BlackBerry Limited (BB) Q1 2022 Earnings Report, Transcript and Summary

Good afternoon, and welcome to the Blackberry First Quarter and Fiscal Year 2022 Results Conference Call. My name is Jesse, and I'll be your conference moderator for today's call. During the presentation, all participants will be in listen-only mode. We’ll be facilitating question-and-answer session towards the end of the conference. [Operator Instructions]. As a reminder, this conference is being recorded for replay purposes. I would now like to turn today's call over to Tim Foote, Blackberry Investor Relations. Sir, please go ahead.

Thank you, Jesse. Good afternoon, and welcome to BlackBerry's first quarter fiscal 2022 earnings conference call. With me on the call today are Executive Chair and Chief Executive Officer, John Chen; and Chief Financial Officer, Steve Rai. After I read our cautionary note regarding forward-looking statements, John will provide a business update, and Steve will review the financial results. We will then open the call for a brief Q&A session. This call is available to the general public via call-in numbers and via webcast in the Investor Information section at blackberry.com. A replay will also be available on the blackberry.com website. Some of the statements we'll be making today constitute forward-looking statements and are made pursuant to the safe harbor provisions of applicable U.S. and Canadian securities laws. We will indicate forward-looking statements by using words, such as expect, will, should, model, intend, believe and similar expressions. Forward-looking statements are based on estimates and assumptions made by the company in light of its experience and its perception of historical trends, current conditions and expected future developments as well as other factors that the company believes are relevant. Many factors could cause the company's actual results or performance to differ materially from those expressed or implied by the forward-looking statements. These factors include the risk factors that are discussed in the company's annual filings and MD&A, including the COVID-19 pandemic. You should not place undue reliance on the company's forward-looking statements. The company has no intention and undertakes no obligation to update or revise any forward-looking statements, except as required by law. As is customary during the call, John and Steve will reference non-GAAP numbers in their summary of our quarterly results. For a reconciliation between our GAAP and non-GAAP numbers, please see the earnings press release and the supplement published earlier today which are available on the EDGAR, SEDAR and blackberry.com websites. And with that, I'll turn the call over to John.

Thank you, Tim. Good afternoon, everybody, and thank you all for joining us today. The main headline for this quarter is that we have organized our software and services business around our two biggest market opportunities, mainly IoT and cybersecurity. In the past few years, we have done a good job in product development. Last year, we launched 59 new products, and the year before, over 30. Later, I'll discuss about - more about the XDR product that we have launched this quarter, and I will provide you an update on BlackBerry IVY. As you all know, at the same time, we were delivering our products, many of you know that we've been investing in our go-to-market as well. We have in a number of -- last number of years, especially the last two quarters, we have turned up the noise on our marketing, expanded our channel and partnerships, and invested more feet on the street. Now we are pivoting the organization more heavily towards the market by creating two business units: cybersecurity and IoT. By aligning the cybersecurity and IoT business units to the main market opportunities, we will drive more focus and accountability. We will also improve our agility, being able to react to the fast-changing needs of the market. And needless to say, both of those be used at their own business dynamics. In order to accomplish this, we have recruited a number of new talents, especially with deep IoT experience. This includes Mattias Eriksson, who joined us from HERE Technologies, as the President of the IoT business unit. Mattias brings over two decades of relevant industry experience and with focus on strategy, operation and driving growth in the IoT business. Blackberry President and Chief Operating Officer, Tom Eacobacci, will focus on leading the cybersecurity business unit. Tom has deep enterprise software experience, and he's the perfect person to engineer the growth of this new business. From a financial reporting perspective, beginning this quarter, we will provide revenue and gross margin by business unit as well as other selected metrics. We believe that this additional color will help investors gain better understanding of the underlying performance of the business units, ultimately driving shareholder value. So let me start by reviewing with you the IoT business unit. This business reaches primarily QNX, but also includes IVY, Certicom, Jarvis and Radar, had a good strong quarter. Revenue came in at $43 million, which represents an increase of 48% year-over-year. Of course, a year ago, it was a pandemic hard-hit quarter. Gross margin was 84%. IoT ARR was $86 million. This growth in both revenue and AAR was achieved, despite the fact that we have global chip shortage. The shortage continues to a significant factor in the auto market in the near-term and it's no doubt currently impacting the production-driven revenue of QNX. The scale of the impact varies by region and by OEM. The impact looks to be greatest in North America and less so in Europe and Asia. One of our largest customers in North America has indicated that production in Q2 will be impacted or may be impacted by up to 50%, but others are less severe. Generally speaking, Q2 appears to be the low point, with Q3 improving and Q4 further so. The impact also looks to be smaller than that of the pandemic last year. So currently, we don't see a need to change our revenue outlook for the year, but we'll continue to assess the impact with our customer, and we'll update you again next quarter. Just as a reminder, our IoT revenue outlook remains at $180 million to $200 million for the fiscal year. In contrast to the production-based royalties, however, revenue from design activities, i.e., the development seat and professional services is strong. Unlike Q1 of last fiscal year when COVID was becoming a major issue, confidence in our OEM appears high, and we've seen a lot of design activities in progress. We are particularly pleased with two design wins this quarter. First one was with Volvo Group, who selected QNX RTOS and hypervisors. That stands for real-time operating system, sorry, guys. And hypervisors on a whole truck basis, meaning that our technology will power multiple ECUs throughout the truck. Second, we further strengthened our position in the EV market with a design win with the Shanghai-based WM Motor. The QNX OS, the outlaws -- the QNX OS and hypervisor will power their all-electric W6 SUV vehicle. In total, this quarter, we had 28 new design wins, with 17 in auto and 11 in GEM, the general embedded market. In auto, along those we just mentioned, notable design wins also include Bosch and Visteon. These design wins span hypervisor, digital cockpits, multiple-socket ADAS, platform and high-definition maps. On the GEM front, the wide range of application won in the quarter include surgical robotic arm, industrial 3D printers as well as nuclear product station. So design wins, such as Volvo, demonstrate two key secular trends that QNX business benefits from. The first one is the consolidation of lower compute power ECUs towards a few numbers higher-power chipset, such as the ARM version v8 and the x86 and 64-bit chipsets. It is on these higher-power chips that QNX operates, and as this consolidation continues, it gives QNX ever more opportunity in the car. Secondarily, there is a -- the second point -- second trend, that is, there is a trend of increasing software content per vehicle, particularly in safety critical systems, such as ADAS, gateway and digital copies. This is, of course, where QNX shine with the highest level of safety certification and has the stronger competitive -- strongest competitive advantage. Our strategy to focus on safety critical system, which we put in place a number of years ago, has allowed the business to benefit from these trends, ultimately leading to higher average revenue per car. This strategy is delivering higher value design wins and will benefit the royalty revenue backlog. The backlog metrics is calculated using contracted price and future production volume estimates provided by the customer when the design is awarded. It's important to note that this is a customer's estimate. The backlog increased from $450 million last Q1 to $490 million this first quarter. This is a 9% increase year-on-year despite the pressure on new auto designs over the last 12 months. Strategy Analytics, a leading independent research firm, recently published that QNX software is now embedded in over 195 million vehicles. That is up from 175 million confirmed the year before. Now for a brief update on progress with IVY. Driving the IVY opportunity forward remains one of our key priorities, and we're working closely with AWS to achieve this. Product development remains on track and in line with the road map. We remain on target for the early SS version to be available in October and for the production versions to start shipping next February. Customer discussion and workshops are continuing, and we remain positive about how things are progressing. This quarter, an additional 5 automakers engaged to explore IVY. This means that we're now engaged with most -- almost all of our major QNX customers. We recently announced the launch of the IVY Advisory Council. Industry leaders on a number of key verticals have signed up, including Telus Telecommunications, one of the Big 3,000 companies in Canada; GEICO Insurance, you see a lot on TV with their commercial; HERE Maps; and Cerence, which is the voice recognition auto business. Development of relevant and exciting new use cases for IVY platform remains a key priority, and we believe that the council could greatly assist us with this. Delivering relevant maps and experience provide a higher engagement model with both the consumer as well as the enterprise. Last quarter, we launched the IVY Innovation Fund, established to invest in start-ups adopting the IVY platform. Since then, we had a great response from the market, and we have reviewed over 200 prospective customers. We recently announced our first investment in an exciting start-up called Electra Vehicles. Unlike those other start-ups in the battery management space, Electra aims to not only analyze activity, but to also actively manage the battery operation using artificial intelligence. Vehicle sensor data on IVY will feed their AI-driven platform, dynamically determining factors such as driving behavior and environmental condition to optimize battery performance. In summary, IVY is progressing well, and we remain very focused on the various elements needed to make this a strong growth business and succeed. Now let me move to our cybersecurity business unit. This business unit includes our Spark endpoint security and endpoint management product, UEM, as well as AtHoc, the critical event management software, and Secusmart, secure voice and text product. GAAP revenue for the quarter was $107 million. As mentioned during the last earnings call, we now switched to a revenue -- GAAP-based revenue only. Gross margin was 57%, ARR was $364 million, and dollar-based net retention was 94%, 9-4. Over the last couple of years and prevailing -- the prevailing narrative has been that detection and remediation is the most important part of cybersecurity. However, the founding principle of Cylance and one of our main reasons that we acquired it is that prevention is far better than cure. And that's why we're a market leader in EPP. Stopping threats before they execute and start doing harm is clearly a better strategy than trying to shut them down afterwards. This quarter, we demonstrated the strategy clearly with our next-generation AV product named Protect, blocking the DarkSide ransomware, believed to have been the cause of the Colonial Pipeline cyber incidents. In fact, the 2015 version of Protect also blocked most of the variants of the same ransomware, obviously, 6 years ahead of its time. We do have the most mature AI engine in the space and the ability to block ransomware years ahead of time without the need for update net connectivity. This shows the power of our prevention-first strategy. Protect was also shown to prevent other high-profile threats, such as the Conti ransomware, Mobillion REvil and others. In addition to large enterprise customers, this AI-driven automated protection also resonates with small and medium-sized customers that don't have the resources to establish their SOCs, meaning the security operation centers. We see strong sequential growth in SMB new business pipeline of around 18%. In the quarter, we announced 2 significant new products, both of which are part of the extended detection and response, or XDR strategy, kind of the latest evolve market from EDR. The first product is Blackberry Gateway. With employees based remotely and not in the office as well as mobile becoming more prevalent, the traditional moat-and-castle model of network assets is no longer efficient or effective. In fact, VPN users, once authenticated, often has access to the entire network, including on-prem and other SaaS applications for the length of their session. Blackberry Gateway is a zero trust network access product that uses the Cylance AI to continuously authenticate network activity. The cloud AI evaluates over 30 risk factors, or we name it factors, such as downloading behavior, DNS, query, time of day, et cetera, to determine unusual activity. The second product released this quarter is Optics 3.0. It's our latest version of our endpoint detection and response market, or namely EDR. With this new version, the AI-driven engine remains at both the edge and in the cloud, allowing new real-time responses, both offline and online. This continues to be a differentiator for us. However, importantly, this new cloud-enabled product will allow even data to be stored centrally in a cloud-based data lake. This, together with a new search engine and a query language allow threat hunters to gain greater visibility. Switching to the sales front. UEM revenue in Q1 was down year-over-year, in part due to the work-from-home ramp up that we absorbed last year but didn't repeat. Let me reassure you that the UEM remains an important part of our cyber business, and we remain fully committed to it. In the quarter, we continued to secure business with our highly regulated customers. Let me start with financial services. In financial services, including Mitsubishi UFJ, Bank of China, Bank of France and the Union Bank of India. In the government and health care sector, we conduct business with the government of Canada, the U.K. NHS Health Services, University Health Network Canada, the United States Department of Energy, and Department of Commerce, the Netherlands Ministry of General Affairs, the Australian Department of Environmental and Energy, also the White House Communications Agency, U.S. Department of State, Department of Treasury and the United States Department of Defense. Also in government -- in the United States federal government, we have increased the number of AtHoc cloud FedRAMP users by 6% sequentially. From a market perspective, this quarter, we gained new business through partnership we recently announced with Verizon, Vodafone as well as Telus. With Microsoft, we have integrated our Critical Event Management product alert with Microsoft Teams. Further, as we've communicated in the past, our cyber suite, our UES platform, is on target to integrate with Intune by the end of August. This quarter, we significantly stepped up our sales hiring. The market for high-quality talent is competitive, and it has taken a little longer to increase our headcount, but we currently expect to end Q2 with around 23% more sales reps than at the start of the year. This expanded reach will help Blackberry to be in more competitive bake offs, where our product stands up well. With the recent increase of sales hiring, many of which start during Q2, billings goal is likely to be more heavily weighted to the back half of the year. Therefore, revenue is likely to be at the lower end of our $495 million to $515 million range that we gave last quarter. Moving on to licenses. Revenue for the quarter was $24 million, which is better-than-expected because some business came in early. Gross margin was 75%. The negotiation for the sale of a large portion of the patent portfolio are ongoing and good progress has been made. In fact, we have started negotiating the definitive agreement. Revenue for Q2 is likely to be in the range of $10 million to $15 million for the IP, as stated last quarter, so this has not changed. This is due to the monetization activities being limited by the ongoing negotiations. In terms of the full year outlook for the licensing business, should the sales not complete, we expect revenue to be around $100 million. Let me now hand the call over to Steve.

Thanks, John. So my comments on our financial performance for the fiscal quarter will be in non-GAAP terms, unless otherwise noted. Please refer to the supplemental table in the press release for the GAAP and non-GAAP details. As John mentioned earlier, starting this quarter, we are no longer adjusting GAAP revenue for deferred revenue acquired. This means that GAAP and non-GAAP revenue will be the same going forward and comparatives have been conformed accordingly. We delivered first quarter total company revenue of $174 million. First quarter total company gross margin was 66%. Our non-GAAP gross margin excludes stock compensation expense of $1 million. First quarter operating expenses were $138 million. Our non-GAAP operating expenses exclude $32 million in amortization of acquired intangibles, $6 million in stock compensation expense and a $4 million fair value adjustment on the convertible debentures, which is a noncash accounting adjustment largely driven by market conditions. The first quarter non-GAAP operating loss was $23 million, and the first quarter non-GAAP net loss was $27 million. Non-GAAP earnings per share was a $0.05 loss in the quarter, and our adjusted EBITDA was negative $6 million this quarter, excluding the non-GAAP adjustments previously mentioned. I will now provide a breakdown of our revenue in the quarter. Cybersecurity revenue was $107 million and IoT revenue was $43 million. Software product revenue remained in the range of 80% to 85% of the total, with professional services comprising the balance. The recurring portion of software product revenue was approximately 90%. Licensing and other revenue, as John mentioned, was $24 million. This is a little higher-than-expected as deals came in early. The monetization activity remains limited while negotiations for the potential IP sales continue. Now moving to our balance sheet and cash flow performance. Total cash, cash equivalents and investments were $769 million at May 31, 2021, a decrease of $35 million during the quarter. Our net cash position decreased to $404 million at the end of the quarter. First quarter free cash flow was negative $35 million. Cash generated from operations was negative $33 million, and capital expenditures were $2 million. Bear in mind, the first quarter of our fiscal year typically has a higher cash requirement due to payment of annual bonuses and other demands at this time. That concludes my comments, and I'll now turn the call back to John.

Thank you, Steve. Before the Q&A, I'd like to update everybody on a few things. Although we have organized along the go-to-market lines, there are a number of future high-growth opportunities that our factory lab is working on that actually harness the power of all the entire technology portfolio. The first is supplying our AI/ML engine in IoT. One good example of this is using Cylance in a car. You may or may not remember at CES a couple of years ago, we demonstrated an early version of how our Persona technology that identifies inappropriate access from the user behaviors can be applied to drivers of vehicles. We also demonstrated how our protected EPP can be used to protect the connected cars from cyber threats. They are just 2 the number of potential use cases that we are currently looking at. The second is our data lake. Drawing data for an ever-increasing number of sources allow for greater visibility and determination of the real level of risk across an organization. Essential -- this is obviously essential to zero trust applications. This applies not only to XDR, but also the increasing sensor-rich auto environment, autonomous drive and smart cities. This centralization of data and insights through our data lake can enable a whole new business model in the future. The third area is related to the recent U.S. SBOM, stands for software bill of materials, the executive orders that aim to secure the software supply chain. This comes in light of the recent incidents, including Solar Winds and the Colonial Pipeline threat -- intrusions. Combining products from our IoT products, including our Jarvis code-scanning tools, our QNX-embedded operating system and our Certicom technology with our prevention-first AI-driven cybersecurity product and services, means Blackberry offers a comprehensive approach to this issue. We have begun working closely with various government and standard-setting bodies. So before we open the line for Q&A, I'd like to summarize the key messages again. We have organized our software and services business around our key market opportunity, strengthening our management team in the process. QNX made solid progress this year -- this quarter, sorry. We're pleased with the strong design activities and the pipeline of new design wins that saw royalty revenue backlog increase year-over-year. We continue to demonstrate real progress with IVY with tangible steps forward, such as the launch of the Advisory Council as well as the first investment by the Innovation Fund. We launched 2 new important products that expand our XDR strategy, and the AI-driven prevention-first approach continues to be our focus. We're also increasing headcount -- sales headcount, and pipeline is growing, particularly for our new UES products. Our main focus is on growing the top line, and therefore, we'll continue to increase investments in both our software business units as we see double-digit billing growth this year. Finally, we remain optimistic about a successful conclusion to the negotiation of the patent portfolio itself. And with that, I would like to ask Jesse to open the line for Q&A, please.

Thank you, speakers. Participants, we will now begin the question-and-answer session. [Operator Instructions]. Speakers, our first question is from the line of Daniel Chan of TD Securities. Your line is now open.

Hi, Dan.

Hi, good evening. Hi, John. So you stated that your QNX royalty revenue backlog increased to $490 million from $450 million last year. Over what period of time do you expect that backlog to be recognized over?

Typically, the highest -- it usually is four to seven years. And typically, it's peak at 4, and then it starts moving down towards the end of the life cycle of a car. Sometime it extended more beyond that.

Okay. That's helpful. Thanks. And then you also talked about the IVY Advisory Council. Can you talk about the level of commitment partners have agreed to as part of this council and whether you plan to include major OEMs on it?

Yes. It's a great question. But before I answer that question, let me make one more comment on the backlog because I have also gotten some feedback regarding that our backlog number is very conservative. I would tell you that it is on a conservative side, and we get it from -- directly from the OEM when we win the design win and they gave us the estimate. We also have not included professional services backlog and developer seat backlog. So in the future, when we have a very solid methodology so that we just don't kind of do much of the guessing and we get a very grounded set of math, you will see that backlog number to go up, and show you all that we're going to include that. But that may take a couple of quarters. And to go back to answer your question regarding the Advisory Council. They're there to help us to define use cases, particularly in the vertical that they operate in, that the IVY could be of great help. And I don't want to exclude any OEM, but I don't think OEM would want to do that. They tend to do it one-on-one with us directly because this is obviously value add that they don't want to share. So our proprietary to themselves. I hope that makes sense.

It does. Thank you.

Sure.

Next question is from the line of Mike Walkley of Canaccord. Your line is now open.

Hi, Mike.

Hey, John. Thanks for taking my question. I was hoping you could update us on Blackberry's UEM strategy. I know there's some tough comps because of the pandemic from last year, but could you just update us on the strategy? Is it still a large piece of your cybersecurity business unit?

Yes. Yes. That's a good question. So let's see. Our spot platform is composed of UEM and UES. And UEM is very strategic to us because it is our gateway to a lot of our major customers who are completely reliant on us on security. So our strategy is continue to expand our footprint in the regulated industry, and we're on the more price-sensitive and kind of the nonregulated industries. We want to make sure that our UES platform, which is our endpoint security platform, also connect and run on it. And obviously, one of the largest installed base outside of my space here -- outside the regulated is Microsoft Intune. So this is why that we're excited about the fact that we'll have Intune released -- connected release in end of August, I believe, yes - end of August. So basically, the strategy is continue to expand the footprint that we have in the vertical like financial, healthcare and government. That's very important to us. With the UEM, with its road map. The road map is highly geared towards security and certifications and compliance and so forth. And then to bring your own device of BYOD environment. And that's the kind of the road map that UEM focused on. And then the UES is, of course, expanding on all the cybersecurity antivirus stuff. So that's our major -- that's our strategy of how we approach the market.

Great. No, that's very helpful. And just my follow-up question, just on the gross margin by division. Thanks for the updated business metrics. How should we think about gross margin trends for the businesses over time, particularly on the cybersecurity business? Where could those gross margins get to over time as the business ramps? And any reason why it might have fallen a bit sequentially?

Yes. Yes. I think the best way to answer the question is that, especially with cybersecurity, we're trying to go to the enterprise software timeless model, and so we have not deviated from that. So the gross margin ought to be maybe highly competitive, but they have the high-volume growth, somewhere between 75% to 80%. I think that will be a very good target to shoot for, for the cyber business.

Okay. And what needs to happen to maybe get there from where you are today? What would be that timeframe you think?

Timeframe, I think probably a year out, if you want me to guess that, like based that on, because if you recall, we actually have a lot of increase of headcount feet on the street this quarter. In fact, our quarter ends in August. In fact, some of them have -- has committed to sign-on and is yet to start. And with that, if I give them the time for 9 to 12 months, 6 to 9 months getting up to speed and at the same time, cultivating the pipeline to make the sales cycle work, I think about a year out, I should see some good results from this class of incoming team members here.

Great. No, thanks helpful. Thanks for taking my questions.

Sure.

Next question is from the line of Paul Treiber of RBC Capital Markets. Your line is open.

Hey, Paul.

Hi. Thanks and good afternoon. First of all, follow-up question on sales. But you mentioned in the outlook or the prepared remarks that you expect bookings -- double-digit bookings growth for the year. How should we think about the ramp or the trajectory over the year?

No, I actually didn't get it. Paul, I missed some part of your words. Do you mind...

Yes, bookings growth. How should we think about bookings growth over this coming year? How should we think about the ramp over the year relative to where we are now?

Yes. As I said earlier, also, we just recently had a lot of increase of headcount in sales. So the booking need to be back-ended this year and then continue on for next year, obviously. So I don't know whether that's the question you're asking.

And the rate of growth there, like where you expect it to go to?

Yes, yes. We do -- on bookings, we do expect that towards the end of the year, we do expect a double-digit percentage growth.

Okay, that's helpful. On cybersecurity revenue, in -- for this quarter, based on the numbers -- the historical numbers and GAAP numbers, I did take a step down. I think you mentioned UEM. Can you just elaborate on what you saw customers doing? I mean, I imagine they purchased last year. Did they churn off? Did they -- can you just elaborate on what happened there?

No. I think it's quite -- in general, it's quite steady and stable. We didn't see the growth that we're hoping for, but it will be forthcoming because we just released the EDR products. We talked about cloud -- the cloud version, the latest Optics 3.0. We just released all these new products a couple -- a quarter ago. Actually, a quarter ago. So we're seeing that pipeline being built up, and it's looking for them to come into being billings and business. So we don't -- I don't see any major movement one way or the other. But people are interested in EDR. I believe that they should be interested more in Protect. That's our job to make sure that, that message comes across. And the benefit of that could be demonstrated. And I definitely could demonstrate. One thing you can look at the Blackberry-Cylance product combo, none of these major viruses, yes. None of these major viruses have actually hit our user base, touch wood. And so -- anyway, that tells you the power of our product.

Next question is from the line of Trip Chowdhry of Global Equities Research.

Very good execution on the product front. Two questions.

Thank you.

First, regarding the battery management system. I was wondering, this is definitely an incremental market for you. There are three parts of the business model, they way I look at. There could be a design win, there could be a production part of it in the software, and there could also be a subscription part to the software that is running and managing the batteries. So among these three things, is it all the three components? Or is it only the software and subscription regarding the battery management software that QNX is running?

Yes. So thank you, Trip. So first off, it's a little early for me to answer the question. I have a preference. The preference is a usage-based revenue or a monthly subscription-type revenue. That will be my preference. Of course, that will have to be in agreement with the OEM. So demonstrating that in IVY use cases -- so it's one of the most important things that we need to do in the next 3 to 6 months. There is a demo being put together, and it will not be available until probably the end of this calendar year, as we -- as full engineering team just started working on it. So -- and in the meantime, we'll try to figure out the question -- the answer to the question that you posed. Again, I have a strong preference for this to be either usage-based or monthly recurring based.

Wonderful. The second question I have regarding your exceptional machine learning models you have, and definitely, currently, your Cylance, machine learning AI models work with only your products. Are you exploring? Or do you think it makes a business sense to open up your machine learning models to, say, other OEMs or to other ISVs, and then charge for connection or charge per second APIs? Because your product, which is gateway security. I think that is very normal. And again, that's another incremental revenue opportunity you can get over a period of time. So I was just thinking, since you have the best training models available, just licensing them or any other business model that can give more revenues to you. Your thoughts on that would be really appreciated. And again, very good execution on the product front.

Thank you. We haven't thought about licensing those models to other application, maybe I would say that. However, we are licensing that -- or licensing is the wrong word, sorry. We are embedding the lightweight agent in IoT devices, including like medical equipment and industrial equipment. So -- and some of those other technology we have, like the mobile threat detection and prevention, also uses the model. So it's being used in a different way. From a business perspective, we didn't think about doing licensing and where I could explore that, but we are more focused on doing embedding in endpoints.

Next question is from the line of Paul Steep from Scotia Capital.

Can you maybe -- either -- this one may be for both you and Steve, and I'll just make it one question. You can parse this up as you like here. Can you give us some context around the cost base? Obviously, you disclosed last quarter that you had 3,497 employees globally. And then earlier in this call, you talked about increasing the number of reps by 23% at the end of Q2. So I'm just trying to square up how we'd want to think about your cost base maybe going forward. Whether you've just incrementally shifted resources or is this like net new adds that we should all be thinking that are temporarily going to get added and then come to productivity, as you pointed out earlier?

Yes. We have not done any major or even minor reduction in force. We have moved some resources around, more for functional investment reasons, not for reduction of people. So it's probably best for you to think about it as incremental.

That's helpful. Maybe just the last one as well. In terms of new cybersecurity products, obviously, you're talking about giving the team time to ramp up. But maybe talk to us a little bit about what you're seeing from inbound client interest because you've launched a significant number of products, been on a bit of a role here in terms of new product launches.

You're assuming you're talking about the cyber side.

Yes. Sorry, cyber.

Okay. Right. Probably most of the conversation center around the Protect product. I would say, if I think about the larger opportunities and sites that we have won, the key winning product, it is the Protect. So this is why you heard me say a number of times on this call today, we're going to try to double down on the Protect side because it's a differentiator for us. And in addition to that, the AI/ML model that we have could be embedded and it could embedded without having to be updated, it's been valid for a very long time. So that's probably the largest opportunities when you think about large installation. And now, what we're trying to do is to position the XDR product. We talked about the new one called Gateway, and that provides zero trust architecture, so enterprise, especially like government, are extremely interested in those two areas.

Thank you, participants. I'll now turn the call back over to John Chen, Executive Chair and CEO of Blackberry, for closing remarks.

Thank you, Jesse. Thanks, everybody, for joining us. I know it's late in the East Coast, so I want to just -- hopefully, you are doing well. And thank some of you who have attended our annual shareholder meeting yesterday, and I'm looking forward to speaking with you folks soon. Have a great evening.

Thank you, speakers. That concludes today's conference call. Thank you all for joining. You may now disconnect.