Edgio Inc (EGIO) 2020 Q3 法說會逐字稿

Good afternoon, and welcome to the Limelight Networks Third Quarter 2020 Earnings Conference Call. (Operator Instructions) Please note, this event is being recorded.

I would now like to turn the conference over to Dan Boncel, Chief Financial Officer. Please go ahead.

Good afternoon, and thank you for joining the Limelight Networks Third Quarter 2020 Financial Results Conference Call. This conference call is being recorded on October 22, 2020, and will be archived on our website for approximately 10 days.

Let me start by quickly covering the safe harbor. We would like to remind everyone that we will be making forward-looking statements on this call. Forward-looking statements are all statements that are not strictly statements of historical fact, such as our outlook for 2020 and beyond, our priorities, our expectations, our operational plans, business strategies, secular trends, product and feature functionality announcements and the impact of COVID-19.

Actual results could differ materially from those contemplated by our forward-looking statements, and reported results should not be considered as an indication of future performance. For more information, please refer to the risk factors discussed in our periodic filings, including our most recent annual report on Form 10-K.

The forward-looking statements on this call are based on information available to us as of today's date, and we disclaim any obligation to update any forward-looking statements, except as required by law.

Joining me on the call today is Bob Lento, our Chief Executive Officer; and Sajid Malhotra, our Chief Strategy Officer. We will be available during the Q&A session at the end of prepared remarks.

I would now like to turn the call over to Bob Lento.

Thanks, Dan, and good afternoon. Q3 was another very strong quarter for us as we grew top line revenue by 15%, announced several new edge deals and made significant progress across our strategic imperatives. Revenue in the third quarter was $59.2 million, up 15% year-over-year. It marked our highest third quarter ever and second highest in company history. We also successfully executed $125 million convertible debt offering, which further strengthens our already strong balance sheet and allows for flexibility in accelerating our growth initiatives.

There is a digital transformation underway, and we've seen acceleration of this transformation in 2020 from COVID-19 and the rapid adoption of the direct-to-consumer content model. The increase of a remote workforce on a global scale and the dramatic increase in content consumption have created sustained demand, along with new and exciting use cases for low latency delivery and edge services. Limelight is well positioned in these dynamic times to provide the scale and services our customers now demand. Our investments in extending our global capacity, deepening our video capabilities and expanding our suite of edge services are some of the most significant accomplishments in 2020 that have not only driven our growth but also allowed us to achieve differentiation.

As a specific example, in India, this year alone, we have doubled our capacity to help meet the growing demand in the region. Online education companies are scaling largely driven by COVID-19. 3 of the top 5 leading providers in online education in India are our customers, including BYJU, a company that was recently valued in excess of $10 billion.

I'll talk in a moment about specific progress in the quarter, but let me briefly revisit our strategy to set the stage. Limelight is doubling down on our focus to serve the video, media and gaming spaces. We have a robust base of existing customers, including 3 of the top 5 global media streaming companies. According to IDC, the public CDN space is expected to experience healthy growth through 2024 with video, media and gaming, increasing most rapidly. With our robust product offerings, global private network and unmatched customer service, coupled with our execution, we are in a strong position to benefit from this trend.

Now let's talk about a few of the specific highlights from Q3. As we've discussed on past earnings calls, our strategic imperatives in 2020 center around 4 pillars, which are: extend capacity; expand control; proactive management; and enable innovation. Each imperative is designed to impact 1 of 3 areas: revenue or performance gains; drive down cost; or most importantly, deliver improved experiences for our customers.

Last quarter, we announced the launch of our new developer community as part of our expand control imperative and developer engagement has been strong. Further, our efforts to extend global capacity remain a priority with steady progress during the quarter. Advancements in the proactive management of our global network through automated traffic engineering were significant in Q3. While this is not new to us, we deployed several new capabilities to better manage the complexities and high volume of changes we see on a daily basis. We expect to see full benefits from these advancements as we move into 2021.

Product innovation at Limelight in 2020 has delivered at a rapid pace and is driven by our strategy to deepen our video capabilities and expand our edge services offering. Last quarter, we released Live Push, which is a video service that allows customers to directly push live video content to the Limelight CDN. It simplifies delivery, protects the customer's origin servers and reduces bandwidth costs. One of our long-time customers, BBC in the U.K., is one of the first users of our Live Push service and reports that it is helping them meet their business goals.

As we all know, edge is a very hot topic in our space and one that is early in its maturity and broad market adoption. While content being pushed to multiple edge locations is not necessarily new, the innovation in how content is delivered and what you can do with it is rapidly evolving. In Q3, we were first to market in our space with the launch of EdgeFunctions, our new serverless compute capability that puts flexibility and power in the hands of our customers to manage, deploy and run their functions at our network edge. It leverages Limelight's global network and takes advantage of our direct peering connections with more than 1,000 ISPs and leading public cloud providers. With the release of EdgeFunctions, we now provide one of the most comprehensive suites of edge services available today for our target customers that ranges from flexible edge compute options like bare metal as a service to serverless compute and application capabilities. Several customers are actively implementing our EdgeFunctions and seeing quick benefits. One of our customers is leveraging EdgeFunctions to deliver optimized images for any device. Another is using EdgeFunctions to execute forensic watermarking algorithms embedded in high-value video content to detect piracy in record time.

Further, this year, we expect almost double edge revenue over last year. We secured several new edge deals this quarter, including those we formally announced with SimulTV and Katapy, a global OTT provider and online video platform company.

Our next evolution is in development now and focuses on providing tools to manage complex workflows at the edge. Additionally, we are weeks away from releasing the next-generation of our real-time streaming service. For organizations doing live streaming that requires near real-time performance, it is a solution that provides subsecond delivery worldwide. Unlike others in the market, real-time streaming is integrated with our global CDN, providing unparalleled reliability and scale. We are currently in limited availability with multiple customers moving into production across a range of industries, including live events, gaming and sports.

Organizationally, we made some changes in Q3 designed to better align with our strategy and foster additional solidarity during these challenging times. We consolidated our development and operations group into 1 DevOps organization and consolidated our edge services group into our product and marketing organization. We are still very much in the middle of a pandemic, and our employees continue to work remotely. As a result, how we work together is evolving. With these changes, we believe we will optimize team collaboration and be even faster and better focused on delivering customer value.

In closing, it was another outstanding quarter. We sit in an opportunity-filled position surrounded by multiple ways we can extend our business and add value to our customers. While all the options present opportunity, great companies know who they are, who they serve, the value they provide and translate it through unparalleled execution. Limelight has invested heavily this year to extend our network, optimize our performance and expand our capabilities through edge innovation and new product introductions.

We are well positioned to serve the video, media and gaming spaces and have tremendous optimism for the future.

Thank you. I'll now turn the call over to Dan to discuss the financial details.

Thanks, Bob, and good afternoon. As Bob mentioned, our Q3 revenue of $59.2 million is our highest reported third quarter revenue in history and second highest quarterly revenue ever. Due to seasonality, the third quarter is generally the lowest revenue quarter of the year to have grown revenue sequentially each quarter this year, further demonstrates the strength of our strategy and execution as an organization.

As noted earlier, our growth is being driven by an acceleration of direct-to-consumer streaming services as well as growth in our edge services products. International customers accounted for 35% of total revenue in Q3 compared to 36% a year ago. Approximately 10% of our third quarter revenue was in non-U.S. dollar-denominated currencies. Average revenue per customer inched higher this quarter and is approximately $110,000. This is our revenue divided by our customer base, and we believe our average revenue per customer is the highest in the industry. This is an important metric for us because compared to others our size in our industry, our financial model has been focused on fewer and larger customers. This allows us to accept a slightly lower gross margin, but against a much lower operating expense profile, resulting in higher operating margin, profitability and flow through to EBITDA.

Cash gross margin in the third quarter was 46.4%. During the second quarter, when stay-at-home mandates were more widespread, we delivered record traffic volumes. We continue to see elevated traffic levels throughout the third quarter and have added capacity on an accelerated timeline to allow us to deliver this traffic. As we add capacity in new geographies and expand capacity in existing PoPs, we see some fluctuations in gross margins. We believe selling through the continuous capacity additions as well as ongoing work to optimize traffic flows within the network should have favorable impacts to overall cash gross margin.

Operating expenses increased approximately $600,000, primarily due to onetime costs related to organizational realignments. We reported a GAAP loss of $4 million in the third quarter or $0.03 per basic share compared to a GAAP loss of $2.8 million or $0.02 per share in the year ago quarter.

This year's third quarter includes $1.7 million of interest expense or almost $0.02 per share as a result of the successful execution of our convertible debt offering in July. Approximately $800,000 was cash interest expense and $900,000 was noncash amortization of debt issuance costs and the debt discount. Excluding these adjustments, Q3 net income is in line with Street estimates.

EBITDA was $3.7 million, up a very strong 49% over the third quarter last year. Adjusted EBITDA was $5.6 million this year.

On to the balance sheet and cash flow. As briefly mentioned earlier, during the third quarter, we completed a $125 million convertible debt offering, bearing interest at 3.5%, with a conversion premium at 27.5%. With an overnight execution, we were able to successfully price the convertible offering to limit the typical downward pressure on our stock price. We also purchased a cap call for $16.4 million. This instrument increased the conversion premium to 100% or $13.38. We purchased this instrument to limit the dilutive impact of conversion over the premium based on the confidence we have in our future performance.

Operating cash flow in the third quarter was a strong $7.7 million, and we paid $7.2 million for capital expenditures, primarily building out capacity to expand our footprint. As a result, we ended the quarter with cash and cash equivalents of $124.8 million.

At the end of September, DSO returned to historical norms at approximately 57 days after expanding to 70 days at the end of last quarter. We anticipate continued DSO performance within our normal range of 50 to 60 days.

As of September 30, we had approximately 122.8 million shares outstanding. Total employee count at the end of the quarter was 620 worldwide. We are leaving our full year guidance unchanged except for GAAP and non-GAAP EPS, solely due to the impact of our convertible debt offering and the additional cash and noncash interest expense. Interest expense will have a $0.03 impact on GAAP EPS, adjusting to a $0.03 to $0.13 loss, and non-GAAP EPS is adjusted to a $0.02 loss to an $0.08 gain.

Annual cash interest expense from the July convertible debt offering will be $4.4 million, payable in February and August.

Noncash interest expense for the fourth quarter of 2020 will be approximately $1 million. However, due to an expected change in accounting rules for convertible debt offerings effective in January of 2021, we expect annual noncash interest expense thereafter to be approximately $600,000.

We recently kicked off our internal 2021 budgeting process. As we have done for the past several years, we will provide guidance for 2021 at a later date.

I will now turn the call over to the operator for questions.

(Operator Instructions)

Our first question comes from Lee Krowl with B. Riley Securities.

Wanted to just start out on the Q4 implied guidance. Back of the envelope math kind of implies at the midpoint flat year-over-year. Maybe just kind of walk through the upside and downside assumptions to Q4, just from where I sit, it seems like visibility has improved since the last time we talked. You have year-over-year contribution from edge services, some gaming launches and obviously, the acceleration of streaming video. So I want to just kind of understand how you guys think about Q4 against that backdrop?

Yes. Thanks for the question. This is Dan. So we raised our guidance in January and then again in February, when many of our people in our industry were withdrawing their guidance. And then we had 1 competitor actually bringing down guidance for this quarter. And so we felt comfortable throughout the whole year with where we are moving as a company and the volumes we are seeing in the business and where the edge business is going and so we continue to focus on the full year numbers. And this year will be our best year ever in terms of revenue growth. And we expect that to be over 15%. And so with where we are seeing the revenue -- the volume numbers coming in as we approach the end of the year. We just felt comfortable with leaving the guidance where we're at. We'll see where we go from here. The fourth quarter is generally our highest volume quarter. With all of the new releases going out and the people staying home for the holidays and just watching more of online content. And so we felt comfortable in where we left the guidance here for the year.

Got it. And then second question, just obviously, on gross margin, last quarter was kind of 1 step forward, and now we're kind of taking 2 steps back again to kind of the Q1 levels. When do you guys kind of expect to see the gains or realize the gains from the network automation processes and kind of, I guess, better utilization of the capacity you guys have brought online?

Go ahead, Dan. And then I'll make some comments on that.

Yes. On the margin, we had a really good Q2 with the lockdowns being in full effect as a result of COVID. And that just helped us on the mix that we are seeing. The margin is not an easy answer to just say, okay, here's the reason for it. There's geographic implications at play. There's product mix at play. And as we continue to add capacity throughout the quarter, we added over 8 terabits of capacity in the third quarter alone, and generally, the third quarter is seasonally the lowest volume quarter. And so there's a little bit of all of those things that play here. So that's what caused the margin change quarter-over-quarter. But we expect margin to generally trend up as these auto traffic engineering things come into play. And I'll let Bob talk a little bit about that.

Yes. So the first version, I mean, obviously, we've had automation in our network for years, but the new development work that we've spent the last couple of years on started to go into production, the end of July. And we're again, very careful and very deliberate about how we use that. And how quickly we allow it to take over more and more of the network operations. And so I would say that, that version will be completely rolled out probably by the end of this year. So we'll get the full benefit in '21. But the real improvement will come with the next version of that, which we anticipate will start rolling out late first quarter next year. So I think we'll see some improvement in terms of asset utilization and our bandwidth management starting early next year, but really accelerating through the second half.

And then to Dan's point, there are also other dependencies. What is our product mix in terms of our edge services versus content delivery, within content delivery, what parts of the world are moving more versus others because margin tends to vary. But we are very confident that on a full year basis, next year, we will be higher. As you know, we don't give guidance on gross margin, but we believe that the improvements will start to show in a very reasonable way next year as we look at year-over-year.

Got it. And then third, just a quick question. In Q3, I'm curious how many 10% customers you had, was it the same as Q2? Or was it plus or minus?

Yes. No. It will be in the Q, but there's just 1 10% customer for the quarter.

The next question is from Jeff Van Rhee with Craig-Hallum.

A few from you guys. I want to circle back to the prior question on gross margin. So the revs were up roughly 700,000 sequentially, COGS were up 3%. And I think coming into the quarter, you had thought crudely speaking, we'd be roughly flat. So just dive in there a little deeper. Was this -- when we see the Q, is this bandwidth that really drove the sequential increase in COGS or kind of -- 2 questions there. I guess, what drove the increase in the COGS? And two, what was the variance to your expectations?

Yes. So I'll let Dan answer the question, but just a reminder, when we talked a little bit about in the opening of the realignment we did in our development and operations organization. And some of those costs hit -- we created a DevOps group, that combined team. Some of those costs hit in our COGS, so there's some of that in there. And then there's structural things beyond that, that Dan can speak to.

Yes. And so yes. Again, we don't guide to gross margin quarter-over-quarter, and we actually don't guide to it for the year. But we know that's a point of emphasis, and it's what drives a significant amount of flow through, given our operating structure. And so that's what we're really looking for is continuing to drive that revenue top line number. And get that flow through down all the way through the bottom number. But again, as far as the individual components within that COGS line, there is a little bit in Colo as we continue to expand that footprint, adding 8 terabits of capacity, you need to add the rack space and the power. So there is a little bit of that. And as we put out those new locations in new geographies, sometimes it does take a little bit longer for that traffic to ramp up. And so you got -- you have those expenses on the front end. And as that traffic ramps, we expect to get the better asset utilization and for the margins to improve.

Hey, just -- Sajid here. One more thing, right? I mean, we are trying to run this business as a whole across every line on the P&L. And so yes, while gross margin wasn't perfect, I think the OpEx line was very good and drove to a really good outcome. I think the business mix and how that was driven was really good. The top line was exactly where we would like it to be. And the bottom line was exactly where we would like it to be. I mean, of course, the headline number kind of -- I read the news, somebody commented that we missed that see, but that's all below the line related to kind of the onetime convert. But overall, we were pleased with the quarter. I mean, let's there be no doubt about it. And inside the company, as we look at these results, yes, the geography of the company -- the numbers did not line up perfectly, but the overall outcome was a really good outcome.

And we think, as others in the industry will report, for us to have sequential growth in the third quarter, for us to have a third quarter that was our second best quarter ever, that goes a very long way for us, and this is a very different company. And we are now focusing on how to improve the mix and how to invest in the things of the future and how to get to a better outcome.

While working completely remotely, while putting out some of the most significant products we've ever done, while reorganizing a big part of the company.

Yes.

So while you're on that reorg, you touched on a few of the logistics of what you did, but why did you do it? Talk about the reorgan? What was the impetus?

So the edge services, I'll start with the smaller one, which was moving edge services under our new Chief Marketing Officer, Christine Cross. We had that as a separate group to provide the amount of initial focus and attention that we felt is needed. But we're really pleased with the product set that we have, both in terms of what we can sell today in development. We're pleased with what we're seeing in terms of revenue momentum. And we felt that the timing was right to integrate it into the larger organization because it's a bigger piece of what we're doing. And I don't have to worry about it getting lost and the fact that it was special was becoming more complex than there was value there now that it's a larger piece of our business. So that was the driving force there.

In development and operations, and in creating a DevOps organization, we're not the first in the industry to do that in the broader technology industry, it's -- I wouldn't say common, but it's not uncommon to do that. And what we felt was that with the amount of changes that we were making and the implications that those changes had on our operations group that a closer alignment would help us basically be better and faster. And so we have the group under the single leadership of Dan Carney, who reports to me, and that we were able to not just -- Dan's group now is in just as existing reports plus new ones, we were able to redesign the organization so that the teams had everything they needed within their own groups and then a shared services and then shared services across the team. And so we believe not only is it going to be more efficient, but we can get product developed faster and at higher quality. And we felt the timing was right for that.

I guess, last for me. Just intra-quarter, how did overall traffic volumes behave month-to-month to month?

Pretty steady. I would say.

Yes, pretty steady. And then Q2 will probably be the highest -- well, we don't know what Q4 will look like. But in the first 3 quarters, Q2 had the highest amount of traffic, but we didn't see the kind of a normal pullback in Q3 despite weather being great in a lot of places and people still taking holidays or vacations even though they may not have gotten on planes. And the other thing in Q3 was -- and this has been widely reported, a lack of new content, right? And so typically, there's a fair amount of new content that starts coming out in the fall. I think this year, it'll be less than normal, but in Q3, that was really at a low. And then there are other factors like Sony has announced its Playstation's new version of Playstation that will start shipping next month. As we met with them early in the year, they expected this to be a light year up until that announcement because of game manufacturers focusing on the new console in terms of consumers waiting to buy new stuff until the new console came out. So there's a mix of things. That are going on in this industry at all times. This year, the mix has just been a little less predictable and a little -- the swing is a little greater than what might otherwise have been without a pandemic.

The next question is from Tim Horan with Oppenheimer.

Just on that point on the volumes, are you seeing kind of a typical pickup here in the fourth quarter of volumes like we normally see with the weather getting colder?

Yes. So it's a little early. Most of that seasonal pickup starts around Thanksgiving and runs through the end of the year. And so there are -- some of our customers that are launching in new territories. And that starts in the middle of November. We know Sony, for example, is coming -- as I've already mentioned, coming out the new Playstation, but that's in the middle of November. And I think that traffic is definitely up year-over-year in October so far, but the real volume tends to come in, in the last 6 weeks of the year and then through the first week of January. So that remains to be seen. But I think that so far, the trend is looking about the same.

Yes. We have no reason to believe it will be any different. I think the games are launching, the new shows are coming on. And traffic patterns are generally healthy as whether gets warmer, people get -- weather gets colder, people get back at home and all the new shows come on board.

We're certainly moving more in live sports in the last couple of months than we did prior to that. So that's a little helpful. So -- but definitely more wildcards this year than in past years.

So it sounds like you're running the network at a lower utilization for like 3 or 4 different reasons, there's new automation tool, new geographies and correct me if I'm wrong, but I guess, how much better can utilization be next year? Can it be like -- can you operate 1,000 basis points better utilization? Just kind of some rough idea, which is what I think you're saying.

Yes. We are saying that we think we can operate it better. It's not 1,000 basis points. It's -- but we think we can get maybe several hundred basis points and that all depends on what happens on the pricing side, too, which is a big component of gross margin. But assuming that the compression on price stays within the range of historical averages, we believe that through better automation and better management with our new DevOps Group that the change can be in the several hundred basis point range.

So Tim, just -- let me try and help you understand that because I know you've talked about utilization a few times. So the delta based on our assessment of us and what we think best-in-class is not 1,000 basis points, right? So there may be a couple of hundred, more than a few hundred basis points of difference between what we think is best-in-class and us, and we will continue to move up. But even small movements in utilization have huge impact to the bottom line results to gross margin to flow through, et cetera, et cetera. So I'm with you, utilization is a big driver of the business and business results. But I also just want to be sure that you understand that there isn't 1,000 basis points to go get.

No. I understand.

The only other comment I'd make is like, we weren't sort of blind to this. Many of these initiatives started over a year ago, right? So we identified this as an area of opportunity for us. As you can imagine, it's fairly complex to create the software, and then we want to be very careful that we don't negatively impact our customers in any way by creating short-term -- we can't afford to create short-term pain but long-term gain. Our performance has to stay at the highest level. And so we're more willing to suffer a slower increase in margin than we are willing to tolerate lower performance for our customers.

But I mean, this quarter, you came in probably 100 basis points below trend line utilization given the reasons you described, and you might be a couple of hundred basis points above next year. Is that fairly accurate?

It's in the range of outcomes.

Okay. I think -- I'm sorry, last -- last, I'm just getting asked a lot of these questions. I guess the main concern on the gross margin, I know you kind of brought it up as a price competition. Did you see pricing out of the ordinary? That's when everyone sees the gross margins decline so much. It's always the main worry, right?

Yes. I would say, overall, when we look at -- because we are negotiating contracts every month. When we look at the entire year, I'd say it's fairly in the range of what we've seen. But one thing that you know is that Limelight, our top 20 customers are a fairly large portion of our total and so while on average, it can be in line, if we get, like we had a few years ago, 1 large customer that renegotiates and resets pricing from one quarter to next. That can have a short-term impact that it took us, what, 2 or 3 quarters to get back up to normal. So we're not immune to something like that happening. And there's always an example I can point to where I think the price that our competitor is given a customer is irrational. But if we look at the average, I think it's within historical averages range. But from quarter-to-quarter, 1 large customer does a complete renegotiation that resets for the next year, it may take 1 or 2 quarters for our margins to stabilize back to where they were prepricing.

Yes. And it's our job to go back to our vendors and renegotiate that pricing we get from them based on that increased volume that our customers are seeing will get from them as a result of these negotiations. So...

And so did that happen this quarter?

Happens every quarter.

Every quarter. I mean, there was -- in the last quarter, though we're not blaming the margin " issue " on that, it was more -- we're bulking up on capacity. We had some realignment costs. And we don't have the automation fully in there. I'm just saying that it can be, given the size of the customers we deal with and the impact any one of them can have on short-term results, that, that is another area that can have a short-term effect on us. I'm not blaming anything like the last quarter on any 1 customer.

And Tim, this is part of the reason why we stay focused on full year results. So for us, the big achievement is that we are on track to have our best year ever, right? Top line, bottom line, et cetera, on lower CapEx than we did last year, right? Delivering better outcomes, so better free cash flow and better utilization of assets. Now that said, within the year and within the quarters, you have times when customers are repricing. You have times when vendors are getting repriced. You have times when efficiency metrics from our R&D efforts are going into the network. You have times when they may be breakage in the network on some competition. And we see spikes in traffic. You have incidents that are occurring. You have live traffic that comes on, you have onetime events that get loaded on and loaded off. So there are so many things to manage across. I mean this is -- people kind of say that it looks at a simple commodity business. It is not. There is complexity in the business. And our job is to manage through that complexity. And I think that with the results that we are reporting and the progress that we are making, that is commendable in terms of the overall achievement that we are achieving right here.

Okay. And we'll come back if you have more questions, I'd just like to move to the next question.

The next question is from James Breen with William Blair.

Great. Just with respect to the guidance ranges, you maintained guidance for the full year. Can you just walk us to how you get comfortable around the EBITDA number, midpoint would be just over $30 million, it implies an absolute increase, obviously, in the fourth quarter of a few million dollars from where you were this quarter. And you talked about adding capacity this quarter. Was that added in anticipation of the fourth quarter, and so there won't be a need to do more of that in the fourth quarter given what you expect in terms of traffic volumes?

What was the last part? I'm sorry, we added what in the quarter?

Yes. So you added capacity this quarter.

Yes. Yes. Yes.

That was in anticipation of fourth. So that will help from a margin perspective in the fourth quarter.

I got you. I got you. Yes. So first question is on Q4 EBITDA. So if we get the revenue growth that we believe that the capacity that we've added throughout the year, and we can get that traffic to come through. Given our financial model, a lot of that will flow through to our EBITDA number. So I mean, EBITDA in the quarter was up 49% year-over-year, with revenue being up 15% year-over-year. So I mean that's demonstrating the fact that this does flow through to the bottom line. And so we need to continue driving that revenue number. And we believe we'll get the results that will get us to those ranges. As far as the capacity adds, we're always adding capacity. We talk to our customers every day and see where they are looking to go with their launches. And as we add new customers and new locations, we add capacity accordingly.

Yes. But the fourth quarter, we do lock down for the last 6 weeks. We try not to -- there can be an exception, but we try not to make any changes in terms of adding capacity in the last 6 weeks of the year.

The next question is from Eric Martinuzzi with Lake Street.

Yes. I wanted to focus on the operating expense this year. I've got the adjusted OpEx at a little over $22 million in Q3. Is that -- I know your head count did shrink a little bit from Q2 to Q3, but is that OpEx number a good assumption for Q4?

Yes. You can keep that there.

Okay. All right. And then when you were -- the last time -- the last earnings call, you talked a little bit about the excitement over edge services, and you had just signed a large customer. This was back your commentary on July 21. And you also talked about hoping to get a couple more edge services customers in the door in the next 30 days. Can you give us an update on those 2 large prospects?

Yes. So let me just start with the overall view. We mentioned earlier that we've doubled the revenue for edge services year-to-date versus last year. And so we're not going to break that out separately, but I can tell you, we were in sort of the mid- to high single-digit millions last year. And so we're starting to get into some material numbers, which we're happy about. In the July call, the customer that we were hopeful that we were on the 1-yard line of closing it, actually hasn't closed and may not close until middle of next year due to some geopolitical things that have gone on with this company. So that's the bad news. The good news is a deal that was on our radar screen that we did think would close that early accelerated. And so kind of the way the deals work sometimes. And so because of that, we were very pleased with the quarter, the names were different than I thought they were -- would be 3 months ago. But we've made the progress that we were looking to make. And more importantly, I'm really pleased with the initial reaction we're getting from our customers that are using Live Push, which is a service at the edge. Our EdgeFunctions is being really well received and our real-time streaming, which obviously needs to be at the very edge of initial customers that are using that are reporting really good feedback to us, which is very different than what we got with our first version, which was somewhat of a miss for us. So we're really pleased with what we've accomplished, but more importantly, where we think we can take this over the next coming years.

The next question is from Colby Synesael with Cowen and Company.

This is Michael on for Colby. Two questions, if I may. First, in line of the capital that you raised via the convert in July, can you give us some color on the expected uses of the proceeds raised? Should we expect any headcount additions, M&A or higher CapEx? Any color around that would be great. And then second, one of your competitors have noted that they've seen lower usage from a number of customers, including one that was impacted for geopolitical reasons. Broadly, did you see any changes in the allocation of traffic or your market share in the third quarter?

Yes. So let me just speak to market share. So from a market share perspective, what is absolutely true is that in our largest customers, we are not the exclusive provider to any of them. Maybe there's 1 or 2 exceptions, but I don't think so. They're large enough that they need diversity from a supplier standpoint. And so we are still in the mix in all of them. Largely across all of them, they have, in most cases, automated tools to shift traffic based on where they can get the best quality. And so things tend to shift around from day-to-day. We've had a fiber cut in some country, and our competitor didn't. And the next day they did, and we didn't. So there's always shifts like that.

Generally speaking, in our customer base compared to the second quarter. I don't think there's any name that stands out in my mind in terms of a loss of market share from second to third quarter. And I think there's more opportunity for us to gain share as we're well positioned with some of the largest names in the industry, and they're doing pretty well. So not to say that it couldn't happen. It just hasn't. And we are spending a lot of time. And again, going back to the whole idea of putting the DevOps organization together, really looking for ways to change the way we perform for our customers globally and drive higher consistency, higher performance. And we think with that over time, comes more traffic, not less.

And getting back to your first question on the converts and what we're going to use that for. So as our stock price has crept up over the last year or 1.5 years, we were consistently approached by various entities looking to do some sort of financing. And we felt that it wasn't the right time. And we continued to operate our business at the levels that we wanted to with the cash that we had on hand over the last, whatever, 4 to 6 quarters. And so we were able to do everything that we wanted to do. But as we got into the second quarter last year and the pandemic really decimated industries that were strong and strong brand names, and those companies had a ton of cash. And that really spooked us a little bit. And we were -- we are beneficiaries of this environment that we're in right now. But who knows what could happen. And so with that convertible market operating efficiently as it was, we felt that it was appropriate time to take advantage of that market and then do something with that cash. And whether that's just going to be internal improvements, added capacity, accelerated capacity, moving into adjacencies internally with added R&D expenses and sales expenses. I think it's going to be a combination of all of that. And then it gives us the wherewithal if something comes around, that we want to go out and pick up on an M&A basis that we have that flexibility in order to do that.

The next question is from Robert Majek with Raymond James.

You touched on some of the traction you're getting with real-time streaming in EdgeFunctions. Can you just talk about the per unit pricing of these offerings in relation to the rates you charge on core video delivery and the impact this might have on your gross margins?

Yes. So for sure, the margin is higher, the way we price is completely different. So with EdgeFunctions, for example, it's a per minute or millisecond of CPU usage, which is different than RTS, which is per stream, concurrent stream, which is different than our core delivery business, which is largely gigabytes transferred. So they're completely different. The margin profile of what I'd call advanced services like EdgeFunctions, like Live Push, like RTS, is higher than our core delivery business. And so the plan is, we've just introduced them, but to grow them as rapidly as we can, both with our existing customers and new and have that be a larger portion of the total. And so in addition to the automation tools we've talked about in terms of better managing the assets, we believe a different mix can also help that story improve over time.

And if I can add one more question. Just can you share any color on your pipeline of potential new business that might help explain the recent investments in capacity?

So from a pipeline perspective, I mean, the -- one of the negative size of COVID from a business perspective from us. For us has been a slowing down of the signing of new deals. And so we had actually a good third quarter, but the second quarter was terrible. So on a year-to-date basis year-over-year, we're behind where we were last year. But as I look forward, the deals that are in the pipeline are larger than historical. And we feel that we're well positioned in those deals. In fact, the deals that we've done year-to-date, on average, have been higher expected total revenue deals, contract revenue deals than the previous year, even though the number is down. So it's a mixed story there. But as I look forward, the pipeline looks very encouraging, especially for some of the edge -- advanced products that are driving our edge services revenue.

Next question is from Mike Latimore with Northland Capital Markets.

Yes. How much capacity do you expect to add this year? And can you give any early sort of view into how much capacity you're thinking about adding next year?

So in 2019, I think we said we added about 30 terabits per second. The plan this year was a little bit less than that, and we'll end up missing that plan by a little bit, largely because of supplier issues in Q1 and Q2. And so we'll end this year somewhere between 20 and 25 of additional capacity. And we're constantly adjusting the number as we learn more from our customers about their plans. But I anticipate that next year will be between 20 and 30 terabits per second of additional capacity.

Okay. Great. And then did you have a fair amount of incremental revenue from live events in the third quarter relative to the second? So was it a few million material?

Dan?

Yes, yes. I would say we definitely add incremental revenue related to live events. It's hard to break that out, though, just because a lot of the live events are with customers that also have video-on-demand. And so we don't break it out by customer, but just anecdotally, we believe that there was incremental revenue.

I mean, we can break it out. We just don't.

But Mike, just keep in mind, for us, live events, are kind of the series and seasons that are long-lived, right? So English Premier League or Bundesliga or MLB or NFL, right? These constant events that are multiple games, et cetera. When you have those seasons in play, rest assured, we are participating in those, and then that translates into good business for us, right? If it is just capturing Kentucky Derby, live event, yes, great deal, but it doesn't translate into much revenue even though delivering that traffic and delivering it well and better than anybody else matters.

Got it. And just last on the guidance. Kind of a wide range implied for the fourth quarter. The lower end would have you guys down a couple of million sequentially and year-over-year. I guess what would have to happen to hit the low end of the range?

Yes. Yes. I mean we -- what would have to happen is...

Things that we don't know about, right? For example, and I don't want to be quoted as saying anything bad about my customer. But like just hypothetically speaking, Sony misses the production date or they find a bug in the system that halts them from shipping it. And not only do you have any new consoles, but there's no new games to the old console. I mean, that's a major event. If there are things like that, that can happen, or we get hit with some major repricing event for a major customer mid-quarter after we've talked to you and before we report. So there are things that can happen. I mean, as you know, largely, we're pretty good in terms of meeting the guidance we give for the year. And we're not immune to things like that happening, and we're not always perfect, but we always do a pretty good job of working with you to give you numbers that are reasonable and achievable.

And by the way, those same hypotheticals exist on the positive side, right? So you can have...

That's why the range is so big.

Yes. So we feel good about the range of outcomes and onwards.

The next question is from Rishi Jaluria with D.A. Davidson.

Just 2. First, I wanted to start, Bob, in your prepared remarks, you talked about some of the traction you're getting in India with online education. I was wondering if you could talk a little bit more about what you're doing with them? Is it specifically video delivery? Is it also the software download component? Is there some edge in there? And then maybe broadening that question, specifically to the environment in India because I know you've been making investments there for a while. And now with the pandemic, we're obviously seeing a huge uptick in OTT consumption there, not just Netflix, Amazon, but even homegrown stuff like Hotstar and Zee5 and Eros now. So maybe let's talk about that, and then I've got a follow-up.

So India is interesting. A few years ago, where we went from almost nothing. I mean, we -- I think we talked about growing our capacity 10x in India. And that was the first time we really paid a lot of attention to that continent. Since then, we've grown every year. And this year, we've more than doubled the capacity in India. But in addition to just pure capacity, we've also invested in the team that's on the ground in India. So in the past, it was largely our customers in U.S. or in Europe, that were providing content for the India market. As we sit here today, that is still true, but we also have many customers that are in India based on the work that our team on the ground has done. So we feel very good about that market. We talked a little bit in the prepared remarks about just one segment of it, the online education. But obviously, and you've mentioned video-on-demand, live events. I think the amount of India content will continue to grow the availability, accessibility and quality of Internet will grow for the people in India. And we feel very good about our prospects of growing there.

And another market that we are now focusing on in a hyper way is the Latin America market, and we're seeing some of the same characteristics there, driven by more content for the market and an Internet service that's more widely available, more affordable and higher quality than it ever has. So when I look at, of course, we're growing North America capacity and traffic. Of course, we're growing Western or European traffic and capacity. But the 2 areas of focus for us are Latin America and India. India, we've been at it for years now, several years. And I would say, in Latin America, starting end of last year, beginning of this year, we're really trying to build a lot of capacity into that marketplace.

All right. Great. And then I wanted to touch a little bit on live events without breaking out specific revenue. I mean, we've seen that ratings for all the major leagues, right, NBA, NFL, MLB, NHL, all down year-over-year, in some cases, really significantly so. Just wondering, is that having any impact? Or is this a case where OTT consumption of those may actually be up year-over-year even if cable consumption is down and dragging that average down?

Yes. And I haven't seen enough numbers to sort of try to pretend I'm an expert on that. But I can tell you from what we've seen and some of the gains we've looked at, where we're looking at year-over-year numbers. On the low end, some of the less popular games are about the same volume as we've done last year and some of the more popular events are even higher than they were last year. So I think, as has been reported, that the cable ratings are generally speaking down, I would say that the over-the-top or streaming activity has been the same or higher.

This concludes our question-and-answer session. I would like to turn the conference back over to Bob Lento for any closing remarks.

Thanks, and thank you for your questions and your participation. It was another incredibly productive quarter with advancements in several important imperatives. We look forward to continuing to add value to our customers, provide opportunities for our employees and create shareholder value.

If anyone has any additional questions, we're always available for you, and that's either me, Saj or Dan. And once again, thank you for your participation.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.