Welcome, everyone. Thanks for coming to the 2019 Extreme Networks Analyst Day. Glad to have you here. We're also webcasting this event. If you're on the webcast, please go to our investor relations website. We have the live presentation as well as the audio feed. I just want to remind everyone, we'll be making forward-looking comments today, and the non-GAAP reconciliation to some of the financials will be available on our website, along with the presentation that you'll see today. We'll post that online on the website. You're going to have the ability to hear from our key executives today. Please reserve your questions for the Q&A sessions. I encourage you to look at the demos. We have some great solutions out in the hallway there during our breaks.
Without further ado, I'm going to open it up to our President and CEO, Ed Meyercord.
Thanks, Stan. Good morning, everybody.
Good morning.
I just want to thank everybody for being here. It means a lot to have our investor community. I just want to welcome and acknowledge that we have a few members of our board, Kathleen Holmgren, Edward H. Kennedy, Raj Khanna. If you guys want to stand up and wave right quick, we have a few of our directors here. We also have a few customers, as Stan mentioned. It's always nice. It's one thing to hear from us, it's also nice to have an opportunity to hear through our customer lens. We have a team from New Jersey Transit. Thank you guys for joining. A customer interesting use case. With Verizon, talking about some of our data center solutions and wireless solutions. We always talk about the NFL, the stadium franchise that we have, and we have Chip Suttles here from the Seattle Seahawks.
Chip, where are you? All the way in the back there. They'll be in our customer panel, we appreciate everybody coming in here. It's kind of interesting because the team has been here for almost four years. This is the new Extreme team, it's kind of interesting to have a title of a chart that says we're just getting started, right? If you've been here for four years, it feels like you've been at it for a while. The reality is, this is how we feel. This is how we feel at Extreme. Coming off a leadership meeting last week in San Jose, 70 of the top leaders of the company globally in all the cultural issues about concerns about the Avaya fabric not being developed because the Brocade team wants an IP fabric or cultural infighting, just gone.
The new talent in the room, impressive. We're attracting amazing people at Extreme. The issues of making several acquisitions in a very short period of time that we have to deal with, it's all on the Extreme platform. We all have it behind us. What you're going to find out today when you get to hear from our team, I think you may be pleasantly surprised to gain a deeper understanding of where we're going with our technology, where we're going with our roadmap, how that's going to position us, to hear about our go-to-market, and how we envision the future here at Extreme. Again, thank you. I'm going to jump in and cover some high-level things about where we are at Extreme today.
Ringing the bell here was fun, it's hard to believe that Extreme is going to be a NASDAQ-listed company for 20 years as of April. In the morning comments, what I said, there's never been a better time to be at Extreme on many levels. We got into this, if we look at our customers now, we have a huge base of customers. We have a very robust portfolio now, we have more opportunities than ever before to work with our customers to help them with digital transformation. Digital transformation is the buzzword. We hear about it all the time. What does it really mean? At the core of digital transformation is networking, because everything runs over the network. We view the network as being very strategic.
We view ourselves as being a market leader in networking infrastructure, which today is really about software, less about hardware, is where we're very well-positioned, and we have a really good opportunity. We're over $1 billion in revenue with all the acquisitions. We have 3,000 employees globally. We have a partner network. Our partners are excited about all the opportunities they have to work with Extreme today. Echoing the refrain is there's never been a better time to be with an Extreme, comes from our partners who are finding more ways to deliver Extreme solutions to their customers. The pieces are in place. We're no longer talking about the acquisitions. We're not going to talk about the old logos and the brand names because now it's all come together into the one Extreme where we're much stronger from a competitive position.
People know about Extreme in the marketplace. A few years ago, people might have said, "Oh, Extreme, are they still in business? Is the brand still relevant?" Absolutely. With what we've done, Extreme is clearly on the map. The Gartner rankings help us with that as well. We have critical mass, we have critical scale. At $1 billion, you're running a global company, service depots, sales teams all around the world. You need to have a certain level of revenue to be competitive, and we're there. Was that number 800? Was it $1 billion? We're there. We feel $1 billion puts us where we're at a level playing field with larger competitors where we can compete, and we can absorb that fixed cost. That's there. I talked about the solutions and the culture at Extreme is a winning culture.
I think you'll get that from the team that we have on board. Everyone is committed to winning, and winning for us is growth. I just thought, this is a little gratuitous, but I thought I'd just step back. When I became CEO four years ago, almost, we reported the third quarter, and this is the second quarter that we just reported. I think the bottom line is this team has made a lot of progress. We think we're just getting going from here, but if you look at more than doubling your revenue to where we are. While doing that, expanding our gross margins. Rémi's going to talk to you about what we've done with the acquired assets that came in at much lower gross margins than anticipated, where we've been very disciplined and very focused.
Software-driven solutions, when we're leading with software and positioning our technology in that way, we sell at a higher margin. That higher margin is going to be driving our operating margins going forward. You can just see the difference it makes from where we were to where we are now, and we're just getting started. We talk about Extreme transformation. Our customers are going through digital transformation. We're talking about the changing landscape of what's going on in retail and what's going on for stadiums and stadium experience, what's going on in manufacturing, et cetera. Well, Extreme, we had an infrastructure. We made a lot of acquisitions. We migrated systems, we migrated data, we pulled in people, but we've completed that. Now we look inside the company, and we see a lot of opportunities to modernize, to update our systems and processes.
When we talk about digital transformation, you'll hear me say it's about what's the experience of a customer, what's the experience of a partner, and we're pulling friction out of the business, and we're doing a lot of things that are going to facilitate and drive our growth in the future. That's what we mean. We have an integrated roadmap. I think you'll like what you hear from Nabil in terms of watching us, how these portfolios have come together. You'll hear from Eric how we view the landscape changing from operating system-centric to more cloud-centric delivery of services over the network and how well we're positioned there. Software development, we talked about the shift from hardware to software. Yes, it's 96% of our R&D spend is in software. How do you monetize that? We've not done a good job monetizing our software. I think everybody knows that.
That's a big opportunity for us. I think you'll get a feeling listening to these presentations, you'll start to see how we're going to be able to do this over the future. Training was a big opportunity for us. We've been enabling our field. When you bring in lots of technology from lots of different portfolios, there's a lot of work that has to go into educating your teams on how to position the technology in front of our customers. We've done a ton of work there. We're not finished. I don't think we'll ever be finished here, but the progress that we're making and the steps that we're taking are having an effect, and they're in motions. We've redesigned our partner plan. You're going to hear Bob talk about that with what we're doing with partners and focusing our partners.
The partners that focusing at their Master Specializations in our solutions pillar in technology, they're doubling and they're tripling their business with us. Remember that partners with Extreme can make a lot more money than they can with the other vendors. Finally, digital engagement. If you look at our properties on the web as far as our website, the tools that are available, self-service that's available, and a lot of what we're doing with our own infrastructure, making it easier for people to engage and do business with Extreme. This is going to help drive us forward. I mentioned the board of directors. We don't usually talk a lot about our board in investor meetings, but I thought I'd just highlight the experience of the board at Extreme.
I mean, the bottom line is from the chairman all the way through, if you look at each of our directors, has a lot of operating experience and a lot of operating experience in technology. I just want to acknowledge what that means to our team. We're very transparent with the board. This is a board that is committed and dives in deep with our teams. They invest a lot of time in understanding our market, how we're going to market, our execution, and they're investing a lot in us. It's very helpful to have a board that has a lot of operating experience because they understand the kinds of things that we're going through, and they're very helpful with us. I just wanted to acknowledge our board. The same thing is true, I talked about the leadership team and the strengthening of the leadership team.
We have a core team that we put together a few years ago in terms of me coming in as the CEO and Bob coming in as the chief revenue officer. Eric is CTO, taking on engineering. Norman picking up several operating functions in addition to the marketing and biz dev. In addition, we've made strong adds to the team. Nabil, who you'll hear from Nabil, very strong on the product side and from a PLM perspective. We got him from the Brocade acquisition. Rémi, you'll hear from, great new add to our team, really happy at the contributions that he's already making, and you'll hear how he's going to be focusing as he gets there. I also would like to acknowledge Katy Motiey. Where are you? Oh, Katy's all the way in the back. Katy's our general counsel.
As we are in deal mode. She is very experienced on the M&A front, a valuable member of that team, and then just overall how we're driving the company, major contributions from her. Deneen Chabrier also. Dean just said she's our chief people officer. She's helping us build and drive leadership management. Some pretty exciting things that we're doing with our teams that are helping us drive performance at the company. Digital transformation. We talk about this. For us, it means changing the human experience within a workplace, a retailer, what are they focused on? They want to improve the customer experience, and they're doing a lot of innovative things. They want to improve their supply chain. They're doing a lot of innovative things there. What does it rely on? What does it run on? It runs on the network. We're right there.
We're right there, and we're able to partner, and we're doing some really innovative things across our industry verticals, and we are viewing these vertical and our customer relationships as partner relationships and opportunities to help them drive digital transformation. Bottom line, explosion of devices coming on networks. I think we all know that. Every single vertical that we talk about, you can look at some facts here. Gartner says by 2020, 95% of all devices are going to be connected to the network. What happens when a device connects to a network? Why? Why is it connected to the network? It's reaching out to an application, right? It's going to report or it's going to connect to an application. Where is the application? The application is not just in the corporate data center anymore, it's in a cloud. It could be in a hybrid cloud environment.
We're focused on building the tools at the edge around automation. The functions that companies used to perform are no longer relevant because you need software tools to drive what's happening on the edge of the network with devices. Also what's happening in the hybrid cloud and being able to manage applications running in many different cloud environments today. This is where we're focused. Some really interesting things you'll hear from our customer panel, but we have so many different examples of. I'll reference some of the examples and some customer case stories in a minute, but very innovative things where people are leveraging technology and changing their business processes. We always like the Gartner ad. We're the only player that's been moving up to the right for the last four years.
Gartner has recognized our technology and our technology differentiation and putting us into the leadership quadrant in the wired wireless LAN. That's been a steady move. I will note that our larger competitors, Cisco and HPE, have actually been moving in the opposite direction, so we're converging. If you look over at the data center, we're in the challenger quadrant, noting that Extreme Networks is the only player that has end-to-end solutions that's in the top half of this quadrant. We're the only one. If you talk to third-party analysts, they'll bang the table and say, "Why aren't you hammering this home more? Because you're the premier alternative to Cisco." We talk about our competitive advantages and how we compete against the different players in our space. What's obvious when you look at this chart, obviously being number one in customer support, which consistently comes back.
The surveys that go out to our enterprise customers, they always come back and say they love the intimacy that I have with Extreme. It's a different kind of relationship. I can make a difference. We like working with them. That's been consistent since I've been involved with the company. The single pane of glass. This is a common database for all the network elements and all the devices connected to the entire network, end to end, all the devices on the edge, through campus, core, into the data center. We're using this technology. We're using all of our technology, and this is a differentiator. No one has that. You'll hear more about that and our applications. We're going to be very focused on software. We'll talk more about that.
We like to highlight total cost of ownership from Cisco because with all the acquisitions that Cisco's made, and if you look at all the technology in their portfolio, they have clearly more solutions than anyone in the market, but it's really complicated, and it's really difficult to deploy, and it requires a lot of professional services. Ultimately, customers who are working with us are going to wind up saving a lot of money because of the ease of implementation and ease of use. You'll hear more about that from us. Applications. We're going to be talking a lot about applications today. What's in store? This is going to be what's in the Extreme store, what you're going to be able to buy. Our vision is Extreme applications and solutions that are delivered over the network from the cloud, where we have our own app store.
It's not here today, delivered today in this way, but it will be in the future. What's in store in the future are going to be more applications coming from Extreme being delivered via containers over the network, and we'll talk about that. I talked about some of the customer use cases and references for us. Massive web hosting company, exclusively Extreme. High bandwidth, that's critically important for the data center. Fully automated using our automation tools, non-blocking technology. These are things that matter, and these are things that we're providing data center customers. We're very active, and we're very relevant in the data center. Hyperscale cloud is a different animal. It's a market we're not necessarily going for, but we do have really interesting applications on this side of the market with customers like Servers.com. People that you would know. Forsyth County Schools, leveraging wireless technology.
If you look at our Smart OmniEdge, how many people remember going to school where you have roll call, then you would go and you would sign in as a form? Okay. They're leveraging facial recognition technology, leveraging our Wi-Fi at the edge. They know all the students who are in the school just because they came in, as they can see them. That's how they're tracking roll call. They're using geofencing to understand who's in the environment who belongs, who's in the environment who doesn't belong. All of this is new technology. This is digital transformation. This is a different way to leverage technology. We're working with customers like this, K-12, also universities, for specific vertical applications here. People know Stop & Shop. You might not recognize Ahold Delhaize. You might recognize Stop & Shop, Giant stores, et cetera. Big retailers.
Some of the most exciting digital transformation is happening with our retailers. I talked about Kroger being one of our larger customers with cameras throughout their stores, and they're looking at consumers when they're walking through stores using facial recognition technologies. What are they looking for? They want to know if they're happy or they're sad. That's it. Whenever there's sad faces around the deli counter, they're going to go figure out what's going on. Do they have a systemic issue? There's so many different ways that our retail customers are leveraging the new technology to transform customer experience. Skoda is a great story for us. Huawei has been in the news. This is the national auto manufacturer in Prague. They love our fabric technology. We'll talk more about the fabric, but hyper segmentation, the security that it brings, the ease of deployment, ease of use.
Huawei was in almost 50% less on their bid. They're owned by Volkswagen. Procurement Volkswagen, they want the lower price. The team travels from Prague. They go meet with the Volkswagen team, convince them it's worth the higher price to go because of the savings and the ease of use in the deployment of this technology. Here's a quote. These are the kind of things that customers want the software-driven solutions that we have across the portfolio, in this case, the campus solution and manufacturing. Skoda is just one example. This is a recent win. We have these kinds of wins around the globe. I talked about our service differentiation in terms of what customers say about us. This is not us blowing our own horn saying we're great at service. This is what customers say about Extreme, and we believe it really matters.
We are a smaller company. We can provide a different level of touch, and it's consistent with what we're hearing. We're just getting started. I'm looking forward to today. I'm looking forward to feedback at the end of the day about what you think about our team as we dive a little bit deeper. At the end of the day, 32,000 customers, blue-chip customers, more opportunities now than ever today with differentiated technology to help them through digital transformation. The team's stronger, and all this is going to produce growth, and strengthening margins, stronger financial performance on top of an already strong balance sheet. I'm going to pause there, and at this point, Eric is going to get up and he's going to talk about our technology. He's going to talk about our vision. Eric just completed his fifth year at Extreme.
We recognize his five years of service, but he's done a phenomenal job. Eric is a agnostic CTO. He's not in love with any of the legacy technologies from any of the portfolios. He has a very objective view. He's very tuned in to what's going on out in Silicon Valley, where markets are going, all the different players. He's a critical team member in guiding our view of technology in the landscape. With that, Eric, I'm going to turn it over to you.
Thanks, Ed. Good morning, everyone. How's everybody doing so far?
Awesome.
All right. My name's Eric Broockman. I'm the Chief Technology and Engineering Officer for Extreme Networks. Over the next 22 minutes, what I'm going to take you through are the technology investments we're making. I'm going to tell you how we've taken the various acquisitions and put them together into a cohesive story for our customers to deliver a platform that's agile, that we can deliver new solutions quickly and make sense to the customer for an end-to-end solution from the edge to the cloud. As Ed mentioned, many of our customers are looking to do digital transformation. When we talk to our customers, there's a series of different categories that they look at to implement their digital transformation, and they tend to fall into one of these categories. Whether it's cloud related to hybrid cloud or trying to access as a service.
Whether it's mobility, such as in a retail environment, because everybody's going through the store with their phone, trying to track people, trying to help their customer experience. Whether it's analytics and better understanding footfall traffic in a store, for example, or application usage in an NFL stadium. Whether it's IoT and healthcare trying to protect all of the devices like MRI machines, blood infusion pumps, et cetera. Cybersecurity, we're not going to be a security company like a firewall company, but we can do a lot of things with regards to the infrastructure we build to greatly enhance the security posture of our customers. We're making a lot of investments in that area. It's always one of the top three things in every one of the executive briefing sessions that I have with customers that they ask about. Edge computing, it's a big buzzword of the day.
You're going to see some investments we're making there that are putting us at the forefront of enabling edge computing. Every customer wants to know about investments in machine learning and AI, and how that's going to simplify their network, make it easier to manage, and make it easier to run their network, so that they can focus on the applications in their network and the customer experience, and not on the operational details of what's going on inside the network. Automation is always key. In one of the slides coming up, you'll see that it's on the top of customers' care abouts for buying. Branch solutions. All the customers are seeing a transformation in how they deal with their branch offices, whether they are a really small office like in a Regus office, or whether it's a really big clinic that's remote from a main hospital site.
All of these areas are ones that we're investing in. This slide is an important slide for a number of reasons. First, it's an illustration of a digital transformation that happened within our industry, and it's another illustration of a fundamental strategic change that's been happening in the networking industry as well. If we had all gotten together 15 years ago and talked about Google or Amazon or Microsoft, and said, "They're going to build the world's largest data centers, and they're going to do it by themselves, and they're not going to use the industry's two largest networking companies to be the foundation of what they are doing," everybody in the room would have said that was crazy.
When they did that, and they invested in the technologies to build the hyperscale data centers, what happened was everybody said, "Well, that is a point thing that's just the hyperscalers. That's not really relevant." What people did not recognize is that was the beginning of a classic technology adoption curve, the gold curve in the middle of the slide. That was the beginning of a new paradigm, where the largest networking companies such as Cisco, were no longer the drivers of the networking ecosystem. If you look at what's happened in technology today, the technology from hyperscalers has gone into financial services, whether it's a unicorn like Uber or a classic financial service cloud. You've seen that technology now starting to influence all the telecom carriers.
Our strategy from the edge to the cloud is to leverage the technologies from that ecosystem and prepackage that and make it consumable by the classic enterprise customer. Why did the hyperscaler guys, why were they successful? They had one user experience, one interface, common management across their entire data centers. They automated the provisioning. They had centralized orchestration. They made it programmable. They gave great visibility to what was happening in their network so that they could easily remediate what was going on. They made resiliency, and they also made it multi-vendor.
They went from a world that was CLI-based, or if you're old enough to remember the DOS command prompt, they went from there to something that was a user interface, much like going to a Mac graphical user interface, so that they do this, and they added orchestration so that they could use three people for 10,000 physical servers in a data center, not a whole army of people. This is also important because an example of the technologies, Node.js is a language, Python is a language, Go is a language, ubiquity of Linux in the data center. All of the chips that are used in Ethernet switching are born in cloud applications, and then they trickle down into enterprise.
Everything that you can think of from a major ecosystem standpoint in networking today, those are all coming from the ecosystems that's developed around applications in the cloud or the hyperscalers themselves. That's an important way that we can leverage technology to be competitive with the largest players. Business transformation. The fundamental recognition from a networking standpoint is that the network OS, as a function of time, is less and less important for the digital transformation of a customer. Over time, what's more important is the driving better business outcomes through software, and in particular for us, focusing on delivering network service applications and edge computing applications. A familiar paradigm that we all have. Some of you might remember that there used to be a Motorola Razr flip phone or a Nokia flip phone. June 29th, 2007, Steve Jobs introduces the iPhone.
If the iPhone, the only thing it did was let Bob and I talk about the latest customer opportunity, and we could do a FaceTime, that would've been nice, but that's not why we bought the iPhone. The reason we bought the iPhone was because of applications. Applications gave us the reason to buy a $1,000 phone. From a network standpoint, where we're going from a paradigm is a similar paradigm, where you're going to be able to deploy applications on infrastructure that assists infrastructure, that helps business transformation, and you're going to be able to do it from a mobile device, a tablet, a laptop, a console, a network operation center. That's the paradigm that we want to deliver to enterprise, because what they care about is simplicity as they focus on their customer. We talk about applications. The question is, do the customers actually care?
If you look at Gartner's NetOps 2.0 study that they did, 49% of all the users surveyed said that the primary thing that they were looking for to buy over the coming year was software related to network automation or tools to improve the customer experience and improve the efficiency of their network. We're well-aligned with what the customers care about. Let's step back and look at the trend of major network technologies in a larger framework for a minute. If you look at speed on the left side and business value outcome on the right side, you can think of time early in networking, people did their own ASIC, that's the ASIC era.
That became the merchant silicon era. Now merchant silicon is entering a new era that I'll call runtime era, where that silicon is really programmable so that you can change how it forwards packets. That adds a lot of new capability over time. If you think about servers, we had the bare metal era, we had the VMware VM era, now we're entering what I'll call the container era. So from a software standpoint, you had the network OS that evolved to software like network management systems and other pieces of software that assisted in running a network. What happened is in the VM era, you heard of things like virtual network functions, like a load balancer or other types of things that you used to think about as a piece of hardware that now started getting delivered as software.
You heard things like service chaining and so forth in the service provider and likewise in many of the leading data centers. Now we're on the verge of going to a container era. Ultimately, that even adds a new technology that you'll hear about, and you'll even see a demonstration of shortly with our Alexa demonstration, that's serverless. How are containers doing? 35% growth rate. There must be something to containers. Why is that? They're really fast. You can spin up a container in less than a second or a few seconds. VMs take minutes. They're efficient. They use less memory, they use less CPU, they use less energy. That's really a great thing for leveraging the value of the physical assets the customer has or a cloud has. They're simpler, dramatically simpler to make a container than to generate a VM image.
They natively enable hybrid networking. They natively allow you to put an application on-premise, in a cloud, or any other place. That transportability dramatically changes the friction of being able to build a hybrid cloud application for a customer as they adopt container technology. A lot of things on this slide. Let me first of all start from the top. In many of your minds, you have, we brought a lot of companies together. The question is, how do we bring all those pieces of technology together and make them efficient, versus a set of individual silos? This is the recipe that we've used over the last 15 months to do that. First, we settled on the Extreme Management Suite of software so that we have a single pane of glass that goes across all of the portfolio.
It doesn't matter which part of the portfolio, one user interface that hides the complexity of the individual details of all the technology underneath. You don't get on a CLI and pound out something for each piece of the portfolio, you use one pane of glass. A key part of that single pane of glass is our network access control and associated suites. If you go to Gartner Peer Insights, they recently published a piece three weeks ago on what do customers say when they were interviewed about our technology versus others. If you look at it, you'll see that we were the second ranked in so far as capabilities with way more respondents than the one that was a private company that's first. More importantly, it far outdistanced the Cisco ISE capability.
Customers who engage with us and try our single pane of glass, we can enter a new customer and say, "You know what? You can use part of one person to run and do access control and management in your network," versus you can't quite figure out the 17 different tabs on ISE and make that work. That's important because ISE, frankly, is the basis of where Cisco DNA is going, and customers say, "Hey, you're delivering what the other guy's PowerPoint is talking about. This is great." That's the way that the sales team is ever-increasingly engaging to win new logos. Oftentimes it's a small amount of hardware, it's our software, and then we come behind that software and can deliver any other solution into their network. We've invested on the left side in taking all of the network OSs and now making them applications.
Those applications run on standard Linux, and we've commonized all the hardware going forward. Nabil's going to take you through the portfolio. Any piece of our hardware can run any of our network OSs, and that makes us much more efficient in the hardware development side of what we're doing. There's always some unique high-end hardware, like our super big chassis is really a high-end piece of hardware, and there's a unique piece of hardware on the other end of the network, like the Defender that you'll see. Everything else is really common, and any of our software can run on top of it. On the right side in the middle there is a very large source of differentiation for us on a going-forward basis.
For our premier tier of products and the tier right below that. We're introducing products from the SLX family to the VSP family to the EXOS Edge family that all have an x86 server in it. It runs server class Linux, the same Linux that's run 67% of all workloads that run in the cloud. Anything that can run on Linux, anything that can run in a cloud, we can run within the constraints of CPU or memory, those same applications inside of a switch. There are some very compelling value propositions to customers to being able to do that. This includes, one of the things this allows us to do is to deliver new capabilities to a customer that look like they're augmenting the network operating system without having to touch the network operating system.
If you went to your IT staff at whatever company you worked with and said, "Hey, I would like to be able to introduce new capabilities into the network every month?" They would tell you, "You obviously don't know anything about networking and get out of my office." If you said, "I would like to be able to introduce new capabilities into my network on distributed servers," they're going to go, "Oh, well, we do that all the time. What's the problem?" That changes the paradigm on how we can deliver value to a customer without them having to think about network upgrades. This is an incredibly powerful paradigm shift. Likewise, there's a new set of edge applications, and those edge applications not only in a wiring closet, any place where the network comes in contact with something outside of it, that's an edge.
There's an edge in the data center as it talks to the cloud. There's an edge on the campus core, and there's an edge in the wiring closet. Lastly, we're not going to be the only company that writes applications for networking that runs on a Linux server. There are thousands of them out there, and we can run those inside our switching infrastructure, or we can run that on our appliances. It dramatically reduces the engagement friction for partnerships. I'm not going to take you through this whole slide here on application solution catalogs. It's showing a set of applications we already have, as well as ones that are coming and sort of the categories that they're in.
As Ed mentioned earlier, maybe we haven't been ideal at monetizing all those yet, but we have a focus on that going forward that's going to allow us to make some pretty dramatic accelerations in that case, both for delivering customer outcomes as we kit everything together, and likewise, landing and expanding in new customers. This is just an example of some of them, our ExtremeAnalytics, Management, AirDefense for wireless, Location Services, the new Security Analytics, which is coming shortly. You'll see a beta of that software. That's an example of what I was talking about, where we're investing to take capabilities in the network and deliver a superior security experience for a customer to protect their infrastructure. 70% of all of the hacks that happen in a network move laterally in the network. They've already gotten through the firewall.
They've already gotten into your endpoint devices. Now they move laterally. We have technologies that can detect that, alert that, and lock that down. We have technologies that limit, through hyper segmentation, the ability of things to move laterally in a network, which makes a huge difference to customers, especially in healthcare. Then you'll see a demonstration of our Defender for IoT so that you can. When you have an MRI machine, that MRI machine is probably running Windows 95 unpatched. Think about that the next time you might have to go to an MRI machine. You really hope that that hospital healthcare facility is putting something in front of that MRI machine because you really don't want to be in the toaster oven, when a virus or malware is going off. This is the kind of device that allows us to do that.
Back to this slide to make it a little more real. Applications that we can already do today are packet capture, application analytics, automation applications. We can bring up a 5-stage class fabric in less than two and a half minutes using an application that we can add in a container on the side of this guest VM. We're the first and only networking company that can support the AWS Greengrass for IoT. We're the first to be able to do Azure IoT Hub, which allows some very interesting new applications. You'll see an example of that coming up. Likewise, there's a long laundry list of third-party applications that we could run on the switch as well.
It could be Splunk, it could be Logstash to work with various Elasticsearch, et cetera, perfSONAR for network performance, Security Onion, Bro IDS, which are open source security, Wireshark for packet capture. How's this going to look in 2020 as we roll this out? You'll be able to see in the demo center sort of the user interface and what the beginning of this looks like. You have your graphical user interface, whether that's your phone or your tablet or your desktop. It's going to be using Kubernetes to distribute the containers, Istio for a service mesh.
You'll have your applications that we talked about a few slides ago. Then for the enterprise, as I mentioned in my opening slide, you really have to have an easy button that makes this consumable and simple, whether it's a drag and drop paradigm, but something that makes it simple for an enterprise to be able to deploy this in an easy manner. Then we're going to have Extreme automated apps deployment, because what a customer wants to be able to do is deploy on one of the major clouds. They want to be able to potentially consume it as a service from Extreme SaaS cloud, there at the bottom, or they want to be able to deploy an edge application on Smart OmniEdge, which is our wired wireless infrastructure, our campus core, our Agile Data Center, or in appliances.
That's an important capability, and container technology allows us to do that. One of the things that you might be asking me, well, make this real for me with some kind of an example. We have a company we're working with where they want to be able to do real-time in-store dynamics. We can help them respond at the edge of the network dramatically faster than if they had to wait for the 300 milliseconds for something to go up into a cloud, get crunched on, and come back. Because they have a server that is nanoseconds away from the switch fabric inside the switch with a 20-gig pipe back and forth between that edge application server.
They can then capture video data, and they can take the essence of that, then they can run that up into a cloud to do the video recognition, and their application then can redirect what they have the network doing and dispatching because of how real-time processing they can do right at the edge of the network. We're in the early stages of doing that with a number of retailers. That's just one simple example of where having edge computing right there is important. Another application where that's important is security, where as soon as you have anomaly detection, you want to be able to lock down a set of ports, block a set of users before a virus starts trying to move laterally in a network. There's a broad set of meaningful applications you can do.
Technology summary, cloud-inspired technology, where we have all of the ecosystems on the side of cloud-inspired ecosystems behind us. Our effective R&D team is much bigger. We're very focused on open cross-platforms, very fit to purpose, whether you're the edge of the network, the core of the network, the data center, accelerating applications and making it easy to consume on-premise and cloud. We're the only company that can make that transparent for the user experience. Last, multi-vendor land and expand strategy that uses our software to get a footprint and then follows behind that afterwards. I hope that was helpful to everybody. I look forward to talking to you on the break. With that, I'm going to turn it over to my great friend, Nabil Bukhari.
All right. Thank you, Eric. I think there is a housekeeping announcement that if you want to get on the Wi-Fi, the guest Wi-Fi, the username and password is EXTR. Did I get that right? Okay. Let's get started. Portfolio strategy. We have a really broad portfolio, from Wi-Fi to data center, from IoT to cloud. There is no way on Earth that I can walk you through the portfolio in the next 22 minutes. I can't even give you a summary of that portfolio in 22 minutes. What is it that I'm going to do? My aim here is to walk you through the portfolio strategy and highlight four centers of gravity that are important to us. They are right here, and we'll walk through them one at a time.
Now, obviously, it goes without saying that integrating the portfolio has been at the top of our mind for the last 18 months, ever since we acquired all of these companies. Now, during these acquisitions, we got technologies that were very specific, custom-built, targeted at various portions of the network. The opportunity was to bring them together, bring them together in a way that our customers can deploy them, wrap their minds around them, and essentially really use them to transform their networks from one edge of the network to the other edge of the network. That is essentially what we have been doing for the last 18 months. We have brought all of these together, all of these technologies together through our portfolio, through our solution pillars, the Smart OmniEdge, the Automated Campus, and Agile Data Center.
You heard from Ed, a few examples of our customers who are using one or multiple of these solution pillars to bring them together and really accelerate that transformation in their business. Those customers belong from multiple verticals, be it transportation, be it manufacturing, be it healthcare, be it retail, be it government. Our customers are using these solution pairs everywhere in this network. Now, I could give you multiple different examples of how we have brought the portfolio together. I thought that nothing really exemplifies it better than the user experience across these solution sets. We are a customer-driven networking company, and user experience is at the core of everything that we do. Let's take an example. Let's take an example of Smart OmniEdge.
Smart OmniEdge is a solution that combines IoT, Wi-Fi, and wireless all together into a single solution, a single solution that you can deploy in any way you want. On top of that, you can manage it, you can orchestrate it in any way you want. You can orchestrate it using our orchestration and our applications from our cloud as a SaaS, or you could deploy it in your network through ExtremeCloud Appliance. You can bring these innovative technologies like Greengrass all the way down into your network and actually use those. All of those things are possible. I can talk about that for a day and a half and still have more to talk about, let's just not go there. Let's just show you how this actually works. I asked my colleagues here. No, we'll just stick to this one. Yeah.
You have a screen? Okay. You're going to switch the screen. All right. I asked my guys here that, "Hey, Eric is going to talk about Greengrass, how about we just download Greengrass onto a live network that is running in San Jose and try to show that in front of everybody live?" They did get scared, that is the power of the technology here. All cool demonstrations have a voice-activated thing, we got to do that. How about we use Alexa? Smart OmniEdge.
Okay.
How many sites are you managing?
You have three sites in your system.
List the sites.
You have following locations in the current deployment: Toronto, Ontario, San Jose, California, Times Square, New York.
Where is Bob?
I couldn't find a username, Abba. Either I did not get it right, or Abba is not here.
Alexa's problem. Where is the user Bob?
I don't know what to do. What can I help you with?
All right. Alexa, bye.
Goodbye.
All right. Obviously, Alexa is a little temperamental. The idea there is that this is a network that is running in our lab in San Jose. It is running on our Extreme Cloud Appliance. It is actually controlling multiple different sites around the globe, like the three sites that we talked about. What it's trying to do is that it has downloaded the Amazon Greengrass onto the Extreme Cloud Appliance, and that's how we're able to actually expose all the capabilities of our own cloud through Alexa. Of course, nobody's going to go control their network through Alexa, but that just shows you the power of that container technology, the power of bringing down Amazon Greengrass or whatever else you have into your network, and then use whatever way you want to use it. Your imagination is actually the boundary for this technology.
Let's try one more thing for this. Are you ready for this? All right. Alexa, Smart OmniEdge.
Okay.
I should ask it, right? Okay. Alexa, where is Bob?
I don't know.
Alexa.
Bob located at.
All right, we'll just stop this. This is not listening to me very well.
I don't know what to do.
Alexa, bye.
What can I help you with?
Alexa, bye. Okay. All right. Anyway, that gives you an idea what we're doing.
I don't know what to do.
Can we
Goodbye.
See, my guys were like, "Don't use Alexa, use Siri." I was like, "No, let's just use Alexa." Okay.
What were you going to do with Bob?
The idea there was that if Alexa ever understands my accent, we could actually find a user anywhere in the world. We could move them into quarantine. We can do whatever. We can change their access. We can upgrade the network. I can actually sit from here and upgrade my network back down in Bangalore, and it would automatically schedule it. It will sign a request, like a change request. All of those things. The idea there is not that, hey, Alexa can do it, but the idea there is that you can deploy your network anywhere in the world, and you can manage it from anywhere in the world.
When Ed talked about it, when Eric talked about it, that single pane of glass, I wanted to make sure that you don't think about a person sitting in front of a screen with some blinky lights on front of it. Those days are gone. This is the single pane of glass where you can actually manage your network, you can control your network, be it wired, wireless, IoT, from anywhere in the world in whatever way, form, or shape you want. By the way, this product is GA. Alexa integration is not. Don't worry about that part, but this is something that a lot of our customers are deploying.
Sorry, I'm not sure about that.
Would you have this woman stop? All right, we'll move on. The next center of gravity for us is stratification. Now, what do I mean by that? We have a lot of products, and big portfolios can be very complex. We decided that we actually want to make it easy for our customers. We are stratifying our portfolio. What does that mean? That means that we are taking our solutions, the right capabilities, the right technology, the right price point to all of the customer across the spectrum. From the customers who are price sensitive and customers who want to buy the best that money can buy. There is a solution for each and every portion of our customer base. Now, this stratification is not only just for products, this is also for solutions.
Imagine, what AP do I want, what Wi-Fi AP do I want, and what switch do I want if I want the best possible performance? There is a premium category for you. Well, my requirements are actually not really high, and my budget is a little bit modest. What category should I buy? Well, you have entry. Of course, you can move between them, but what it does is it simplifies a complicated portfolio for the customers to understand and our teams to sell. That is the part of stratification. Now, as we are stratifying our portfolio, as we are simplifying it, we also need to refresh it. As we acquired a lot of products from our acquisitions, let's just say that not all of them were up to date as far as the refresh cycles were concerned.
Now we are in the middle of this refresh cycle, powered by some of the technologies and some of the optimization that Eric talked to you about. We are into this refresh cycle where we are going to refresh 70% of our portfolio in the next 18 months. What you're seeing on the screen here, this is just examples of some of our portfolio. We have a massive portfolio, and we are able to refresh 70% of that in 18 months with 1/3 of the R&D that you would have expected had it been any other company or those three companies individually. That is the power of the technology that we are building together. Now, what does this refresh do? Obviously, this refresh brings new capabilities, top-of-the-line chips, new differentiation for our customers, but it also brings higher gross margin potential for us.
As these new products start showing up in our product mix, it has the potential to give higher gross margins to our businesses. Now, we talked a little bit about how various customers in different verticals, they are putting our solutions together. Well, they're already doing that, but we want to accelerate that, and that's the verticalization part of our portfolio. Now, if you look at it on the left-hand side, we have our three solution pairs or pillars, and on the right-hand side, we have all of these applications. Some of them Eric mentioned, some of them Ed mentioned. We have tons of them here. Now, the idea there is that you still require a network that is purpose-built, that is the right fit for your business, and then you add applications on top of that for orchestration, for security, for cloud management, for analytics, for security analytics.
These things come together. What they do is they deliver a vertical solution. They help you get to that business imperative that is very specific to your vertical, the customer, the vertical that the customer belongs to. Here you can see how different pieces come together for healthcare. Here you see how different pieces come together for retail. We also have horizontal solutions. For example, data center. We bring hardware, we bring fabrics, we bring applications together for data center solutions. Data center is horizontal. It applies to multiple of our verticals. That's really another center of gravity for us to really accelerate the verticalization of the portfolio. Let me just take an example. In the verticalization I have or we have two different prongs of our strategy. Number one, refresh and expand.
Obviously, you want to refresh the portfolio, you want to expand the footprint of the solution. Let's just take an example. In healthcare, we have tons of healthcare customers out there. We deploy our campus fabric, highly differentiated. People love it once they start using it. Hyper segmentation, plug and play, super easy to operate, super awesome for that vertical. Lots of deployments out there. Our strategy, let's go refresh the products that are part of that. Let's add some more applications to it. We are bringing analytics to our campus fabric. We are bringing access control to our campus fabric. We are bringing S and C to our campus fabric. By the way, when I say we are bringing, we've already done that. On top of that, what's the expand part? I think Eric briefly mentioned branch.
I know all of you have been asking me about branch. Here is a branch in healthcare. What's a branch in healthcare? It's a clinic. It's a little doctor's office. We can extend the campus fabric from your main hospital all the way down into your healthcare branch, that clinic, that doctor's office, so you can have the same security, you can have the same exact access control, and you can actually have the same user experience that you have in your large hospitals. That's refresh and expand. Of course, we build on top of that as we move into calendar 2020. Let's take another example. Our data center solutions, of course we have a best-of-breed data center fabric. We have border routing. We have data center interconnect. What's next? How do you refresh that?
First, in data center, bandwidth is always at the top of the mind. 100 gig, 400 gig spines, you bring that in. You bring in new technology, new ASICs that bring in new capabilities in it, and we refresh that. We also bring in the application portfolio that we acquired from other companies. Analytics, automation, all of those things are now in the data center solution as well. What we are doing, we are refreshing, we are expanding. As you move forward, start looking towards the next year, 2020, then you bring in the cloud management part. That's where you bring in hybrid cloud networking and stuff. Again, the strategy is refresh the current solution and then add to that solution. We also have a second part of our verticalization strategy, and that is increase the wallet share.
Why is that important or why now? As we brought the portfolio together, we realized, and it was pretty obvious, our customers told us that they were using one part of our portfolio but really interested in the other part of our portfolio. For example, in healthcare, they were using our fabric, our switching fabric, our campus fabric at the edge. They were like, "Wow, you have really good Wi-Fi technology as well." What do you do is you don't just hand it to them like, "Hey, here's Wi-Fi, here's Switch." You bring them together under single management. You bring them together under user experience. Now our customers can take that Fabric Connect that I talked about, that you can bring it to the campus, from the campus down into the branch.
You can also bring it all the way back into your Wi-Fi. It starts becoming a solution. If you add IoT and wireless on top of that, there is more potential to get increased wallet share, and it also has a positive impact on our gross margins. You add technologies on top of that, like analytic, ExtremeAI, the security part, the IoT security part, and then you can increase the wallet share even further. Let's take another example. Retail, the same thing. If you look at it, we've been selling Wi-Fi, we've been selling switches, but now we are selling the Smart OmniEdge portfolio that brings everything together. Good thing none of our retail people actually use Alexa integration. Otherwise, they will be very upset at me.
I'm not sure about that.
I knew it was going to do that. Okay. That's an example of an increased wallet share there. The one that is the closest and the nearest to my heart, the fourth center of gravity, and that is the application space. I'm not going to go through all the applications that we actually have. Before I can talk about the strategy for application, I do need to draw out the canvas because there are so many applications. You have applications running on your phone, you have applications running on your laptop, you have applications running at home. What applications are we actually talking about? Obviously, we are not going to get involved in the consumer-side applications. We are talking about applications that are in the network space or in the space that we play.
They go all the way down from foundational networking, all the way up to artificial intelligence, machine learning, analytics, autonomous reactions, and everything in the middle. This is the space in which we play. Remember I showed you a lot of blocks, red blocks on there, and we said, "Hey, we acquired a lot of products, and we needed to stratify that, and we needed to refresh that." You know what? The situation is not that different on the application side either. If you look at it, if I take the wheel off of that previous slide, you see that we are actually present in a lot of these areas already. There's a lot of technology that we already have, and there's a lot of technology that we acquired from the acquisitions that we have. Now we are present in all of these places.
You could ask, well, if you have that many applications, why isn't your revenue on the application going so high? I think the other gentleman kind of mentioned it as well, that now we have the technology. Technology is not really our issue. Where we want to focus, where we are focusing, where we are investing is in packaging, delivery, and monetization of these applications. That is the focus for us for a little while. Attach that to the go-to-market changes that Bob is going to talk about or he has instituted. We do believe that it would be an area of massive strength for us in the portfolio. Again, four areas of gravity for us. We already integrated the solution. We are going to go stratify and refresh that. By the way, that process has already started.
If you were following us, we have already GA'd. We have already launched multiple of these new platforms in the last one quarter, from 100G in the campus core and the border router in the data center and so on and so forth. That refresh cycle is already there. Then lastly, the application space. Here too, let's just take an example. You've heard quite a bit about the Defender for IoT. What is this? Defender for IoT is essentially an application. It runs in our ExtremeCloud Appliance. It also has a hardware component at the end, and the reason for that hardware component is very simple. If you're already running our networks, well, first, good for you, and the second is that you have these capabilities inbuilt into our fabric. Through our hyper-segmentation, we can already secure your IoT devices and stuff.
A simple fact of life that not everybody is running end-to-end Extreme Networks. What about those people who are not? How do we help them? How do we help them bring that capability into their network? That is where the Defender little dongle comes into play. As always, we like to list a risky live. Instead of showing you slides, we're going to try and show you a live demo for this. We don't learn, do we? All right. What they're going to do is that they're going to shift the screen, and they're going to show you what is essentially the embedded software that runs in most of the heart rate machines that you have in the hospital.
When you actually go, and I think Eric kind of mentioned it, when you go into the hospital, if you're plugged into all of those devices, they are checking your heart rate, they are checking your temperature, they are checking your oxygen levels, and so on and so forth. Now, all of those devices, there are two problems with them. First, as Eric mentioned, that they. You guys are going to bring that up? Okay. One of the problems is that they're running very old code. The second problem is that you can't really run your endpoint protection on it. There's no antivirus for your heart rate monitor. There's no firewall for your MRI device. That's just what the reality is. Now all of these devices are plugged into your network. What happens?
While they are bringing it up, trying to figure out technology in the back, the idea here is that now imagine that there is a hospital where you're plugged in and all these devices, and you're taking all these parameters for a patient. There is automation built into that too. Based on the heart rate, there are medicine pumps that are attached to that inject more or less medicine in there. Just to give you an example. There are alarm systems built around that and so on and so forth. A hacker comes in. If they are able to switch the screen to that thing, what we were planning on showing you. Oh, there it goes. Once that comes up.
The idea is that there are all these hackers sitting around, and then Yuri here is going to play the role of that hacker. What he's going to do is that he's going to hack into this live system. He can either change the values live, so it looks like the system is actually working, or he can just make it go flat while the person is still alive. That sounds like a little prank, but that's not really a prank because walk with me into this thought experiment. There's a hospital where patients are sitting, and all of a sudden, your heart rate or your parameters, your vitals are going all over the place. What's happening is that the hospital is trying to react to that. There are nurses coming and responding to that. There's things that are happening in the hospital.
Are you ready, Mr. Hacker, for this? All right. This is actually live heart rate from Stan. We thought that his heart rate. He has leads connected to him. We thought that his heart rate would be elevated, so it would actually show up really nicely here. By the way, you should exercise more, man. Okay. Go ahead and Yuri is going to hack that system and look at it. The values are all over the place. He can make Stan appear dead. Stan is obviously not dead. The idea here is that if a hacker can connect in and do these kind of things, what really happens, jokes aside, this is a denial of service on your healthcare system. Imagine that every single monitor in your healthcare system is having the same exact problem.
There is no healthcare system in the world that can respond to that. I wanted to end it on a little bit of a somber note, because it's a funny little demo, but it really underlines the real risk that we have. Now, go ahead and plug our Defender app in there. When he plugs this Defender app in there, the Defender app automatically onboards itself to the ExtremeCloud Appliance, which is in the network or in the cloud. It automatically downloads all the policies, and it applies those policies right there, no matter whose network is in the middle. It could be a competitor's network, it could be the worst network in the world, it will secure it instantly. It'll block all of those hacker communications from outside.
Not only that, if this device is something that is compromised, it cannot now go and compromise other devices in the network. We have stopped that denial-of-service attack on the healthcare right there. That is the power of applications. That is what happens when you take applications, marry it with some good hardware, and then convert that into a solution. Just to give you an example that now Stan is back alive. I cannot hack anymore. Yeah, and Yuri is still trying, and he can't hack that. These demos will be available outside here as well. My time is running up, I wanted to end on this side. This is the power of software-driven networking. This is the power of applications.
This is the power that we bring from one portion of the network, one edge of the network to the other edge of the network. As Ed rightly said, we're just getting started. Happy to talk to you more in the break. Thank you.
We're going to move to Q&A. We're going to invite the speakers that have presented already up on the stage, and we'll do Q&A. We have some mics going around the room, so just wait for the microphone, announce your affiliation, and then limit yourself to one question and one follow-up. Okay. Let's go with Alex here.
Mike.
I'll repeat it.
Okay, sure. You put a box up there that had a lot of red in it and a lot of green in it for your 2018 product. You put another box up that had a lot fewer boxes in it, implying a significant reduction in the number of products you're selling, although they're refreshed. The simple question is, now that you're flattening your product line and converging your product line, have you also undermined some of the products that might have been in the field that people were using that might have wanted-
Great
to continue and therefore put some risk into your model?
Now you ask that question. Did everyone hear that question?
No.
No. Okay. I'm going to repeat the question, which we were referring back to the product slide that showed 2018, you saw all those boxes, okay, in terms of product. As you fast-forward, we thin out the number of boxes. In other words, what we're doing is we're streamlining the boxes into solutions. The question is, I think that the question is, are we losing capability or are we abandoning legacy customers perhaps, or legacy technology that's not moving forward?
The question is simply, are you undermining any of your existing customers that may have been reliant on a portion of this technology that might now no longer be supported?
Are we undermining the customers that have invested in our technology that we may not now support going forward?
Right. Great question. A straight answer to that is absolutely no. That is why we took 18 months for that. If you look at it, that's the beauty of stratifying the portfolio. What it does is that if you looked in 2018, and that, by the way, was not the whole portfolio, but if you look at it, we were heavy on some areas, and we were not that heavy on other areas. The way we do it is that as we have taken all of the different portfolios, all of the different products, and put them into these pillars. Think about Smart OmniEdge. Smart OmniEdge has products from Extreme, from Avaya, from Brocade, similarly Automated Campus, and similarly Agile Data Center.
When you bring them forward, the idea is that every single solution. Not only that, but all the way from the value tier to the premium tier has refresh cycles which are drop and refreshes. Any customer who's using any technology from even the companies that we acquired, has moved forward with their existing networks, no rip and replaces. That's why you have to stratify and refresh it as part of the solutions. The quick answer is absolutely no. Nobody would end up in a space where they'd be like, "Oh, I bought it from XYZ companies, and this is the end of life, and you are not going to give me that capability anymore." That's absolutely not true.
This is kind of a related question. Nabil, I think you made the contention that Extreme is able to do R&D much cheaper than something like all other tech companies or most other tech companies. I don't exactly understand, in plain English, why that would be. One of my fears has always been that it's so difficult to compete against Cisco, that's a zillion times larger than you, and Huawei. Did you really say that? And if so, why?
Let me answer it, too, is first I'll clarify what I actually said because. But you're right. What I'm saying is that we are doing it a lot more efficiently compared to the companies that we acquired. That's the argument that I made. Say, I'm going to take these names here, but if Avaya, Brocade, and Extreme independently refreshed this portfolio, and it took them 100 engineers to do that, combined between them, we can actually do it in 30 engineers or 40 engineers. That's based on all of the things that Eric talked about bringing the pieces together, developing it in such a way that our pieces, our components, be it software components or hardware components, can be utilized across the board multiple different times in multiple different solutions. That's really the point that I was trying to make.
Right.
Eric, do you want to add to it?
No, that's a good summary. Yeah.
Is that one-
Hi. Ed. Yeah, I just want to continue on that R&D side. My question is, you've got a lot of operating systems that are now running on Linux. Are you going to continue investing in all of those operating systems? If not, it gets back to the first question. If you are, it kind of gets to the second question because you've got all those operating systems, you've got a refresh cycle, you've got some security products that you're bringing out. You've got a whole lot that you're fighting off. Do you have the R&D people to take care of all that?
Let me handle that. From an R&D standpoint, each of those network OSs or operating system you were talking about, well over half of the development cycle of any network OS from one generation to the next, historically, has been related to porting and rewriting half of it, because it was using a very old version of Linux, and you had to touch a 2.6 Linux and take it all apart and put it all back together again. When you convert a network OS into an application, now that application can move from generation to generation, and the speed at which you add new features is directly related to the market demand for new features for that network OS.
I don't have to put in this huge effort to allow it to move from one generation of hardware to the next, to the next, to the next, which has been the case in the past. We can throttle the speed at which we deliver it based upon the feature velocity required for that market segment. That makes a huge difference in the R&D efficiency. Second of all, you heard Nabil talk about, in some sense, places in the network. We have the edge of the network, we have the campus core, we have the data center. We have a flagship in the data center, and that's where we put most of our investment for data center. We're not putting those data center features in the other network OSs, which is what the other network OSs were trying to do.
Likewise, on the edge of the network, the edge of the network is adding all sorts of interesting edge computing, IoT, 90-watt PoE power, things that aren't applicable to the data center. Now we can focus the development on that piece of software to that part. If you think about a buyer's journey, they buy either from the edge in, or they buy from the data center the other way. We focus the R&D for that place in the network, and we can run the speed of it based upon the fact that now that network OS is an application. It really gives us a huge performance advantage in the speed at which we can crank out new products versus the historical way that network OS has been done in the past.
Nabil's going to add something to that.
That's from a technology engineering point of view. I do want to reiterate from a product point of view that, and this comes back to a little bit to the question that you asked, is that, yeah, it would be very easy for us to say that, "Oh, you know what? We acquired four OSes. Let's just kill three of them, and let's just take one forward." Now, that would be the most horrible thing for us to do for our customers. And we, as Extreme, we just focus on the customers, we'll never do that. What is the alternative? The alternative is to take all of them forward but become really, really efficient in doing that. And that is exactly what Eric told about focusing the OSes in their areas of strength, number one, reusing components, whether they are software components within the OSes or hardware.
Today, when we create one hardware box, we release it under multiple OSes. Yesterday, we had to build three different boxes to do that. Automatically, our engineering effort on that is 1/3 at this point. That's where we're becoming efficient, very efficient and smart, street smart, if you would, to really focus on the areas where it is required, because leaving behind a customer is not in our DNA.
Paul, do you have a question?
Yeah, I do. I apologize. I know these were product technology discussions, in trying to measure the success or the degree of success of your new product portfolio and your integration efforts, how many customers have you added? How many have you lost post. I'm not talking about the ones that came over by virtue of the inorganic acquisitions. Following those acquisitions, how many have you gained? How many have you lost, whether for Avaya-specific or Brocade-specific or for Extreme classic products?
Yeah. I think everybody heard the question. It's hard for us to go back and quantify and say exactly how many customers did we lose. Based on when we're acquiring a customer, we're looking at a pipeline of opportunities, the question is what's in that pipeline? They may have a very large installed base.
Yeah, I apologize for the interruption. They view one of your customers today and had signed once, you're going to lose this customer.
You would think.
You would hope.
You would hope. That's not actually as clear as you might think. That was part of the learning that we had when we acquired the businesses. The Avaya business sat in bankruptcy for over a year. Okay, what happened during that year? Okay. It was a very challenging environment. From a selling perspective, if you look at the networking business, the networking business at Avaya was an overlay. They didn't have direct teams, quota-bearing people selling. It was an overlay because they were focused on voice solutions, and they had contact center solutions. In what we acquired, there was a lot of learning because the voice teams that may have been selling, switching is a derivative. They're quota-carrying people, okay? The people that we brought over didn't have a number. There was some confusion around that, so we had to retrain.
We literally had to train sales teams on how to use Salesforce, how to create an opportunity. We were relying on the Avaya Salesforce instance, in which one day we opened up Salesforce, and all the opportunities disappeared. Somebody wiped it out as part of the TSA. Well, that's a little troubling. That escalated right up to the CEO. Magically, they turned it back on. These are some of the things that we deal with, the realities of things that we deal with from an M&A perspective that don't seem obvious from the analyst seat. Now, we've managed through this. We've migrated everything onto our platform now. What I can tell you is that from a bookings perspective, that $200 million that we're targeting is over $200 million today.
Because of the discounting in the portfolio, we've added 12 gross margin points to that business, and it's driving cross-sell because of the power of the fabric, hyper-segmentation, the security, the Internet of Things, the IoT Defender. This is driving cross-sell. It took a while for our field to embrace it or to believe it. It took a while for customers to believe that we're going to invest. Are they going to abandon? Are they going to leave us behind? Well, now the field and the system engineers, partners, and customers recognize we're not leaving them behind. We're not abandoning them. In fact, we're providing them with a clear path forward with their technology. That's what we're excited. We came out of the shoots, $40 million a quarter a year ago. Good Lord, do they have control over the revenue? Now it's all the way back.
We're over the number. We've strengthened the gross margins. It's driving cross-sell. I don't know how we can go back and quantify all the different Avaya customers that may or may not have had a switch. That's not how we're looking at it. We're looking at how we're driving those customers forward today. The data center side, the same thing. If you look at the Brocade assets, it was in limbo for a year. The company was being split into 11 different pieces. In terms of the teams that we got from a selling perspective, many of them overlay, many of them going direct. In terms of who we got, there was a lot of people that weren't necessarily sellers, so we had to completely rebuild that.
As we mentioned earlier, from a leadership perspective, we had to make some changes so that we could put our teams in where we had our visibility, account mapping, et cetera. The good news for us, what we're so excited about today and we're so ready to get going here, is that the technology is great. The differentiation, the solutions that we have are very powerful, very strong. The execution issues that we experienced are issues that we've addressed. That's what gives us confidence as we go forward. It's not a quantification of the answer, but I just wanted to provide color to the question that you're asking.
Let's do one more question. We move on to our customer panel.
All right. Erik Suppiger, JMP. Traditionally, networking customers haven't paid a lot for applications associated with the network. How do you convert them into being willing to pay? What kind of incremental opportunity do you think of the applications representing for you, maybe as a %, as a proportion of the overall deal size? How much can these applications start to generate?
Let me handle it from a high end. I'll get these guys to jump in. If you think about it, when I make the statement that 96% of our R&D is in software, you look at our revenue, you say, "This doesn't make sense." It goes back to the whole OS system where all our features are built into an operating system that are part of a box. We're selling the box. We're not recognizing the revenue. If you look at Extreme in terms of how we're set up from a licensing perspective, we haven't been set up institutionally to sell software.
As part of our digital transformation is completely rebuilding, if you will, or building state-of-the-art licensing entitlement management system that's going to allow us not only to sell software perpetual license but also term subscription as well as consumption-based software models. It's early stage, Erik, we're not at the point yet where we're making a prediction. It's more visionary as to where we're going at this stage of the game. We are investing in that new, what we call the LEM, the Entitlement Management platform. It's going to give us the flexibility now to charge and to position solutions and sell solutions on a consumption-based, as well as traditional SaaS-based model, where typically it's been a perpetual license that we've sold internally. Structurally, we're going to have the foundation that's going to enable us to do that over the next 12 months.
We're targeting these different applications and different ways to monetize them. We've made key hires into the company, people who have track records of doing this, on a meaningful way. Rémi comes from a software business. We're bringing in a lot of people in the company that have this software focus. We believe we're going to be able to monetize going forward in what is more of a cloud services future, where we see the industry going.
You want to add something before I jump in?
We're out of time.
Yeah. One last thing. A lot of the things that you're talking about have traditionally been associated with network management and a network management console. It was very difficult for an equipment company to sell a piece of network management software because a customer felt like, "Hey, I can't really use it without that." That sort of drove that beginning. If you look at the kinds of applications that we're talking about, whether they're cybersecurity applications or automation applications, you can see there's a market already for those kind of pieces of software. There's an opportunity for us to benchmark against those opportunities and be able to deliver some interesting value because of the unique insights we have of running a network and working with really big customers who help drive the definition of those products.
We can talk at the break, thanks very much.
Thanks. Let's welcome Norman Rice.
All right. Thank you. Let's welcome the customer panel. Come on up, Gerry. My name's Norman Rice. I'm the Chief Marketing Development and Product Operations Officer for Extreme. That means it's a long title with a lot of words, I look after and I lead our business and corporate development activities, our marketing, our IT, supply chain, and product operations. Very pleased to have our panel and let's start. You guys go ahead, take a seat. We got this gray couch. Everybody's wondering when we were ever going to use it. Let's start, Gerry, with you. Do you want to just introduce yourself and where you're coming from?
Yeah.
He's thinking back to his chair there.
Yeah, really.
Yeah.
Gerry Sgari. I'm the CTO for New Jersey Transit. I've been with Transit for 32 years now, I've started doing networks probably before most of the people in this room were born. The first network was a point-to-point 300 baud, if anybody knows what that is.
Sure.
We've come a long way.
Thank you. Welcome, Gerry. You came by bus today.
Came by bus.
That's ironic, right?
Well.
It was on time?
He was worried about the train being on time.
It was on time, one of the things that a lot of people don't know is we carry more people into the city by bus than by rail.
Really?
It's about 60/40.
Wow. Interesting. I'm going to go to Mike Peck. Mike.
Hi, my name's Mike Peck. I work for Verizon. I'm on the network team. Specifically, I run a program nationally for in-building and sports venue coverage solutions. Obviously, you guys know about, you've seen the big red maps on the commercials, but when you're doing something in a venue in a high-density area, it's a much different egg to crack.
Yeah. Verizon really is a partner and a customer. Internally, as a customer, Verizon leverages our technology for the enterprise side of their business. We work with Mike's team on smart cities, high-density venues where we partner and bring our technologies together, which we'll talk about in a moment. Chip.
Thanks for having me, Norman. I appreciate it. I've got the fortunate privilege of having a couple of my partners here. Verizon's been a great partner of our organization as well. My name is Chip Suttles. I'm the Vice President of Technology for the Seattle Seahawks, the team, and the CenturyLink Field stadium. My responsibilities include the IT and security, technical operations of the stadium, and our mobile platforms and applications and features.
Yeah. Some of your background, it's pretty interesting. You worked with the Olympics.
Yeah, I've been fortunate to have a career in sports technology. Started off with the Atlanta games in the 1990s. I was able to do Atlanta, and Salt Lake City Olympic Games, Torino, and then Vancouver was my last games involvement. Various IT roles, a little bit of broadcast roles, supporting the Olympic movement.
That's great.
Yeah.
This is a rare moment because on stage you have Norman, and Chip's real name is Norman. He goes by Chip, but Norman and Norman right next to each other. What are the odds of that?
Yeah.
Very small.
I'm not Norman.
You're not Norman. You're Gerry.
Yeah.
Gerry. You're not Norman. You can just go from there.
Right.
Let's talk about working with Extreme, and you work with us as a customer, a partner, and Gerry, I'm going to start with you. Which aspects of the technology do you work with Extreme, and how do you use it?
Well, we first started with Extreme with the Super Bowl that New Jersey hosted with the Seattle Seahawks and Denver. Our network is pretty complex. Most are, but we do many diverse things on our network. One of them is surveillance. We have about 7,000 cameras on the network now, fixed cameras. At that time, we were rolling them out, and we needed a way to bring video traffic back to multiple locations throughout our network. Because we don't only use CCTV for surveillance, we use it as part of our operation. We use it to count buses coming into Port Authority. We use it to monitor what track a train is coming in on in Penn Station. We use it for many things. We didn't like the Cisco solution for multipoint.
We made a decision back in 2013 to go to Avaya's solution, and it was met with a lot of consternation, especially by my group, because nobody ever got fired for suggesting Cisco. This was a bold move at the time.
Right.
Once we got through the Super Bowl. How we did it was we're a cash-starved state agency, so we had to be very careful about how much money we spent to bring this technology to the right people, to the police, to operations, to customer service. We placed the hardware in the remote locations where it was needed to get the end-to-end video. After the Super Bowl was over, which it was quite a success, the way this was deployed, we had people coming to me and to my boss saying, "When are we moving off the Cisco network? When are we moving?
Because the performance of the video just was not up to par with what they had experienced during the Super Bowl.
at the control centers and the emergency operations centers. We have, since that time, deployed Extreme, and we're in the process of going out to most of our locations. We have about 250 locations on our network. The equipment in the field is beyond PCI compliance. We also run a large point-of-sale operation. Instead of upgrading with the existing vendor, we are deploying Extreme out to the edge to handle all the applications that we support.
It was the fabric technology. The campus fabric, the hyper segmentation, the embedded security aspects that give you autonomy.
Well, the one thing.
Yeah
With the segmentation, we're looking at the network, as I mentioned, is pretty complex. We run a point-of-sale network, a surveillance network. We have public address running on it. We have voice over IP. We also operate several radio systems. We have radio over IP on it. We're moving along with the segmentation to now add more industrial control and SCADA applications.
which we have traditionally kept separate. Now we feel that we have a technology that we can trust to start bringing the boilers and ventilation systems and the actual signaling for the rail and light rail systems onto the network that we use. At the same time, leverage, probably Rio's Verizon, our infrastructure in the ground, our own fiber, we're running along the right of way to take advantage of that as at the same time maintaining segregation of these critical operational systems.
That's great. Thank you. Chip, how about you? Working with Extreme and why you selected us and your experience.
Yeah. I just completed my seventh year with the Seattle Seahawks organization, and I was brought on in 2012. I was asked to address the immediate concern, was connectivity at our stadium. It was becoming not only a health safety matter, but it was degrading the fan experience. We put in a neutral hosted DAS, and immediately following that, we installed the Wi-Fi. It's a unique setting. We're not just a football team. We host 45-55 events a year where we have anywhere from 40,000-68,000 fans and guests in our building. We didn't go about it lightly. We try to associate with best-in-class practices. We spent about 18 months in researching, trying to find who we thought we could best partner with to solve the high-density Wi-Fi solution.
We're happy that we were able to find Extreme Networks and establish this partnership. It certainly is delivered as designed. We're actually maximizing it. I think I just saw a stat, and a little bit jealous of, at the Super Bowl, there was over 24 terabytes of data transferred, and I was fortunate enough to attend, and I think solely that reason was because the game wasn't much to watch. People had to do something. I know I was sending out quite a few pictures as well. If we had that capability, and I think we would do the same. I think we push about 4.5 Terabytes per game on a Seahawks game day. We're really happy with the way it's performed to date, and it's certainly becoming more and more reliant all the time.
Your colleagues touched on a couple points that are key to us is ability to monitor the traffic real time, and to take those analytics and maybe make decisions based on those analytics during a game. It's important for us, and it makes for a better fan experience.
Yeah. Monetization. As you look at analytics, whether you're learning about what the fans are doing, there's different ways to monetize it. You either look at sponsorship opportunities or the hot topic today is gambling.
Right.
Right? That's what's hot in the NFL and in the
Yeah
collective psyche.
Yeah. I'm not going to touch on gambling too much because I think it'll be a long time before Washington State will approve those laws, but we're certainly following what the NFL's doing, what the other teams in other states are doing with the teams. I would like to touch on a partnership with two things that where we're able to monetize the use of the Wi-Fi network a little bit is we offer in-seat food ordering in our club as an amenity for our club patrons. It's been a great successful program to just be able to deliver beers and select food menu items to that select group of high-end customers. Recently we partnered with CLEAR, if you're familiar with the airport security company. We did a two-year partnership. We just finished year one.
What that is enabling us to do is not only we have a couple entry points that if you're a CLEAR member, it expedites entry, and gets you into the stadium. It's still a ticket required, mag and bag scan for security reasons. What we really like, and we're really excited about the idea is that we put CLEAR in four of the point of sales around the stadium, and we have permission from the Liquor Control Board to use CLEAR as a age verification. It speeds up the whole process. You go up, you put your fingerprint down, a green light comes on for age validation for the attendees there. Because your credit card's on file, you're able to utilize your CLEAR membership to make payment.
That whole transaction goes from a normal transaction of about 7 seconds to, with a CLEAR transaction, we can do it in 2.5 s- 3 seconds. Speed of service is important, but just a frictionless experience is kind of important, too. It's been a great partnership, and I think the Wi-Fi's going to help us be able to expand that next season.
You're also looking at Wi-Fi 6. You're in the upgrade cycle.
We're looking at an AX deployment. We probably change the model to get more access points into our environment. We're limited, we're going to look at an under-seat deployment. Yeah, we're excited about the opportunity to be able to do that in the future. Yeah.
Great.
Yeah.
Just to rattle off some stats from the Super Bowl, there was 24 terabytes of data, 49,000 fans were connected, 69% of the people that were there were connected at some point during the event itself. It's a 25% growth year-over-year, 100% growth over the last two years in terms of amount of data being consumed. It's exponential where it's going, where it's coming from. I'm going to traverse, Mike, to you. Verizon, you're a partner in a lot of these venues, not just in the NFL, but also in collegiate athletics, which is an untapped opportunity. It's really the beginning of that in the marketplace. Why don't you talk a little bit about our partnership and why you work with us, and we'll ask the 5G question after.
One of the things about, I mentioned the high-density solutions. Chip knows there's a lot of people in the venue trying to do a lot of things. If you think back to Field of Dreams, if you build it, they will come, and they're going to bring devices, and they're going to bring their kids and their devices and their significant others and their devices. That's the problem that we try to solve for at Verizon. High-density Wi-Fi with Extreme is a great opportunity to find another means for those devices to connect. Probably preferably outgoing pictures, look at all the fun I'm having, Instagram, tweets, et cetera. Also just downloading and doing whatever anybody wants to do. The Wi-Fi partnership with Extreme is a significant enabler to be able to do that. There's only so much that we can do with our licensed spectrum.
Through our partnerships with the NFL and Extreme, it again is an enabler for that connectivity. I think the value of connectivity really can't be overstated. The value of our partnership with Extreme is really unique because we might have a partnership with the Seahawks, who want to bring in Wi-Fi, Extreme supports us as a direct customer. It's a unique situation. We also, of course, are a direct customer. In situations where the customer is Chip, we're on a peer-to-peer level that we have access to all the support mechanisms, and all of the innovation that we do together is really top tier.
Yeah. I think one thing we like to highlight, and we do talk quite a bit about the NFL because there's brand cachet with that. The going forward, which was our investment thesis when we went after this business, we thought about the smart city, the coming of the smart city, and the idea of the concentric circle. Meaning, hey, we could go into a major Division I university and try to go head-to-head in IT where there's incumbents or challenges, or we could go in through this green grass, this open territory, which was a venue. We felt that if we establish our cachet with the NFL, we could go into those venues, and then the concentric circle would mean that it influences the university, it influences the healthcare facility, it influences the local K-12, it influences local government. That's happening.
Verizon's doing, the smart city race is on. We're seeing that exact plan playing out. This is the tip of the spear, this broader opportunity. It's great to have you as a partner. I think a lot of people in this room probably want to hear 5G. You're going to cut me off before-
I'm going to start the conversation about 5G because-
Good
with our Verizon partnership, proudly to say that we're the first stadium in North America to have a couple 5G antennas. Verizon put those in to help us do some testing and to showcase some of their technology as the standards get defined more for rollout. You're probably saying, "Why are you talking about 5G? We're supposed to be talking about Wi-Fi." It's the same conversation I had with management. My management team hears about 5G, sees all the marketing and advertisement, and said, "Why do we need to roll out another Wi-Fi deployment? Why don't we just wait for 5G?" It's coming, but it's not here yet. Certainly, I think they complement each other, and they will complement each other in the future. We need robust Wi-Fi until things are standardized and products are on the market.
Yeah. Even with that, 5G, we ask this question, Mike, 5G, does that mean you need less Wi-Fi, the same Wi-Fi, or more?
Yeah. Adding to what Chip said, I'm not here to talk about 5G and our roadmap.
Right.
I will say that it's a complementary design for whatever we're doing, whichever G we're on. When we look at designing a stadium, obviously, there's going to be limitations to license spectrum. Wi-Fi is another spectrum that we can use. It's something that we have used for our connectivity solutions, and we're going to continue to use it, regardless of the G we're heading towards.
Gerry, back to you. One of the stories you told me was how the video cameras were able to, and your search capability, were able to identify from a terror incident.
Yeah.
In New York.
We have various analytics that we run on the network for video. Facial recognition is a little bit rough, and it really comes down to where the camera's mounted. If the camera's mounted at head height, it's flawless. If it's this high, it's probably less than 60%. We have about 7,000 cameras on the network now, and another 7,500 or so mobile cameras that are also on the network. Back to the Super Bowl, we were able to key in on an individual and actually track travel time from Penn Station, New York, to Secaucus Junction, to the Meadowlands, to the stadium, and be able to come up with some statistics of how long it's taking customers to get there. We've also used those analytics to get the Jersey City bomber. We caught him on our system.
We were able to track him through our system and work with the various police departments to apprehend him. We also, recently, within the last several months, an autistic child that had gone missing was found because we were able to pull up the image of that child at one of our train stations. From there, we were able to go in and recover him. We use our cameras and the analytics as best we can. We use them more and more for many things, like I said, other than surveillance. We look to see if the platforms are getting overcrowded so that we can react to that. We count trains with them. We do many things. We're counting on you guys to help us continue with that because the 7,000 or 7,500 fixed in cameras is about to explode.
There's going to be a lot more of them.
Yeah. First of all, thank you, all three of you, for giving us an insight, we have a lot of opportunity together going forward, whether it's Smart City, it's Wi-Fi 6, it's the expansion of your deployments. For folks here, we just want to give you a snapshot just to hear it from the customers, not us telling you the stories. These three gentlemen are available at lunch and during a break. Please feel free to ask them questions. They're here, obviously, in an open forum. We also have lunch coming up next. I'm the only thing standing between you and food, and a break. We have a demo area outside. We have different technologies being showcased. A lot of what Nabil and Eric were talking to are out there. I encourage you to check those out.
Finally, I'm going to put a plug in for our user conference. Last year, we did our first user conference. We brought everybody together from all the various organizations, and it was very successful. The type of numbers we're seeing, people that participated in the event, their investment with us on an apples-to-apples basis is up 15%-20%. If they've invested time in terms of getting training, their numbers are up substantially higher in the 50+ range in terms of increase in what they're investing with us. We're building it bigger, badder, better. It's in Nashville this year. We expect to double the number of attendees, and we'll have a good time. Anyway, just want to make you aware of that. With that, I think it's lunch, Stan.
It's lunch, let's all come back here around 12:15 P.M. We'll get started earlier. We'll have more time for Q&A at the end.
Thanks, everybody.
Thank you.
I've got one question about that last slide. What's the home run through here?
Yeah.
[Break]
All right, folks. I think we're going to get started with our next speaker. I just want to make a comment before we begin. We have three industry analysts in attendance here. We've got Bob Laliberte from ESG Group. Bob. We have Brad Casemore from IDC, and we have Alan Weckel from 650 Group. You can listen to us, but if you really want to get the industry background on us or the industry, I encourage you to speak to those folks when we go into break next time. Without further ado, I'm going to kick it off to Bob Gault.
Thank you, Stan. Was that loud or was that me? How's everybody doing? I drew the straw where everybody's-- That food coma is starting to work in. As you're shaking that loose, let me just spend a minute on this slide or so. Before I just jump into the go-to-market, I think it's really important to highlight the context as what's driving our approach. You saw a couple of data points on Ed's slide, which is, believe it or not, we're a 20+ year-old company. We have a heritage in quality. We are a pioneer in Ethernet switching. As you saw, we are the first in market with a one Gigabit Ethernet switch, first in market with a 10 Gigabit Ethernet switch, which is all great. At the end of the day, the team was selling a box, right?
When you're selling boxes, it's good for the top line. At the end of the day, it ends up becoming the law of diminishing returns, because even when you win, when you're selling a box, you could end up losing because you're going to compete on price. When you're competing on price against companies like Cisco and HP, and when you're Extreme, it's a very difficult battle. As a result of that, 3.5 years ago, we needed to step back. We needed to transition our company. We needed to transition so that we were focusing on software, focusing on our services, because it's the software and the services that are going to address a customer's business problems. The hardware is going to be a means to an end. That transition led us to a vision.
Our vision is to be a recognized leader in software-driven networking solutions from the enterprise edge to the cloud. What's important is that is the strategy to execute against that vision, which is really leading with software and services. If we can lead with software and services, that will accelerate the adoption of end-to-end wired and wireless networks from the edge to enabling the cloud services in the data center that you heard Nabil talk about. That's extremely important to us because we're going to go to market in specific targeted verticals as well. This is all about changing the conversation with our customers so that we can ask more questions, ask more business questions, and based on what they tell us, we can then deliver a solution to them that's based on the software and the services.
We then align our go-to-market to what our approach is. As you can see here's our go-to-market. When you build a go-to-market, there's certain things, certain factors that you need to take into consideration when you're building the go-to-market. This pyramid is just the obligatory sales go-to-market pyramid. Everybody uses these. It's simply an easy way to explain how we're going to market. The things we need to take into consideration now, we're now a $1 billion company. We have 46% of our revenues in the Americas, 44% of our revenues in EMEIA, 10% of our revenues in APJC. We have 275 AEs, 275 SEs. Those AEs and SEs, they have similar skill sets. They have different skill sets. They're covering different geos. They're covering different accounts. They come from different backgrounds.
Some are legacy Extreme, some are legacy Avaya, some are legacy Zebra. You have to take all these things into consideration when you're building the go-to-market. You can't solely focus on that, because based on experience, where I came from, so many companies organized on the inside out. We have to build an organizational structure that's from the outside in. What's the right thing for the customer? What's the right thing for the partner? What's the right structure that's going to deliver the best possible experience we can to our customers and partners? We're going to market in verticals. We have teams that are dedicated to specific verticals, and they're dedicated to verticals based on the geo and the presence that we have. As an example, in DACH, we have dedicated teams that are tied to manufacturing. We have dedicated teams in UKI that are dedicated to healthcare.
It makes sense based on our geo and based on the presence that we've got in those regions. In APJC, it makes no sense to have a dedicated team tied to retail. Our presence in that geo doesn't make sense for that to do that, so we have resources at the geo level versus in the region there. Right. We have dedicated resources tied to specific verticals. You heard what our verticals are earlier today. If you're a team that's not tied to a vertical, you're in a territory. It's that simple. Then we have another layer down, which is our channels. Our channels are helping us not only with the high-end customers, but they're also helping us with the long tail of business. They're helping us fill spaces that we necessarily can't touch from a dedicated resource perspective. Then we have inside sales teams.
If you inverted this pyramid, it's the same look from a service sales and a service delivery perspective. We are one of the only networking companies in the industry where sales and services rolls up to the same organization. Not a great thing for me at times, but it gives our customers one single throat to choke. They know we want their business. They're not going to give us their business unless we resolve their problems. They know our sales teams are not only interested in getting a PO, they're also making sure that once we get the PO, that their experience with Extreme is better than nobody else's out there. Our team supporting this pyramid, it's pretty interesting. We have our second-level management. We have about 42 second-level managers, and they have 8.2 years of experience with Extreme. 8.2 years.
The dedicated team, my team that reports directly to me, 4.5 Years with Extreme. What does that mean? It means that if you're part of our team, if you're our customer, you're our partner, you're going to respond differently because you know that person's going to be there. Something you said the first year you were at the company, did the customer or partner necessarily listen to you? Maybe. More than likely, they're probably going to say, "You're not going to be here," because this has been a revolving door. It has not been a revolving door over the last 8.5 Years for a second-line manager, and four years with our direct-line managers. Our direct-line managers have great tenure, have great experience, and great diversity. Our direct-line managers have experience from legacy Zebra, legacy Avaya, legacy Brocade.
It's a good thing because it brings a different way of thinking, a different way of acting in the organization. What's been interesting as well is over the last four years, our ability to recruit talent has changed dramatically. I came from a company where I thought it was going to be easy to try to bring some people over, and you're bending over backwards trying to convince people to come to Extreme four years ago. Now, I literally have people bending over backwards to come into Extreme. It's changed. Examples are, we just hired a great leader in the federal space. We just hired a great leader in the service provider space. These are A players that are going to recruit A players. In the U.S., the federal team and the service provider teams report directly up to me. Right.
That's a part of the organizational structure. Channels. Channels continues to play a critical role in our overall go-to-market. We have over 4,000 channel partners. We have about 180 distis, distributors, in the marketplace, and you can see the way they do business. Two-tier means a partner's buying from a disti, the disti's buying from us. 65% of our business is two-tier. We have another 15% of our business that's single tier, which just means that they don't want to do business with a disti, they want to buy directly with us, which is fine. We have business with the OEM partners buy directly from us, and then customers who truly believe they're going to get a different experience if they buy from a partner. They want to buy directly from Extreme. We're fine with that. Right?
About 80 %+ of our business goes through channel partners. We don't expect this to change. We're fine with this because we need our channel partners to bring things like capacity to us. Capacity meaning we don't have 10,000 account executives out in the field. We have 275 account executives out in the field. How are we going to cover that territory where we just can't afford to cover them with a direct model? Our channel partners are helping us with capacity. Our channel partners are also helping us with capabilities, and this role will continue to play a more critical role for us. Because the capabilities means that they can not only sell our stuff, they can install, they can deliver our stuff. They help us with competency.
Competency means that they made the investment to build a practice around Extreme and a certain specific solution tailor or all of them. We have a channel team, a channel infrastructure team that consists of channel account managers, channel system engineers, distribution account managers, trainers. We have a full complement team, and the role that they've been playing over the last couple of years is rationalizing. They've been rationalizing our partner community. They've been rationalizing our distribution community, our programs, our trainings to make sure that we've got a structure in place that can align to growth. They've been busy doing that over the last 1.5-2 years. They made fantastic progress here. This slide is really how we're approaching the marketplace. We have multiple levels of partners. The level of partnership is really based on the investment that they're making in Extreme.
That investment's value-based and is volume-based. Regardless of the role that they have, all are equally important to Extreme. On the left-hand side is how we were going to market with our partners. Think of that as a one-to-many model. It was, I'd say, more of a reactive, tactical overlay support model that you're used to within the channels. On the right-hand side, we've moved from a one-to-many model to a one-to-few model. The one-to-few model is basically meaning that we've got about 400 partners that represent 65% of our business. Why wouldn't we put dedicated coverage on those partners to make sure that we're getting as much wallet share as we possibly can? It doesn't mean that the other partners aren't important. They absolutely are, but we'll cover them differently. Maybe we'll cover them through distribution, and we'll cover them through inside sales.
We've made investments here to make sure that we can do that. We've invested $50 million into our channel program. That's including programs, marketing, tools, people, and those incremental people are helping us cover more partners. We're seeing success here. You heard Ed talk about the Master Specializations. We have over 120 master specialized partners now. Over 120. We had none 1.5 ago. Master specialized partners just basically saying, raising their hand and saying, "I am going to invest deep and broad into a solution pillar." That's what they're doing, 120 of them. The 120 master specialized partners, it's really easy for us to have a conversation with them as to why to be a master specialized partner, because they're growing 48% year-over-year. 48% year-over-year, so they're growing faster than the non-master specialized partners.
We have 550 specialized partners, 550, they have generated almost $1 billion in pipeline for us so far. We also have a new tool called Deal Registration. Deal Registration basically means it's that science and the art between partners and direct selling, who delivered the incremental sale to Extreme. Because the Extreme people will say, "We always did it." The partners are not playing a role there. Deal Registration allows us to help quantify that. Because the Deal Registration basically means there's an opportunity in the existing customer, or there's a prospective opportunity, and the Extreme account executive wasn't aware of it. We have hundreds of millions of dollars in the pipeline around Deal Registrations. It's helped us drive incremental business so far this fiscal year. What's important to understand here is that we have the same number of partners. We have the same number of partners.
The coverage model's different because we're focusing on the partners, less is more, that can help us drive incremental business. We're seeing the data. It's working. Our Black Diamond partners, which is a designation for partners that says we don't have one master specialization, we have two, those partners have driven the preponderance of our cross-sell business within the company. Regardless of how we go to market, we have to make it easy for our sales teams, for our customers, for our partners to do business with Extreme. Because if it's not easy doing business with us, why do business with us? They can do business with somebody else where it's pretty hard to do business with them. Digital transformation is helping us do that. You heard Ed talk about it.
We are investing tens of millions of dollars in the digital transformation to help us move key parts of our business from manual to automated. As an example, sales process automation. This is where we have teams of people within our organization that are spending 20% of their time just helping our other teams with quoting, pricing, BOMs, literally 20% of their week. That's one day a week that they're spending just doing that, which means they're not helping us build pipeline. We're fixing that. We're fixing that. We've already begun making that transition. We're through phase I. In March, we'll be through the final pieces of that transformation, where we can give 20% of the time back for our inside sales teams, where they're spending a lot of the time doing administrative stuff and less time calling existing and prospective customers. That'll help.
That'll help drive pipeline warm leads to our account executives. We're also moving towards a worldwide sales channel portal, where 80% of our business goes through channels. Yes, our channels are leaning on the Extreme account executives and Extreme system engineers to help them with things like pricing, configuration, quoting, BOMs, which means our account executives are not calling new customers. They're helping our partners. If we can take that out of the hands and give that responsibility to our channel partners, first of all, they're looking for that, we can put more productivity back into the hands of our partners. That will be available in another month as well. We're moving towards order process automation, which is just going to facilitate an order through the system that much quicker.
We're doing all this so that we can take the friction out of the sales process, and we can give more selling time back into our partner account managers, channel account managers, account executives. This is paying big dividends for us. Then you have sales enablement. Sales enablement is just another way how we can reduce the friction, and we can reduce the frustration that our sales teams have right now. Frankly, this is all about how do we simply get our teams to sell more? How do they understand what they're selling, and how to go about selling it? We are launching a comprehensive sales and technical program. We're making investments in our technical teams, our training teams, and it's for the sole purpose of investing in our customers' success, our partners' success.
Just last year, we've doubled the number of training sessions that we did with our partners in 2018. We have also invested in technical training so that our customers and partners, our technical labs, so our customers and our partners can actually get their hands on the equipment that you heard Nabil speak about. We have broadened the number of training partners we have so that we can deliver training in nine languages. We're a global company. We need to start doing that. What we're really excited about is we introduced Sales Dojo. Sales Dojo is a training program, a sales and technical training program, where it's martial arts belts-based, and so it's where we're providing our teams and our partners with knowledge and information around Extreme. What's the Extreme story? Foundational. How do you go tell the Extreme story? Here's the questions that you ask.
If a customer asks this type of question, here's how we would recommend answering that question so everybody is on the same playbook. Everybody can tell the new Extreme story. We're moving to a next belt, and that next belt's really commercial training. It's all about our programs and our services, so everybody's familiar with that. You go to another belt. That other belt's all around solution selling training. Everything's getting a little bit heavier. All the solution pillars that you heard Nabil speak about, well, how do we educate our team so that they know how to sell this stuff? The first two belts, we have 100% of our field sales teams, they have their yellow belt and they have their green belt. Solution selling belt, which is a little bit heavier, we have 60% of our folks who have their green belt.
Then we move to a purple belt. A purple belt's basically advanced solution selling, which means it's 5 days out of the office. We have 20% of our people who have their purple belt. We have people that want to leapfrog the blue belt to the purple belt. You can't do that because you have to test out. We have to know that you're not only watching the material, you're actually paying attention to it. With our blue belt, you have to take a video. You have to take a video, and you have to prove to us that you can sell that in front of the customer before you go and get your purple belt.
We will then move to a black belt. The black belt's basically, it's belt-based, you're sensing, which means you not only know this stuff, but you can train other people within the organization. We've added a ton of horsepower into the team as well. We went from two people to 15 people in training. I'm just telling you that because we're committed to doing this. You heard about all the transformations that the company's going through. If we don't enable our teams so that we're all aligned, we're not going to be able to grow the business. We're committed to doing this. What we're also committed to doing is making sure the teams are getting their belts. Equally important, that we're pushing all the same belt strategy into our partners.
They're an extension of our sales teams. We all speak in the same language in front of the customer. We feel like it's working, that all the enablement we're doing, it's translating into results. Our teams are now able to have a different conversation with our customers, where they're able to sell the broader portfolio in front of our customers. We have sold tens of millions of dollars in cross-sell. What's interesting is that there's 12 different combinations of cross-sell, 12 different combinations. Every single one of those combinations have been sold throughout the fiscal year. Every single one. Equally interesting is how the different cross-sells are being adopted within each geo. In the Americas, basically how you read this is that you have the Avaya legacy base, networking base in Americas. They're buying a lot of Extreme. If you think about it makes sense.
We inherited a lot of healthcare customers. A lot of those healthcare customers had campus switching, campus fabric. What they were really struggling with was visibility and manageability of that. We can help them. What they were also struggling with is how do they thread that fabric all the way to the edge so they can get the same experience. We can help them with that as well. That's happening a lot. What's also happening is in EMEIA, you have a lot of the Extreme legacy customers. The fabric story, they absolutely love it. They're embracing the fabric story, Extreme customers, moving not just to what used to be Avaya switching, but also management and policy and also the wireless story. It's also a great transitional approach for us for our chassis story, moving to more of the legacy Avaya product line. This has resonated.
What you see is in the Fed and the SP space, and also in APJC. It's the Brocade customers, legacy Brocade customers who are leading the way. Regardless, because of the success that we're having here, a little bit of piggybacking on the question earlier, our go-to-market over the last 18 months has been cross-selling into 32,000 new customers. Every single one of them are warm lead for us. New logos are really, really important. They will never stop being important. That's how you become a healthy company. At the same time, we have 32,000 warm leads that we're going to continue to cross-sell into each one of them over the next 12 months. And these are just some of the companies that we've been cross-selling into.
This is a who's who of large and commercial enterprise customers, each one of these customers are using an element of our software suite so that we can provide them a great customer experience and deliver a great vertical outcome. What do I mean by that? In the healthcare space, no matter where I travel in the world, no matter who you speak to, the one common feedback that you have from a healthcare company in Australia, it could be in Germany, it could be U.S., they're all looking for secure, roles-based access to medical records. That's what they're looking for. They want to have the physician to have access to one set of records, the nurses have access to another set of records, and by the way, the patient having access to their own records on a secure mobile device. Man, wouldn't that be ideal?
If they get to deliver that, it might be easier for them to attract better doctors, better nurses. If they can deliver that to their patients, man, that would be a great experience. It's all about delivering a great experience. You saw all the boxes on the Nabil slide. I appreciate it. What we're doing in the sales organization is I challenge every one of them, go into a conversation and never talk about a box. Don't talk about a box. Talk about the customer's business problems, if you can figure out what the customer's business problem is, more than likely, we'll have a solution to be able to address that. More than likely, it's the software that's going to be able to address that.
The way we implement it, our wired and our wireless and our policy and our management, it's able to give policy to healthcare clinics so they can give secure, roles-based access to medical records. If somebody's going to get a CT scan, they want to know the results right away, they can pull up their mobile device, regardless of what the results are, and they can get them. Good experience. Hey, I like that healthcare clinic. They're probably going to speak to other people about that, if their doctors are associated with that, they may attend. We fixed a business problem, right? Just use another example, you take a look at logistics. FedEx is a big customer of ours. Believe it or not, during the holidays, they ship over 500,000 packages a day. Guess what their biggest challenge is?
Man, let's not lose a package. We cannot lose a package. They lose 1% of their containers. I don't know how they do that. I don't know how that happens, but in that container is a bunch of boxes. If during the holidays, you happen to be in one of those boxes, one of the presents you're supposed to receive or one of the presents you're shipping, it's a bad experience. Somebody's not going to get it. I may go DHL. I may go UPS. If we can help them with our software and our services to help them resolve that problem, reduce the number of containers they lose by a certain %, it's a better experience for them and for their customers, and they keep those customers. That's just some of the ways that our software and services are helping address business problems without talking about the hardware.
I thought I'd share with you our compensation plan because it's always important to understand what's driving behavior, why our teams are asking the questions that they're asking. This is just to give you a sense for what we're doing. Last year, we thought we cracked that compensation code. For those of you who were here 18 months ago, we said we were going to pay our teams on cross-selling, and we were going to set up certain gates so we can take the technology religion that came with them off the table, and they can make sure that they're performing the right design for their customer, regardless of the solution. We went to market in gates. We figured out through all the integration of the tools, processes, people, that that was a little bit harder than we thought.
Now as we move into fiscal year 2019, cross-selling, it's expected. Everybody's doing cross-selling. It's table stakes. We're paying everybody to cross-sell across every one of the solutions. What we're also doing is emphasizing software. We are emphasizing software, we're paying everybody double commissions. We're not just paying people commissions. That's also going towards retiring quota, which is very unique in the networking industry. Many times people pay, but you can be a rich person without a job because you didn't hit your number. Now it's going towards retiring quota, and you're getting compensated on that. Gross margins. They're extremely important to us. We not only have to grow the revenue, we have to grow the gross margins. You saw we are doing that. We want to continue to do that.
We have a material piece of the team's compensation plan tied to gross margins, it's working, right? It's putting its self-discipline in the field around discounting. We're doing all this in addition to paying a dollar for dollar on services. We don't want this to happen, if somebody has a $5 million quota, they can retire their number selling services. 75% of our business is product, 25% of our business is services. We want our teams to continue to think about new maintenance and renewals as well as professional services, premier services. We will pay them like that's a data dollar. We're doing all this so that we can harmonize our teams, so that we can get our account executives, our channel account managers, our services people all working in the same direction, right?
For the sole purposes of focusing on a differentiation, which is software and services, we can leverage that to grow the business. When you think about can we compete? Well, with our end-to-end portfolio, with our increased market awareness, we are touching thousands and thousands of new customers and partners with Extreme now over the last 18 months. Our position in the Magic Quadrant, well, we think we have enough to compete with our competitors, and our customers are telling us. If you take a look at Watson Clinic, Watson Clinic had an old architecture. They were looking to move to a next-generation platform. They also needed to do that within a consistent cost model. It was their key objective. Upgrade the network, do it within a consistent cost model that we know that we can afford.
With that, we asked a bunch of questions, tried to figure out the problems they're trying to solve, and we moved them towards ExtremeWireless, ExtremeSwitching, and professional services. We also did this with a five-year cost model, right? A flexible business model. We not only provided them with a great technical solution, we provided them with a great business solution, which we also believe is a differentiator for us. Watson now is the first healthcare clinic in the county to be able to offer up a 64-slice CT scan to their heart patients. Their doctors and their nurses can now provide non-invasive robotic surgeries for their cancer patients. It's a great experience for everybody, a better experience for everybody, which again, is going to attract better patients, better doctors. If you think about Bowen Center had huge communication challenges throughout the clinic.
They had big throughput issues. They had teams that were spending 30% of their time working on unplanned outages, trying to fix the wireless network, inputting documents. It's just very non-productive. We sold them Extreme Management, ExtremeWireless, ExtremeSwitching with our services. As a result of that, they became more collaborative. They have more clinical video chats. They have more video meetings. The wireless switching is more reliable, which is providing them a better experience across the healthcare clinic. The challenge was, when you make three acquisitions, you had great customer service. You were known for your customer service. Can you keep that customer service up after the acquisitions? People didn't think we could, that it's really difficult to do that. The number one thing you need to do is our strategy is to do everything in-house.
We needed to take all the different acquisitions and how they were outsourcing everything and bring it in-house, and that's what we did. You can see here, we've been rated number one in the industry. This is after the acquisitions, and not only providing a box sale, but a solution sale to a customer, and we're still able to provide world-class support to them. A number that's not up here is 94%. We are able to deliver a 94% first-person call resolution. That's unheard of in the industry. Basically, what that means is your call's going right to level 2 TAC. It's bypassing level 1, going right to level 2. That person who fields your call, 94% of the time, they're going to resolve your challenge. That quote that you see at the bottom, a customer sat next to me and said that.
They were an Extreme customer. They ended up going to another customer, and they forgot how good the service was after we bought Zebra WLAN. They had some challenges, and they were just like, "I forgot. I just forgot how good the customer service was." Just means that once you get a customer on board, we keep them. Just wrapping this up, this fiscal year 2018 has been a transitional year for Extreme, but not just Extreme. It's been a transitional year for networking in general. There's been lots of conversations around cloud and multi-cloud. Lots of conversations around 6.0, around virtualizing the campus. Our customers have more alternatives than ever before to make a decision, not only what networking vendor they're going to go with, but what they're going to buy from that networking vendor.
Sitting number 3 in enterprise networking, it puts us in an enviable position because we have the ability to be able to lead these customers through a transition, through the next transition that's going to happen. I sat in at Americas CAB, Customer Advisory Board. We sat in the EMEIA Customer Advisory Board. Our customers literally unsolicited said, "You have to become more aggressive. You have to become more confident that you can compete with the big boys because you can. You have the solution that can do that." That's what's exciting about this. We have to get our teams confident knowing that they can compete, and we have the solution to be able to offer a customer to address almost any problem. That's what we're doing with our teams. We're reinforcing this. It's an exciting time. Always been an exciting time to be part of Extreme.
No more exciting time than right now for our customers and partners because we're turning that corner. From our view, we're just getting started, and that's what the exciting thing is about this. Thank you.
About three months ago to the day, I was looking at the offer letter I got from Ed, and I had a great conversation with him where he was very straightforward and candid about his expectation. There was one thing, however, in that conversation that he forgot to mention, which is that in any public event, I have to wear this tie.
I've got to tell you, had I known back then, I may have reconsidered. For the record, this is courtesy of Norman Rice, who in addition to the long list of roles, happens to be our chief fashion design officer. To his credit, he's spending more time on the other roles than on this one, which is good news for us. For the next 20 minutes, I'm going to ask you to please look at me in the eyes. Don't spend too much time on this tie. We're going to focus on the numbers. I think Ed mentioned the journey that we've been through over the past five years. It's always nice to see when you have these type of bar chart, the revenue doubling, but at the same time, a steady improvement in the gross margin.
That growth that we achieved, which was
A combination of organic and inorganic, as you all well know, was certainly not achieved at the expense of gross margin. If I look at the operating margin, which is the line at the bottom, it peaked at 9.8%, which is a little frustrating for me back in fiscal 2017. We're not quite back at 10%, but as I will discuss later, there's absolutely no reason for Extreme not to generate a double-digit operating margin. Turning to the balance sheet and cash flow, it's also a pretty impressive journey. Cash doubled from $76 million- $141 million. Accounts receivable also doubled. We were able to take our DSOs down six days. I'll talk about cash conversion cycle. I think we can do a much better job. We maintain our inventories despite doubling the revenue at the same level, which means that our days went down almost 10 days.
Days of inventories actually went down from, sorry, 121 to 62. We've cut in half our days of inventories. Our debt, gross debt did double as a result of some of the acquisition that we made. If I look at our net debt, we had a cash position back then of $7 million. Today, we have a net debt of about $44 million. Needless to say that those assets that we acquired were acquired at very reasonable multiples, and we paid off some of that through the cash flow generation. Turning to cash flow. For the whole of fiscal 2015, we generated an operating cash flow of $37 million. Last 12 months, ending in Q2 of 2019, that number was $66 million. We did have quite a bit of investment in CapEx.
We mentioned the digital transformation, we also invested a lot of money in our San Jose site, where we consolidated several companies into one site, invested big time in a data center. The free cash flow is impacted by this CapEx. The ability to generate operating cash flow is well reflected, I think, in these numbers. We're spending an awful lot of time, and that was part of the expectation when I joined the team from Ed, on gross margin and operating expenses. This is where I'll be spending more than 50% of my time in the quarter going forward. You see the yellow line going up. That's the service gross margin. We're already north of 60%. We're doing a fantastic job there. The white line, that's the product gross margin. It took a dip with some of the acquisition we made.
We expected it to recover. The issue around the tariff that we discussed during our Q2 earnings mean that recovery is not happening as fast as expected. I will talk in a minute about what we're doing to make sure that line goes back above 60%. As far as operating expenses are concerned, the blue bars here represent the absolute dollar spend, and some of that is seasonally driven. What I really track is the expense as a percentage of revenue. For the most part, we've been above 50%. That means that if our gross margin is in the high 50s, operating expenses at around 50%, you get a high single-digit margin. I see no reason why that number should not come below 50%, and I'll talk about that in a minute. Ed mentioned this.
When we acquired those assets, obviously they were part of bigger companies, and there was not a huge focus on their ability to generate high gross margin. We've been able to turn this around dramatically. Zebra, Avaya were both in the mid-40s. They're now both in the high 50s. Brocade was already in the low 50s, but we managed to increase that to the high 50s, and some products are actually in the low 60s. Finally, we also did a great job with what we call Heritage Extreme, which is the former Extreme plus Enterasys, and took that from the low 50s to high 50s. Again, we're spending a lot of time, and I'll talk about it in a minute, on trying to improve that. I think Bob talked about the opportunity that's in front of us.
I just wanted to put numbers, those numbers obviously coming from the industry analysts, the total market for which we have products is currently $33.4 billion, and that's growing at a rate of about 4%. The part that we can effectively address, because either we have a product range or we serve in that geography, there's markets like China, for example, where we don't play, is actually $22.2 billion, and that's growing at a rate of 3%, according to the analysts. The part where we feel we have the strong position in terms of market share, which is the combination of wireless LAN plus campus switching, I've added two of the rows above, is actually $17.4 billion. For our part that we service, $11.3, and that's growing at a rate of 5%.
the point here is where we have the strongest market share is also where there's a big opportunity in terms of growth. I just want to highlight a point that for us, it's all about gaining market share. The leverage is quite substantial. When you service a market of $22.2 billion, if you're able to get just one point of market share, that's additional revenue of north of $200 million, which at a gross margin of 60%, even if you may incur additional operating expenses to be able to service that $200 million in revenue, is a significant incremental contribution to earnings. I'm not saying that it's a walk in the park to get a point in market share. As Bob and some of the other speakers mentioned today, it's a fight of every day around your best go to market and having the best product.
I just wanted to provide that sensitivity analysis to say that as we grow market share, we're going to generate significant incremental revenue and earnings. This slide, I saw a lot of you took notes frantically during the presentation before, is really a summary of what was said. I can now try and translate that into numbers. You heard both Eric and Nabil talk about the product refresh. 70% of our product either will introduce new products or in some cases of existing product, we'll do what's called value engineering or redesign to cost, which consists in continuing to sell that product, but with improved motherboards or componentry or a chip set inside. We're also talking about the importance of software. Today, that's less than 5% of our revenue. As we grow this, obviously, it generates additional revenue, but also higher margins.
We talked about greater wallet share by providing end-to-end solutions where we can potentially displace Aruba at our customers or Cisco in other areas. Bob mentioned the cross-sell and up-sell. He mentioned the importance of education, both of our salespeople as well as our business partners. The industry recognition that we're getting through being part of the Magic Quadrant, for example, is also definitely helping us. One point I also want to mention in terms of gross margin is we see a path to a 62% gross margin. Again, very difficult. It will require flawless execution and a strong focus on every element of this waterfall. The combination of pricing, just to give you an idea, every time we're able to increase our net selling price, so list price minus the discount we offer our customers by one percentage point, that's an additional 1.5 percentage point in gross margin.
I mentioned the refresh of the product and the value engineering. That's potentially another one percentage point. We're having huge efforts in improving our supply chain, and that consists not only on having best-in-class manufacturing and supply chain, but also better demand planning. The better we're able to plan demand six months ahead from now, the better we're going to produce products that either business partners or end users are going to take, the better we're going to be able to ship our products by sea instead of by air. That aspect that we're working on is also one percentage point. We talked at our Q2 results about the impact of tariff. We took a hit in Q2. We're going to take another one in Q3. That's the one point that you see there. We're not making any assumption for the mix.
The mix can play both ways, depending on what regions or what products or the mix between software and products. At this stage, we're assuming that's neutral, but it may play both ways. Finally, as we grow, and I'll talk about our expectation in terms of growth, the volume impact enables better absorption of the part of our fixed costs that go into the cost of goods sold, and that's potentially another point. That's 62%. Obviously, what is not shown in there is the unknown. If the folks that are currently in Beijing trying to negotiate with the Chinese come back empty-handed and we end up with a 25% increase in tariff, that would have an impact on us. If, for some reason, we have to be more aggressive to defend certain customers from Huawei, that could potentially have an impact.
We're not saying that we'll necessarily get to 62%, but at least we see a path to get there. I also want to talk about operating expenses. First of all, let me be specific in saying that getting from 8.1% operating margin on a non-GAAP basis in Q2 to our goal, which is 10%, will come largely from the operating leverage. In other words, if Extreme, which is currently a billion-dollar gross of $1.2 billion, $1.3 billion, $1.5 billion over the coming years, and we're able to maintain our fixed cost where we are, that's going to take us to 10%, 12%, 15%. Having said that, there's still a number of things we can do to improve our fixed cost basis, our operating expenses as it stands today. I wanted to highlight some of that.
Part of lifecycle management, I'm spending a lot of time with Nabil on looking at where products are at a various stage of their lifecycle and defining a set of non-financial and financial metrics that would basically enable us to say a product which is now on the decline could have an R&D to revenue ratio, I know Nabil doesn't like that, of 5%, as opposed to a product which is growing, where that ratio has to be 15%, 20%, 25%, just to give one example. Engineering hubs. Most of our sites tend to work on several technologies. However, we've really tried to gather expertise in each of the sites. Toronto is a center of excellence for wireless. Salem tends to focus on XMC as well as campus. Raleigh, North Carolina, would be where we have the most expertise on Smart OmniEdge.
Finally, if you go west in San Jose, you'll see a lot of the data center expertise that has been gathered in that site. Product platforming is another great example. We try to have common hardware that can run multiple pieces of software, but we design the hardware once and for all. Bob talked about some of the initiatives that he's having in terms of trying to have a frictionless process. We were particularly pleased last week that for the first time, we had some orders that came through our business partners or distis in our portal, where we had an automatic DA Discount Approval that was processed, and this went straight into our supply chain, and the product was shipped without any human intervention.
The reason this is so important is 80% of the bookings that you see every quarter are what we call run rates. That run rate really consists in a customer who already has an Extreme install base buying additional switches, additional power cords, ventilators, and we have a lot of salespeople that are involved in that process. They should not be. They should be focusing on the remaining 20%, which is getting a new network, getting a new customer, or selling additional stuff to an existing customer. That is going to drive significant reduction in our sales and marketing expense. Finally, in G&A, which is where I tend to focus a lot, if you look at our G&A to revenue ratio, it's best in class. We can still improve that. In finance, we've introduced, with Automation Anywhere, robotics and process automation in how we collect from our customers.
We intend to roll out RPA to other fields of finance in the months to come. There's a number of things that I'm looking at in terms of our real estate footprint, for example, where we can get additional savings. One thing that we haven't talked a lot, although we do talk about DSOs, DPOs, and days of inventory, is our overall cash conversion cycle. When you actually do the math, it does not look that great. The total at the end of Q2 was 78 days. Some of our competitors, HP, for example, are at a much lower cash conversion cycle. Part of the reason that you see this high number is the introduction of vendor-managed inventory, where we were able to reduce our inventory because our partners for our supply chain are taking some of that with them.
The offset of that is that when they pass that inventory over to us, they expect to be paid within 15 days. That's the reason why you see DPO where they are. Over the longer run, however, I believe we should be paying our suppliers the way we're paid by our customers. You should see DSOs and DPOs both in the 55-60 days, and we should continue to improve our days of inventories down to 45-55. That would take the overall cash conversion cycle to about 45-55 days. Each 10 days reduction is $10 million in free cash flow. There's a significant opportunity there as well. I wanted you to get a feel of how we think about capital allocation and the outlook. Under our authorization to buy back shares, remember we had $60 million.
We used $15 million, 15, one five of that in Q2. We have another 45 to go, which will be basically potentially spread over the next year and two quarters. Our current debt to EBITDA ratio is well below two, well below some of the financial covenants we have with our debt. We still have the capacity to bring additional debt of $140 million between the incremental term loan that we have and the revolver. That's something to keep in mind. We're very happy with our current cash on hand, which is $140 million. We'd like that number to pretty much stay there. You should think about CapEx as about 2%-3% of revenue. We had some high investments around digital transformation, the facility in San Jose, but those are now behind us. The normal run rate is 2%-3%.
In general, as we think about how we're going to grow the business, we're going to have to have a balanced view between organic growth and inorganic growth. If there's any opportunity that comes in front of us, we'll certainly look at it. We talked a couple of weeks ago about the guidance for Q3. I just want to remind you of what we expect, but I'm pretty sure you're all familiar with that. I thought I could skip directly to what we see, even if we take a longer-term view. Let's talk about the next two to three years. We think that with everything we described today, this company should be growing at a minimum of 5%. It's not going to be easy. Not every quarter will be there. That's certainly the target that we have in mind.
We talk about a path to 62%. There's always some unknown factors. We saw that we didn't get the gross margin we wanted to have in Q2, and Q3 is going to be at that level. We're providing a range of 60%-62% for our gross margin. We're definitely going to take the operating expenses below 50%, which is currently where they are, through a combination of operating leverage and some of the actions that I described. Overall, I see absolutely no reason why Extreme should not be generating an operating margin of 15% on a non-GAAP basis. I think those are the three key metrics that I wanted to mention. With that, I think we have a Q&A session. Thank you very much.
I want to invite all the members of our executive team on the stage for Q&A. Then, as we kick it off, please limit yourself to one question and one follow-up. We have two mics.
Yeah. Is this on? No. Are we on now? How about now?
They don't want you to talk.
She's Bob's guy. I could use this.
Hey, Chris, it's not working.
I just wanted to acknowledge a couple of things. First of all, I think Rémi's been a great addition to our team, so we're excited to have him on board. One of the things that we have at Extreme, we have a diversity initiative. We have a women's council. We've made a lot of progress with that. It's making a dent in the company, and we're excited about that. When we were embarking on the CFO search, we always take the view that we want to hire the best athlete, but from a gender perspective, if we can hire a female, we would like to hire that best athlete. I spoke to the board about the CFO search and let them know that's what I wanted to do, and we ended up with Rémi. Sorry. Someone who looks a lot different than I look. Yeah.
That's true.
Seriously, we got the best athlete with Rémi. It was quite a search process. I want to recognize Matt Cleaver in the back of the room, who's our interim CFO. Matt did a great job, and Matt's an integral part of our team. Integral part of our team, still is. Anyway, just strengthening the overall bench and the finance team. We'll open up for questions. Paul.
Just a simple question for Rémi. In terms of driving the margin improvement, at the end of the day, does it just simply come down to volume improvement in terms of revenue? That's what you need to get the leverage on your OpEx, and I assume your gross margin, notwithstanding what you laid out, is going to be also highly tied to revenue improvement, or is there something more?
There's a number of things. I start with the gross margin. There's a very delicate balance between the improvement in gross margin and growth. Bob and I regularly have constructive conversation about certain deals where we're being challenged by some of our competitors, and the question is, do we want to actually go after these deals because they make sense strategically or just let them go because they're going to further delay our target? Now, obviously, one deal is not going to make a difference, but every quarter, you're probably going to see half a dozen of these deals that we have to go in at gross margin that are significantly lower, and we always have a good discussion about whether or not we want to do that. That's one aspect.
Second aspect is I don't think we're best in class in terms of demand planning. We have a good view of what's expected in the next three months. If you ask me what we're going to ship in terms of products for the next six to nine months, it's harder. Therefore, we must, as we work with our manufacturing and supply chain partners, really think hard about what products we're putting out there, because if you have the wrong products, that you end up shipping 95% by air, which costs you a lot. That's a second work stream that we're going to address. I think the third one is every aspect of the bill of materials that goes into the product.
The 70% refresh that Nabil talked about is super important as well, because not only does it bring products that are more attracting to our customers, it also enables us to significantly reduce that BOM. If we're able to maintain the net selling price or even reduce it less than our bill of material is reduced, that is a significant driver. That's gross margin aspect. On the OpEx, I think I went through it. It's a question of operational excellence to reduce our cost base without reducing the number of feet on the street in terms of sales or the number of lines of codes that are written. There's a number of things that we're doing. More importantly, it's going to be the operating leverage. Our fixed cost base is pretty healthy.
It's just that it's basically adapted for a company that's bigger than 1 billion. We need to grow that.
If one tried to isolate the impact of the 70% refresh, if we assume no incremental revenue. I'll repeat the question.
That was the one point I was showing.
That'd be one point?
Yeah.
Even in a no revenue growth environment?
Correct.
Okay.
Yeah, because I showed the volume impact later on the right.
Paul, I think the other point that we have to raise as it relates to gross margin is Bob's comments, I think it's worth Bob commenting is, again, as it relates to discounting. Discounting is so important. When we are leading with software-driven solutions, we typically have lower discounts and higher gross margins, and typically, we're over 60% today. The issue comes to the competitive situation where it's a box selling war and we have pricing pressure, and these are the kind of things that we fight in the field. It's the overarching move towards software and software-driven solutions that can have a bigger move. I think Rémi's isolated very specific, tangible things as far as the refresh, the portfolio, stepping stones, if you will, to get us to 60%.
Ultimately, it's what happens in the field every day with all the deals that happen in the quarter. Bob, I think it's worth you commenting.
I think you're right. Old habits die hard, and we've had people who've been over-discounting, whether it's through our partners or directly to the customer, for some time, and maybe because we had to. We've really pulled that back. We pulled that back a lot. We saved some discounts there. I also think that we put more balance back into our business, whether it's vertical or geo. 75% of our business happens through the verticals right now, so we're not so reliant on one geo or one vertical, which impacted our gross margins in the past. If you think verticals like K-12 or education, we're not so reliant on that anymore. Through the acquisitions, we also picked up other verticals like federal, where we're much stronger in today, and other geos like Japan, which will help balance out the gross margins across APJC.
I think it's a combination of all of it, as well as just putting focus on it from a compensation. You can't overemphasize that. If you're paying them to be duly diligent around discounting, they're going to be sensitive to it.
We have a couple of follow-up questions. We'll get back to you.
Alex, we'll come back to you. Eliza didn't want you to have the mic.
Just a clarification, the 1.5% of margin benefit from pricing, is that conclusive of the discounting improvement, or is that just the price increases? Thanks.
That's assuming no additional discount. The way we think about it is that if we do a three percentage point increase in the list price based on an average discount, that leaves us a one percentage point improvement in the net price. We take one and a half points to gross margin. If for any reason that is offset by additional discounting, that will obviously reduce the improvement.
Thanks for the clarification.
Alex had a question there.
Just a couple of quick ones. How long does it take you to get out of China so that you're no longer exposed to the tariffs, assuming that you're moving your production out of China? You've talked about pricing step up that you've announced, but you've also said that you're going to make that part of your ongoing business model. What should we think about in terms of annual price increases? Could you give us a little bit more granularity on the software percentage today versus where you think it'll be two or three years out?
Okay. I'm going to divert the first question over to Norman, if you want to talk about supply chain and shifting from China.
Sure
Taiwan.
When everyone was notified mid-September of the tariff, it happened relatively quickly. People had to react and make plans. Our production for U.S. products, call it 80% of our revenue comes from 20% of our SKUs. That production is shifted by the end of March for those SKUs for the United States. Think of it as a TAA-compliant production. Those will be predominantly made in Taiwan, and it's a shift of the legacy products to that market. Our future products, a number of our future and go-forward products, are being made in the U.S. or Mexico, or a combination therein. Our next generation super spec platform that both Nabil and Eric talked to, which is called the 465, it's being made in Mexico. That was something we were thinking about long term in terms of our strategy and moving closer to the source.
There was another question relating to software. Today, it's less than 5%, just less than 5% of our revenue, and I think over the next couple of years, as we look at it, we would see that number climbing over 10% for sure. That's below the internal goal, but that's a number that I would put out there.
Pricing annual?
I'm sorry?
Annual pricing.
Oh. The annual pricing, it was interesting. When we raised price, customers came to us and said, "Your competitors raise price all the time. You never do. Why don't you raise price every year and maybe at a lesser rate?" We said, "That's a great idea. We'll do that." What is that going to be? If that's how the market's conditioned and it's a point or a two-point increase in pricing, that's something we're going to do that going forward. We just should. It's good business practice. It's something that not only Extreme, but all the companies we acquired just didn't do. The reaction from partners and customers was pretty interesting on that front, and it was sort of an eye-opener for us, and I think good business practice, you'll see that from us going forward.
In the waterfall I provided, we're assuming one price increase of 3% on the list price, which is the 1% on the net price that I described. We're assuming once. If we have an opportunity to do again, we will, but we're assuming once.
Mark.
Yeah, just a clarification on the last question first. Tariffs go up, you're anticipating you'll be out of the way of that, or will you get hurt by that, or will you raise pricing if you can't get out of the way of that? I just wasn't clear on your timing there.
Yeah. First of all, we issued a price increase, took effect November 1st. It was 5% globally, 7% for the U.S. The 7%, the uplift of 2% was directly tied to the tariff itself. The global price uplift was related to other component costs and other things that have gone on, really kind of stabilizing our cost basis for the products. Going forward, the anticipated tariff, which I think you're all familiar or may be familiar, the latest news is it's still in conversation. I believe Xi Jinping is meeting with the trade leaders on Friday. The anticipated outcome is that they'll delay the March 1st date by another 60 days. In effect, saying the 10% tariff that's in place will stay in place for call it another 60 days.
That said, what have we done in terms of preparation for the 25% or the additional 15% uplift? Well, we shifted production. That's number 1. Number 2, we've already communicated to our distributors we have a requirement of a 30-day notice. We issued that 30-day notice in December. If and when that tariff ever moves into effect, we have an automatic price uplift that's already in the market. It's already a known thing. We did that because we're not sure about lead time in terms of when the notice will come in and when that tariff will take effect. That's already been communicated. We will uplift prices if the tariff does go up.
Great. That's helpful. Switching topics, the K-12 vertical. You just mentioned a few minutes ago that you've done a great job diversifying away from that. You were heavily dependent on that. It may actually be on a refresh cycle. We're three years later from where it was. Are you seeing that? Are you devoting resources to capture that? Just an update on K-12.
Yeah, let me comment. Then I think, Bob, it'd be great for you to jump in as well. I think that the K-12 market is sometimes it's a blessing and a curse because of the E-rate spending cycle. That has opened the doors. It's been a good thing for schools to build out their networks. The volatility of the funding of the project has created volatility in revenue around that stream. This past year, we've been hurt because it was a very soft E-rate season last year. This year we're