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

All right. Well, welcome to everyone for joining us for this next session of Communacopia. This is an audio session. So if you're trying to find a video, you can stop looking and just listen. I am Brett Feldman, the firm's U.S. telecom analyst, and I'm happy to have joining me on the line right now, Dave Schaeffer, the Chairman and CEO of Cogent Communications; as well as Sean Wallace, the Chief Financial Officer of Cogent. Dave, Sean, thanks for joining us.

Brett, thank you for hosting us. I'd like to thank all the investors for their time today and thank Goldman Sachs for a great forum to communicate.

Well, we're happy you're here. And more importantly, we're looking forward to hopefully having you here in person next year.

But I want to start off by talking about COVID. You've stated that you would expect that Cogent could be a net beneficiary of stay home -- of the stay-at-home trend that's been brought on by Cogent. How would you describe the current backdrop for the services you offer? What are the key tailwinds that have been brought on by this pandemic? And also, what are some of the headwinds that the business is working through as well?

Yes. So there are both headwinds and tailwinds. I do think in aggregate, we probably do have more tailwinds than headwinds. So let's talk about each segment of our business separately. Our corporate segment, which represents approximately 70% of our revenues benefited positively by seeing our customers needing larger connections. So in the roughly 960 million square feet of office space, 1,800 buildings that we are connected to, Cogent has approximately 25% of the businesses buying services today, 75% of the businesses in that footprint buy their Internet access from another provider. For us to win business, there usually has to be a catalyst. Those catalysts could be a customer moving, a current service provider failing or a change in IT infrastructure. There's now a fourth catalyst, which is the COVID catalyst. It's really customers needing a bigger connection to support their work-from-home employees. So when an employee works from home in most companies, they are building an ad hoc VPN into the company's firewall. All that traffic goes in and out of that firewall. And those firewalls need to make sure they have enough bandwidth. Cogent's traditional product has been a 100-megabit-per-second non-oversubscribed or non-bought corporate connection. We've seen a significant acceleration of new customers buying 1 gigabit connections at higher ARPUs and also existing customers upgrading and paying approximately $200 a month upcharge to increase their bandwidth tenfold to support work-from-home. So that has been a positive. The negatives to our corporate business are branch offices. So Cogent traditionally sells to its customer at that initial site. And from that relationship, may sell additional sites, both on-net and off-net. And in those additional sites, we will be selling both dedicated Internet access and a permanent VPN solution over-the-top using either SD-WAN or VPLS. We have seen a dramatic deceleration in sales to secondary locations. Many of those branch offices are shuttered. They don't have the same firewall requirements that the main office has, and that has resulted in a deceleration in our corporate sales, mostly in the off-net portion of the business. So Cogent's corporate growth rate has been about 11.5% per year for the past 15 years. Our corporate on-net growth rate decelerated in the last quarter to 9%. And that's still against a backdrop where GDP declined at an annual rate of 32%. So we think we're doing quite well, but we saw a slight slowdown, where the bigger shortfall was, is in our off-net business for those branch offices, where we actually saw a year-over-year growth go negative, resulting in a total corporate growth rate of about 5.1% on a year-over-year basis, which is a deceleration from the long-term average of 11.5%. In our NetCentric business, we actually saw tailwinds that accelerated that business. Our NetCentric business is selling bulk Internet connectivity in one of nearly 1,300 carrier-neutral data centers that we operate in, in 47 countries around the world. The primary application that drives unit volume growth is over-the-top video or streaming. We today connect to about 7,100 access networks that buy their upstream connectivity from us and about 5,000 content publishers or aggregators. That business is a metered business, so as volumes increase, revenue increases. Cogent has historically grown twice as fast as the market because of our aggressive pricing strategy. That has been true since the pandemic hit. We saw our sequential traffic growth rate accelerate and our year-over-year traffic growth accelerate. As a result, our NetCentric revenues turned positive on a reported basis year-over-year at about 3.5%. And when adjusting for constant currency, as roughly half of that business is outside of the U.S., we actually were up about 5.1%. So virtually identical to that of our corporate business, resulting in our entire growth rate being 5.1%. Because of the mix shift to a greater percentage of on-net, we actually beat substantially on margin. Our EBITDA margins expanded on a year-over-year basis at 290 basis points, roughly 80 of those basis points came from an accounting change. But even at a lower growth rate and netting out that accounting change for the maintenance on an IRU that needed to be capitalized, we still exceeded our long-term average EBITDA margin expansion even on lower growth rates. So net-net, when I look at the -- both puts and takes, I think COVID has been a positive for our business.

So if we just stick with corporate there, and you gave some great color about how that segment has been impacted. Are you continuing to see the increased need for connectivity into your customers' core locations? And I'm just going to extend that question. Have you found that other tenants in buildings, where you're under-penetrated, are calling you to take advantage of your pricing as they need to amplify their bandwidth needs into those core locations? And then how do we think about the impact of these branch offices going forward? Is it we had a run rate where it's not a headwind anymore, it just is what it is? Or could it be more of a decelerator as we move ahead?

So 2, 3 different questions, Brett. First of all, this does become a fourth catalyst for decision making. I think we saw a period of almost panic buying from mid-March through early May. We then saw a period of uncertainty set in as customers were apprehensive about what the future would bring and were not ready to make a new buy decision. I think that apprehension has subsided at this point. I think most customers understand that we're in a protracted period of work-from-home, and therefore, they need to make decisions to be able to support their remote employees. And that has resulted in both better uptake from existing customers upgrading those that did not originally upgrade; and two, it's another reason why our core locations where we're selling to corporate customers, our penetrations are improving, and we're continuing to win some of those other tenants who have historically not chosen Cogent, and we expect our penetration rates to continue to improve. We also know that our buildings are less negatively impacted due to the recession that has been caused as a result of the pandemic. We do not have significant exposure to hospitality, to restaurants or small retail. While there is some exposure on the first 4 of these buildings, the vast majority of our addressable market and our customers are businesses with much better credit rating. And as a result, our bad debt remains very low, below 1% of revenues. And our DSOs actually came in last quarter, even though a lot of businesses, smaller businesses were struggling. For the branch offices, I think the jury is out. I don't think that they're going to return immediately, but I do expect over the next year, as a vaccine becomes available, as therapeutics improve, that people will return to the office, both in the central business districts and these branch offices. I think the office will look different. It probably will not be staffed at the same level every day. There'll be a hybrid model for some period of time. And I also think that most businesses will revert back to a more traditional office layout as opposed to an open floor plan, which will actually mean a higher amount of square footage per capita per employee. While there will be some continued work-from-home, there will also be many employees anxious to return to the office. So I think for the next few quarters, until there is a vaccine, we will probably see the secondary location sales be slower than your historic average. Over time, on a unit basis, we typically sell about the same growth rate to on-net and off-net customers, meaning branch office and primary location with off-net typically having a higher ARPU and lower margin. But for the next few quarters, we will be at a lower branch office growth rate.

Okay. Got it. So I want to ask about one more thing. In the past, you've noted that one of the key ways you're able to increase your penetration in your buildings is when there's moves. So a tenant is moving in. They have to make a decision. That's when you can kind of get them and make them realize the value proposition. I would imagine that moves are lower than they've been. How significant is that going to be as we think about bookings opportunities over the remainder of this pandemic period?

Yes. So traditionally, in Class A office space, tenant turnover is about 7% per year. There is not good third-party data available for post-pandemic lease activity. I can speak from personal experience as a commercial landlord, and I operate mostly in Class B buildings, and I have about 450 tenants. I'm actually continuing to sign new leases at about the same pace I did pre-pandemic. Now I've been a little more exposed to poor tenant credit quality. So I don't -- I'm probably not a perfect barometer for the Class A space, but I do think businesses are looking past the pandemic. Typically, when you move your office, it is a multiyear process to plan, to build out and then relocate. And I know, for example, there's been a couple of new leases announced at Hudson Yards, which is a new development in New York, just in the past couple of weeks. Now that's anecdotal, but I do think the turnover rate in our Class A buildings will remain about the same.

Prior to the pandemic, you had been growing the corporate sales force pretty rapidly actually. And you had noted that it does take a little bit of time for your sales reps to be up to full run rate from a booking standpoint. How do we think about the way you're managing the size and training of your sales force now, knowing that the market opportunity is somewhat abnormal and hopefully temporarily abnormal?

So we continue to hire throughout the pandemic. First, we had to pivot almost instantaneously in mid-March to a work-from-home as opposed to work-from-office environment. Virtually 100% of our sales activity is done over the phone, or via email. We do have 38 physical offices around the world that house 68 discrete sales teams. All of those individuals in mid-March went to work from home. We quickly modified our management systems to be able to monitor call activity, CRM activity and e-mail activity for that remote workforce. And we did a pretty good job. The second thing is our outbound telesales model has only been fairly prone to high rep turnover. In fact, in pre-COVID times, we were churning 5.7% of the sales force a month. It is a very hard job. We had to quickly adapt to be able to hire remotely. And that sounds easy, but it's actually pretty difficult because you have to get an individual, you have to drop-ship them a laptop, a cellphone, headset to be able to set up their home office. And then you need to walk them through that setup before they can even log into the corporate network. We developed that remote training materials for the initial on-boarding. And then secondly, we modified our training program to support a complete remote training experience as opposed to a hybrid one of in place and remote online learning. All of that went fairly well. And I think we did a good job plus we saw an unprecedented number of applicants as unemployment rates spiked and as our competitors laid-off individuals. Where we probably failed in the second quarter, and I think we're doing a better job now, is remote terminations, that is managing out those underperformers more quickly. It's easy to do when a manager is physically with someone and can monitor them. It's a little harder to do remotely, and we've added some additional tools to help us do that. We feel that we still do not have enough reps to call on our addressable market. And we do plan to continue to grow our sales force through the pandemic at the same pace that we expected prior to the pandemic, which is a 7% to 10% net growth in the sales force on a year-over-year basis. We're at the high end of that at the end of Q2 as compared to last year. We also saw our average rep tenure decline slightly as our hiring had accelerated at the end of 2019. I think we'll revert back to a more normalized rate of hiring, and therefore, our rep tenure will end up reverting back to normal. We think that our rep productivity will continue to improve and we will probably over the next several quarters return to our long-term average of 5.1 installed orders per rep per month. We are below that at this point.

Thanks for that color. I want to pivot and talk a little bit about the NetCentric business. As you noted, NetCentric data traffic accelerated in the second quarter, increased to 49% growth year-on-year. It had been a 36% year-on-year growth in the first quarter. But I think that you had actually pointed out that you were exiting 1Q, you were even above 49%, which would imply that at the front end of lockdowns, there was actually a very significant spike in the NetCentric traffic that you were handling. So I was hoping you can maybe provide a little bit of context around that. But really, the question everyone has is, what have we settled into? How fast is traffic growing now? And then maybe we can get into some of the factors that are behind that?

So let's talk about the key driver of traffic growth, which is streaming video. We have been a beneficiary of that trend. We have seen a broadening in the number of streaming service providers. We provide bandwidth to virtually every streaming service globally. We also are the primary upstream for most regional access networks from the very largest like China Telecom to some very small rural cable operators. So we benefit on both sides of that streaming equation. We saw a few things happen: one, more choice, which meant people were going to choose new services; two, we saw an acceleration in cord cutting and migration away from linear, so they had more minutes to stream; and third, people were home more, and we saw the period of peak utilization, which had historically been between 7 p.m. and 10 p.m. in any times in the developed world to broaden out, and we've seen that peak window now stretch from 3:00 in the afternoon all the way to midnight. I think that's a result of people being able to work-from-home and have some additional flexibility in their schedules. Now our growth rate did accelerate at an accelerating rate for about 6 weeks. We have now reverted to a more normalized growth rate. We continue to see traffic grow. We also know that in the summer months, where people spend more time outside for the past 20 years since Cogent has been in business, we have seen a slowdown in sequential growth in the summer months. This year was no different. May have even been a little more pronounced as people were probably tired of being cooped up in their houses and shelter-in-place and wanted to enjoy some fresh air. But we have seen a resurgence in traffic growth starting in mid-August, which is a traditional trend. It's usually driven as students go back to school. We've seen that pick up. Even though many school systems are not physically open, the students are now online and more automated behavior patterns are in place. So we expect to continue to grow traffic at about twice the rate of the market, and we also expect our NetCentric revenue growth to continue to improve, as it did last quarter.

And just to sort of get into the basis of that improvement. Is that specifically because of your outlook for traffic? Because in the past, you've also talked about the importance of pricing and how average pricing trends typically improve as the base of where your traffic come from -- comes from is broadened.