Want users to watch more video? Simple .. just make them auto play!

Have to say Fairfax Digital are really testing users patience at the moment.

They have their extremely annoying FDTV video popup that plays videos with sound uninitiated over the top of other ads and articles. Yeah it’s annoying.

Now they’re autoplaying their not so awesome video news clips embedded in stories.

Example – http://www.theage.com.au/national/train-passenger-nearly-loses-eye-in-station-attack-20090915-fp2i.html?autostart=1

Why not just make these user initiated?

Oh … I know. If you make the users initiate them their won’t be as many streams OR ad plays.

FD – I can understand you have ambitions to be a player in video. Great – good luck with that. But can you do it in a way that shows the user some respect?

Maybe I’m a cynic but does FD really think their alleged ‘saavy, high end, educated and financially well off’ audience will cop this?


Short term, opportunistic and embrassing.

Give consumers control over whether they want to play a video. We’re not stupid – we can work out how to click a play button.


11 responses to “Want users to watch more video? Simple .. just make them auto play!

  1. Very smart hein?!
    OK, so you’ve gotten a potential buyer to click a link to your website while they are at work. Once you site loads the audio comes on and starts playing its little song loudly! Guess what is going to happen?

  2. joelyrighteous

    I completely agree, I have written numerous posts over the last few months about “publishers” who are focussing on short term dollars through invasive and frankly ridiculous ad formats such as auto-initiated video with sound. In almost every instance I used the term “publishers” Fairfax was top of mind.

    I know we’re in a tight economy and that they have targets to reach, but if they continue to grab for TV dollars by sticking TVC’s in these horrible units they are seriously going to damage their proposition moving forwards.

    Not only will the advertisers realise that this is ineffective and annoying to their consumers (potentially resulting in negative brand perception) but they will also steadily lose the audience they claim is so cherished by advertisers as they move to sites where they can read their news without interruption. (I personally don’t visit SMH any more if I can avoid it simply because I cant stand the FDTV product)

    I’m also interested to know how this level of advertising will sit with users once they are asked to pay for access to content that is, lets face it, pretty much run of the mill.

  3. When I open up multiple browser tabs with all the aritcles I’m interested in and one of them starts auto playing – it’s SO annoying

  4. Apart from the annoyance to the user, there are also implications on the metrics.

    Do we REALLY want to count an autoplay video as a stream? What about the issue of page auto-refesh – what implications does this have on page views? I’ve also heard that someone is counting the close button on pop-ups as a click-through … could that possibly be right?

    To paraphrase … if it looks like a cowboy, walks like a cowboy, talks and acts like a cowboy … then it’s a cowboy!

  5. hey john – good observations as always (keep them coming!)

    In answer yo your questions. Autoplay shouldn’t count as a stream … it’s like the TV suddently changing channel without user action and the network counting them as a viewer. Auto refresh is an issue but not a big one (Neil Ackland has a good post on the Sound Alliance blog on this). Close button being counted as a click through on popups … my senses tell me this happens all the time. ie … when a publisher reports 73,000 click throughs on an OTP but you only get 6,000 people on the tracking tag on the clients website … something ain’t quite right.

  6. Ben, I agree re auto-play that it shouldn’t count as a stream. This has implications regarding audience measurement vs website analytics. This is why (on behalf of the MFA) I have been insisting that we have panel-based data as an integral part of the planned hybrid audience measurement system. With a panel there is no click and therefore no ‘credit’ given to the auto-played stream, whereas in the web-traffic it will be counted. Ever wonder why server-side data is always higher?

    Tommy also hits onto an interesting point as well. He becomes aware of the auto-play in the multiple browser tabs because of the sound. What about all the other tabs – are they all viewed? Again, with panel-based we only count the tab that has the focus, whereas with server-side all pages are counted whether viewed or not. This in particular affects time-based ‘engagement’ metrics, as while all tabs maybe have been served and rendered, only one can have the focus at a time. Accumulating time for pages without knowing whether they have the browser focus is just plain wrong!

    And on the issue of auto-play, what if a publisher changed their auto-refresh rate from say 15 minutes to 10 minutes? Wouldn’t their page views go up suddenly even though the unique audience didn’t? I’m not suggesting that anyone would do that though … would they …

  7. john – what are the odds the IAB/MFA would not endorse server side measurement (ie like the current site census product) as you can’t be sure some publishers won’t try and manipulate the data … and move to only endorsing the panel netview numbers?

    MI/Site Census had its time (and was somewhat useful for some publishers when Netview had its holes) but now it must be time to revisit whether it’s serving the interests of the wider industry and especially the advertisers who are responsible for the revenue.

    I hate to sound like a cynic but some publishers are making it really hard to trust claims around readership and engagement and volume etc … you constantly feel like you’re being mislead.

  8. One of the first discussions way back between the IAB and the MFA revolved around what we SHOULD measure. It was clear that as buyers we needed people, consumers, targets – whatever you want to call them. We did not want counts of computers, files or cookies. The IAB agreed and accepted this is an over-arching principle. This led to the IAB BluePrint, which led to the IAB audit of the Nielsen Online panel. This had 16 findings which when addressed led to the launch of the new NetView panel.

    Just a word on new NetView. If any cynics out there say that that the new panel was just to get the NetView numbers higher then they are wrong, wrong, wrong. A serious “hole” was found in the NetView panel in that the “at work” usage of “home” panellists was being missed, thus understating the market. As everyone can appreciate, getting metering into businesses ranges between “damned hard” and “impossible’ (banks, defence, govt). The collaboration provided by the newly formed IAB Measurement Council with Nielsen Online was able to fill in this gap that was miniscule when the panel started, but had grown to be significantly large over time – hence the need for action.

    However, panels DO have problems. One of the biggest relates to measuring small sites when there simply isn’t sufficient sample size. This is where traffic data comes back into play.

    I think we all recognise that server-side measures are also problematic. One only has to look at MI’s “total market” of over 50 million users a month in Australia – a market of 22 million people of around which 75% go online in any given month – to realise that cookie-based measures are poor longitudinal measures. (And yes, this issue also plagues BT – are you REALLY getting what you’re buying?).

    The goal of the “hybrid”, in layman’s terms, is to use a COMBINATION of traffic data AND panle data. Think of it as converting the volume of server-side traffic into consumer-side people and then providing demographic profiles for audience measurement (it is a LOT harder than this, but if you get that concept you get the drift).

    This leaves the prickly issue of ‘endorsing’ measurement systems. From a purely personal perspective (i.e. not official MFA policy), I think we need ONE CURRENCY for audience measurement. We need one set of numbers on which we trade that $2b to avoid misleading claims. We have it in every other media and I believe we need it online. This however goes against my grain of wanting open competitive markets. Hence I favour (i) currency contracts with KPIs and SLAs that are continually reviewed (ii) a 100% open market on website analytics, whose job it is to ‘vertically’ analyse sites, rather than the more ‘horizontal’ measures provided by an audience measurement system (AMS) currency. This of course means that a “unique” in the AMS will be a different number than a “unique” in website analytics systems.

  9. I absolutely hate un-initiated sound, the video part isn’t great but it’s the sound that really annoys me (and I’m sure most users). Let’s ignore the fact that the digital media world has the ability to target, segment, engage, etc and just emply the ‘ram it down their throat’ mentality. To be fair, Fairfax isn’t the only ones doing it and I believe the ninemsn video product (mainly on their network sites) was the first to market with this.

    I also cannot believe that people are actually counting closes as clicks… surely someone has to call them out on this??

  10. Hey Ben – either Fairfax have changed their stance since you’ve posted this, or it just doesn’t happen for everyone. I get no autoplay on the article you’ve linked, using Firefox 3 on Vista…

    That said, their auto-playing slider video ads (the ones that pop out of the right side of the page) that autoplay with sound still drive me nuts.

  11. weird – i get it on the new firefox on mac and also the new IE on Windows.

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