Home > Distribution > Streaming Major Events

Streaming Major Events

Tim Siglin over at StreamingMedia.com has done a thorough review of NBC’s live stream of the Super Bowl.

Super Bowl Streaming Fail

As you might have guessed from his title, Tim does not have  ton of nice things to say about the effort to deliver one of the world’s most-watched sporting events over the web. There’s a lot there and I strongly recommend you read the whole thing, but here were my key takeaways from the article:

  • The quality of the stream was poor for large parts of the broadcast
  • The technology was limiting for many people on some widely used platforms
  • Cluttered display did not enhance the viewing experience
  • The time lag for online viewing grew to unreasonable levels

Naturally one would ask some questions about the event, and I think there are some important things to think about before streaming one’s own live event.

I imagine none of us will be streaming an event with a TV audience of 111.3 million (117+ million by the end of the game), but that doesn’t mean our audiences of several thousand (or even just a few dozen) want to struggle with a difficult or poor user experience. Presumably NBC was not looking to do a poor job, but one does have to ask if they had thought this all the way through before they opened the stream up to viewers.  So what should we consider before going live?

1. Have we tested, re-tested, then tested again?

The one live streaming event I was involved with was aimed at the organization’s Board. That by itself was reason enough to not only take it around the track and kick the tires, but to run the living daylights out of our setup daily for a month. There’s no question we would’ve been more confident if it was something we did more regularly, but you simply never know what’s going to blow up. By the time of the live, three hour event we had been testing for 8 hours a day for a long time. At the very least we had an audit trail to show we did our part. While there’s no way NBC could’ve tested the volume of users they probably saw yesterday (and I can’t at the moment find any numbers of online viewership), it’s important to make certain all your pieces are working as best as you can make it. Don’t wait for gametime to find out your quality is unusable.

2. Is our technology ready for the load?

In NBC’s defense, delivering this much data over the public Internet pipes is tricky at the best of times. They have to depend on infrastructure they don’t own, and compete with completely unrelated traffic. This is much, much easier for a smaller event if it’s being broadcast over a closed, self-contained network. You have the benefits of multicasting, restricted access, and control over every piece of the puzzle. That said, it’s incumbent for the manager to familiarize themselves with the network tools, and more importantly, the nice IT people who manage those tools. Is your network multicast enabled? How many people will be watching this? Do the IT people understand the parameters of what you’re doing and its impact on their network? Are they onboard to help you deliver the event smoothly and efficiently? Are your end users properly configured to receive the stream without specialty configurations? Are their local IT support people available and prepared to help the users out if needed?

3. What’s our backup plan?

If something goes wrong during the live event, is there anything you can do? Do you have failover systems you can re-route to? Do you have dedicated resources in place to address a problem immediately? Can local personnel facilitate an alternative discussion or portion of the program while everything is brought back online? Can you switch to an audio-only format until the video is back online? Viewers’ patience is limited, and the department and the organization’s reputation is at stake. Make sure some kind of contingency plan is in place.

4. Have viewers’ expectations been set properly?

As I’ve said elsewhere, viewers are increasingly more comfortable with online video, and YouTube and the like have set a level of expectation. We can watch our favorite videos on any device, at any time, and most of the time without any technical issues. A live, corporate event, on the other hand, may be a new experience for them. Corporate restrictions may prevent them from using their preferred device or format. They may be used to interacting with other users in real time or via a message board, functionality you may not be including in these circumstances. The lag time for online streaming is normal – do they know they may be getting the video many seconds or even minutes after it’s happened live? It’s very important to set the expectations correctly, and communicate it clearly and often. Especially where international events are in play – you don’t want to wake people up ridiculously early on the other side of the world and then deliver a tiny, tinny event surrounded by extra windows of information – unless you’ve warned them in advance.

5. What does success look like?

Failure is pretty easy to spot – complete and total breakdown of systems, phone/email ringing off the hook, steam coming out of the ears of the Executive VP. But what does success look like for your live event? Absence of failure? Thousands of comments on a social site? X thousands of viewers? Bits & packets delivered as expected? Any or all of these may be part of your success matrix, and it’s important to identify those factors before the event, and find a way to measure them during or afterwards. It should be more than bean counting – presumably the entire exercise is meant to communicate messages to the audience, and somehow the impact of that communication has to be measured. Clearly a happy CEO is a sign of success, but you and he may have different goals, and it’s necessary to figure that out before you ever start.

In the end, NBC and the NFL did a lesser job than I’m sure either party wanted. They have time and opportunity to fix the problems, and I would expect they will at the end of it all. In the interim, the rest of us can learn from their experience.

  1. Tim Siglin
    February 8, 2012 at 9:57 pm

    Thanks for the synopsis of the article; now that the numbers have been released, I’ve conjectured that the average second-screen viewer lasted just a few minutes, as the lag between TV and streaming was untenable. See the new article at http://www.streamingmedia.com/Articles/Editorial/Featured-Articles/Super-Bowl-Streaming-2.1-Million-Served-80491.aspx

  2. Dan
    February 13, 2012 at 10:59 am

    Thanks, Tim. As you may have already seen, I did catch your follow-up posting and blogged about it here: https://streamingenterprise.wordpress.com/2012/02/08/super-bowl-numbers/

    Keep up the good work!

    • Tim Siglin
      February 13, 2012 at 11:17 am

      Yes, saw that after I submitted the f

    • Tim Siglin
      February 13, 2012 at 11:18 am

      Yes, saw that after I submitted the first comment; thanks for following this interesting and still fluid topic.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: