• Follow Curiously Persistent on WordPress.com
  • About the blog

    This is the personal blog of Simon Kendrick and covers my interests in media, technology and popular culture. All opinions expressed are my own and may not be representative of past or present employers
  • Subscribe

  • Meta

Google’s Think Video event

Yesterday, I attended the Google/Youtube hosted event “Think Video”. It was an event primarily aimed at marketers (research seminars tend not to have goodie bags), but I found it an interesting – if not groundbreaking – session.

Below are some of the notes I scribbled down during the talks. I’ve linked or embedded the presentations, where they are available.

Quantitative overview of the market

After an introduction from Dara Nasr, Dan Cryan of Screen Digest gave the first talk with a speedy overview of the current state of the online video market, and where it is heading.

Some statistics and projections I noted include (note that I was having to write quickly to keep up with Dan, and so there may be some errors in the below)

  • 90% of online views will be user-generated content in 2014 – about 2.5bn viewing hours
  • In 2014, ad-supported revenues will be around £180m, transactional VOD £40m and download to own £10m.
  • At this time, online TV will still only be 2% of the total TV market
  • Currently, 67% of online video viewing is short form, since “work time is prime time”
  • At the moment, 80% of the UK online video revenues are with the broadcasters. In the US, it is only 57% but Hulu represents the majority of the difference (Hulu looks like it is losing a lot of money at the moment, since 70% of its revenues goes to the broadcasters.

I found the most interesting part of the talk to be about distribution strategies, and whether to affiliate and syndicate or not. It would appear to be a good move, given the increases in reach that such a tactic produces. Quoting Comscore figures, Dan showed that in the US

  • ABC has 4.7m monthly site users but 9.9m monthly video viewers
  • Youtube has 98.2m visitors but 120m viewers
  • Hulu has 8.4m visitors but 38.1m viewers

Within this distributed model, there are two main options. Hulu operates the “player as the platform” strategy, where it allows its player to be embedded on sites such as MSN and Yahoo!, enabling it to retain control over advertising sales revenue. The alternative – which Channel 4 and Five appear to be pursuing, is “content as the platform”, where they licence their content to other video sites to include in their players.

The power of social

Neil Perkin of Only Dead Fish followed this talk with his keynote on the importance of social and data in online video. The presentation can be seen below (RSS viewers may need to click through)

Some of the key points I took from the presentation were:

  • Businesses should get their hands dirty – Neil has learned about the space by participating in it
  • Cisco say that in 2013 90% of data will be video, while 70% of data will be created by individuals. For businesses to succeed, they need to derive value from data – of which metadata is the fastest growing category. I’m a big believer in this.
  • However, businesses also need to understand social. Media brands are now less defined by the platform and more by their community (I think this is true, and we are increasingly seeing companies seeking to “own” a particular audience – whether the Guardian, Channel 4 or NME)
  • Attention is no longer the only key metric – we also need to consider participation, content and interaction
  • Distribution should be wide, with content scalable and portable. Slippy, not sticky
  • Companies need to loosen their grip on the creation process and let the community interact in pre-production, actual production or post-production.
  • Ultimately, convergence is about content flowing across channels.

Youtube’s development

The final presentation came from Bruce Daisley of Youtube, who coped admirably with technical issues that prevented his presentation being visible for the first five minutes of his talk. His presentation can be viewed via Google Documents here.

He mentioned how Youtube are looking to reinvent the experience for premium long-form content on their site. They want Youtube to be seen as a revenue opportunity by content owners, not a threat. He believes Youtube’s strengths come from 4 R’s.

  • Reach – it’s not necessarily as big as TV, but it can help build audiences. He cited Britain’s Got Talent audience figures rising from 8m in episode 1 to 12m in episode 2, fuelled by SuBo fever (of course, this can’t be blamed solely on Youtube)
  • Rights – he used Monty Python as an example of how content owners were licencing their material, and creating a viable business model
  • Research – Youtube can be seen as a virtual research community (Bruce called it an online focus group), with different creatives compared in terms of views or ratings
  • Revenue – whether through pre-rolls, in-video adverts or the front-page masthead video ad. He showed that searches for Avatar on Youtube actually peaked during the masthead campaign, and not when the film opened a few months later.These can also be integrated with Facebook, so that you can “like” an ad within Youtube.

He then quoted a range of research and statistics to highlight some of Youtube’s strengths

  • Social – Two thirds of users say they’ve sent a clip onto someone else. 10% of video views are from embedded clips on social media websites
  • Reach – a medium weight TV campaign of 160 GRPs aiming for 1+ cover among adults would reach 54% of the audience on TV. On Youtube it would reach 9% – 2% being incremental reach. This figure is higher among younger and more affluent audiences.
  • Mobile – Although mobile and TV only represent 2% of total views, iPhone users will look at an average of 3-4 clips per day, and Android users will look at 10.
  • Engagement – Ipsos have created an engagement metric across different media channels, accumulated from “around 50 metrics” (I’d be interested in hearing more about this). Of the channels shown, Youtube had the highest proportion within the high engagement band, and the least in the low (followed by BBC brands, then Channel 4, Five and ITV).

Youtube are now working with more content owners and broadcasters, with material ranging from films to the Indian Premier League. Bruce mentioned that the first company they negotiated with – the BBC – were very cautious and were arguing for removing comments, ratings and ability to embed, but gradually companies are warming to them.

He closed by talking about Youtube Bubble – a Youtube channel accumulating the best of advertising on the site. I would imagine the Evian babies dancing to Rapper’s Delight (with over 58m video views worldwide) heavily features.

Panel discussion

A Q&A session was chaired by Matt Brittin – the Managing Director of Google in the UK – who was very engaging (and puntastic). The speakers were joined by someone from Channel 4’s sales side (Ed someone, I believe) and Ben Chesters from Mediavest.

Some disjointed points I noted down during the session were:

  • Channel 4 were wary about giving specific figures in viewing across different channels, but did say that the total reach was more important than what individual channels made, since they controlled the advertising across each.
  • Ben Chesters mentioned that advertisers default to BARB for cover and frequency metrics, and that an incremental reach figure would be the killer feature needed for advertisers to trust online video
  • The first TV ads were like display advertising but with a voiceover. Currently, online video adverts are mainly repurposed 30 second spots. Gradually, this should change.
  • Neil Perkin talked about how transmedia storytelling across platforms should be emphasised. Content shouldn’t be copied onto different platforms, but should be unique to take advantage of different strengths.
  • Neil Perkin also mentioned that a big challenge for online is that it lasts forever – it is not campaign based and you can’t control who sees your spot when and how.
  • Channel 4 are trialing teaser 2-3 minute clips of forthcoming comedies to gauge reaction.
  • Answering a question from the audience, Dan Cryan mentioned online video was seeing a renaissance of the soap opera – particularly in the United Stated where sponsored and branded content is increasingly prevalent
  • And answering a question from Nick Burcher regarding the culling of Downfall mash-ups on Youtube by the studio despite the approval of both the director and the audience, Dan Cryan said “you would have to be a brave man to bet against the anarchy of distribution”. It may not always be timely or the best quality, but people will always find a way to watch what they want.
  • A final question concerned the film market online. Dan Cryan mentioned the primary barrier was the windowing of films – films make most of their money on DVD within the first 12 months. During this time, they won’t be available to view on ad-funded sites, but will gradually transfer to parallel availability on pay per view or download to own sites – though this would be more likely to be situated on a TV than a computer.

From a personal standpoint, more focus on the metrics of success would have been welcomed, but as addressed in the Q&A, this is a very thorny issue.  I would also have liked to have seen more from Youtube on how they are breaking down the distinction between TV and online and innovating in advertising formats – this Wario advert is one of my favourites – but nonetheless it was an interesting afternoon, with some useful tidbits from each of the speakers.

sk

Reblog this post [with Zemanta]
Advertisement

The four types of Online Video – which is best to advertise around?

This has been something I’ve been pondering for some time. It is still a work in progress, and feedback or suggestions are welcome.

I believe that online video can be categorised into four broad categories:

Reference – Largely, this is the how-to guides such as Videojug which relay advice and practical tips. They will have a steady, but most likely small, stream of visitors looking for specific content. For how-to, specific content is related to the genre or topic but not necessarily the producer.  I am also grouping the long tail of video on demand into this category. Arguably it could be a fifth type as the content producer is now key, but I believe the specific nature of the search is enough to group it within this category. People that watch Buffy on the WB (US only) will go there specifically looking for Buffy – few viewers will arrive there via another method with another show in mind

Scheduled – Web series and TV catch-up fall into this category. The Secret World of Sam King had a new video every weekday; the latest episode of Spooks (UK only) arrives on the iPlayer shortly after it is aired on BBC1. Similar to reference videos, people will seek out scheduled videos with the specific content in mind. The key difference is that time is now as important as a mindset, and viewers are more likely to visit after an external prompt – such as a TV guide for catch-up, or email reminder for a web series – rather than a simple desire to view.

TopicalThis differs from reference videos because they represent the long tail, whereas topical videos are the short head. TMZ will get a spike in traffic whenever a celebrity has a “moment” (such as Mel Gibson or Michael Richards), and the news sites will see growth whenever there is a major story such as an election (NB: the link refers to unique users, but the trend holds for video). Topical videos continue to get small levels of traffic in the long-term, but nowhere near the levels of when the story is breaking. Portals and news sites will be the primary vehicles for this type of video as their superior resources will ensure the fullest coverage.

Viral – Youtube’s bread and butter. A viral video can be attempted by anyone, but success is far from guaranteed. They may be corporate (Nike has a good track record at viral videos), user generated or a combination. If you ignore the fact that Youtube is now the primary mechanism to consume music for free, the top videos on the site include a stand-up show (Evolution of Dance) and a home video of a newborn and his toddler brother (Charlie Bit My Finger). Not content people would necessarily have predicted to have enjoyed the success they have done. Virals don’t seem to have many rules – they can break instantly or after bubbling under a surface; they can come and go or they can hang around.

But like Creative Commons licences, these categories aren’t mutually exclusive. Examples where content can straddle multiple categories include:

Scheduled/topical – Liam’s death in Coronation Street had three alternative endings uploaded to ITV.com after the episode showing the chosen fate aired. It received 650,000 views over that weekend

Reference/viral – the Japanese art of T-shirt folding has circled the Internet on more than one occasion

Viral/topical – political out-takes such as the Sarah Palin/Katie Couric interview or John Prescott punching a protester.

Viral/scheduled – Web series such as Kate Modern that encourage interactivity

Viral/scheduled/topicalTina Fey-lin got the short head, but the long tail shows no sign of abating

Of these four broad types, which would be the best for video advertisers to target? The best choice will be campaign dependent but each format has its advantages and disadvantages

NB: A previous post of mine details the advertising options available around online video. Given the swift evolution of the medium, I may need to write an updated version soon.

Viral – these tend to have the biggest numbers but success or failure cannot be legislated for nor accurately planned. It would therefore be best for open-ended campaigns, but even then the quality or content of the viral video needs to be carefully moderated

Topical – these will be short-term so the campaign needs to be perfectly timed with an immediate call to action

Reference – likely to be special interest and so the audience will be more targeted and efficient. Good for niche brands, but the numbers may not be there for those with a more mass appeal

Scheduled – using the traditional TV model, these can be planned in advance to a greater degree of accuracy. But TV flops show that predicted audiences are an art rather than a science, and the large growth in online gives an extra degree of uncertainty compared to the gentle fragmentation of TV audiences. Scheduled content means quality can largely be vetted in advance and so the advertiser has reassurances of their investment, but this sort of model may come at a premium.

However, this premium can be justified. When asked to choose, the vast majority prefer professionally produced content to amateur work, and people are also more accepting of advertising around this content.

As an employee of a TV owner making forays into online video, I am biased but I do believe that for the most part advertising around scheduled content is the best method to use. There is nothing preventing multiple video formats being utilised (after all, each has unique advantages) but in most situations, I believe scheduled content should be the primary focus.

Viral clips may provide mass reach, but scheduled content has the advantages of

  • Easy incorporation into a media plan
  • Assurances over the quality of content
  • Acceptance from viewers willing to sit through ads in exchange for free, premium content

A “best practice” ad model is yet to emerge, but there are interesting experiments going on, and it will be fascinating to see how this develops.

sk

Image credits: http://www.flickr.com/photos/atencion/ and http://www.flickr.com/photos/pagedooley/

Reblog this post [with Zemanta]