Cogent Communications Holdings, Inc. (CCOI) Earnings Call Transcript & Summary

All right. Thank you very much. I'm Sami Badri at CS. And we have Sean Wallace, the CFO of Cogent Communications. Thank you, Sean.

Good morning.

So Sean, one thing I wanted to discuss with you is one of the big things about Cogent is looking at the KPIs and then also tracking those KPIs on a quarterly basis and how those KPIs are either correlated or not correlated to your business. So one of the biggest ones that actually is a leading indicator to your corporate business actually is the key cards and office building access. Where are we in -- from like that metric perspective even after 3Q and going into 4Q.

Yes. Well, we've discussed this, and there's really 2 things that we look out -- really 3, it's key cards, and that has been steadily improving. It's a little bit different on -- in different parts of the world. If you're in the Southwest, if you're in a place like Dallas, it's pretty close to normal. And some of the southern cities, it's actually above normal, where it was sort of pre-pandemic. In a lot of other cities like New York, Boston, Philadelphia, San Francisco, LA, it is still well below where we thought it would be. And that correlates significantly with our sales in the corporate side. The other thing we look at very, very closely is subleasing and leasing activity. We had in the third and fourth quarter of last year a significant amount of retail space that came on the market. Vacancy rates, pre-pandemic were like 7%, 8% in places like New York and Boston, they're now up in the mid-teens. So we have customers who are going out of business, and we don't have a lot of activity bringing new tenants in. We're optimistic that when the world comes back, January 3, maybe it's January 11. That subleasing activity and leasing activity will go up. And when new tenants come in the building, that will hopefully be good for us. When we set out a proposal, we win 40% of the time. We know those buildings exceedingly well where we have salespeople who are now thankfully can go back into those buildings. They know who's in there. They know tenants are moving in, and they were typically all over those opportunities.

And then win rate of 40% of the time when you lose that other 60% of the time, who are you losing to? And what are the usual features in those deals?

It's a good question. It's typically a smaller capacity. So it's 100-meg or less. So it's a little paper -- a little bit pricier and don't need mission-critical high SLA, high capacity lines. It is sometimes where someone has 15 different locations. And maybe we're only 3 or 4 on Net, and we can't reach some of the others as effectively as we can. It is when some folks do an integrated package and put everything together, and some folks do security or other things, and they sort of throw that in. But we typically don't lose when sort of our really good spot, sweet spot, a 1-gigabit circuit, we're not losing very, very much.

Got it. Got it. I did want to address some of the salesperson turnover. But before I get into that, could we just kind of recap the dynamics in NetCentric? And I know you're starting to lap some pretty difficult comps. What's the latest kind of on NetCentric?

We had 4 just amazing quarters in that business. And it is a function of the fact that Dave Schaeffer and the team have done a phenomenal job of positioning us for that streaming business. What does that mean? We're in 1,300 data centers or more data centers than anybody else around the world carry neutral data centers. We are in now 50 countries. We're also in Hawaii. That footprint enables streaming companies to basically effectively say what is the best network I want to build. And because we're in such greater places, we're going to be a first choice for them. Other pieces that sort of get us there. We have over 200 people, professionals who sold on life to sell this product. They're really smart about it. They know all those. They find those types of customers, and they do a phenomenal job around that. And we not only have ubiquity in terms of footprint, we're very aggressive on price. We will not lose on price. And so, when you add all those things together, this is a very much a commodity sale that's done at the cross-connect in a data center. We have more cross-connects. We have more sales than selling it, and we have the best price.

Got it. Got it. One thing I've been trying to triangulate is you actually had a very significant salesperson turnover dynamic in 3Q of '21. At the same time, your growth rate didn't actually take that big of a hit or a big of a deceleration. What is the bridging factors that explain this, right? Like how did you -- the growth rate did decel quarter-on-quarter, suddenly, but relative to the salesperson turnover, which was actually extreme, how do you kind of like connect all those dots together?

I really look at the world in a sort of pre-pandemic post pandemic. So on March 15 of 2020, our entire sales force and indeed, our entire company sort of went home. It's very clear, and Dave has been -- our CEO has been pretty honest about this that it's just not as -- we're not as effective. It's a telephonic sales force. We're just not as effective in managing and mentoring and nurturing and bringing along salespeople in that environment. As a sales manager, you're going to have 20 or 30 salespeople in an office. We have them in D.C. where I work. And they can go along and they can stick their code and they can hear their salesman, how they're doing, understanding how they are [Technical Difficulty] not finding a challenge, hiring people. We're optimistic that we're now in that in situ office situation that we'll do a better job in monitoring, nurturing and mentoring these folks, and we hope productivity comes back, but it's been a very difficult environment because of that.

And then this is going to be kind of one of my more fun questions I ask people. Assuming you have more kind of turbulence with the salesperson in Flywheel. Could you still grow 4% or would growth start to deteriorate if you did not backfill those seats for sales?

Mitch McKenna is a guy who runs our training and sales force stuff. And I like -- I call them up and I say, okay, Mitch, from -- Mark has called up every week, say, how are we doing? This is not a problem to hire people. We can get people. We sometimes have competitive comp and everything else. We're optimistic that we can now train people in-house as opposed to do it virtually. We're happy that they're in-house. And I think, candidly, if you look at our numbers, and this is on our presentation, which is on our website, you can see our productivity went down the last 6 quarters. We're optimistic that in a situ environment, we are going to be much more productive. And hopefully, we'll get back there. It's -- so your question is, can we grow 4% and...

Even with no backfilling of like productive salespeople?

The answer is yes, because I think it's going to be a much easier environment. Remember, we have close to 600 people in our sales and marketing organization who are already there. We continue to hire people and everything else. So that's at the margin. But it's been a very difficult environment in the corporate side. You can't go into the offices. You can't find people. You have clients leaving. We now, hopefully, are going to see new tenants who are subleasing space who are taking lease space. And I think when corporates get done with getting people back to the office. They're going to look at their configuration, say, you know what we haven't upgraded in 18 months. We have a lot of people who are now working more disparately. We need to upgrade the systems, and we'll be hopefully going to benefit from that.

Got it. Got it. Well, I think what's like very impressive is with the amount of turnover you guys saw and 4% growth, I think cable companies would die to have that kind of mechanism in place. So maybe shifting gears a little bit more and going to 2022 and 2023, what are the -- other than just office key cards and tracking and volumes of people going into the actual offices, what other metrics should we be tracking to detect a pickup back into your business?

Vacancy rates. So you should be looking at that. Colliers has pretty good data on that by city. You should probably focus on more northern cities and California cities. Those are the area -- where we have more -- I mean, we are, in many ways, I don't like to say this, but we are highly correlated to the return to office because we're in these cities, most of our -- over 60% of our buildings are in cities where they've been sort of hurt the most, Chicago, New York, LA, San Francisco. If those cities return, we are optimistic for those 2 big pieces. One, you should look at subleasing and leasing activity. It's all of the brokers deliver that. And you can look at that on a quarterly basis. It is coming back. And you could look at the net absorption of space. If the net absorption space begins to get negative again, which is a good thing, we should be performing well. And then just anecdotally, our companies forcing their employees come back to work.

Got it. Got it. I want to shift over to margins because the margin story is equally as impressive as the durability of growth with excessive sales [indiscernible].

Yes.

So I guess, like what I asked, Dave, is the scenario analysis. You grow 4%; margins expand 50 bps. If you grow 6%, they grow 150 basis points. What if they grow 10%, which is actually the long-term guide of the company? What is the margin expansion or contribution margin profile?

Dave talked about this. And that's really -- I'd like to say that's a sort of -- that's what we hope to achieve. It's not a projection in any way, shape or form. And Dave is very consistent and saying we're going to hit those numbers. We'll see. It's just math. So when we sell a product that's on net, 90% to 92% of every dollar falls to the bottom line. So as we grow that base of customers, we're going to continue to grow that margin. I'm not going to give projections on sort of how that will perform. What's really interesting to me is in a pretty anemic environment where corporate is basically -- we've been losing like $1 million to $2 million of revenue per quarter. We've obviously had real benefits of real dynamic growth in the NetCentric side is that we've been able to continue to get those margins up. We were up 60 basis points year-over-year on the EBITDA margin. And I guess, I got to tell you, I've been at Cogent 1.5 years. We have a very smart, disciplined management team that is constantly looking for systematic ways to improve our cost base. And I just -- last summer, we implemented this thing called Compass. It's a CRM system. We had a software provider that we're spending $300,000, $400,000 a year on that. We now are probably our out-of-pocket cost by $100,000. So there's $300,000. In December of this year, we hope to finish a 1-megawatt solar array on our facility out in California and Pasadena. We'll probably -- if -- depending on electricity rates and everything else in usage, we'll probably save another $300,000 a month in SG&A. We went and found a -- we looked at every loop circuit that we have. We found out that we were -- sometimes we'd have customers who would turn off, and we wouldn't turn off those circuits. We're still paying for them. So we created the process, and we found another couple of hundred thousand dollars in expenses. This is just a really well-run company that systematically looks for ways of improving the bottom line. Combined that with 90% on net contribution margin, we will get those margins up over time.

Got it. Got it. I think historically, I recall the entire network running on one specific large-cap networking vendors' equipment.

Yes, sir.

But have you guys upgraded that, have you refreshed the network, or are you guys still running on the former -- I mean, not that there's anything wrong with it. It's just maybe more of a question of [indiscernible] rates?

Pretty [indiscernible] story. So the network is this big living breathing thing of 9,000 nodes and the 2 big pieces of equipment we have are routers and DWM. DWM shoots delight on the fiber. The router basically takes in the packets when you type in something and looks at a table and then sends it on its way to the next. What we typically have done is when we have -- we get to capacity in central parts of the network, we take that equipment. We send it back to a facility in Northern Virginia. It gets cleaned up and they get sent back shrink wrapped to an edgier part of the network. That big company today said, look, we're not going to support that equipment anymore. You've had it too long. So our CapEx have come up a little bit because we've preordered stuff, and we've had to go and take stuff out that, candidly, we couldn't -- it couldn't be supported anymore.

And the Cogent network is already superior than most networks, right? Because it's one layer, one operating system, fully distributed. So it's not like you guys need to upgrade?

Here is a great question I asked Dave said, why did you choose IP Ethernet? And it was very simple. You could have done time division multiplexing around 2,000. There were thousands of networks out there. There were millions of lands. And so Dave felt that IP and Ethernet was so far and so superior in terms of its cost base because of the installed base that every network would want to be an IP Ethernet network and things. It's just our cost -- we are so much further down the cost curve in terms of manufacturing and R&D versus those stuff. And indeed, a lot of those protocols that you're mentioning of other companies, which they can't get rid of. It's just too difficult and too costly. The cost of those networks is actually increasing on a per bit basis.

I have an off-script question for you. These can be a little bit scary. So one thing that people don't do is associated Cogent with this concept of the metaverse. But if that is a killer application, which it looks like it's turning into one, what would be the implications to a Cogent?

Of the metaverse?

Yes. So metaverse web traffic.

I don't know enough about that metaverse.

Actually. So the bigger question is, Dave always talks about the next big killer app. And that's going to be a significant step function up in a web traffic or just overall processing of bits. In the event that there is this big step-up function of traffic, Cogent would be one of the key beneficiaries of something like this.

Correct.

Right. Right. Yes. So you never know what the off-script question.

I mean that's an off-script question. I appreciate. I mean, the things that we -- if you look back 30 years on the Internet -- sorry, grown 35% a year, it's amazing. Things that sort of are candidly are issues of content distribution, it's clear that some of the traffic that would have been on IP transit has moved there. Those modules are moving faster. But at the end of the day, the Internet just continues to innovate and grow. And we were looking at -- I was looking at my daughter works at another investment bank. And I was just looking at -- she's an analyst, I was an analyst 25 years ago, and I was looking at what she does on a daily basis and the amount of data that is just there. And now we're just looking at their presentations. Our presentations were, for the most part, black and white. You might have 1 or 2 things. It is amazing the amount of data on every page and the amount of analytics and other things that are required to produce that page versus where it was 25 years ago. And that's across. George?

Touching back on corporate -- I think most investors are pretty comfortable in NetCentric. How -- it's early days. What have you heard customers on the new variant? Is that affecting office plans, or is it too early to say?

You can -- I think what we're -- what we typically see with these variants that they become much more contagious as this one is and much less lethal. And what the real estate people, and these are folks who are talking to companies all the time about what they're doing with everything else. What -- the question is no longer about COVID anymore. It's about their lifestyle. People have become used to working remotely. They don't like -- they're happy that they don't have to commute. People who are -- have long commutes are thrilled to do it 2 days or 3 days a week as opposed to that. So I don't know -- I think it's too early to tell. Candidly, I think people are -- I think everybody is tired of COVID. It's unclear how lethal this really is. It looks like it's not. And if it is, more people will get it, particularly young people who will probably have less of a chance of having mortality. So I think this is just a bump in the road. I don't think it's the delta variant.

I just want -- but I'm not an expert. So yes. Maybe just one follow-up. Can you reiterate how the corporate segment can benefit from these smaller footprints and offices?

So there's a law firm in Texas, a buddy mine is a senior partner. They have 3 floors in a building. They're going to go to 2 floors. And I said, what you can to do with the other floor? It says we're going to sublease it. We have -- we're -- we have 10,000 square feet in that foot plate, and we're going to lease it to somebody else. So we average about 51 tenants per building. If you talk to JLL or CBRE, they think that the footprint, the core footprint in Class A buildings is going to shrink maybe 10% or 15%. Do we go from 51 to 60 tenants per building? I don't know. If that happens, that's a good thing for us. But we're not counting on that. We haven't changed our models or we wouldn't change guidance on how we're going to grow.

I want to split the regions and different growth rates. Europe, I guess, has moved a little bit slower in terms of reaction, reopening, things going back to normal. Part of that is a supply chain issue. Part of that the U.S. just kind of moves faster sometimes. How is Europe progressing relative to the U.S.?

So we don't sell any corporate. We sell it in North America, it's Canada, a couple of cities in Canada and the US. So our customers in Europe are primarily to folks that are NetCentric. So streaming companies, ASPs, all these folks who are pushing out content and then, we have phone companies, cable companies, ISPs. So the impact of people going back to office, it really has not changed and everything else. And as you look at our revenues, revenues are accounted where contracts are signed. And so, we might have a global customer who might be having all of its network out in the rest of the world, like HBO Max uses AT&T. No surprise there in the United States, but uses other networks outside of the United States. So it's a little bit misleading. And obviously, there was no corporate business in Europe. So we don't have that impact of back to office.

Got it. I wanted to go on to one of your newer businesses that you're trying to start up, which is the peering business? How is that progressing? What is the latest? And maybe if it's not progressing as planned, what has come up as maybe some of the things that you have to focus in on?

It's needless to say we're disappointed. We've had conversations about this. We thought that, that business would be further along. It's a virtual business. So we've got to get folks on that network, using it and making it part of their toolbox. It's likely that we're going to have to get a large player to join that and be part of that. We're in discussions with some. Their pricing expectations are probably not where Dave would want them to be. So we've -- we started that last fall, so almost about a year ago, I would say, on a scale of 1 to 10 in terms of performance and where we are versus expectations, it's a 1 or 2.

Okay.

We're not shutting it down. We're going to continue to try to grind through, and we are optimistic that over time, it will just -- it will cascade and get there. And then the incremental cost of just powering that up almost 0. It was Dave's idea in July of last year. It was out in October. We had to create a new product code. We had to train some of the salespeople. There were some software, but 0 cost.

Got it. And then for building that business up, is it going to require a lot of salespeople like salesperson investment?

It's -- those 200 salespeople doing NetCentric are probably the folks [indiscernible] most of it. Candidly, I think it's going to require and there's debate internally about what we need to do. We're literally going to have to probably give some big customer, either you discount or free the time to get them to go on. And once they go on, the flywheel will sort of turn.

One other question I forgot to ask you earlier, this is about the corporate customer base. A lot of corporate customers actually upgraded their Internet speeds in the beginning of COVID, right, from 100 meg to 1-gig, some even higher. Have you seen any reversions to slower speeds or have they -- for the most part, once they upgrade, they're there and they stay?

It's...

Actually, it's a complicated question because even I remember when Dave did this, there were -- I think it was contingencies or features in the contracts of upgrading. So the price was compelling, but the duration of the contract was longer and locked. So maybe you could just talk about, like, have there been any changes to those contracts or negotiations?

The March sales bonanza, March, April, May of last year was a blip in a very long trend of -- in 2015, I've gone back and looked at all our products, I look at the entire general ledger. And there are about 8 products, which are the most important products in the company, they're about 80% of the revenue. And you can see -- I mean, first of all, people thought Dave Schaeffer was out of his mind for selling 100 megabits around the turn of the century. It is now that's the buggy web. And so I would almost look at it as an MPLS product that has been around for a very long time and it's declining. And it is -- it was declining in 2015. It was declining in 2018, 2019. In 2020, that decline accelerated due to COVID. Folks going out of business as well as sales to get there. That product base is declining around 20%, 25% in terms of count every year. And so, it's now getting into that asymptotic range. It used to be our largest product. It's now our fourth biggest product. And it will continue to decline. And Gigi, which was, at the time, like our third or fourth biggest product, is now our biggest product. What's really interesting to us is that it's more difficult for a lot of our competitors to provide 1-gig symmetric service. They like to sort of say, you're going to get 100-meg, but you won't get 100-meg all the time. This is another leap and another more pressure on their network. And so, we're optimistic that we can go to customers and say, look, you don't want to have any time where you don't have blockages. We will give you 1-gig for basically just a little premium over 100-meg that you get from somebody else. It's a compelling sales opportunity. And I think when we see more tenants, when we see folks restructuring their network, we're going to really benefit from that. And that is -- and FE will go down and Gigi will continue to go up. And I think I've said this to you, we're sort of -- in 2020, we were like a duck on the water. Corporate was sort of going flat, but there was a lot going on. And a lot of that transition from FE to Gigi has already been done.

Got it. Got it. One thing I'd like to ask people that are maybe with investors is, what do you think is the most overlooked thing about the company at this point? When investors ask you questions, they might ask you specifics, things that may seem borderline, non-relevant?

You do really good numbers. Maybe George does more of the numbers than you -- sorry, George.

We'll give Georgia the credit.

Will give George the credit. And if you -- the big issue I ask is an arbitrage issue. So we have senior secured notes. Our -- the other side of your house helped us raise $500 million [ or ] the 3.5%. So that's trading at 3.25%. And our stock, which is junior in the capital structure, let's call it, 4.75%. So why is that happening? Basically.

4.75% dividend yield.

Dividend yield.

Yes. So unique arbitrage.

So what is the market saying? And the market is saying that it doesn't believe that Cogent can continue to grow that dividend. And it will slow down. And once it goes ex growth, there are other telephone stocks. I'm not going to pick on, but they're yielding 8%, 10% that have similar dynamics. Because of the things we were talking about earlier, which is the margin, the 90% margin. It doesn't take much for this company to continue to grow EBITDA, EBITDA margins and pay those dividends. And I think if you look at the numbers of what you really need to achieve, it's not a material growth rate for this company to continue to grow. So I sort of look at those 2 numbers. And the market has really made a decision that, let's call it, we're a 5% yield going to 8%. When in reality, it doesn't take much for us to get to that yet.

You brought like a very good, not covered idea. What's even more interesting is maybe if we went back 2 or 3 quarters, the trajectory of revenue growth was higher, so free cash flow was higher, which means that the leverage was actually stabilized or locked, right? Did this arbitrage exist back then, or is it because you guys renewed the debt and therefore, now the arbitrage exists?

It sort of did. But now we had debt that was coming close. It was -- it's yield to call paper. So it really doesn't -- didn't have that. But this is a 5-year instrument, not 4.5 years that's yielding a little over 3%, and we're yielding at 4.75%. I'd also say, and this is public data. If you go back and look at Dave's purchases of stock from 2015 on, 2 standard deviations are 25 basis points around that 4.75%. So you have this interesting dynamic where if you run the numbers and you guys have and you say Cogent grows 4% to 6% a year, which is not a big leap. It will easily be able to pay that dividend. You also have a CEO who has historically bought the stock sort of at that level. And you have got to -- you have an interesting arbitrage between what the bonds trade out and what the stock trades up.

Interesting. Well, Sean, I think that was -- I was actually my most interesting insight that an executive mentioned about like sometime people overlooking. But yes, thank you very much for pointing that out. And thank you for spending time with us.

Great. Thrilled to be here. Thank you very much for the conference. It's very exciting to be here, and good luck. Congratulations on your [indiscernible].

Thank you. Thank you.

It's awesome.