Posts for 'Hulu'

continue reading

continue reading

Last week while I was in LA I had a chance to sit down for an extended chat with Albert Cheng, EVP of Digital Media for Disney-ABC Television. Aside from general catch-up, I wanted to dig into a comment I'd heard Albert make at the recent NATPE conference - that full-length programs on ABC.com are now achieving "DVR economics."

The comment caught my attention because, as I've written a number of times, I've been concerned that the broadcast networks' streaming initiatives (and Hulu specifically) could be undermining their traditional business models. The main reason for this is that the ad load in streaming programs is a small fraction vs. what it is in on-air. If all online viewing is incremental to on-air that wouldn't matter. But despite certain research that suggests online doesn't cannibalize on-air, for some viewers who have long since transitioned to time-shifted consumption, it surely does. More importantly, as convergence devices that link broadband to TVs gain penetration, the choice for viewers of how to watch a particular program - via online or via on-air - gets even more pronounced, putting further pressure on on-air.

Albert explained that ABC has been closely following the economics of programs' different viewing methods and recently concluded that it was more appropriate to compare online's economics to DVR's economics than to on-air's. Their reasoning is that because online is a "catch-up" medium it should be weighed against other comparable opportunities, not against on-air. Importantly, ABC "windows" the online release of its programs by 4-6 hours, so that hard-core fans who have to watch immediately will skew to on-air, rather than waiting. (Of course the question arises - in our increasingly on-demand, time-shifted world, how sizable is the "must-see" audience for all but the most popular programs like "Lost?" But that's a question for another day.)

When looked at this way, ABC believes online delivery compares favorably to DVR. No surprise, Albert would not disclose ABC's revenues or research, but he did give me a wink-and-a-nod when I shared my estimate that the on-air revenue per program per viewer is in the $.50-$.75 range (of course specific programs and specific episodes are above and below this range). To be clear, this only means the revenue generated is in this range. Because of bathroom breaks, channel flipping, viewers chit-chatting, etc. obviously not all of the ads are actually viewed.

Estimating the revenue per program per viewer range for DVR playback, given its attendant ad-skipping, is more complicated. Ad-skipping is surely high, but it's unclear exactly how high. For example, last Nov, the NY Times reported Nielsen research that somewhat remarkably showed that 46% of viewers age 18-49 still watched a program's commercials when in DVR playback mode. A different story is told by TiVo, which released data last Sept saying that for the programs that won the top Emmy awards, somewhere between 55-83% of the audience viewing these programs in DVR mode skipped the ads.

Just to round off, if we say that 60% of the ads in DVR playback are skipped, then DVR economics - and therefore ABC.com's economics - are in the $.20-$.30 range on a per program per viewer basis (i.e. 40% of $.50-$.75). Even on the low side of that range, that's better than my previous estimate of $.15 per program per viewer for Hulu in particular (which in reality was probably a little high anyway).

Further, Albert said that there's plenty of room for improving online's economics. One key focus is increasing the ad load, possibly to as much as double the current 5 ads per program. ABC.com has experimented with this and its research shows that neither the viewer nor the advertiser experience is diminished. As a result, ABC is inclined to increase the ad load to continue improving online economics further, but is somewhat constrained by advertisers' desire to minimize clutter and their own desire to remain consistent with non-ABC sites' ad loads.

Online distribution of full-length programs is still in its relative infancy. Yet as consumers hunger for it, broadcast networks have little choice but to provide it. The key is how to make this new delivery method profitable and also not harmful to the traditional network P&L. The use of windows for example, seems like an effective tactic insofar as there exists an audience intent on watching a program the moment it's shown on-air. Based on last week's conversation with Albert, along with prior ones, it seems like ABC is balancing things well - taking steps to pursue online, but doing so in a well-researched and analytically sound manner.

What do you think? Post a comment now (no sign-in required).

If you want to get a sense of how significant an inflection year 2009 was for online video, have a look at the chart below.

As you can see, according to comScore data, while Jan-Dec growth in 2007 (up 2.8 billion views or 39%) and 2008 (up 4.5 billion views or 46%) were impressive by any standard, the Jan-Dec 2009 growth of 18.4 billion views, up 124%, completely blows them away. Growth was so significant in 2009 that I think years down the road it will be pointed to as the year that online video really turned the corner.

But if that's the case, the question begs, "Why did growth accelerate so much in 2009 vs. prior years?" That's what I've been asked several times by industry colleagues since posting "comScore Data Shows 2009 Was a Blistering Year for Online Video" 2 weeks ago. It's a great question and though I don't have a really precise answer, here's my best sense of what happened.

No surprise, the most important contributor to the year's growth was YouTube. It zoomed from 6.3 billion views in Jan '09 to 13.2 billion in Dec. '09. That increase of 6.9 views accounts for 38% of the 18.4 billion delta between Jan and Dec. So what did YouTube do to generate such significant growth? Part of the reason is surely organic; more people uploading, sharing and viewing YouTube videos. But in 2009 YouTube also made strides in professionalizing the content on YouTube, broadening its value proposition to users. For example, its "Content ID" program, which lets media companies manage and monetize user-uploaded videos, has largely addressed the copyright infringement concerns from past years (the Viacom suit is a notable exception).

In 2009, among other things, YouTube also signed up Disney/ESPN, Univision and others as content partners, began implementing FreeWheel's ad system so 3rd party content providers could better monetize their views, engaged a number of leading brands to use it as a promotional platform, and with "YouTube Direct" engaged news organizations as partners. In short, YouTube continues to immerse itself into the fabric of the Internet. Whether users are viewing videos at its site or through its wildly popular embeds, YouTube has become omnipresent. YouTube now also claims to be the 2nd largest search site.

A second, but distant contributor to 2009's growth was Hulu, which saw its views increase by over 763 million from Jan to Dec, accounting for about 4% of the 18.4 billion increase in total views during that period. Hulu's mindshare leaped following its 2009 "Evil Plot" Super Bowl ad featuring Alec Baldwin and the subsequent ones. No doubt the addition of ABC programs throughout the year, plus other new content partners, also helped generate more viewership, along with the hugely popular SNL clips.

Once you get beyond these top 2 sites, the individual contributions to 2009's growth are more dispersed. The comScore data shows that across all video sites, usage intensified significantly during the year. For example, the number of videos viewed per viewer increased from 101 in Jan to 187 in Dec. The number of minutes watched jumped from 356 in Jan (almost 6 hours) to 762 in Dec (more than 12 1/2). There were also 31 million more U.S. Internet users watching video in Dec vs. Jan (178 million vs 147 million).

Looking beyond the numbers and thinking more qualitatively, it's also fair to conclude than in '09 online video reached a certain level of awareness that made it almost ubiquitous. There is just so much video online, and it is shared so widely, and highlighted so frequently by mainstream media, that it is unavoidable, even for the least technically-savvy among us. People are increasingly entertaining themselves with online video, but they're also finding new uses for it in their daily personal and professional lives.

I think it's unlikely we'll see the same level of growth in 2010 as in 2009, but I do believe the growth curve over the next 5 years will be very steep. The primary contributor will be convergence devices (e.g. game consoles, Blu-ray players, Roku, etc.) that are bridging online video to the TV where longer-form consumption will be the norm. Another key contributor will be TV Everywhere services, which are just now getting off the starting blocks. Lastly, I think growth in mobile consumption will be another important contributor. Add them all up and the 33.2 billion videos viewed in Dec. '09 will look relatively small 5 years from now.

What do you think? Post a comment now (no sign-in required).

Daisy Whitney and I are pleased to present the 49th edition of the VideoNuze Report podcast, for February 12, 2010.

This week Daisy and I dig into the 2009 comScore data that I detailed in my post on Tuesday (slides available for download too). It was a blistering year for online video, with total streams growing from 14.8 billion in Jan '09 to 33.2 billion in Dec '09. All the other relevant metrics also recorded strong growth. I share more details on the numbers and what they mean, focusing particularly on the top 2 sites YouTube and Hulu.

Then Daisy discusses her takeaways from the recent iMedia conference she helped organize. She talks about how brands are trying to break through the clutter, and the role of online video ad networks. Finally, she also discusses recent interviews she conducted with Facebook executives.

Click here to listen to the podcast (13 minutes, 55 seconds)

Click here for previous podcasts

The VideoNuze Report is available in iTunes...subscribe today!

Last Friday, comScore released its Dec. '09 data for online video usage. I've been tracking comScore's data for the last 3 years and Dec put an exclamation mark on what many of us already knew: 2009 was a blistering year of growth in online video consumption. Below are graphs of the most important data (Click here if you'd like a complimentary PDF download of all of the slides.)

The first graph shows total online video views more than doubled from 14.8 billion in Jan '09 to 33.2 billion in Dec '09. The historical growth is even more impressive. Just two years ago, in Dec '07, comScore reported 10 billion video views.

Online video usage is now nearly ubiquitous in the U.S. According to comScore, in Dec '09, 86.5% of all U.S. Internet users watched online video, up nearly 10 percentage points from the 76.8% in Jan '09. That translates to 178 million people watching video in Dec '09, up from 147 million in Jan '09. Back in Jan '07, there were 123 million viewers.

Those users are watching a whole lot more videos as well. For Dec '09, comScore reported that 187 videos were watched per average viewer, up 85% from 101 in Jan '09, and more than triple the 59 watched in Jan '07.

As well, those viewers spent a lot more time watching online video. In Dec '09 comScore said that the average online viewer watched 762.6 minutes or 12.7 hours, more than double the 356 minutes viewed on average in Jan '09. Here's the really incredible stat: back in Jan '07, comScore pegged this number at just 151 minutes or about 2 1/2 hours, meaning average viewing time has more than quintupled in the last 3 years.

I've talked many times about how YouTube is the 800 pound gorilla of the online video market, and 2009 only further cemented this. Videos viewed at YouTube surged from 6.3 billion in Jan '09 to 13.2 billion in Dec '09. To put this in perspective, Google closed its acquisition in Nov '06. In Jan '07 (the first month comScore publicly released online video data), YouTube notched 1.2 billion views. That means that in the 3+ years that Google has owned YouTube, it has grown more than 10x in size. More amazing is that even with all the growth by other sites (particularly Hulu), YouTube has kept up its approximate 40% share of the overall online video market, starting the year at 42.9% and ending at 39.8%.

Speaking of Hulu, in its first full year of operation, the site surged from 250 million views in Jan '09 to 1,013 billion views in Dec '09. Unique viewers increased from 24.4 million in Jan '09 to 44.1 million in Dec '09. But if you look at the red line in the graph below, you'll see that uniques jumped to 41.6 million by Mar '09 which I believe must be due, at least in part, to a likely measurement change by comScore. Since Mar you'll notice that uniques hovered right around 40 million each month, dipping below during the summer and then bouncing back in Q4.

A few months ago I speculated that Hulu's relatively flat pattern in uniques could suggest that, in its current configuration, Hulu may have saturated the market for its content and user experience (for example, contrast Hulu with YouTube, which grew its uniques by 33% in '09 to 135.8 million by Dec '09). I'll be looking to see if Hulu can notch more noteworthy increases in uniques during '10; if not, then I think my thesis will be proven correct.

Nonetheless, Hulu's viewers clearly love the site, with average number of videos per viewer more than doubling to 22.9 in Dec '09, up from 9.8 in Dec '08. Users are spending more time on Hulu, increasing the amount of total minutes on the site from 58 in Mar '09 to 132 in Dec '09. What's remarkable though is that the average minutes watched per video (the yellow line below), has stayed virtually constant at around 6 minutes each month. That shows that while there's plenty of long-form consumption happening at Hulu, clips are still very popular too.

comScore is a great source of month in and month out online video data, but as always my caveat is that no third party can ever track usage as closely as the sites themselves, so take these numbers with a small grain of salt!

Click here if you'd like a complimentary PDF download of all of the slides.

What do you think? Post a comment now (no sign-in required).