticals, Being able to tune and deploy models using our own data has huge advantages for cost, efficiency, accuracy and AI safety. Turning to M&A and building on what we've said last quarter. We are excited that our previously announced agreement to acquire LivePerson is advancing. We recently secured all key regulatory clearances and continue to expect the deal to close before the end of the year. Bringing the two companies together will compound the winning strategy we have proven. We've now done this with SYNQ3, Allset, Amelia and Interactions, and it has become a repeatable formula for us. With each, we get faster and more efficient at integrating the strength of the business. Our first acquisition took about 2 years to fully turn around and our most recent is demonstrating early signs of acceleration in just a few quarters. What makes us confident that we can do this again and even faster with LivePerson is OASYS. OASYS is built specifically to unify the technology from every company acquired onto one platform. Harvesting the years of innovation SoundHound has developed, coupled with the great innovation these companies created. Put simply, this is an amazing opportunity to bring true end-to-end omnichannel conversational AI to some of the most recognizable enterprise brands in the world and combining decades of customer relationships and data with the speed and innovation OASYS unlocks. We are proud of the strategy, energized by the team we are bringing on board and confident this positions SoundHound for its next chapter of scale and leadership in Agentic AI. I already spoke of a number of customer highlights this quarter. But let me list out a few more of the key deals we have across a diverse number of verticals, including financial services, insurance, telecommunications, health care, pharmaceutical and restaurants. They included Rakuten Securities, one of Japan's largest online brokerages, serving over 6 million accounts. The company offers purchase, sales and brokerage of securities throughout the world, a global asset management firm providing investment management and research services worldwide to investors. A major international financial services organization headquartered in Canada, that offers life and health insurance, wealth solutions and asset management, an insurance company that offers individuals, professionals and businesses, casualty insurance products, an American health care customer that provides technology, pharmacy care and direct health care services globally, a leading provider of practice management and electronic health record solutions customized for the eye care industry, a leader in in-home nursing services, pediatric therapy, enteral nutrition and ABA therapy Healthcare. A rehab facility offering nursing care for short-term rehab, respite care and long-term care facilities. A global biopharmaceutical leader and one of the world's largest generic drug manufacturer, a British multinational telecommunications company, operating networks in 15 countries, a large QSR specializing in seafood to adopt SoundHound drive-thru solution. A major QSR known for American-style Mexican food. A sushi restaurant known for its rock music and concert theme menus. Also in restaurants, Ruby Tuesday, Habit Burger, Red Lobster, Torchy's Tacos and Lazy Dog. We also continue to make progress with channel partners. We entered into a partner agreement with a large global IT services and consulting provider specializing in comprehensive enterprise digital transformations. We also entered into a multiyear partnership with a Latin America-based company to deliver SoundHound technology to their vast network of customers spanning over 20 countries. In closing, we had an exceptional Q2 and our results demonstrate the incredible momentum SoundHound is building achieving a strong revenue performance, disciplined cost management and industry-leading platform validation. The demand for our solutions is at an all-time high. Enterprises, in particular, are looking for high ROI voice and agentic AI solutions and our OASYS platform delivers just that. Our expanding IoT ecosystem and in-house model innovation position us to lead in this new era of enterprise automation. We are excited about the progress we are seeing with the planned acquisition of LivePerson, which we expect to be in the second half of this year. With that, I'll now turn the call over to my co-founder James, to talk about our financial performance, key growth drivers and business outlook.

Thank you, Keyvan, and good afternoon, everyone. In Q2, we had $61.9 million in revenue, up 45% year-over-year. With sustained high demand for our AI solutions, it's become more evident each quarter that SoundHound is leading the charge as a pure-play conversational AI vendor. The launch of OASYS, our self-building and self-optimizing agentic AI platform has fueled excitement and accelerated deals. This high-performance platform, which draws upon the collective technical strengths of SoundHound and our acquired businesses, appeals to businesses looking to automate and scale reliable, consistent customer service resolutions while reducing costs. We are seeing traction across all industry verticals and in all pillars of our businesses, building on this OASYS momentum. As Keyvan mentioned, one of the verticals we are seeing SoundHound pull ahead of the competition is health care, where we provide automated patient services powered by AI agents that facilitate appointment making, care updates, prescription refills and more. We had seven deals in health care alone with one hitting 7 figures, and we continue to execute with manufacturers, bringing AI agents to vehicles and physical devices, particularly in Asia, where we had our largest deal of the quarter. The pieces are coming together, and we continue to execute on the strong pipeline we built earlier this year, resulting in an exceptional first half of the year. With that, let me discuss the second quarter financial results in more detail. As I mentioned earlier, Q2 revenue was $61.9 million, up 45% year-over-year. The growth was driven across multiple verticals such as health care, financial services, technology and automotive. And our broad-based expansion once again enabled us to realize strong customer diversification. Our enterprise AI business performed particularly well and continued to be the largest contributor to revenue. In automotive, we continue to accelerate our Asia business, growing substantially there 5 quarters in a row. In Q2, our GAAP gross margin was 45%, up 6 percentage points year-over-year. Our non-GAAP gross margin was 58%, which adjusts for noncash amortization of purchased intangibles and employee stock compensation was flat year-over-year, but up 8 percentage points sequentially. We continued to drive efficiencies by modernizing infrastructure, optimizing cloud spend, consolidating legacy systems and improving the efficiency of our core platforms, including shifting from third-party solutions to our own in-house ones as well as realizing synergies from acquisitions. We are being prudent on costs. While acquisitions have temporarily impacted our gross margin, we continue to explore and harvest synergies and expect to see our gross margin exceeding 70% in the future, similar to when we went public. R&D expenses were $27.1 million in Q2, up 5% year-over-year, largely due to acquisitions and related head count and development costs. Sales and marketing expenses were $16.6 million in Q2, reflecting a 5% year-over-year increase, primarily driven by acquisitions. Outside of that, the bulk of our investments here continue to be go-to-market efforts via direct and indirect sales as well as customer success to increase retention. We continue to elevate our brand and market presence continues to generate demand and convert more opportunities into pipeline. G&A expenses were $26 million in Q2, reflecting a 43% year-over-year increase primarily driven by various legal, advisory and other costs related to our acquisitions, including increased head count. We also continue to add additional resources to support necessary functions as we continue to grow. We had noncash employee stock compensation of $21 million and depreciation and amortization including the amortization of intangibles of $11 million in Q2, all of which are included in our GAAP results. Our GAAP operating loss for the quarter of $43.3 million improved by 45% compared to the same quarter in the previous year. Adjusted EBITDA was a loss of $9.6 million, an improvement of 33% year-over-year. GAAP net loss was $42.8 million and GAAP net loss per share of $0.10. Non-GAAP net loss was $9 million and non-GAAP net loss per share was $0.02 in this quarter. This adjusts for items such as noncash depreciation and amortization, M&A transaction costs and stock-based compensation. Our balance sheet remains strong with cash and equivalents at quarter end of $203 million with no debt. Now let me discuss our financial outlook. Q2 was clearly a strong quarter for us as the demand for our solutions accelerated. We are moving fast to capture the underlying market shifts, which continue to be in our favor. With the strong first half of the year, we are increasing our revenue expectations to be in the range of $230 million to $260 million. Overall, this outlook affirms our expectation of another year of very strong growth. We are pleased with the cost synergies we have realized from acquisitions and the targeted investments we have made in go-to-market, such as in the channel and in technology with OASYS. We will continue to stay aggressive and capture market share by expanding within our existing customer base and attracting new customers with our rapid innovation. In closing, it's clear that our momentum is real. We are excited by the strong interest we are seeing with OASYS which is a testament to the category defining technology we continue to deliver to the market. We will continue to be disciplined on costs as we drive our business towards achieving profitable growth. We have a lot of opportunity in front of us. And we are operating in a way that reflects our optimism. With that, we'll now move to Q&A. [Technical Difficulty]

And pardon me, everyone. We are experiencing technical difficulties. Please stand by.

Okay. Then we'll continue and take Thomas Blakey with Cantor Fitzgerald.

Congratulations on the record results here in 2Q. I guess maybe to start off to talk about OASYS and especially building your own models was very interesting Keyvan. Could you just maybe talk about what you're seeing in terms of 2Q trends with Amelia and Interactions installed base and these new technologies and try to parlay that into what you're seeing for potential successes with LivePerson's installed base as you see it and potential to grow that very large revenue base in the context of these new technologies and what you're seeing in terms of 2Q trends.

Yes. Thank you for the great question. So OASYS represents really decades of innovation from several companies that have come together by merging with SoundHound. One of the things that we were really mindful of when we design OASYS was to be able to integrate the innovation of these acquisitions very seamlessly because every company that we have bought that has been pursuing AI for customer service for 20-plus years, has done innovations that are unique to them, and we absolutely wanted to bring these under the same platform. So we designed OASYS to be able to capture the innovations very seamlessly. OASYS has a lot of strength and promises. One is just absolute high-quality conversational AI and agentic AI for businesses, low latency, high accuracy, naturalness, flexibility, and kind of a better-than-human experience that our customers have been waiting for, for a long time, and it's finally here. One more thing that is very prominent in the design of OASYS is time to value. So it has a feature called AI builds AI and then self improvement after it goes live, we used to sign up a customer, for example, a health care provider with complex use cases, hundreds of APIs and would allocate a team to them to work with them for a month before they could go live. Now what used to take us months with a large team is now taking us minutes by just telling the AI what to build, we upload the documents, we give it all the APIs, and we described the use cases, sometimes we are even able to upload previous calls with humans, the transcription and it just goes and builds itself. And it shows you on the screen what it's doing and you can follow it. And in a matter of minutes, you have something that again used to take us months. And then once it goes live, it learns how to improve itself. For example, if it's missing an API, if it's missing document, if callers are asking about certain things that are not in the original design of the use cases. And it even programs the enhancements and then it allows human moderators to review those enhancements and approve them with one click. So it's -- the time to value goes down. We don't need months and months of expensive professional services for our customers. So it makes it easier for them to choose us and faster to go live. You asked a lot of questions, I'm going to answer them one by one. The next one is building our own models. So SoundHound is very unique in the sense that we have our own foundation models. We were a pioneer in voice AI. We started working on voice in a dorm room at Stanford University, a long time before even the big tech -- a lot of the big tech players are even thinking about voice. A lot of the innovations in the space were invented by SoundHound concepts like speech-to-meaning that combines speech recognition and language understanding together. And that gives us the advantage of owning the core technology compared to some of the new players that putting the pieces of the LEGO together by signing up for different APIs. We have our own models that gives us flexibility, ability to enhance things, respond to complex use cases. It also -- our Polaris foundation model for speech outperforms the other providers by a large margin in both speed and accuracy and cost. But we also, in OASYS, we integrate with the frontier models because we want to promise our customers that they get the best experience no matter where it comes from. In most cases, it will come from SoundHound, but if one frontier model does an innovation that is very unique and it delivers a better experience for their callers, we don't hesitate to bring that in. And that actually makes it more interesting for them to choose SoundHound versus going directly with the frontier model. Because if a customer goes directly with the frontier model, they are making a very risky bet that, that frontier model provider is going to be always the best and that doesn't necessarily hold. We've seen some beats others every quarter. They publish results that beat each other. So by choosing SoundHound, they are actually reducing or removing that risk by being able to tap into multiple frontier models. So we are investing in our own specialized LLMs and text-to-speech. We'll have a lot more to say about that in the near future but our smaller customers are now running entirely on SoundHound stack. So there's no use of frontier models for our smaller customers, and we are starting to now convert the bigger customers, one is about to go live very soon because they're getting much better results with our own models. And over time, we expect that the majority of all of our customer service interactions will be powered by OASYS and will be powered by our own models, which also reduced our costs at a higher quality to our users. And then your last question, if my memory serves, is what happens to the customers of the various companies you're acquiring, for example, LivePerson hopefully coming soon in the second half of this year and the companies that we have purchased in the past, like Interactions, Amelia. So all the new customers we expect will utilize OASYS. There's no reason for them not to. The legacy customers, we are converting them at a pace that we are comfortable with, right? Because some of these customers have been on a legacy platform for many years, and they are very loyal to it. They are very happy with the results. The last thing we want to do is to disturb that. So we are going -- some of them are very eager to switch and some of them will take a little bit more time. So it's really going to be at a pace that they are comfortable with it, but OASYS is ready to take them on now, especially with the AI builds AI and the self-improvement we have a lot of reasons for them to be convinced to make a switch.

Very insightful, Keyvan, and it certainly sounds like your installed base and even potential LivePerson customers will have a lot to look ahead and upgrade to. Just as a quick follow-up on the -- it's very impressive to see you continuing to execute and signing large deals intra-quarter. I'd love to get an update there in terms of the pipeline. And if you wanted to double click here on any kind of the recurring component of these large deals, I think, would help investors kind of from an outlook perspective.

Yes. Our pipeline has never been this big and our win rate has never been this good. We are very bullish and excited about our prospects. And we're not the only one that thinks this way. Gartner just published a report on the total addressable market for agentic AI is going to approach $1 trillion by 2030. So that's just a few years away. Every quarter is going up exponentially. It's 50% compounding. So -- and SoundHound has emerged as a leader in their Magic Quadrant. So you can expect a lot of the big logos are relying that to make a decision. So we are very excited about our prospects. We -- our win rate has never been this good. We win in demos, we win in RFP, we win in pilots, because of OASYS. And really, our results speak for themselves. It was beyond our own expectation, the quarter, and we think we can repeat that. You asked about large deals and repeatability. We've had large deals every quarter for the past few years. So there are components of it that may not be recurring, but they have been absolutely repeatable. So just maybe the size of it is not as recurring as one might be able to build a model around it, but the fact that it happens every quarter should show that our pipeline is strong and it's just getting better.

[Operator Instructions] our next question will come from Gil from D.A. Davidson.

Let me follow on that conversation and ask about the 8-figure deal. That's a pretty significant jump forward in terms of the types of deals that you're doing. So if you wouldn't mind, without talking about who the specific customer is, talk about what the components are for those -- for that deal what's the scope? What are the types of products you're selling into that customer to make it such a big deal?