Last week, Akamai Technologies reported a very strong first quarter of 2010, with revenue of $240 million, up 14% year-over-year, and net income of $40.9 million, up 10% year-over-year. The company pointed to 3 main drivers of its accelerated growth: developments in cloud computing, video distribution over the Internet and online advertising. To learn more about what Akamai is seeing specifically among its media & entertainment (M&E) customers and in online video, late last week I spoke to company president and CEO Paul Sagan. Following is an edited transcript of our call.
VideoNuze: Akamai just reported a very strong Q1, which as you explained on the earnings call, included accelerated growth in the company's large M&E business. What are the key trends Akamai is seeing from its M&E customers?
Paul Sagan: The most important thing M&E customers are focused on is how they can build a sustainable business online. For video it's all about engagement - how to get people to stay longer and make the model work. We believe the big driver of that is quality and more specifically HD. A number of recent things have happened that help HD - first is pervasive and consistently strong broadband in the last mile. Second is variable bit rate streaming. Third are all he new convergence devices connecting broadband to the TV. The key for customers is trying to get TV-like quality with interactivity. Just broadcasting an HD signal over the Internet isn't enough because TV works well already. It's the interactivity - things like multiple camera angles and instant DVR - that make the difference.
VN: You said on the earnings call that HD is driving double or higher engagement by viewers. That's an impressive data point.
PS: Yes, with live events where you can do a true A/B test - we're seeing roughly double the viewing time when delivering at 1-2 MB or higher. That's offering a big potential lift in time spent viewing for our customers.
VN: Does that mean Akamai's M&E customers using HD are also doubling their revenue as a result?
PW: I'm not sure they're doubling just yet, but HD delivery is making content into something that can be monetized more strongly - possibly through sponsorships as well as advertising and paid models. So it's not as simple as saying you sell double the number of banners. What HD also does is push people toward longer-form. An issue with some of the shorter-form content like 2-3 minute clips is that you just can't put that many ads in or it will be worse than TV for users. I'm not suggesting we should see 8 minutes of ads in a 30 minute show, but you can certainly do more.
VN: Speaking of business models for high-quality video - what do you hear from your customers as the emerging standard - ads, paid or a mix of both?
PS: It's a mix - some ad-supported, some per event payments or subscriptions, particularly with sports. Some movies will be more subscription-oriented. Online delivery and HD are unlocking a few different models, yet it's still early days for all models. Clearly some have struggled to date which is no surprise when you consider it's taken 15 years to get online viewing to just a 1% share - which is obviously pretty small. But given everything that's going on, I'm sure it won't take 15 more years to double again.
VN: On the customer front, Akamai is announcing Magnify.net as a new customer this morning, which has an interesting "video curation" model. Can you say more about how these kinds of non-traditional distribution models like Magnify's fit into the online video landscape?
PS: What we're seeing across a wide number of sites is a strong desire to add rich media. We're also seeing sites think about programming in a non-traditional way. The goal is how to keep the user experience compelling. That means adding audio and video when users expect it. That in turn drives higher engagement and monetization. We've evolved from a time when there was a "priesthood of 3 networks" who produced video and nobody else could, to today when there are lots of ways to produce video - including millions of hours of UGC. The curated model is so important because it helps sites get relevant content.
VN: How mature is the idea of curating online video from the web and non-traditional distribution models generally?
PS: Well, I haven't even figured out the etymology of "curation" - in the old days we used to call it "editing." But that was about journalistic sites. Many of today's sites are not purely "journalistic." So the video added isn't always "news," though it still has to be highly relevant. For example, biking video belongs on biking sites, not on hockey or baseball ones. How do we make those sites more compelling through video? That's what curation does. And Magnify's trying to make that easy. I've know (Magnify CEO) Steve Rosenbaum for 20 years and I'm thrilled that they appreciate how Akamai's quality, scale and reliability can be central to delivering the experience they want to achieve.
VN: Shifting topics to CDN pricing, which is of course widely discussed. Can you say more about Akamai's approach to pricing for its M&E customers - on balance, does Akamai try to keep prices stable or is it continually trying to push them down?
PS: I've gotten a question about CDN pricing every day for the 12 years that I've been here! My view in general is that unit prices in technology always come down and in this area they need to come down a lot because we're trying to enable our customers to deliver a lot more data. So we've been relentlessly driving the unit price of delivery down for years. For us it's not about keeping prices stable and reducing our costs solely for our own benefit. Rather, we've been driving the unit costs down every year and sharing that savings proportionately with our customers. That's worked out well in generating more traffic on our network every year. We plan to continue doing that because it creates a virtuous circle of ever-more traffic and reduced costs. VN: Regarding Akamai's costs - can you describe what the principle components of cost savings are?
PS: There are several areas: what it costs us to get network capacity and buy the servers we need plus how efficient we can make our software in delivering customer traffic. Because of our massively distributed model we're able to source huge amount of capacity with low fixed infrastructure costs and route to most efficient place. As I mentioned on the call, with one sporting event we delivered last quarter we used 500 different delivery locations. Nobody else has anything close to that; the ability to serve thousands of customers this way. When you sum these things up, we get a very differentiated delivery.
VN: Finishing up - can you say more about live delivery - is that primarily for sports or do other content categories benefit as well?
PS: Sports is certainly the big one for live. There are others as well, but sports has had the most success.
VN: You mentioned "convergence devices" earlier - what impact will they have on online video growth?
PS: A huge impact. If you go to the first inflection point of online video 5 years ago, it was more about broadband connections. But now it's going to be about all these 3rd party devices that are joining the network. We're going to see tens of millions of homes with "broadband TV." I call them "trojan horse set-top boxes." But over time, most people aren't geeky, they want things simplified. Things like WiFi-enabled TVs are going to be interesting.
VN: What happens with last-mile bandwidth in the future? And what role if any, does net neutrality play?
PS: There will be more and more broadband from wired and wireless providers (like 4G). Broadband networks need to stay competitive and the government will insist on it. Networks will remain open. If the FCC can't do that, then Congress will, by stepping in and saying networks need to remain open and transparent. Users need to be able to get to anything they want.
VN: Wrapping up, when you look ahead 1-2 years what's different in online video?
PS: What's different is that people will be shocked at how much HD online video there is, plus the interactivity available. The default setting will no longer be standard-def, but rather high-def. That's going to be very exciting.
VideoNuze is the authoritative online source for original analysis and news aggregation focused on the burgeoning online video industry. Founded in 2007 by Will Richmond, a 20-year veteran of the broadband, cable TV, content and technology industries, VideoNuze is read by executive-level decision-makers who need to get beyond the standard headlines and achieve a deep understanding of online video’s disruptive impact.