Level 3, Limelight Displacing Akamai at Netflix?

November 9th, 2010 by · 9 Comments

That’s the word according to Dan Rayburn, and it matches what I’ve been hearing since my post last week about Level 3’s CDN push.  Multiple sources that are almost certainly independent of Dan’s have now told me that the new content customer won in October for whom Level 3 Communications (NYSE:LVLT, news, filings) is spending $15M in capex is in fact Netflix (NASDAQ:NFLX, news, filings).  But that Akamai (NASDAQ:AKAM, news, filings) may be the one taking it on the chin is a new one on me, though I doubt we will hear for sure until it’s time for Q4 earnings and 2011 projections.  I somehow doubt that Akamai is out entirely, but who knows.

The amount of bits pushed by Netflix’s streaming business has grown so rapidly that it’s easy to forget just how big they are – it seems like just a year or two ago they were giving streaming a try (oh wait, actually it was).  In fact, I’ll bet that its contribution toward Akamai’s fantastic revenue performance this year is not minor at all.  And given Netflix’s continued subscriber growth, its traffic needs next year will be even bigger.  For Akamai this is not the end of the world, they are the giant of the sector with or without this particular contract.  But for Level 3 and Limelight, it could mean a very material change to their CDN fortunes in 2011.

One thing we can say is that Limelight and Level 3 may be adding substantial amounts of Netflix business, but whatever it is that they won appears to be asymmetric somehow.  Whereas Level 3 is forecasting a huge burst of $14M in capex in apparent response to the contract, Limelight is not expecting anything similar.  In fact, Limelight has not spent that much in a quarter on capex in any quarter I have checked, and when it comes to CDN capex alone I suspect Level 3 hasn’t either.  This suggests to me that either the two were tapped for different things, or that Level 3 is building network infrastructure while Limelight will be leasing it.  The two would have similar expenses if it were merely thousands of more servers in similar quantity.  

Probably it’s a mix of both actually.  If, as Dan says, Netflix had issues with Akamai’s performance although nobody else has mentioned such problems, then Level 3 may have promised to build its CDN into new geographies to improve that performance whereas Akamai was perhaps not so willing.  In other words, Netflix may have wanted better performance into markets that are further out in the internet boonies than most content customers have traditionally cared about.  A network operator with both network and colo already in place can do so much more readily for less popular markets where prices for space and bandwidth aren’t as competitive.  Netflix’s traffic needs may now be big enough to make that a worthwhile effort.

If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!

Categories: Content Distribution · Internet Backbones

Join the Discussion!

9 Comments So Far

  • carlk says:

    Although TWC remains a “wholesale customer” for (3), and “Mouse Lady” seems to have deduced the wrong data on the last call, I hope (3) is moving aggressively to help Time Warner with their CDN, too, because my stream being delivered by their t.v. system is a laggard!

    Moreover, I see their new offerings are attempting to “buy” more time from people abandoning their packages by providing new customers another “promo year,” including one year commitments on “premium channels” like HBO and Showtime, for example.

    On the other hand, I’ve read certain posters in more nebulous communities, referenced by a pseudo poster here, Enron, one who is up to his eyeballs in xoHO, are attempting to give some “advantage” to the telecoms in the last mile for internet, video and voice “bundles,” at this time. They believe people are fleeing their cable packages in lieu of the telecom package.

    In the big scheme of things, to me, this is absolutely wrong headed!

    Why? Because the larger CABLES do own as well as produce some of their own “CONTENT” in their own “STUDIOS.” As the shift to using the internet connection for streaming content to end user continues, this should reduce overall “net” costs to their customers whilst moving towards an ala carte purchasing menu instead.

    As partial proof, Comcast’s acquisition of NBC/Universal rings loud and clear.

    However, these types of costly acquisitions will inevitably become an “Achilles Heel” to these cable monsters as consumers, tired of seeing overpaid promoters of “self,” like Olberman, or O’Reilly, for example, flea traditional brands for better access to unobstructed, untainted, unbiased “data.”

    Stated differently, media will continue becoming very specialized or personalized in accordance with individual needs, as opposed to mass audiences models which have developed in order to control mind share historically.

    Untether the people from your DUMB EM DOWN BUNDLES!

    Free the people from your POP CULTURE PRISONS!

    This is the key to advancing Americans specifically, into the 21st century information age, one where individual expression, creativity and education, becomes focused and meaningful for all aspects of their work and play lives. imo

    • en_ron_hubbard says:

      ” I’ve read certain posters in more nebulous communities, referenced by a pseudo poster here, Enron, one who is up to his eyeballs in xoHO”

      waylong– just curious– why describe me as a pseudo poster? I am real and do post here and under the same nom de plume as elsewhere

  • VPG999 says:

    Of course AKAMAI will lose video streaming business, unlike Level 3 and Limelight they do not have fiber connecting super-RAID arrays of servers, they are JPG focused on suited for video.

  • Nice win says:

    Speaking of carlk’s wandering on TWC – you heard any rumblings on L3 winning the Sprint business that TWC is unwinding/bringing in house? It seems that would be much more material than a TWC CDN.

    • Rob Powell says:

      I have not heard anything along those lines lately, nor have such revenues seemed to show up in their results lately. Maybe next quarter 🙂

  • carlk says:

    If I can confirm En_ ron_ hubbard as your real name, you are not a pseudo to me. Until then, you always are.

    As a matter of fact, the name containing Enron, even as you choose to write it with “hubbard” at the end, keeps conjuring up thoughts of “accounting fraud” and “enterprise corruption” in my mind. That’s just a side note for credibility purposes.

    That’s a scary proposition to be affiliated with though, at the same time you attempt securities analysis.

    Is Rob Powell a pseudo to me as well? yes

    Why? Because I have never met him personally, nor seen him opine in a public domain video, in order to believe who he says he is when I see his image here. I can’t be sure his picture he is using on this site is who he says he is from my own perspective.

    Same goes for carlk or carlkj, waylong(3), valuecarl or whoever the hell else it is who I think I am!

    Just don’t call me JOHNSON, though! 🙂 imo


  • enbrodo says:

    All that Netflix traffic and bandwith consumption and the lost revenue for Akamai per Dan Rayburn’s estimates is only $10M-15M? Maybe Akamai just withdrew their commitment for more profitable biz. and use of resources. I laugh when I see the peanut gallery speak of Netflix buying their own CDN service. Why do this when their cost is only peanuts (or should I say popcorn) to them?

    • en_ron_hubbard says:


      below is an AKAM analyst’s estimate of the revenue impact– seems to tally with Rayburn’s view:

      Several media outlets (BusinessofVideo blog, Barron’s, Silicon Alley
      Insider) are reporting this morning that Netflix is moving its CDN
      services for streaming movies away from Akamai.
      Key Points
      We think it’s likely Netflix (NFLX, $169.13, Buy) will move all its
      traffic to Limelight Networks (LLNW, $7.70, Buy), at the expense
      of Akamai. Recall that Akamai signed a contract in late 2009 to
      deliver 51% of the traffic for Netflix’s streaming services, with the
      incumbent Limelight retaining 49%. Research by our colleague
      Yousef Squali indicates that Netflix has experienced several outages
      in its streaming network over the last 6 months, which the company
      apologized for. The public nature of these outages makes it seem
      likely that the media reports of poor performance by Akamai are true.
      Limelight’s network was purpose-built in 2002 to stream rich media
      and the company has long enjoyed a technological advantage for the
      various types of streaming, which include live, progressive, Flash and
      file streaming.
      A $15mm annual revenue impact (~1.5% of total revenue) seems
      reasonable, as this translates into about 63GB per month committed
      usage of Akamai’s network. While Netflix does not provide much
      detail on the size of its streaming services, we know from the past
      that Akamai’s largest customers have committed to upwards of
      267GB per month. Below is an analysis showing the potential
      committed usage at $10mm, $15mm, and $20mm in annual revenue
      to Akamai, assuming an ambient per GB price of $0.05.
      So while the Netflix contract is most likely immaterial to Akamai,
      the psychological damage of losing a marquee customer could result
      in a dip in AKAM shares today. The company is scheduled to host an
      investor day on December 9th in Boston. Maybe management will
      provide some color then.

  • Homer says:

    So… who is Blockbuster using to deliver its on demand content?

Leave a Comment

You may Log In to post a comment, or fill in the form to post anonymously.

  • Ramblings’ Jobs

    Post a Job - Just $99/30days
  • Event Calendar