Extreme Networks
>100% ROIC on tangible capital
Little debt
Trading at <10x FW PE
Various tailwinds ahead
What I try to look for is the holy Trifecta of value investing: Revenue growth, margin expansion and multiple expansion. I think this stock has got all three of them. I think the company has reached an inflection point in the past few years in terms of product and service offering with SaaS revenue of $100m and growing 50% YoY. Overall service and subscription revenue being 30% of revenue, up from 24% 3 years ago. With an overall very sticky revenue base with high switching costs.
Overview
Extreme Networks (EXTR) has been a ship of Theseus in the past 10 years. It used to be a hodgepodge of poorly coordinated networking hardware solutions selling mostly to the government. But it has since 2015 transformed itself into a very competitive and focussed enterprise LAN and WLAN networking solutions company. And is now taking market share from the likes of Cisco with a record hardware backlog of $425m as of last quarter.
It has sold off due to supply chain issues constraining revenue and earnings growth, which management expects will be mostly resolved by early calendar 2023.
The company does not try to compete on having the latest fanciest hardware with exotic features that are rarely used anyway. They seem to excel in security and ease of use. Proof of success in security is that the CIA is one of their customers. And going by some Reddit scuttlebutt here, here and here, network admins are a fan of their ease of use too:
“I’ve installed a few over the past year. The learning curve was a bit steep, mainly due to the leak of training available during the pandemic, but once it clicks it does make life much more straightforward and it just works.”
Another commenter agrees:
“ditto.
I consulted on a large CCTV project. 1500+ cameras and 10 viewing stations. The integrator wanted to use Cisco or Juniper. 7 years earlier I would have agreed with them. But I told them to check out SPBm. They were a bit hesitant but tried it out.
They finished the network config in one day. ~50 switches. They budgeted for two weeks.
The best part. Because they finished so fast they started to go live before all the fiber between the MDF, DC, and IDFs were complete. They were expecting outages when they completed the fiber connections for all the redundant paths. Once they realized they could mess around during the day without disruption it saved them more time because they didn't have come back on evenings and weeks.
the technical lead from the CCTV firm also told me that it was the first time he didn''t have to troubleshoot a Multicast issue during an install.”
Another quote from the first link:
“I changed jobs to primarily deploy Extreme Fabric Connect as a pro services engineer. If that doesn't tell you how much I enjoy working with the product I don't know what will.
I think that for all but the largest environments it's a wonderful fit and seriously eliminates so many headaches that come with traditional networking.
The reason I think it's better than what other vendors are pushing is because it is "implicitly" automated. Meaning that it doesn't rely on Ansible or puppet or chef or whatever the kids are using these days to auto provision and on board new switches. The switches themselves provide the tools to automate the actual network connectivity.
For example, I can unbox a brand new VOSS 5520, plug it into my network (on any VOSS switch port, anywhere in the fabric) and it will learn the parameters needed to join the fabric through LLDP signalling and be up and running. You could then go further and use a radius server to do edge port assignments automatically too.
I could go on and on, but to be honest I really enjoy working with the technology and showing others how great it is to work with. If you have any specific questions or want help understanding more, feel free to reach out!”
The second Reddit link is less ecstatic, but overall still positive (they are a competitor’s subreddit though). The third link also confirms that Extreme scores high in ease of use. And since networking engineers probably have a significant say in the matter of which networking solution is chosen for a company or government, this should bode well for EXTR taking more market share.
Furthermore EXTR is a front runner in WiFi6 technology which significantly increases the amount of traffic that a network can process at once.
Profitability inflection point
There are going to be three drivers for margin expansion going forward. Increased scale, a simplification of their hardware platform and high margin software and software and services being a higher % of revenue. CEO Meyercord summarises it well here at the September 2021 Citi Global tech conference:
“So from a gross margin perspective, we've got this in our investor deck. We say 63%, 65%. We've seen -- we were approaching that range prior to the supply chain issues and constraints. We've taken our guide down as a result of the near-term costs that we're experiencing.
But as we move into the -- I would say, into the second half of calendar '22, and really the first half, we'll see loosening as well. We'll have a -- we'll benefit from the loosening of supply chain and those expenses.
The other thing that's a big driver for us is that we've made a lot of acquisitions over the year, Adrienne, over the years. We've scaled up. We've made a lot of different acquisitions. And with that, we had a very diverse portfolio. And we've been working to rationalize and streamline that portfolio, and you mentioned the universal hardware.
So as we are now consolidating our portfolio and moving to the latest generation -- next-generation chipsets and technologies, this is where we have inherent advantages to our gross margins. So our new products, our universal hardware platforms will make up a much larger percentage of our product revenue, and with that comes higher margins. So we will naturally benefit from higher gross margins there.
The other point is that -- and we've acquired Ipanema. This is recurring, high-margin recurring revenue, 70% type gross margin revenue. And then we're going to be investing and adding more services on top. We mentioned CoPilot, other services that we'll be announcing that will come out that are software recurring services. So as we bring on high-margin software -- cloud-delivered software services through the network, we expect to see the percentage of that recurring revenue grow. That margin is higher margin revenue, and that will help us drive and hit the higher end of that gross margin target that we set.”
This new hardware platform will be a double whammy, it will allow for taking market share and also creates demand for their current customers who might be looking to refresh their network with the latest equipment (with new WiFi6 tech). All at higher margins!
It will require some patience though. I expect Q4 ‘22 (EXTR’s financial year ends in June) to be a bad quarter. Possibly Q1 ‘23 as well, although management expects Q4 to be the bottom. Then when China moves away from zero Covid probably somewhere later this year, or at least figures out a way to not let Covid nuke global supply chains, it is off to the races. Broker estimates for 2023 and 2024 are $130m and $175m in net income on a current market cap of $1.25bn.
Longer term management thinks this is attainable (from Q3 earnings release):
“We expect FY23 revenue growth of 10-15%, accelerating to a mid-teens range, and a non-GAAP gross margin of 64-66% through FY25.”
Assuming operating expenses of $600m (has been stuck at $550m despite growing topline), revenue growth of 12% and gross margins of 64% by 2025, that would imply a whopping $390m of operating income! Or about $300m of FCF.
Easy to see how a 15-20x earnings multiple is possible if they really deliver on this. a 15x multiple on $300m of FCF by 2025 would create a 50% CAGR over the next 3 years. The market does not seem to price this in at all. Historically EXTR has traded around 10-15x earnings (for the years it actually generated earnings) vs the current 9-12x multiple. Given that a portion of hardware sales is replacement sales from sticky customer relationships (a large LAN network is not exactly easy to rip out and replace), I don’t see a whole lot of risk here.
My only sticking point with this stock is the low insider ownership. Directors and officers only own 3.1%, with Meyercord the CEO owning 1.34% of that. Which he has quadrupled since 2016 through his stock option compensation. You would think that if he really believes in that 2025 target, he would not have sold $10m worth of stock in the past 2 years (vs current ownership of about $17m)?
Despite that I think it is well worth the risk at current price. I would not be surprised if this is taken out at some point over the next few years. So I am long a sizable position at an average price of $10.76 per share. As always do your own DD and I may sell at any time!