SoundHound AI Inc

NASDAQ:SOUN

Earnings Call Analysis

Market Leadership and Diverse AI Offerings

SoundHound is distinguishing itself as a market leader in AI applications with real-world implications. Owning the entire technology stack and offering fully automated solutions that are scalable across different business sizes has been a key part of their strategy. Products such as SoundHound Smart Ordering, which has been adopted by chains like Jersey Mike's, and SoundHound Employee Assist provide automation to improve efficiency and customer experience in the restaurant industry. They are also expanding internationally with their technology supporting 25 languages.

Financial Highlights and Revenue Growth

The company has witnessed tremendous growth with an average increase in revenue of over 50% in the past four years, despite global economic challenges. They foresee this growth trend to continue and possibly accelerate, supported by the adoption of their AI solutions across various industries. Revenue for Q4 was reported at $17.1 million, an 80% increase, and a full year revenue of $45.9 million. These figures met previously set outlooks, and were mainly driven by automotive royalties and contributions from new multiyear commitments and IP licensing opportunities. A point of pride is the consistently high gross margin which stood at 77.2% for Q4, implying efficient scalability.

Strategic Investments and Cost Discipline

The company has strategically invested in its administrative functions to support long-term growth. They have also emphasized cost discipline, reflected in reduced R&D and sales and marketing expenses. Through calculated investments and a focus on high ROI areas like demand generation, the company expects to bolster its traction and market presence. Evidence of fiscal prudence is also seen in a 41% year-over-year decrease in Q4 R&D expenses, paired with a 34% decline in sales and marketing expenses.

Outlook for 2024 and Beyond

Looking towards 2024, SoundHound is optimistic about driving strong growth, expressing confidence in its ability to grow the business by roughly 50% year on year each quarter. They expect to surpass $100 million in revenue by 2025 while achieving adjusted EBITDA profitability. The company plans to leverage generative AI and large language models as part of a technology paradigm shift, with ambitions to position conversational voice AI as a ubiquitous tool for accessing information and services through a variety of consumer products and devices.

Earnings Call Transcript 2023-Q4

[Operator]

Good day and thank you for standing by. Welcome to the SoundHound Q4 2023 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 speaker today, Scott Smith, Head of Investor Relations. Please go ahead.

[Scott Smith, executive]

Good afternoon and thank you for joining our fourth quarter and full year 2023 conference call. With me today is our CEO, Keyvan Mohajer; and our CFO, Nitesh Sharan. We will begin with some short remarks before moving to Q&A. We would 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 discussion 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 definition, 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 the 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 would like to turn the call over to our CEO, Keyvan Mohajer. Please go ahead, Keyvan.

[Keyvan Mohajer, executive]

Thank you, Scott, and thank you to everyone for joining the call today. Once again, we broke a new quarterly record for revenue this quarter, which surpassed $17 million and was up 80% year over year. We are winning new customers and expanding with existing ones. We continue to create value for our customers, their customers, and their employees while delighting users with our technology. At the same time, we are also driving efficiencies with an 80% year-over-year improvement in adjusted EBITDA.

At SoundHound, we've made 2 important predictions for the future of AI. The first is that AI customer service will be as necessary for every business as WiFi and electricity. Second is that speaking will become the primary way we interact with the devices around us. We believe voice and conversation is the most natural way to interact with technology, and fortunately for product creators, they can bring their products to life by simply adding a tiny, inexpensive microphone. These predictions are the foundation of our 3-pillar business strategy.

Let me start with our first pillar where we power products such as cars, TVs, and IoT devices. We have a mature history in Pillar 1. We went from 0 to over 20 automotive brands in just a few years. These brands represent over 25% of the automotive industry, and we've added several brands just in the last few months. We powered TVs from legacy giants such as VIZIO to exciting new innovators such as Telly. We are in millions of devices and process billions of queries from cars, TVs, and other forms of IoT devices. Our customers choose us because they believe our technology is the best, and we help them protect the brand, users' data, and because we partner with them to differentiate and innovate.

Then something incredible happened. The introduction of large language models and generative AI created a positive disruption in our field. While the big tech companies were disoriented and have been scrambling to make their own models, which could take years, we were able to swiftly integrate multiple generative AI models within a matter of weeks. Our integration was clever due to our years of experience and technology foundation and that introduced our SoundHound Chat AI product with generative AI. SoundHound Chat AI can integrate with real-time domains and can go back and forth between multiple models seamlessly in a single conversation session, and is able to utilize even multiple models for a single query. For example, the driver of the car can ask for real-time weather, request sightseeing recommendations that match the weather conditions, ask follow-up questions, then request to start navigation to one of the recommended destinations by simply referring to it, such as, can you navigate me to the second option you recommended?

SoundHound Chat AI has seen rapid adoption. Launched in early 2023, we are one of the first companies to show how generative AI could be integrated into a voice assistant. While others are still only making broad claims to generate headlines, we've been building applications for generative AI, thanks to our proprietary technology, which helps mitigate hallucinations and other issues that are inherent with unmanaged generative AI models, our applications are already live in production and in the hands of consumers across the world.

Let me highlight a few examples of our momentum and progress with SoundHound Chat AI. Immediately after it became generally available, auto brands saw the potential to increase in-car voice AI usage and improve the user experience using our product. That's because this is not just a simple API integration, but a seamless back and forth between large language models, content domains, and the onboard features of the car, such as the navigation system.

Throughout last year, we piloted our technology across a range of high-profile OEM brands. We found that by adding generative AI capabilities to in-vehicle voice assistants, the frequency with which drivers and passengers engage with it increased multiples and the user satisfaction increased significantly. The enhanced performance of SoundHound Chat AI for automotive has created numerous opportunities, and we are proud to announce that it has become the first in-car voice assistant with generative AI capabilities in the world to go live in vehicles beyond just an experimental pilot.

Last year, Stellantis started a pilot with DS Automobiles, and due to the incredible results, they announced going live in production in 18 countries. Recently, more brands such as Peugeot, Opel, and Vauxhall, have announced their plans to utilize SoundHound Chat AI. Given our strong partnership with Stellantis, we expect many more brands to go live into production this year. SoundHound Chat AI is increasing the product quality and user delight, but we also expect it to lead to an increase in revenue with a higher royalty rate for customers that choose to adapt, SoundHound Chat AI is enabling automakers to add value to their drivers' experiences with new functionality. We believe we are at the very beginning of a refresh that will bring exciting new opportunities for car manufacturers, drivers and SoundHound as we upsell these new features.

In addition to SoundHound Chat AI, we also launched a new incredible product called SoundHound Vehicle Intelligence in the fourth quarter. Drawing on SoundHound's voice AI allows automakers to simply and easily voice-enable the car manual. This means that drivers can use voice and conversation to ask about settings, special features, troubleshoot and more without having to spend time and effort leafing through a cumbersome printed document. SoundHound Vehicle Intelligence is another innovation that we expect will improve product quality and increase our revenue.

We are working closely with well-known EV brands and have been gaining strong traction with luxury brands as well. For example, we won the business for a prominent U.S.-based EV major to voice enable their full fleet of market-leading vehicles. In addition, we significantly extended and increased the volume of our existing committed contract with a large auto OEM until 2037. Creating value in our Pillar 1 category is not just limited to automotive. We also license our technology to smart TVs and many IoT devices. For example, we've integrated our AI voice assistant with Telly, a [ newly-started ] smart TV provider. Another customer, VIZIO, an integrated platform for cutting-edge smart TVs, is a strong opportunity with our monetization model using voice commerce such as ordering food, groceries, and supplies on the TV, which is typically the largest screen in the most social setting of every home.

For IoT devices, the use cases are endless, and we are working with more companies every day to extend or go into production in multiple areas such as with large appliance manufacturers, telcos, airplanes, robotics, printers, wearables, coffee machines, and kiosks, among many others. We are also thrilled that in Q4 we have a notable revenue contribution from a preeminent AI chip company. Although this company has been a close partner for us for a number of years, this was the first time we generated licensing revenue from the partnership, and we expect more value will be realized from this partnership in the future.

Moving on to Pillar 2 where we offer AI customer services for businesses such as restaurants, home services, personal care, and professional services. As we predicted a year ago, AI customer service is now outperforming it. User adoption is on the rise and consumers are increasingly choosing to interact with conversational AI due to its reliability and extended capabilities. We believe disruption in the customer service industry will be one of the first major commercial applications for generative AI and large language models. It is also becoming affordable and accessible to merchants of all sizes. Businesses are beginning to see AI as a necessity, just like WiFi and electricity.

We now have a proven scale in Pillar 2 as well. Today our technology is live in about 10,000 merchant locations. We are fully engaged with over 100,000 locations in our pipeline, and we have over 30 million businesses in our near-term addressable market. We are at the very beginning of what generative AI will bring to the world, and we are emerging as a market leader by offering applications with real-world use cases. There are 2 very clear reasons that we believe we are leading here. First, our solutions are built on proprietary technologies, and we alone own the entire technology stack. Second, our solutions are fully automated and therefore easily scalable and capable of serving brands of all sizes, from the largest brand in the world to your local favorite shop on the corner. We are offering easy-to-use, easy-to-implement solutions that make a real and immediate impact.

I'll share some examples across our various offerings. Our SoundHound Smart Ordering product uses voice AI to support businesses by answering all inbound phone calls, taking customers' food orders and answering numerous questions, allowing restaurants to free up their staff to focus on making food and engaging with customers in the store. Jersey Mike's, a chain with over 2,000 locations, went live in Q4 with Smart Ordering and is rolling out at a rapid pace. Targeted for an initial 50 locations, we've already reached that goal today, and we expect them to expand the partnership based on the solid results to date.

White Castle, our first drive-thru offering, continues to ramp up production. We are on pace to reach our goal of 100 lanes by the end of 2024. So far, we are also seeing our order processing of under 60 seconds, which is significantly faster compared to humans. In Q4, more large restaurant chains chose to integrate our voice AI ordering solution to help automate in-store operations, including Church's Chicken and a large hamburger chain located in the South, which have approximately 2,700 locations we can integrate our solution into.

With these larger deals for our voice AI solutions, we're also seeing a number of customers look to bundle these solutions with our other offerings. For example, we deploy our new SoundHound product called Employee Assist as restaurants look for multiple offerings to gain efficiencies in ordering and with employees. SoundHound Employee Assist is a gamechanger for businesses. It uses conversational voice AI to support restaurant employees like a copilot across a variety of tasks via their headsets. For example, it can provide ingredients and allergen information to customers. It can also train employees with step-by-step instructions to ensure the product is prepared correctly without them ever having to remove their gloves, consult a manual, or distract another staff member, significantly reducing the stress on the worker.

Another incredible SoundHound product, Dynamic Interaction, is our multimodal, full duplex voice interface for kiosks, drive-thrus, or any device with a screen. We are working with a number of partners and customers and our pipeline keeps growing. For example, Krispy Kreme has selected Dynamic Interaction. As you can see, our portfolio of restaurant solutions is taking shape and major brands, from convenience stores to multinational restaurant chains, are taking notice. They come to us to improve their order process, experience and accuracy of the order, while at the same time improving their bottom line. Today we have restaurant solutions live in all 50 U.S. states and now we are expanding internationally. With 25 languages in our arsenal, we believe we have a massive opportunity to go beyond English-speaking countries.

Next, Smart Answering is our phone answering solution that goes beyond restaurants and can be used by any company and any industry. With this product, we are empowering businesses of all types in how they respond to customers on the phone, allowing them to capture every business opportunity while also mitigating [ cost ]. Smart Answering can handle multiple calls at once, 24/7, conveniently filters out spam calls, saving hours of wasted time for any business. It also provides SMS responses, captures leads with intelligent messaging that asks important follow-up questions in addition to verbal answers around policies, [ hours ], products, services, pricing, and more. We are already live with customers with Smart Answering and the feedback is very, very positive. Given the Smart Answering experience is fully automated, we can easily offer this product to smaller businesses that in the past wouldn't have access, and to massively increase our ability to scale and offer this solution with no touch onboarding for any business, we've been piloting a self-service ecommerce approach in Q1 and seeing early adoption.

Let me talk about SYNQ3. Following up on the announcement of our acquisition of SYNQ3, we've begun the integration of their exceptional team with SoundHound. SYNQ3 brings more than 20 national and multinational chains spanning drive-thrus, fast casual, and casual dining segments, as well as convenience stores to SoundHound's fast-growing customer base. Examples include Chipotle, Casey's, Applebees, Panda Express, Papa John's, and 5 Guys. Because our voice AI is fully autonomous and does not use humans in the loop to perform, we are also able to rapidly scale our deployment without any degradation in quality. For instance, with SYNQ3, we recently completed the rollout of voice AI ordering to 2,400 Casey's general stores, which also is one of the largest pizza chains in the U.S. Jointly, we are combining nearly 2 decades of SoundHound AI innovation with decades of SYNQ3 industry expertise and established relationships.

Together, we will accelerate the deployment of leading-edge generative AI capabilities to the industry while ensuring that our technology always works to preserve the best interests of our customers, their customers, and society more broadly. The combination makes SoundHound the preeminent global provider of voice AI for restaurant, significantly extending our market reach. Although we only closed the transaction in Q1 of this year, we are already seeing tremendous momentum with their customers and the ability to offer them additional solutions such as Dynamic Interaction, Smart Answering, and Employee Assist. The synergies are already taking hold, and we are moving at a rapid pace.

As I mentioned earlier, we have 3 pillars: royalty from voice-enabled products; subscriptions from voice-enabled services; and the third pillar is monetization from connecting those services to those products. For example, drivers of a SoundHound-enabled car will be able to use their voice to order food by just talking to their car, and can do so before arriving at the restaurant, or they can make appointments to get a haircut, book a flight, or perform other types of transactions. We believe this will create the future of the voice commerce ecosystem, and SoundHound is in a unique position to realize this vision thanks to our strong footprint in both devices and [ services ].

With our expanded national restaurant reach, we see an acceleration of our Pillar 3 monetization strategy via transactions in millions of cars, TVs and IoT devices. We believe this is a powerful business model which could add significant value to all participants. We would collect a fee for generating new leads for merchants and facilitating transactions, the store or restaurant benefits from the new customers being channel in their direction, and the automakers and the device makers take a share of this economy. Most importantly, the end user will have the convenience of using their voice to get things done. We are excited to see our portfolio of customers using voice-enabled services continue to grow, which should allow us to begin to pilot this new commercial ecosystem.

In closing, our revenue has grown on an average of over 50% over the last 4 years. This incredible growth during challenging times and against global market headwinds is a testament to our strong foundation, demand for our technology, and great partnerships we've built with some of the most well-known brands in the world. Going forward, we expect to maintain this strong growth and see massive opportunities to accelerate it even further beyond this year. The timely AI revolution is creating tailwinds in our favor. We believe that SoundHound has a unique advantage with its own IP, 2 decades of experience building trust with global customers, a trove of data, and a track record of innovation to move fast and invent our way forward. We remain at the forefront of conversational AI, and that's why we have been working on Polaris, our multimodal, multilingual, generative AI foundation model, which could be another important tool for us as we position ourselves as one of the leading and major forces in this new era of AI.

With that, I'll now turn the call over to Nitesh to talk about our financial performance, key growth drivers, and outlook for the remainder of the year.

[Nitesh Sharan, executive]

Thank you, Keyvan, and good afternoon, everyone. Q4 revenue increased by 80% year over year. We finished 2023 with accelerating momentum, and we are seeing exceptionally strong interest from enterprise businesses, most notably within the restaurant and auto sectors. In fact, demand for SoundHound solutions is so high that we now have a healthy waitlist of restaurant merchants, and we're calibrating our resources to ensure we deliver for the customers. While we continue to be mindful of our pathway to profitability, delivering against this demand is paramount.

The exceptional growth we realized was coupled with cost containment. We focused on cost discipline this past year and we'll continue to do so. For the year, adjusted EBITDA improved by over 50% and in Q4 our results were even better with 80% improvement. Excluding certain costs that were necessary to fuel our growth and transformation brings us close to the positive Q4 adjusted EBITDA target we laid out last quarter. For example, the near-term impact of switching to big 4 audit firm PwC midyear while becoming a large accelerated filer and upgrading some of our internal tools contributed to the majority of these costs. We expect these expense pressures to lessen as we go forward.

The underlying drivers behind the remainder of the expenses are positive indicators of our future. First, we saw some tremendous growth opportunities in the second half of the year that we chose to invest in. We saw significant demand from enterprise restaurants in particular. Second, we saw a great opportunity to inorganically accelerate our go-to-market motion and capture a sizable and meaningful restaurant customer portfolio. The acquisition of SYNQ3 clearly positions us as the leader in the restaurant AI space. Lastly, we accelerated investments in our administrative functions, notably around internal processes and controls, to ensure we have an even stronger foundation for the sustained long-term growth we expect to deliver.

We fundamentally rearchitected this company in 2023. We are more efficient, more focused, even more nimble, and aggressively driving disruption and growth across voice-enabled products and services. Accordingly, we are updating a metric we have previously shared in what we call cumulative subscription and bookings backlog, which includes our previously reported cumulative bookings backlog to also now include new subscription revenue streams that we are focused on. Cumulative bookings backlog takes into account the prior quarter-end balance plus new bookings in the current quarter, minus associated revenue recognized. Cumulative bookings backlog is still derived from committed customer contracts and this definition remains the same as the previous one.

Subscription backlog takes into account customers where we are the leading or exclusive provider and assumes a 4-year ramp to fully scale with a total 5-year duration. We have incorporated reasonable assumptions about adoption percentages, with lower percentages applying to pilot and proof-of-concept customers. We believe we can outperform these assumptions given the faster rollouts we are currently experiencing and expectation to work with these partners for much longer. Similar to the previous definition, we do not include expected auto renewals for our Pillar 1 customers. This allows us to combine our Pillar 1 and Pillar 2 businesses into a single unifying metric.

As we have communicated before, our cumulative bookings backlog mainly represented our Pillar 1 businesses. In Pillar 2, we have previously noted that ARR standardizes the annual subscription like potential of these contracts, while also indicating the better stability and predictability building in our financial model. That said, 2 different metrics on 2 different timescales doesn't synthesize our full potential in an easy-to-understand manner, so we believe this updated figure is more representative of our medium-term revenue potential. Ultimately, we are addressing a greater-than-$100 billion rapidly growing market. So this new metric gives you a view of the tangible customer activity we have won within that larger opportunity set.

In Q3, I mentioned our cumulative bookings backlog was $341 million, with automotive being the largest constituent. I also mentioned that separately in the restaurant vertical that at full scale-out we would have 4,500 locations signed up and roughly $25 million in ARR. When we look at the combined potential of our signed-up customers at the end of 2023, across both Pillars 1 and 2, our cumulative subscriptions and bookings backlog was $661 million, up nearly 100% year over year on an apples-to-apples basis, thanks to growth in Pillar 1 and the incredible list of customers we have added in Pillar 2.

Let me now get specific on our financial results for the fourth quarter and full year. In Q4, revenue was $17.1 million, up 80%, and within our guidance for the quarter. Full year revenue of $45.9 million was also within the outlook set at the beginning of the year. Revenue growth in Q4 was predominantly driven by automotive royalties, with strong increase in units offset by slight decreases in average selling prices due to a higher volume of edge licenses that generally have a lower royalty than our cloud licenses. Note that particularly with some of our new generative AI solutions and ultimately with monetization, we think there are meaningful opportunities for unit price expansion.

We also benefited in the quarter by a strong multiyear commitment of minimum guarantee volumes of our edge solution with an automotive partner and a new IP licensing opportunity with one of the preeminent AI chipmakers that Keyvan referenced earlier. Over the full year 2023, we increased auto units by 68% and active cloud users by 55% versus the prior year, and over the last 4 years we have delivered an overall compounded annual growth rate of greater than 50%.

In Q4, our gross margins were 77.2%, up over 600 basis points year over year, largely resulting from the greater scale in our business. This helped drive gross margins above 75% for the full year, also up over 600 basis points year over year, as we increased our revenue sequentially every quarter and at the same time improved our cloud and data center efficiency throughout the year.

R&D expenses were $12.7 million in Q4, a decrease of 41% year over year, resulting largely from our corporate restructuring actions earlier in the year. Despite the expense reduction, we continue to invest in disruptive innovation and expand our existing suite of products with solutions like Dynamic Interaction, Smart Answering, Employee Assist, Vehicle Intelligence, and SoundHound Chat AI.

Sales and marketing expenses were $4.5 million in Q4, a decrease of 34% year over year, also, due to the aforementioned restructuring. We continue to invest in go-to-market and customer engagement. As mentioned earlier, we are seeing tremendous momentum and heightened customer demand, largely resulting from the investments we have been making in sales and marketing. We will continue to invest in high ROI, demand generation, and brand awareness to ensure we further build upon the current traction.

G&A expenses were $7.6 million in Q4, an increase of 3% year over year. The increase in G&A reflects 2 elements, some of which was not contemplated when we guided last quarter. First, part of our spending was related to diligence, negotiation, and closing of the acquisition of SYNQ3. Second, we accelerated investment in financial and nonfinancial processes and internal controls to support requirements under SOX 404(b) as we became a large accelerated filer. In total, these additional expenses, as compared to prior Q4, amounted to over $3 million and were the primary factors not fully encompassed in our previously provided adjusted EBITDA outlook.

Across all operating expenses, noncash employees stock compensation was $6.5 million in Q4. As a result, our operating loss for Q4 was $12.4 million, which reflects an improvement of 57% year over year. Likewise, for the full year, we saw improvements in our operating loss of 35% as we successfully grew the business while maintaining cost discipline. OI&E was $4 million of net expense for the quarter and net loss was $18 million in Q4, an improvement of 42% year over year. This led to a net loss per share in Q4 of $0.07 compared to $0.15 in the previous year, an improvement of 53%.

Adjusted EBITDA, which excludes noncash charges of stock compensation, acquisition costs, restructuring and depreciation and amortization, was a loss of $3.7 million in Q4 2023, which was an 80% year over year improvement and a sizable dollar reduction, down from an $18.8 million loss in Q4 2022. Net cash used in operating activities for the entire year ended 2023 was about $68 million, improving roughly 27% year over year. Our cash position at year end was approximately $109 million, of which $95 million was in cash and equivalents. Given additional actions we have taken in early 2024, our current total cash balance is in excess of $200 million. Our capital position is unequivocally a source of strength and gives us the security and optionality we need to drive the business forward.

With that, let me discuss our outlook for 2024. We are committed to continuing to fuel strong growth, with cost discipline and returns focus. We expect to expand with our existing automotive partners and add Pillar 1 customers. We expect our Pillar 2 businesses to grow meaningfully and increase from less than 10% of our total revenue in 2023 to more than 20% in 2024. We see the overall top line growing to within a range of $63 million to $77 million, with $70 million as the midpoint target. As we look further ahead to 2025, we believe we will cross $100 million in revenue and deliver adjusted EBITDA profitability.

Our gross margins have been in the range of 70% to 80%, providing a strong indicator of our software business profile. Since we just completed the acquisition of SYNQ3, while we work on migrating their cloud and AI infrastructure to our own, we expect to have a temporary decrease in gross margins due to some of these duplicative expenses. In addition, while the majority of SYNQ3's revenue is AI-driven, they also have a legacy call center operation which their team has been gradually upgrading with AI. We expect to further accelerate this migration, which will ultimately calibrate their gross margins to ours over time. As a result, we expect the combined company will show a onetime gross margin decrease in the early part of 2024, with steady improvement towards our historical levels as we get to the latter half of 2024 and beyond. Overall, though, we see this acquisition as roughly EBITDA neutral over 2024 and accretive beyond.

Lastly, let me comment on expected revenue seasonality for the year. Our mix of business will shift through the year, as I described, to a greater mix of Pillar 2, but we will start the year more automotive heavy just as we ended 2023. As such, we remain affected by the seasonality in the automotive sector, which tends to be higher in Q4 and lower in Q1. Again, in 2024, we see our quarterly revenue building through the year and back end weighted. We believe we can grow our business roughly 50% year on year each quarter. Let me close by rearticulating our excitement about what lies ahead. Generative AI and large language models have created a generational technological shift. The old way is out. Customers are increasingly realizing they need partners like SoundHound to help pave the way forward. We are balancing the massive long-term opportunity with the inherent near-term volatility to navigate towards our goals, and we are bringing conversational voice AI to consumers everywhere so they can seamlessly access the information and services they covet through the products and devices they most interact with.

Thank you. We will now move to Q&A.

[Operator]

[Operator Instructions] Our first question comes from Dan Ives with Wedbush.

[Daniel Ives, analyst]

Good quarter. Can you talk about, do you expect more strategic partnerships from a distribution perspective to happen over the next 6, 9 months? And is that going to be a big focus from a go-to-market?

[Keyvan Mohajer, executive]

Yes, definitely, in our Pillar 2, we see that as a big contributor. We have partnerships with, for example, Olo and Toast and Oracle and Square as channel partners, and they help us reach a larger number of merchants. And in Pillar 1, our customers are generally very large. For example, we just signed an extended deal out to 2037 with one of the largest automakers. That's a lot of commitment for a very long time, so it shows a lot of trust.

[Daniel Ives, analyst]

Okay. And then can you just talk about how the conversation with customers is changing? It seems like the vision has been there, but now you're starting to actually see the traction. Does it feel like it's more strategic in terms of conversations you're having, whether it's on restaurants, autos? Maybe you could just compare conversation today to even 6 months ago.

[Keyvan Mohajer, executive]

It's totally different, Dan. The demand is going through the roof, and we are dealing with more demand than ever before. We have, for example, very large customers, like multinational chains with tens of thousands of locations, we actually have to put them on hold to deal with someone who's a little bit bigger, so it's a temporary problem that we're dealing with, but it's a really good position to be in. So customers are now coming to us, whereas maybe 12 months ago, even 9 months ago, we were actually going after them, knocking on their door. They are now knocking on our door.

[Operator]

Our next question comes from Gil Luria with D.A. Davidson.

[Gil Luria, analyst]

I have one product question, one financial question. The product question, chip company, what's the nature of that revenue? Are they incorporating your technology in a chip? And then when they sell it into devices, you're getting a license revenue, or is it a different type of arrangement? And how does that scale once the chip makers incorporated the technology?

[Keyvan Mohajer, executive]

Yes, actually, it was not a channel partnership, it was a direct revenue. So the company paid us for our IP, and we've known this company for a number of years. They're a close friend, close partner, and this was the first time we generated revenue from their relationship, so we are incredibly excited about it, and we expect more positive things to come out of it in the future, but that's as much as we can share today.

[Gil Luria, analyst]

So then maybe another product question. You mentioned a large or high-profile EV maker in the U.S. Does that mean that you can now send the technology over-the-air updates? Because part of your gating factor so far has been incorporating into new cars only as they ship. But I think the EV makers, based in the U.S. at least, the 3 that I'm thinking about, allow for over-the-air updates. They already have a microphone built in. So does this type of deal make it possible for you to deploy your software a little more quickly and to a bigger set of automobiles, as opposed to only the ones that get shipped?

[Keyvan Mohajer, executive]

Yes, absolutely. Especially for our clouds, we could replace an existing solution, or sometimes they don't have a solution, so we could actually go to legacy cars and power them too. In this particular case, it would be for new cars and older cars that have already been sold.

[Nitesh Sharan, executive]

And maybe one thing to add, Gil, I think, a part of your question is about pace of scale. The other thing we're super excited about, as Keyvan mentioned the prepared remarks, is around generative AI opportunities that we're already adding with existing customers. So it's a build on top, and it really unlocks so many different use cases for us, and there's tremendous demand. So I think the opportunity of scaling faster and doing more per unit, revenue per unit, is a big opportunity for us.

[Gil Luria, analyst]

So then on the new backlog metric, I want to make sure I understand the duration. So it sounds like on the subscription side, you define the duration as a 5-year duration. What's the duration on the booking side, on the other side, on Pillar 1 that used to be, I believe, 6 years. Is it still the 6 years or are you now harmonizing, it is also a 5-year duration?

[Nitesh Sharan, executive]

No, great question. The current one is basically because they're committed contracts over the duration. So it's not like we're harmonizing. We're actually based on whatever is in the contract. So, in Q4, we actually had some deals that extended it. So I'd say the average of that -- I think when you take the weighted average of the full portfolio, it's still about that 6.5 years. But the actual device side, the product side did extend because we got some deals that extended the duration of the device for the Pillar 1 product. So the way, like you said, we, on the subscription side, made it 5 years to standardize, but just like we used to do with cumulative books backlog, we just tie to whatever the contract length is. Hope that makes sense.

[Operator]

Our next question comes from Scott Buck with H.C. Wainwright & Co.

[Scott Buck, analyst]

First, I was hoping I could get a little more color on your visibility into '25 and maybe why you're comfortable putting out a target today.

[Nitesh Sharan, executive]

Yes, I think it's a couple of different things. One is we're building and we're making choices to build a multi-year roadmap, and we're not playing short term, quarter to quarter, even 2 quarters. So I think with the pipeline we're building, with the customers we're working with, with the investment we're making to try to scale, we're getting increasing confidence beyond just this year. And we thought it made sense, frankly. We, through last year into this year, were giving mile markers only up until the 1-year mark. And so I think that with the bookings backlog type metric that we were talking about, there's a lot of good visibility. And frankly, it's like where we're sitting today in February, early in the year, and we know there's just a lot more pipeline and traction that we're gaining. So we felt confident enough to give an early read of 2025, which we'll obviously calibrate as we go forward. But I think it's really a sentiment of confidence and commitment we're seeing from our customers.

[Scott Buck, analyst]

That's helpful. Now, does that include any additional M&A or any potential deal you would do in the future would be on top of those figures?

[Nitesh Sharan, executive]

Yes, it doesn't require it. I will say we're super excited about the SYNQ3 acquisition and what it can bring. So from that vantage point, the 1 plus 1 equals a lot more than 2 opportunity we have with them. And then in terms of future opportunities, we'll continue to look. We think it's a very dynamic market with a lot of potential opportunities organically and then even possibly inorganically. So we certainly have nothing on the horizon to talk about there. But the numbers we laid out do not require it. If something structurally changed there, also it, obviously, as you know very well, depends on the flavor and the size and the type of M&A that possibly could be out there. So if something we do changes that architecture of that ramp, we'll let you know.

[Scott Buck, analyst]

And then on SYNQ3, I'm curious if you could give us something in terms of dollar terms, maybe what the upselling/cross-selling revenue synergy opportunity is there, or even as maybe a percentage of their legacy revenue.

[Nitesh Sharan, executive]

Yes. Let me try to think of the best way to unpack it. First of all, they bring this great portfolio of customers. We articulated the likes of Chipotle, Papa John's, Applebees, Firehouse Subs, it's -- just amazing are some customers that we're early days of having some of those conversations. We certainly had a thesis when we acquired them as to what the possibilities could be. And I'll just tell you, early days are even more exciting of what we can do together. To your point, cross selling, our solutions bringing together the different voice solutions, upselling some of the Employee Assist capabilities, Smart Answering, there's so many conversations going on. So I'm going to a little bit punt to later to give you better tangible visibility. But the roadmap we've laid out, we believe SYNQ3 will be a big contributor on that.

The impact I mentioned to margins, and particularly as they've been evolving the call center business to AI, that is an evolution we'll work through together. There is definitely -- anytime when 2 companies come together, there's a lot of work to integrate, and we're in the middle of that. But long term, I just want to clarify again, we're talking a restaurant space that for us is billions and billions of dollars of annual revenue opportunity, and we're just scratching at the surface of it. And together, I think, we can go faster in capturing a greater and greater share. We also think when we look out at the competitive landscape, it's mostly greenfield. We don't think people out there have comparable tech, not certainly in what we're seeing in terms of fully-automated solutions. And most importantly, the restaurant demand, it's exactly what they need. They have labor shortage, they have cost pressures, and they need consistency of service. And this is a solution that fits very well. So in terms of the organic opportunity is tremendous with SYNC 3. We're very excited about the synergies and the upside. And then the revenue outlook and even maybe to your first question, us coming together gives us confidence to give you a, call it, put a flag in the ground on what we think early signs of 2025 are. But I would just tell you, we wouldn't put that out there if we didn't think we could do that and even more.

[Operator]

Our next question comes from Glenn Mattson with Ladenburg Thalmann.

[Glenn Mattson, analyst]

So, I just wanted to talk a little more about the comment that was made that perhaps some customers on the restaurant side have been put on hold for a little bit in terms of because demand is just overflowing your ability to fill it right now. So I'm curious if that means that there's got to be more hiring involved, or maybe it's more of a learning process as you've scaled up from more smaller mom-and-pops to national chains and whatnot. And perhaps there's a learning curve for both sides of the equation whereby the process of implementing the software will get faster over time or that thing. Maybe you could just give some color on your thoughts there.

[Nitesh Sharan, executive]

I'll start, unless Keyvan wants to get out on top. First of all, yes, we're constantly learning. This is a new market. We're disrupting, we're growing, we're improving every day. I think, we're learning a lot, frankly, also with the acquisition. One of the things SYNQ3 has been really good at is scaling very quickly. And I think we can come together and drive improvement coupled with -- one of the beauties of our solutions, being software, it can really scale very quickly once you get the integration with point-of-sale systems, get menu ingestion, some of those things done right. And every time we add a new customer, it really expands the market of other customers we can go to that we could really rapidly accelerate. So there's a lot of positive build effect here that we can build and we can continue forward with. On the investment side, yes, we definitely want to go after the growth. When the demand is there, we don't want to say no to any customer. We want to be thoughtful of the roadmap. So we will absolutely look at our hiring and make sure that we're calibrating it. And I wanted to jump in before Keyvan, just to put my point of view, like it's always going to be cost discipline focused and so forth, and we'll be very judicious. But I think that we know the opportunity there. We're not going to shortchange it by any stretch of the imagination. So we're certainly going to invest to make sure we're capitalizing on the demand.

[Keyvan Mohajer, executive]

Yes, that's all that. And I would just add that the hold that I mentioned, we are talking about days and weeks, not really months or anything beyond that.

[Glenn Mattson, analyst]

And then last for me just on the model side. I think, Nitesh, you said that historically you've been 70%, 80% gross margin. And I guess theoretically, if you get to that north of $100 million 2025 revenue target that you put out today, a lot of the growth will be on the restaurant side or the SaaS side, which theoretically would be, I think, higher margin than the auto. So would you have us model towards the higher end of that range for 2025?

[Nitesh Sharan, executive]

I won't guide gross margin for 2025. I'll leave the guidance as I laid it out, but I will give you some color commentary to dimensionalize it. So, absolutely, long term, we are a software business, and so the margin profile should certainly be well north of 70%, and we think the EBIT margin should be 30% plus. And that's at scale, fully penetrated, I'd say multiple years down the road. We want to make sure we're investing in the near term. And then I would also say to your specific question on gross margin, I don't think it's as clean as Pillar 2 versus Pillar 1, because it depends. Even within Pillar 1, some of our edge solutions are very high margins. So it depends on the mix across both.

But I do think, yes, we'll be marching back to the 70% range, and then part of it will also be, this is going to be new where we bring on the call center business, which we'll continue to -- I think we look at Pillar 2 and the restaurants as the first gate on major disruption across customer service. I think we've said in the prepared remarks, generative AI in particular, one of the major disruptive areas is customer service, with very natural conversations to enable consumers to just seamlessly access information and services that they want.

So with that, and the reason I mentioned that is, we believe that we can scale that really quickly, and we can leverage the call center capability to get the data and the actual production, live customer interactions that will help improve our model. So there's going to be a bit of a journey to extend beyond restaurants into different verticals and finding that right mix between, maybe there is an advantage because the customer really demands a little call center support as if you go from, like, we can do the base 80% of the answering and maybe some part isn't. But those are things as we grow into Pillar 2 we'll calibrate on. So that's why I can't give you a precise answer right now. I will tell you, though, yes, long term, we absolutely should be growing back in the 70% range, but we need to keep going through the work and get closer to that finish line to give you an exact on what 2025 will be.

[Glenn Mattson, analyst]

Right. I know a lot of moving parts. I understand.

[Operator]

[Operator Instructions] Our next question comes from Mike Latimore with Northland Capital Markets.

[Mike Latimore, analyst]

Congrats on an excellent year here. So last quarter you highlighted a lot of restaurant wins like White Castle, Jersey Mike's, Krispy Kreme. What percent of those opportunities do you think you might get deployed by year end? And what are the keys to getting there, whether it's internal work or leveraging third parties?

[Nitesh Sharan, executive]

Yes, depends on the customer. So drive-thru is a little different than phone ordering. We've publicly, last year, made joint announcement with White Castle about our opportunity to scale into 100 locations this year and we're on the pace. And that has some requirements of the drive-thru set up on the hardware side, and so that's a bit of the gating, pacing. They're a company that has full corporate ownership, so that's a little bit easier to fully deploy. Get into another type of customer that has franchisees, again, that's one of the gating points. Jersey Mike's is an opportunity in phone ordering. They have, I think, over 2,000 locations at scale and went out of the gate with 50 locations that we're pushing through. And so this can move really rapidly. So I guess not to [ answer ] your question too much, but it depends on the shape and flavor of the customer. And we are getting faster. Every week that goes by, we're learning and we're pacing it faster and faster. So we're hopeful that this is why maybe one dimensionality of the answer would be the assumptions we used in the subscriptions estimate that we gave 5-year, which we believe conservatively will ramp over 4 years, and we use the 5 year duration. So if you want to use the 4 year that I conveyed as the measure, I think that's conservative, again. We're going faster than that. But that's maybe not to set too high expectations too early. That's probably a safe place to think.

[Mike Latimore, analyst]

Yes, that makes sense. And on SYNQ3, I think they had something like 10,000 locations, I believe. Do you feel like you could apply your tech to all those locations, or is there a percent that's already happy with what they have?

[Keyvan Mohajer, executive]

The solution is already great, but there is an upgrade opportunity by combining what they have with what we have. And then there's upsell opportunities by offering things like Smart Answering and Employee Assist to existing customers. And we're pursuing both of those.

[Nitesh Sharan, executive]

And I'll also add, Mike, there's also an efficiency opportunity for us as we migrate their back end systems, the cloud systems migrate to our technology. So it's also backend not only just our front end, and that's a cost opportunity for us.

[Mike Latimore, analyst]

I guess, just a basic financial question. What do you think SYNQ3 contributes in '24 and what is current share count?

[Nitesh Sharan, executive]

So in '24, it depends on the synergies and so forth, but they have a good book of business that we're growing. The reason we're hedging a bit is some of it is the AI and call center, that we're going to migrate the call center to AI. We're going to upgrade. So there's a few choices through the integration. Again, we're still pretty early in the integration process, but we're excited about the fleet of partners that they have and really mostly about the synergies that we think we could build. But it's all contemplated in our guidance, and we think together we can accelerate even faster than we could have alone. Share count, I think it's probably 270 million-ish, I don't have the number right in front of me, on the Class A side is probably where we are. We have, I think, I mentioned in the prepared remark, some capital raising that we've done this year to get the balance sheet over $200 million of cash. So that is about where we are.

[Operator]

Our next question comes from Brett Knoblauch with Cantor Fitzgerald.

[Brett Knoblauch, analyst]

It was nice that you gave us some info on, I guess, the year over year growth in auto units. I was wondering if you could maybe give us some type of indication for how many units or devices your technology is currently embedded in and if that's something you could foresee giving us on a more frequent basis going forward.

[Keyvan Mohajer, executive]

We are in the millions. We have considered disclosing the exact number. One challenge is that some of our customers don't give us unique device IDs, so then it makes it more difficult to accurately measure. So it might all look like one device, but it might be on many devices. There are ways to interpolate that, but definitely we are in the millions of cars and TVs and IoT devices.

[Nitesh Sharan, executive]

The other measure we do give, and we put it in the press release, but it's the queries, which is another an indicator of usage and activity. And that has been growing. It's been growing. We've been giving that number for a couple of years now and crossing 3.5 billion run rate is an indication of the breadth and also just the usage curve increases that we're excited about.

[Brett Knoblauch, analyst]

And then on maybe the restaurant business, I guess, any updates of how many units you guys are currently deployed in and what you're targeting for end of this year, I think, White Castle you're expecting to be in over 100 by the end of the year, but maybe more broadly speaking, given you expect Pillar 2 to be north of 20% of the business this year.

[Nitesh Sharan, executive]

Yes, I'll break apart. We do with SYNQ3 now have thousands and thousands, right? They brought a big arsenal of restaurants, the likes of customer names I gave a little earlier. And then we're always growing and have these massive deals that are currently we're working on that it's tens of thousands of potential opportunities. And I will say the pricing can range, if you're in the phone ordering, you could be in the hundreds of dollars per location. If you're in drive-thru, it could be more north of thousands. So really on our more advanced state could be well beyond that of tech. So units and location are going to be driven by the integration pace. We believe we are unpacking and addressing a market that has hundreds of thousands of locations. So what we have today is a small fraction. And I'd say in the U.S. alone, it's a million food establishments and globally, multiples of that. So we are going to grow from the 10,000 range by the end of the year quite meaningfully. And then hopefully we'll be soon talking about hundreds of thousands that we're in and the revenue numbers will follow.

[Operator]

Thank you. Showing no further questions at this time. This concludes today's conference call. Thank you for participating. You may now disconnect.