NBC completely failed in their execution of streaming Sunday's Super Bowl on the web, yet they want us to believe that it was an engaging and successful webcast. Do they think we're stupid? Instead of coming clean and saying they had technical issues, especially with the quality of the video and re-buffering, the company is trying to blame these issues on the "last mile" networks and are pointing the finger at someone else. NBC should know better than this and should be ashamed of themselves.

Tim Siglin has a great article at StreamingMedia.com that re-caps all of the problems with the webcast, so I don't need to repeat a lot of what he said, go read his article. What I will add to Tim's piece is that NBC does not get that the success of any webcast is judged by the message you deliver, the experience you provide, and how you engage the user – not simply how many streams you deliver. Would you rather reach a lot of people with a poor experience or fewer people with a good experience that keeps them engaged longer?

NBC was quick to say how many streams they got, but didn't say how many simultaneous streams they did. Most webcasts are measured by simultaneous streams, so why isn't NBC giving out that number? They called the event a record and they are quoted on news sites as saying things like, the event “exceeded our expectations in every way," and that it was a, "tremendous success". Exceed their expectations? If poor quality video with bad re-buffering is something that "exceed their expectations", I'd hate to see how low they set the bar internally.

If the quality issues were as a result of the last mile like NBC claims, why were users like me able to stream videos from other sites with perfect video quality during the Super Bowl webcast, but not from NBC directly? That's such a cop-out on NBC's part and any viewer who played videos from others sites during the same time could easily see it wasn't a last mile issue. In addition, while NBC isn't saying what bitrates the Super Bowl was encoded for, it looks like the average bitrate was about 2Mbps. So how could I stream clips at twice this quality from other websites during the Super Bowl if there were last mile issues? Of course I couldn't, but I guess NBC thinks we won't notice that.

All NBC had to do was come clean and say they had issues, they screwed up and they have learned for next time. While failure is no good, at least no one could claim that they aren't in touch with reality. Also, I find it very strange that as I write this post, Akamai is doing their quarterly earnings call and has referenced the Super Bowl webcast as an example of a "great experience", when it was actually a very poor experience. Akamai was the CDN provider for the online webcast, so I am not surprised they are also saying the same thing as NBC since that's their customer.

If the webcast was such a "success" and was a good quality experience like both companies suggest, why isn't NBC or Akamai releasing any video quality assurance data from a third party like Conviva? Myself and others can show, by example, all the issues we had with the webcast, so where's NBC's data to show it was a last mile issue? Their data doesn't exist.

In Tim's article he says "it's disingenuous for NBC to blame last mile issues" and I'd say it's even worse than that. Streaming media technology is not new, it's not cutting-edge and we've had companies webcasting live events, with success, for over 15 years. So for NBC to encode such a low quality video to begin with, and then try to blame the last mile networks for video quality and buffering problems, someone should hold these guys accountable. So nice try NBC, but there are plenty of us who are not buying it.

I have been invited to host a dinner for buy side analysts, next Wednesday February 15th in NYC. The event will start at 6pm and the location is midtown. It will be an informal event where I will be presenting some of my latest pricing data and talking about Akamai, Limelight and Level 3's recent earnings as well as answering questions about technologies like front-end optimization and dynamic site acceleration amongst others. It will be interactive, so you'll also have your chance to get your specific questions answered.

If you would like to attend, please contact me and I will put you in touch with the group that is organizing the event. Space is limited and currently, it is only open to buy side analysts.

This morning Akamai announced they have acquired Blaze Software, a cloud based provider of front-end optimization (FEO) services. Terms of the deal weren't disclosed, but the numbers will come out, maybe even by Akamai during their earnings call today. Either way, I'll update the post once I hear what value was placed on the deal. FEO is going to be a big push from all the CDNs this year and Akamai needed to acquire someone to add a FEO product to their portfolio. This acquisition is probably also to combat Level 3 which has been working closely with FEO provider Streangeloop and has been having success in taking some big deals away from Akamai as of late.

FEO might sound similar to another subject I have written about lately, dynamic site acceleration (DSA), but it's very different. DSA's focus is to bring network resources closer to the user by pre-fetching or caching files. FEO makes the content itself faster. DSA makes page resources download faster. FEO reduces the number of page resources required to download a given page and makes the browser process the page faster. For example, analysis shows that popular sites like CNN, who already use a CDN, can double current performance by implementing FEO.

There are a lot of small FEO vendors in the space and many of them have now been acquired or teamed up with larger CDNs. While some might think that Akamai's recent purchase of Cotendo gave them a FEO product, that's not something Cotendo was doing. Blaze is now owned by Akamai, Limelight acquired AcceloWeb, Strangeloop is working with Level 3 and Riverbed acquired you also have stand-alone FEO services from Aptimize. Here's a slide that helps to break out where all the vendors fit in, but these lines are quickly blurring as many of the larger companies are taking out the smaller ones and adding this kind of functionality to their product suite.

For content owners, it used to be that the answer to most website performance problems was either to add more hardware, use a CDN, or re-engineer their backend application code. But for a number of reasons, investing in backend optimization is now providing diminishing returns. Google's research shows that for many popular sites it's the front-end that accounts for over 90% of a users wait time. Content delivery networks help to address part of this problem by reducing network latency, but even larger performance gains can be achieved through front-end optimization techniques that streamline the Web page HTML code and resources.

If you want to learn more about how front-end optimization works and what role it plays within a CDN, see my blog post from last year entitled: "Why Web Applications And Mobile Browsing Are Making The Frontend A Major Performance Bottleneck".

We will be talking a lot about FEO and seeing demos of these services at the Content Delivery Summit, May 14th in NYC.

Yesterday, Brightcove filed an amended S-1 as the company prepares to have their IPO in the coming months. The company is looking to raise between $50-$60M from their offering and estimates shares will be priced between $10-$12. After IPO expenses and spending approximately $7M to repay an outstanding loan, the company looks to net about $40M in working capital. Depending on the share price, Brightcove would have a market cap of between $275-$300M, which is about 5x 2011 revenue.

While I saw lots of folks repeating the highlights from the filing in their blog posts and simply cutting and pasting the numbers, I didn’t see anyone asking the questions that have yet to be answered regarding Brigthcove’s revenue or the size of the market they are in. It’s also clear that many folks only read the first few pages to the filing and not the whole document as their are lots of data points that gives one a lot to think about, and question. It’s just another example of many showing that too many blogs only care about publishing as many 500 word posts as possible in one day, rather than actually telling a story, engaging a reader, or trying to explain the impact a company could have on the broader market.

Brightcove had revenue of just over $63M in 2011 which makes them the largest provider for online video platforms, sold direct to content owners. Other video platform vendors selling direct to MSOs like Cisco, thePlatform and Kit digital have more revenue than Brightcove, but aren’t selling into the same vertical with the same solution. Many people want to compare a Brightcove to thePlatform or Kit digital, but they are selling very different solutions and should not be compared to one another.

There are three kinds of online video platform (OVP) providers in the market. Those that sell direct to content owners, those that offer carrier grade solutions to MSOs and vendors that offer video management platforms for enterprise customers who deploy video inside their firewall. Brightcove has been going after the enterprise market as of late, but most of their customers are still content owners and publishers in the media space and I have yet to see Brightcove enter the carrier and MSO vertical as of yet.

But the real question is what’s the breakdown of Brightcove’s revenue and how much of it comes directly from their SaaS business? Through Brightcove’s CDN partners like Limelight Networks and Akamai, almost 9 billion streams went through Brightcove’s platform last year. That’s a lot of bits and since the delivery of those videos are handled by a third party, Brightcove pays the CDNs for those delivery services and marks up the bandwidth to their customers. So one has to wonder what percentage of Brightcove’s total revenue comes from the re-sale of bandwidth? Brightcove doesn’t break that number out in their filing and the company told me they could not comment on those details when asked.

While there is nothing wrong with some of Brigthcove’s revenue coming from bandwidth, considering that the average cost per GB delivered drops at least 20-25% each year, there is only so much of a markup Brightcove can add on top of what they pay the CDNs. Some have told me that it really does not matter as the percentage of Brightcove’s revenue that comes from the re-sale of bandwidth continues to decline each year, but that’s a bad answer. Just because it is declining doesn’t mean the number isn’t material. I don’t know what percentage of their revenue is from bandwidth and I won’t put a number out there as it would be a complete guess on my part. But if the number was low, think single digits, then my guess is that Brightcove would be all too happy to say what it is.

While Brightcove admits they will lose money in 2012 and even possibly beyond, that’s not what worries me. Brightcove is trying to follow the Amazon mentality of “get big fast’ and that’s why the company has raised and spent so much money over the past few years. My biggest concern is how quickly the market Brightcove is in will grow and what the total market opportunity really is. While Brightcove gave out some of those numbers in their original filing, they are so big that they are hard to believe. Brightcove says that by their estimate their “total potential market opportunity was approximately $2.3 billion in 2011, growing to approximately $5.8 billion in 2015“.

Those numbers are far too big. If Brightcove’s entire market opportunity was $2.3B in 2011 and the company did $63M in revenue, that means Brightcove only captured 2.7% of the market. That’s not a lot. Brightcove has some really interesting methodology on how they came up with the total market size though, which you can read in detail on page 69 of the S-1 filing. My other concern with any business like Brigthcove’s is that R&D costs are high. In 2011, 25% of Brightcove’s revenue went towards R&D, which is a lot. Also, while all of Brightcove’s revenue comes from one product and the company is working towards diversifying their revenue, app cloud is an example, they have yet to do that to any large degree.

All of this aside, Brightcove does have a some positive things going for them which gives them a legitimate shot at building a long-term profitable business. For starters, Brightcove invented the market they are in and to date, even with increased competition, has yet to fall behind any of their competitors. They have lead the market for more than five years and while being a first mover is not always a guarantee of long-term success, so far, Brightcove has shown that they can continue to be the leader.

Brightcove had a total of 3,872 customers at the end of last year of which 2,571 of those customers were using the Brightcove Express product which costs between $99 and $499 a month. If those customers are all paying $250 a month on average, that means that about one third of Brightcove’s customers make up about 91% of their revenue, which is very good. At the same time, no one customer makes up more than 4% of Brigthcove’s total revenue which is a nice balance. Another thing I like is that in each of the last eight fiscal quarters, Brigthcove’s recurring dollar retention rate was at least 86%, and for all of 2011, was higher at 93%, which is quite impressive.

And finally, in order for any company to do well they have to have a good management team that has experience in building and cultivating their business, especially when it is still in the early stages of real growth. Brightcove isn’t even a $100M company yet so for them to really get to the next level with their business, management needs to be laser focused and take the necessary steps to make sure the growth happens. If there is one thing I think Brightcove is very strong with, it’s the management team they have in place.

Brightcove’s CEO Jeremy Allaire is smart, he’s easy to speak to and reachable (even if you are not a member of the media) and he doesn’t have the arrogance a lot of CEOs tend to have, which can sometimes hurt the growth of their company. You don’t see Brightcove getting into public spats with their competitors and the management team that Jeremy has built at the company has a lot of experience in building and scaling software companies in the past. That alone does not guarantee that Brightcove will be successful, but it sure helps gives them the opportunity to make it possible and they are going to be closely watched by a lot of other companies in the space who want to follow in their footsteps.

This evening, Limelight Networks announced their new managed CDN offering, called Limelight Deploy, aimed at helping telcos, carriers and service providers build and deploy CDN services inside their network. While this is a new product line for Limelight, they aren’t new to the carrier space and have already built out networks for Bell Canada, starting in 2009, and Asia’s largest telecom provider, Bharti Airtel in 2010. In what I think is related news, I have gotten a lot of calls today asking me about rumors people are hearing about F5 buying Limelight Networks. I have not heard any of those rumors, and it makes no sense, but my guess is that F5 may be working to license Limelight’s CDN technology the same way Juniper just licensed CDN technology from BitGravity. But that’s just a guess.

The licensed and managed CDN business has really been heating up over the past few months and is still dominated by EdgeCast who has the most carrier based licensed CDN deployments and re-sellers in the market. But unlike EdgeCast who simply licenses their software to carriers, Limelight’s approach it to actually help build and manage the CDN component of the carriers network with their professional services group. In conversations with Limelight, the company said carriers were asking for more hands-on help and Limelight’s goal is to help the carrier build and manage the network with the carrier eventually taking over 80% of the operation of the service on their own. This approach also allows network operators to get their CDN up and running much faster since they get to leverage Limelight’s expertise of running a network over the past ten years.

Customer’s who use Limelight’s Deploy solution are not required to have to re-sell or use Limelight’s off-net services, but that’s clearly an option if they want to. Limelight also said they have no desire to create a federation of CDNs, like EdgeCast has done, although nothing would stop Limelight’s customers from exchanging traffic amongst one another if they wanted to. Limelight said their Deploy offering will be offered worldwide and that they now have a dedicated team inside the company, focused just on their managed CDN offering.

There are two primary uses cases for a managed CDN and not ever customer is trying to “reduce their cost” like you hear people say. The first use case is a carrier that is looking to add CDN services and expand their product portfolio with products that are adjacent to services they already offer on their network, thereby increasing their revenue. The second use case relates to the network operators who have the eyeballs and are concerned with the volume of video being delivered over the last-mile to their customers. In this case, most of them are trying to control their costs, improve their quality of service (QoS) and create content services that will allow them monetize the video they are delivering. (think Xfinity) Limelight says that customers for their managed CDN offering range in size and could be asking for help to build out small deployments in the 10Gbps range to customers who have the desire to support 700Gbps within three years time. Not surprisingly, Limelight and all the managed and licensed CDN vendors I talk to say the biggest demand for these services comes from outside the U.S.

The terms “licensed CDN” and “managed CDN” will probably start being used in the industry as if they are interchangeable, but they are really very different solutions. Right now, there are basically three different kinds of licensed, managed or carrier based CDN products in the market. EdgeCast, Jet-Stream and Aflexi offer software based licensed CDN services. Limelight Networks and Highwinds offer managed CDN services which go beyond a software license and vendors like Juniper and Cisco are or will offer carrier based CDN solutions that combine hardware with software, but won’t help the customer actually build the CDN. Alcatel-Lucent is also in the business, but is targeting MSOs more than carriers. Other vendors like Akamai have talked about a licensed CDN product but have not yet entered the market.

Limelight’s Deploy offering is somewhat unique in the market as their platform isn’t only focused on helping carriers build their own CDN for video delivery, but also for website and application acceleration services. EdgeCast also has this functionality in their application delivery network, which can also be licensed by carriers and it is expected this is what AT&T will license when they stop re-selling Cotendo once Akamai completes the acquisition of the company. Vendors like Juniper are focusing more on the video side, but then also offer a transparent caching component that some of the other managed and licensed CDN vendors don’t have. If you’re a bit confused by all of these different technologies, you’re not alone.

In the past two years, it’s my opinion that the CDN market has been pretty stagnant, with not a lot in the way of new services being offered in the market. The term “CDN” usually referred to the delivery of video content and small objects as a service based business, but that’s about to change. Content delivery products and offerings now encompass services like video delivery, licensed and managed CDN, dynamic site acceleration, front-end-optimization (something you will hear a lot more of soon), mobile content acceleration and a host of other services. These days, calling a product or a company a “CDN” offering, could mean a lot of very different things.

The CDN industry is starting to get really interesting again and we’re starting to see lots of new types of CDN services and products coming to the market. We’ll be covering all of these CDN technologies at the fourth annual Content Delivery Summit, taking place May 14th in NYC. You can come see these technologies in-action, talk to the carriers that are using them and hear about real-world deployments.

Realted Posts:

– Updated List Of Vendors In The Content Delivery Ecosystem

– An Overview Of Transparent Caching and Its Role In The CDN Market

– Transparent Caching & CDN Merging: Juniper Licenses BitGravity’s CDN Technology

– Here’s Why The Future Of The CDN Business Is In Mobile Content Acceleration

This morning, California based Conviva, a video quality optimization and analytics platform, announced a series D round of funding totaling $15M, lead by Time Warner Investments. Founded in 2006 as Rinera Networks, the company changed their name to Conviva in 2008 shortly before Darren Feher, former CTO of NBC Universal took over as CEO of the company. The company is not saying how much of the latest round of funding comes from Time Warner Investments, but did say that it was the "majority" investor. To date, Conviva has now raised $59M.

Unlike a CDN, Conviva does not deliver the video themselves but rather works with content delivery networks and content owners as a software based overlay service. Conviva's software monitors the performance of end-users video quality and not only records the experience but can also manage, in real-time, the shifting of traffic amongst multiple CDNs. Conviva has deals with both Level 3 and Limelight Networks who uses their technology and has a nice roster of content owners as customers including HBO, NBC, MLB, Netflix, MTV, ESPN and is also used for a lot of one-off live events like the 2010 Winter Olympics and the FIFA World Cup.

Conviva is one of those interesting companies in the industry because no one really competes with them in the market and as a result, many don't really know what bucket to put them in or how they should be classified. I don't see a lot of companies trying to develop a similar solution to Conviva's mostly because the market size for SaaS based video quality optimization services just isn't that big. That said, if the market grows fast enough and Conviva is the only major player, the company could have a nice exit strategy, getting purchased by a larger entity for their technology. At the same time, Conviva has raised $59M in five years and from what I hear, still isn't profitable. So if the market does not grow fast enough, Conviva could have a difficult time growing their revenue to meet their VCs expectations.

On the surface, I like what Conviva is doing as their whole goal is to improve the video quality for the end-user, in real-time, without anyone knowing what's taking place. Whether it's bad buffering, a frozen encoder or some problem on the network, Conviva's platform brings the intelligence to proactively switch servers mid-stream, change the bitrate and apply policy filters, all on the fly. This is a solution that I really would have liked to have had in the market ten years ago when video quality on the Internet was poor and there were very few ways to correct the problem in real-time. But today, with the quality of video being delivered from the major CDNs getting better each year, I just don't know how much of a demand there is for Conviva's solution, since video quality is only getting better overall, not worse.

Conviva's solution is not affordable for a small or medium sized content owner and that's why to date, Conviva's customers are mostly the larger content brands on the web. That could be a good thing for Conviva if the company can ramp revenue fast enough, but their are also fewer big content brands on the web than there are small and medium sized business, which is probably why Conviva is now working directly with CDNs to have them offer this functionality to any of their video based customers. Content owners I have spoken with who use Conviva's solution like how it works, but many also don't use it exclusively as their only video optimization and measurement platform. For example, Netflix and MLB both have their own internal platforms that offer a lot of the same functionality as Conviva, but Conviva offers them the insight of a third party.

To date the company has done very little in the way of marketing and most of the content owners I speak to haven't heard of the company, don't know what they do, or think they are a simple analytics company – which they're not. The company does much more than analytics and their solution really does work. But Conviva needs to get more brand recognition and exposure in the market and hopefully, they will use some of their new funding to increase awareness with content owners.

I like any solution that's designed to improve the quality of video being delivered to end users, so I'd like to see Conviva be able to become a profitable vendor in the industry soon. But I just don't know how big of an opportunity there is for this kind of service in the market, how quickly Conviva can grow or how much content owners will value this service, and be willing to pay for it. All that aside, Conviva is an interesting company to watch and hopefully soon we'll be able to get a better insight into their business.

Updated: Conviva contacted me to say that the majority of their revenue does not come from live streaming, but that it is mostly recurring revenue from video on demand content.

A few weeks ago Juniper announced they had licensed BitGravity's CDN technology and plans to add the functionality to their transparent caching Media Flow solution. Terms of the deal were not disclosed, but I hear that there is a revenue sharing component between Juniper and Tata Communications, which owns BitGravity. Juniper does not plan to re-sell any of Tata Communications CDN service for off-net delivery and is simply licensing the service management layer of BitGravity's CDN platform, on a non-exclusive basis, for Juniper's own product, expected to be introduced later this year.

As service providers look for products and solutions that deliver more than just one type of content, more deals like this need to take place, especially from the hardware manufactures. Last year, we saw licensed CDN provider EdgeCast team up with transparent caching provider PeerApp to merge their platforms for telcos and carriers and I suspect other vendors in the space are looking at similar partnerships.

The CDN space is finally starting to evolve once again with vendors deploying all kinds of solutions like dynamic site acceleration, front-end-optimization, transparent caching, licensed CDN and application acceleration, which now all fall under the generic term of "content delivery". After year's of the CDN space being pretty stagnant, lots of new solutions are coming to the market that combine many different technologies and types of content delivery, into single platforms. This is exactly what customers are looking and asking for be it inside the carrier network or for off-net CDN based services.

We'll be talking a lot more about all of these different content delivery technologies at the Content Delivery Summit, taking place May 14th in NYC.