Good afternoon. My name is Jesse, and I'll be your conference operator today. At this time, I would like to welcome everyone to the conference call. All lines and please on mute to prevent any background noise. After the speaker's remarks, there will be a question and answer session. If you'd like to ask a question during this time, simply press star followed by the number 1 on your telephone keypad. If you'd like to withdraw your question, press the pound key. Thank you. I would now like to turn the conference over to Maria Lukens, Vice President of Investor Relations. Please go ahead.
Hi, everyone. Thank you for joining our second quarter 2019 earnings call. We have Fastly CEO, Artur Bergman, President Joshua Bixby, and CFO Adriel Lares with us today. First, we would like to outline the format for this and future earnings announcements. We will publish a shareholder letter with our financial results on our investor relations website and with the SEC shortly after market close on the day of our earnings release. Since the shareholder letter provides a lot of detail, we will begin each call with brief remarks before opening the line for questions. Our hope is that this format will end result have greater transparency and provide a more open dialogue. We would also like to remind everyone that we'll be making forward-looking statements on this call. Today, Fastly will discuss the unaudited financials that were distributed on August 8, 2019.
Certain statements on this call include forward-looking statements, including statements related to the expected performance of our business, future financial results, strategy, long-term growth, and overall future prospects. These statements are subject to known and unknown risks, uncertainties, and assumptions that could cause actual results to differ materially from those projected or implied during this call. You should not rely upon them as predicted as future events. All forward-looking statements that are made on this call are based on assumptions and beliefs as of today, and we undertake no obligation to update them. Please refer to our risk factors filed in the final prospectus in connection with our initial public offering and those to be included in our quarterly report on Form 10-Q. Our discussion today will include non-GAAP financial measures.
These non-GAAP measures should be considered in addition to and not a substitute for or in isolation from our GAAP results. A reconciliation of GAAP to non-GAAP measures is included in today's shareholder letter. The shareholder letter and accompanying financial tables can be found on our website at investors.fastly.com. Finally, this call in its entirety is being webcast and will be archived on our website shortly after the call. With that, I'll turn the call over to Artur.
Hello, everyone. Thank you, Maria, and thank you all for joining us for this call. I want to start with a special thanks to all of our employees, customers, investors, and all the developers around the world who use us, for the support, hard work, and got us to this point. Becoming a public company was a big step for us. We're very happy that we've done that. We create a trustworthy internet with some of the best customers in the world. To do that, we really need the right people on our team. We strive to be a company full of kind, honest, passionate, ethical people. To keep growing at a high rate, sustainable rate, we will continue to maintain that culture. We have intentionally grown values first and scaled our workforce services and customer portfolio purposefully.
I'm proud to say that we've grown with integrity and will continue to do so. This afternoon, we issued a shareholder letter with detailed remarks, which I encourage everyone to read. On this call, I'd like to focus on a few key takeaways from our most recent quarter. We generated $46 million of revenue in Q2, up 34% year-over-year. Our dollar-based net expansion rate was 132%, up from 130% last quarter. Revenue from enterprise customers as a percentage of total last 12 months revenue increased to 86%, up from 82% year-over-year. Enterprise customer count increased to 262, up from 190 year-over-year. We have added to our network and now have 64 PoPs online, providing access to 52 terabits per second of global network capacity.
We continue to identify opportunities to drive leverage and expect to maintain our path toward profitability in the years to come. We've built a powerful Edge Cloud platform that fuels modern digital experiences, allows agile development as close to end users as possible. We really are powering the best of the internet. Our platform enables high quality, low latency delivery of applications, content, security, and edge compute capabilities. This Edge Cloud platform is based on three core tenets. Developers must be empowered to innovate, platforms must innovate ahead of market demands while still being reliable, scalable, and secure, and vendors must provide exceptional flexibility and support. We will continue to see increased developer demand for our Edge Cloud platform. Our long-term growth strategies will be driven by two key factors, winning new enterprise customers and expanding usage from existing customers and partners.
I believe that our results demonstrate continued success in pursuit of this strategy, and our team remains focused on building upon this momentum. We'll continue to invest in our Edge Cloud platform. We'll make it faster, more secure, and more flexible for our customers and developers. For example, we have our new Edge Cloud developer tools released this week that make it faster and safer for developers to discover, test, customize, and deploy Edge Cloud solutions with pre-tested, pre-baked solutions. We believe this will drive continued growth throughout the third quarter and the remainder of the year. We're encouraged by our recent progress. We're excited about our future. We look forward to sharing our progress with you for many quarters. Adriel, Joshua, and I would like to reserve the rest of this time for Q&A.
Thank you. At this time, I would like to remind everyone, in order to ask a question, please press star, then the number 1 on your telephone keypad. We'll pause for just a moment to compile the Q&A roster. Your first question comes from Brad Zelnick with Credit Suisse. Your line is open.
Excellent. Thanks so much. Congratulations on an initial quarter here as a public company. Great to see the strong results. I guess maybe just for starters, if I look at your cost of goods in the quarter, it was at least percentage-wise, a little bit higher than we were expecting. You commented in the letter and in your remarks that you're investing ahead of an expected strong second half. Can you maybe just unpack your cost of goods? If we look at them by their various components and then we look to your largest competitor in CDN, I know that they're talking a lot about capital investments ahead of an expected surge in OTT traffic. If we think of your business in terms of traffic mix and what it is that you're pursuing, are you going after that same opportunity?
How should we think about the mix of traffic on the platform? Thanks.
Thanks, Brad. Hey, it's Adriel here, and thanks for joining us on the call, and again, happy to be here. I think, one way to think about it is, we've talked about this before, is we're building basically ahead of the second half. From an allocation standpoint, we're allocating much of this CapEx ahead of time. From a percentage standpoint of one, percentage of CapEx, percentage of total revenue, I still feel comfortable with where we will get sort of a 10% of revenue as percentage of CapEx, percentage of total revenue over the long run. This year, we're going to continue to make progress in that vein. Some of the challenges here is we make these strategic bets around the world.
In terms of what we've seen in previous years, usually the first half of the year is where you'll see CapEx percentage of revenue higher than the second half, because normally the second half, we've now signed up those contracts from a booking standpoint and you see the ramp over time.
Thanks, Adriel. If I could just follow up with another question. At the time of the IPO, at least a portion of the proceeds were intended to invest in building out your go-to-market. Can you maybe just give us an update? I know it's only been a couple few months or what have you, but in terms of the progress in hiring, I mean, the enterprise customer growth seems to be very strong. Was it 36% or 38%? From the leading indicators that you look to in terms of the productivity and the ramp that you're seeing in some of these hires, how do you feel you're doing relative to expectations?
Yeah. On the hiring front, I think we are on our expectations of where we want it to be. We've gotten folks into sales and marketing, especially in sort of the demand generation and account-based marketing initiatives. From that standpoint, I feel like we're on our sort of expectations. In terms of sort of the progress that we've made so far, you're right, it is a bit early, but I'll let Joshua comment a little bit more further on that.
Sure. Thank you, Adriel. Brad, thanks for joining. I think as we've talked about, this is an area that we have historically underinvested in. We are a company that has grown through word of mouth for the most part, and we know that that's an area that, with the proceeds of the IPO and with our intentions long term of continuing strong growth, we have to invest in. It has been what feels like a very short two months since the IPO timeline, but we've made good progress. As Adriel said, we're on track from a headcount perspective. On the marketing, we're on track on the spend perspective. Now, these are large enterprise deals, so as you know, they do take time to work their way through, but all signs indicate really positive results from our investments thus far, but it is early.
Awesome. At the risk of being greedy, I just want to slip in one more for Artur. To us, perhaps the most exciting future opportunity is in edge compute and with a lot of the capabilities that you most recently announced. Artur, can you maybe just talk about some of the excitement, maybe some of the leading use cases, and the confidence that you have in more and more compute and programmatic complexity actually occurring out at the edge? Thanks so much.
Thank you. The demand for use cases of the edge compute kind of environment where you can do just more, range is pretty wide. A lot of it has to do with personalization, content filtering, applying different rules at the edge for different users. You can see this, I can see this. Aggregating data from IoT devices so that you don't have to communicate all the way back to a data center. You can actually do most of the edge and then feed the data back. There are mapping companies that I've talked to that want to be much more creative in how they package the maps that you want to get. These are all use cases that we perhaps could have developed in the past, but they are kind of very specific to some customers in the actual product use case.
We are seeing a lot of demand for customers to just go off and do that final bit of customization on their own.
We're also driven the GraphQL evolving standards around how to interact between clients and servers. You can really assemble API requests at the edge, also drives the kind of demand for this. I think, last but not least, once you can deploy more code that does more at the edge, you can start evolving security introspection functionality that's very customized for your application, and try and prevent attacks as far away from your data centers or your clouds as possible.
Awesome. Thanks so much, guys.
Thank you.
Your next question comes from Will Power with Baird. Your line is open.
Great. Yeah, thanks for taking the question and all the allocated time here. Let me first just start from a high-level perspective. As you look at the strong year-over-year growth, the dollar-based net expansion rate, improvement sequentially, I wonder if you could just talk from a high level as to what the key drivers are. I know video's been a key driver, live video. Maybe just help us break apart what those key secular themes have been that's driving that revenue growth.
Sure. Hey, Will, it's Joshua. Thanks for the question. There are a number of them. I would start certainly with edge compute and security. As Archer said, we continue to evolve edge compute, but we are by far the leader in that space, and we continue to see incredible innovation. We came out, as Artur said, with a developer library that really is a collection of tools that help continue to push our strong leadership on the edge solution space. People are coming to us because they know that they can do more at the edge, and they want that. That is continuing to drive growth. We continue to see, as we've talked about in the past, a strong connection between our growth and the security needs of our customers.
We continue to see underlying growth in the security space, which, if you look at the market in general, that should be no surprise. I think we're also seeing a strong demand in the live video, the higher value side of that market, which is driving us. I think as we start seeing edge compute security and the OTT growth, as well as some of the really nice tailwinds around 5G and other things that are coming forward, we're seeing a lot of important growth vectors. I would say it's across those ones for sure.
Okay, great. All right. Maybe for Adriel, just maybe come back to the earlier gross margin commentary. I wonder if you could just comment on kind of what you're seeing competitively out there, whether there's been any meaningful shifts in the pricing environment, if that's contributed any of the weakness. It sounds like, again, it's more about investing in front of the second half. I guess as part of that, how do we think about the gross margin outlook from here? Is that something that should tick up a little bit in the second half? Any thoughts then over the next couple of years from here?
Thanks, Will. I think to the second part of your question, yeah, you should see some upticks for the second half. In general, I'm comfortable in delivering gross margin leverage on a year-on-year basis. We talked previously in the past that our sort of primary goal right now is to evangelize the Fastly solution, talk about edge compute, talk about all the benefits that you have coming onto Fastly platform. From our standpoint, we're not leading with price. We certainly don't want to lose on price. There's just too much opportunity out there, and we know that once a customer comes on board to Fastly, we have our 98% retention rate, we've got our strong DBNER, and we want to take advantage of just bringing a customer on board.
There is the normal seasonality that we experience, which Q2 is sort of no different, where we're getting ready to build ahead of time for the second half of the year. You should see leverage both in gross margin and as well as sort of an operating leverage over that timeframe.
Okay, great. Thank you all.
Thanks, Will.
Your next question comes from Tal Liani with Bank of America Merrill Lynch. Your line is open.
Hi, guys.
Hey, Tal.
Was there anything that changed since you went public until now that drove gross margin roughly 200 basis points below what you just told us a few weeks ago? Trying to understand the gross margin part.
Sure. One of the biggest differences just from Q1 to Q2 is Q1, as we talked about on the roadshow, had a particularly strong February, which we, up until this point in the company's history, we hadn't quote unquote "broadcast" that particular sporting event in the past. We had actually built CapEx-wise, in Q4 of last year, ahead of that period of time. So we had the benefit of that one-time sporting event, which didn't replicate. The football season doesn't go into Q2. That provided us a bit of a bump. If you think about it from a year-on-year basis over a 12-month timeframe, you actually do see gross margin going up on a year-on-year sort of 12-month basis. So we should continue to see that trend over time. Seasonally, you have another factor in Q2. You have two 30-day months.
You have a little bit shorter amount of days for revenue. Most of those pressures in Q2, that explains really sort of the quarter-on-quarter contraction in gross margin. Overall, in terms of where I think gross margin will be from a longer-term perspective, I'm still confident in.
Got it. Okay. The other question is anything change in the revenue trajectory versus what you thought before?
Everything, at least from what I can see right now, still seems to be on course. We're continuing to win the deals that we need to go win, get the POCs that we need to get out there. We're still, I think, relatively a small player in terms of the numbers of deals that we're in. If anything, we're just not in enough parties to actually go and try to compete for that business, which is why we are ramping up sales and marketing now and into the end of the year. Part of the balancing act that we're trying to do is grow revenue, grow share, while also delivering a discipline in leverage as well. I think we're still continuing to do that, and at least at this point, nothing has changed in terms of what I think we'll actually be able to go deliver.
Great. Thank you.
Thanks, Tal.
Your next question comes from Jonathan Ho with William Blair. Your line is open.
Hi, good afternoon. I just wanted to start out with some of your comments around the streaming business. Can you maybe provide a little bit more color on what drove the strength there? If there's any marquee events, what does that mean for maybe future opportunities?
I'll speak a little bit that way. I look at it from a finance standpoint, and then I'll see if Artur or Joshua have any sort of follow-up comments on that. From my standpoint, I look at some of the marquee events, especially in Q1, and some of them similar, maybe some episodes in Q2 that came to a series end that I can't talk too much about. Nevertheless, effectively all the different providers there who they know who all the different edge providers can be and which ones are succeeding. The more that we win, they become beacons, if you will, of reference points for, wow, they can deliver this, they can deliver time to first view of sub X or consistency of stream without interruption.
All of that really does make a difference, and it enables our current and our future sellers to go out and win more of this business, which then gives me, from a finance standpoint, more confidence from a forecasting standpoint to, in this case, give the guidance that we're giving for Q3 as well as for Q4. I'm sorry, as well as for the rest of the year.
I think on the importance of these events is absolutely a signal that we are now a player that can handle a very large event. That's important not only because of the OTT events, but also when you go talk to large enterprise or commerce customers, because they know how large these events are. For them, suddenly, if you can handle the largest sporting event in the U.S., I just don't need to worry at all that you can handle my traffic. There is a very strong tailwind that that helps with. On the OTT side, we focus on the live streaming and our underlying technology platform with the modern software-defined network, how we've architected it. We are very well-suited for the kind of traffic that livestreams are.
There's some other traffic like that, API calls during election, for example, where a lot of people are just trying to get the same object. Right? We are really good at delivering that with a very consistent quality. As the OTT providers on live, especially in sports, are trying to get the delay from glass down to shorter and shorter, we have a customer who's aiming for 500 milliseconds delay. That consistency in the delivery at really high volume is key to them. Platform-wise, we're very well-suited to it, and it really helps us in all parts of the business to be able to say that we can deliver these events. Thank you.
Got it. Just as a follow-up, I just wanted to understand if you're seeing any shifts in the competitive environment. I know there's a few competitors that had outages during the quarter, is there anything there or any opportunities that that creates?
Yeah, great question, Jonathan. We have not seen significant changes in that. As you know, we're in the enterprise. If you see the number of enterprise customers and the size of our deals, there were certainly outages in the small and medium CDN side of the business. We continue to be laser-focused on the largest customers and those that aspire to be, and that targets directly in the legacy CDN space, and that environment hasn't changed at all.
Thank you.
Your next question comes from Michael Turits with Raymond James. Your line is open.
Hey, guys. Good evening. First, Adriel, on CapEx, came in higher on CapEx this quarter than was in our model. I'm wondering if there's a change in the outlook or prospects for the year on CapEx.
Yeah. Hey, Michael. From a year standpoint, I still think CapEx as a percentage of revenue should be at or near where we were in 2018, and so we're still targeting that. As you can see, some of our investments in Q2 were much more timing-related as we think about what we see in our pipe in terms of our customers that we believe are going to be ramping up, which is reflected in our guidance. Nothing over a year or longer timeframe has changed in that regard. It just might have been more timing from sort of a quarter, Q3 to Q2 basis.
Okay. A question on programmability, which has obviously been where you guys have been winning share. Maybe an update in two respects. One, in terms of your product roadmap there for advancing programmability. Secondly, your assessment of the competitive landscape in terms of whether or not others in the space are making any changes that'll change your competitive advantage.
Hi, Artur here. Thank you. On the product roadmap innovation side, we're investing heavily into a technology called WebAssembly, which is driven by us and a couple of large companies to create a safe and secure environment to run code. Initially, it was to run code safely and securely on your desktop. We have adapted that for server-side use and are involved heavily into the ecosystem, the WebAssembly ecosystem, to make that run very well on the servers and kind of evangelize it. One of the key things with it is that it frees us up to allow many different programming languages, and so customers can use a wide variety of those. To help the adoption of this, we've open-sourced a tool called Lucet, and that's part of our commitment to this technology and just help drive usage of it across developer base. We're very excited.
One of the cool aspects of this technology is that every single request that we get from a customer, for our end user, will run essentially equivalent to its own sandbox. We can really offer a high degree of safety and security for our users, both for eventual potential attackers or if they have problems internally, we can help them detect that and mitigate that. That's on the R&D side. On the competitive side, I'll hand over to Joshua.
Sure. Thank you. The competitive landscape hasn't changed dramatically. I think we're definitely seeing, particularly from the legacy CDN players, a lot more developer-friendly overtures. We're seeing a lot of marketing story arcs. Fundamentally, our customers referred to the legacy providers as the anti-cloud because you have to get professional services and engage on a non-developer-friendly sort of platform. We have not seen any dramatic changes in the competitive landscape in the enterprise space other than some marketing information that's come out. Competitively, we remain the dominant Edge Cloud, and the only one today that really allows you to build the type of programmability that a modern developer needs.
Okay. Thanks, everybody.
Thanks, Jonathan. I mean Michael. Sorry about that.
Close enough.
Your next question comes from Timothy Horan with Oppenheimer. Your line is open.
Thanks, guys. Any particular customer segment or maybe any other more color on applications that are growing faster than what you've seen in the past or yeah, any color around that would be great?
Joshua here. Thanks for the question, Tim. I wouldn't say that we've seen a dramatic shift in the overall base at all. We continue to see strong growth across the core verticals. If you think about areas like e-commerce, travel, high tech, we continue to see a lot of really big brands who often are in a time of digital transformation, and that is often one of the precursors that brings us in. We'll see examples like Ticketmaster's a great example where we've, over the history of that business, as they have evolved their platform, they build us in early and are able to re-platform, which really is less of a shifting out of a vendor and more shifting in a platform. We see that type of evolution in a lot of those core verticals. High tech follows that same pattern.
I think on the platform side, we continue to see growth. As we've talked about in the past, we believe very strongly that the way into the SMB market, the most efficient path, is through platform vendors, where we sell to one vendor, and they're able to sell to hundreds of thousands of customers. We continue to see strong growth. I wouldn't say it's anything outside of the historical norm. They're all growing, and I think the edge delivery and programmability continues to anchor along with the security use cases.
Maybe just a touch more color on edge compute. I know you're saying you're well ahead of your peers, and that clearly appears to be the case. Is WebAssembly kind of an example of this, where you do need the programmability or developers need the programmability and other features and functionality, especially in this new toolbox that you rolled out? I guess, given that this is a brand-new market, do you have a sense of what percentage of revenue this can ultimately get to and maybe what it is now even? I know it's very small now, but yeah, any more color would be great.
I think it's hard to say the percentage of revenue, but a large amount of our customers, to a certain degree, use our edge compute capabilities. So, quite often not very much, but it is accelerating, and I think we'll continue to see that accelerating over time. The use cases are really, as Joshua said earlier, really diverse. What I find really interesting is that even if you take something that ostensibly sounds kind of why would it need edge compute? OTT live streaming. Well, it turns out that if you can personalize the ads and do real-time ad insertion at the edge, that's the most efficient place to do it, the fastest way to do it, and it can drive more revenue to you because you're not showing everyone the same ad. To do efficient ad insertion for each individual user, you need edge compute capabilities.
We're seeing it advance functionality across a lot of most areas of the business.
That's very helpful.
Your next question comes from Rishi Jaluria with D.A. Davidson. Your line is open.
Hey, guys. Thanks for taking my questions. First, wanted to start off by asking about international. Wanted to see what your traction there looks like. We did see the announcement recently about you expanding your Australia presence. I think any color you can give in terms of international traction, be it with local companies out there or with just the international parts of American companies, I think that'd be helpful. Then I've got a follow-up.
The international that we talk about is, we classify the traffic based on where it happens. For revenue, we attribute that to wherever the customer is based. The growth that we're seeing when we talk about international numbers within the roadshow, is from customers in those regions. We're seeing that the strategy of being that we enter a region for our existing customers, that's how we have the justification to build a physical presence. As we build those networks up, we are getting local customers who notice this network and want to work with us. Once we have a mass of those, that's when we start investing people into actually selling into those markets. We have had really good success with this.
International is accelerating, and we're going to keep on, as you said, Australia, where we have a really fantastic network between the 4 PoPs in Australia, 2 PoPs in New Zealand, and so we think that's a large untapped market. Europe is behind the U.S. when it comes to digital transformation, and so what we're taking and doing here is very much applicable around the world. Since they're a little bit behind, we're actually really well-positioned to help them.
Rishi, it's Adriel. Just a quick follow-up. The actual revenue contributed internationally in terms of just where the billing locations are, was 30% in Q2.
Thanks, guys. That was helpful. I just wanted to ask a little bit about the gaming opportunity. We've had some of your competitors out there talking about capitalizing on gaming opportunities and beyond just software and onto things like live streaming, which is probably at the intersection of gaming and OTT and even some of these cloud gaming platforms down the line. Would love to get a sense from you, how are you thinking about that gaming opportunity, and is that something that you think you can probably capitalize on that secular tailwind? Thanks.
This is Artur. Thank you. I think definitely it is a large growth area. As the gaming companies also are adapting to an online world in a way where it's not just I'm downloading a game, and moving to more modern architectures, we do see a lot of opportunity around delivering part of the game experience, delivering the APIs around it, delivering metadata, and enhancing that with edge compute to give personalized experiences through those games. We have some really good gaming customers, and we do believe that there is a large growth area for us and that the evolution of the edge computing platform fits really well with that. We also do help the live streaming of games.
Going back to my comment earlier about the time from glass latency sensitivity, that's an area in esports where they really want to be as close as possible, and being 15, 20 seconds behind is not acceptable. We have customers in that segment.
All right, great. That's helpful. Thank you.
Thank you.
Your next question comes from Walter Pritchard with Citi. Your line is open.
Hey, thank you very much. I wanted to return to the question of CapEx and maybe to the sources of the CapEx spend, because in the entirety of 2018, you spent around $19 million, and just in the first half of this year, you spent $21 million in CapEx, and in most recent quarter, it was $12 million. I was wondering, could you help us unbucket the sources, the kinds of projects that you were invested in, especially in Q2, that contributed towards this larger number?
Sure. As I sort of mentioned earlier, as a percentage of revenue, I do expect over time that that should get to about 10% of revenue overall in CapEx. That's inclusive of around 2% of software capital additions. As a percentage of revenue in 2018, we were probably sitting somewhere around 13%. I think we'll probably land somewhere in that range in 2019.
Keep in mind that in 2018 benefited a bit from a big investment we did in 2017, and there was a large investment we did across the network to upgrade, I believe the protocol, in terms of, we went from a 10-100 gig ethernet, either 100 gig internet ports in the back, which really upgraded the entire network, which that alone, in terms of doing that with the Fastly network, I would find would be a real challenge for any other provider to do that. If you sort of streamline those two years, it would sort of average out a little bit more smoothly, versus sort of the bump that we did in 2017.
Returning back to your question, there was, again, a bit of a front-loading in the first half of 2019. I still expect, for the second half of the year, you won't see nearly that amount of CapEx in the second half of the year.
Right. Just turning to the nature of projects, was this due to geographic expansion? What were you invested in?
There's a bit of it in Q2, with a much more APAC expansion in terms of, as our customers, in some respects, pull us there to serve traffic in those areas, then it allows us to actually sell into those areas more prospectively. In general, it was overall traffic growth signals that we're receiving from regions like APAC and EMEA, as well as just overall signals we're receiving in terms of bookings, in terms of overall traffic growth.
Just maybe as a last question, because what this brings me is probably to one of the key questions on people's minds right now, in that you do compete against very well geographically diversified, albeit clumsy incumbents. The question becomes, does this level of CapEx spend become a really semi-permanent part of your strategy as you are entering this new stage of competition that reaches across geographies?
If I understand your question correctly, given where we're competing, are we seeing that this is driving CapEx dollars as a percentage of revenue? I'm putting this as a percentage of total revenue here. Would that change the guidance, or at least the perception of where we would go in the future? The answer is no. I think that what we're beginning to realize and what you'll see overall in gross margin is many of the investments that are needed to get us to a point of scale, especially in a lot of the fixed cost areas that incrementally are unrelated to bandwidth and some of the servers that are deployed worldwide. We do a lot of catch up to get for enhancing 24 by seven support, customer support, technical account managers, et cetera.
As we grow, we will get scale on those, and we realize scale in those areas. That is what drove a bit of the impact in Q2 as well as the preparation for the second half of the year. There's nothing, at least that I've seen so far, that moves me off of our efficiency in terms of where we need to be long run in terms of CapEx or % of revenue, sort of around 10%.
Thanks very much, Adriel.
Thank you.
Your next question comes from Jeff Van Rhee with Craig-Hallum. Your line is open.
Great. Thanks for taking my questions. Just a couple from me. I think in terms of the pipeline, maybe, guys, just spend a minute and expand a bit more there. I'd be curious if you see any variability in terms of what's working through the pipe. Obviously, you've had some very large, high-profile events. Are you seeing that result in a very notable shift in, say, deal sizes in the pipe, use cases, verticals, et cetera? Just maybe a little more color in terms of any variance of what's coming through the pipe.
I'll take a stab at it first, just from a finance standpoint. I think in general, when we're bringing on a customer of any size, and I think it's really more of the opportunities that are getting bigger. In general, customers who come on board, there is a ramp period. You will see a ramp period, and that hasn't changed. I think the size of the opportunity in terms of what we actually are being invited to take on is increasing in terms of the names of those opportunities, and the scale of those opportunities certainly is growing.
In addition to that, as we expand our solution into areas that then make our solution more attractive for areas like e-commerce or financial services, Many of those areas are dependent on having qualifications like SOC 2 or PCI or security offerings such as web application firewall. As we add more and enhance our solutions, we're able to win those faster and ultimately ramp those faster. In general, nothing has changed other than sort of the size and the scope, which has incrementally changed. I didn't see anything significantly different, although I will say we probably got a lot more inbounds post our little event in Q1. That's sort of my perspective. Joshua, if you have some two cents on your side.
No. The only thing I would add is that the publicity around the IPO was also very helpful. We're certainly seeing larger deals and some change to how that pipe moves in the sense that people can look at us, and they can see our financial position, and that has certainly helped. We are seeing some larger deals. We're definitely, I think, seeing the potential to see, at least over the long term, some of those timelines come down, in terms of the sales cycle. As Adriel said, we're not seeing a dramatic shift in any underlying vertical per se.
Got it. Fair enough. I think in terms of the business in the quarter, any variance with respect to the usage versus committed percentages? One other housekeeper on DBNER, just a little, maybe an update on your thinking about that nice uptick in the quarter. Thoughts next 12 to 24 months in terms of how you think that trends?
Sure. It's Adriel here. I'll take the second part of your question first. On DBNER, we talked a bit about, over time that will likely come down. All of our teammates here at Fastly, we focus on delivering just the best of service to our customers. We feel like we've innovated a lot with partners like Slack. We offer Slack support for many of our customers. From our standpoint, to the extent that we keep doing that, we feel like that puts upward pressure on DBNER. The extent that we can, with the technical account manager, show and educate our customers on all those solutions and the possibility that you can in terms of moving traffic over to us. We think those are all upward pressures.
Although at the same time, as a finance person, I'm well aware of the law of large numbers and the challenges associated with that. I was very happy to see the uptick quarter over quarter in DBNER. Again, I think it's one of those things we work on sort of 24 by seven all the time. Nothing has changed my tune about eventually we'll see some attrition there, but at least so far, we're focused on keeping that as high as we can. In terms of your first question, nothing changed from a split standpoint. Commit versus usage over the commit, still around 50/50. I don't expect that to change too much into the future, and if it does change significantly, we'll certainly let you know.
Yeah. Great. Last one from me then. Just on the sales reps, where are you now rep count-wise? You obviously came into the year and out of the IPO looking to add a lot of capacity. Are you finding the heads? Are you finding them at the pace? Are you onboarding them at the pace? Just maybe a little more color about the sales adds, the ability to find them, and maybe a little early to get a sense of incremental ramp, but any thoughts on ramp as well?
I think, as you can see from our total headcount, at 544 at the end of Q2, in general, everything is going the way that we expect it to, and we still have a target rep count somewhere around 60 by the end of the year, and we still feel like we're on track to hit that number. From that standpoint, I feel like things are in line with what we expect.
Fair enough. Thanks.
Your next question comes from Brad Reback with Stifel. Your line is open.
Great. Thanks very much. Many of mine have been asked, but maybe on the pricing front, Adriel, as you look at longer-dated customers as they come up for renewal, what have been the pricing trends there? Thanks.
Thanks, Brad. Nothing has changed too much on that front. There are still situations where if the customer's coming up for renewal, I sort of generally think about our business as we are still an evangelical sale. Given the fact that our DBNER is 132%, that encapsulates renewals into that metric, I still feel like we're doing a good job there. From our standpoint, it still takes a lot of work, the way that we can reduce pressure on renewals is to continue to not talk to a customer at renewal time, continue to deliver great service and great support throughout the year. From that standpoint, nothing unusual has changed in Q2, as it relates to that.
Great. Thanks very much.
Thanks, Brad.
Again, if you'd like to ask a question, please press star one. Your next question comes from Michael Turits with Raymond James. Your line is open.
Hey, guys. Thanks for letting me take a follow-up. Maybe you explained this and I missed it, I'm sorry, are you saying, Adriel, that in the end, COGS is about where you would've expected it for the full year, and this is just a pull forward, sort of like a CapEx? COGS is going to end up higher than expected for the year? If so, why is COGS higher but CapEx is not? What's the difference in where you need to invest?
I think it's the former, but I would replace COGS with CapEx in the sense that overall CapEx, on an absolute and as a percentage basis, given where we've given guidance, should be about where we expect. It's really more of a pull forward to that. Again, if you pull CapEx forward, you start the depreciation at that time. There may be, what we're assuming there is that over time, as the depreciation starts to kick in, COGS as a percentage, should still hit the numbers that we want to deliver overall gross margin leverage year-on-year. Think about 2019, we should be able to deliver gross margin leverage relative to 2018. What you see in Q2 specifically is gross margin was up year-on-year.
What you saw, the difference there being is that Q1 to Q2, you didn't have that one event in Q1, and then as we get ready for Q2, it sort of depressed it a little bit before we get to the second half. Does that answer your question?
Sort of. I was just wondering if COGS, as an absolute number, ends up higher for the year, and if so, is that just because you pushed CapEx forward and you're depreciating earlier, or because your expenditures outside of depreciation in COGS are higher?
If COGS is going to be up just on an absolute basis, it's because revenue should be driving that up as well. The reason we're then investing the CapEx ahead of time is because we expect the revenue to continue to grow.
All right. Okay, thanks.
Thanks, Michael.
There are no further questions. I turn the call back to the presenters for any closing remarks.
This is Adriel. I want to thank everybody for joining us on our first public conference call. We look forward to seeing you all out on the road. We're going to be at the KeyBanc conference on Monday. We look forward to seeing you out on the road.
This concludes today's conference call. You may now disconnect.