News sees significant fall in video views after paywall, rivals question Nielsen on drop off

Hi,
Steve Browning from News corporate affairs here. This article is somewhat misleading in stating that we have not offered an explanation for the rise. We have an explanation but our tech people are triple checking everything before I go on the record with it. Nic knows this because we have been speaking off the record about this matter for the past few hours.
Cheers
Steve

News and fairfax. Best producers of internet video. Ever.

These newspapers are really showing us all how video is supposed to be done….

Nielsen is a joke! Why have the IAB decided to endorse such a woeful product is beyond me. Given the supposed ‘accountability’ that online offers to have such an archaic, unreliable audience measurement tool just doesnt make sense… Time to re-assess

The IAB have met and discussed concerns about the video census data. Someone needs to tell the truth here.

Anon, I’d like to thank you for your factual, well-researched and cogent contribution from behind your cloak of anonymity.

It’s all fishy – any kiddie could puff the numbers with a simple script running through a TOR browser, refreshing its IP multiple times a second.

Imagine what the professionals could do.

Online metrics cannot be trusted. Period.

Anon, the data referred to in this article was released yesterday. There have been no IAB meetings since this data was released. All major measurement issues are reviewed in the monthly Measurement Council meetings which have representatives from a range of publishers, the MFA and media agencies.

1) @ Stephen Browning

Let’s hear it then!

2) Pending 1) above – let’s hear from Nielsen, we know very well that they read these comments and often reply.

3) Re SMH & The Age – if they went up 3% and 2% then I’d guess 30 articles is more than enough for most people and very few have been asked to pay?

4) Not great news for Guardian to miss this list.

Hi, it’s Steve Browning from News Corporate affairs here again.

The rise in video numbers aligns with our internal data and is, we believe, accurate.

From Dec to Jan it was predominantly due to the fact that we had recently subscribed to VideoCensus (as it was called then) and did a review of the URLs being tagged.
A number of additional URLs were added and our numbers went up accordingly.

The rise from Jan to Feb was due to a number of product and editorial factors such as upgrading our players, embedding more video in stories and making it more prominent. We also activated video autoplay around this time.

Throughout this entire period we have also been generally improving the number, quality and delivery of our videos across the network and have been leveraging FOX SPORTS’ video content.

As is explained in our quote in the story, the recent decline is predominantly due to two things – the launch of news+ and a transition of some of our major sites onto our $60m editorial management system methode. Our internal numbers suggest a decline significantly less than Nielsens and we are working with them to clarify the matter. We are already seeing a recovery in the numbers and expect this to continue.

For the record the article is also disingenuous in stating that we were “unable to say” whether we “genuinely suffered a huge drop in video views, whether the previous data was too high, or whether the new number is now too low” – at no point did mumbrella ask us whether the drop was genuine or if the data was too high or too low.

News Ltd video is fine so I cannot see the reason for a drop in views.

Fairfax video is infuriating and the ‘opt in’ or ‘opt out’ option is a stupid idea given the majority of us always clear our cookies. It puts me right off clicking on any Fairfax video and most especially I do not want to see a video with some hack rabbiting on and repeating what is there in print as well.

Apart from that i wish them both well.

I am completely unconvinced by the reason given for both the increase in stream volume and subsequent decrease. Having worked in audience research for some years and having specifically been involved in developing and testing a system for measuring online video something looks awry.

As my old boss told me ‘if the data looks surprising, exciting or interesting it is invariably wrong’. This has served me well over the years in audience measurement as almost every example of this kind of spike in data that I have dealt with has been caused through measurement error. The critical issue which seems to be ignored is that the news.com.au website stats have remained very stable (as you would expect) despite the fact many of the issues raised should have impacted the site in general as well as the video streaming.

I await the results of the investigation that Nielsen is clearly undertaking. Hasty press releases post rationalising both the increases and subsequent declines are likely to look rather silly in the cold light of day when the true cause of the dramatic data swings are revealed.

So Stephen the explanation for the views falling off a cliff is the news SiXTY MILLION editorial system.. Money well spent it seems!?!?!?!?! Methode or madness????????? Get a grip.

Its the paywall, people have found alternatives and the autoplay, upsets people, especially in an Office environment, when it unexpectantly plays.

Hi, Steve from News again,

In response to ‘Researcher’ – the rise Nielsen reports is in line with our own internal numbers and was expected given the improvements we enacted during that period. As I wrote above, the recent decline was also expected, but our internal numbers don’t align with Nielsen’s numbers. As I noted, we believe the decline should be less and we are working with Nielsen on this. Our numbers don’t suggest the kind of ‘spike’ you talk about. I’m not sure if you are referring to us or Nielsen when you talk about hastily written press releases – as you can read above, I didn’t go on the record re the decline for many hours, until I was confident that the explanation was legit. This is why I was annoyed with mumbrella for suggesting that we didn’t have an explanation, when I had told Nic on background that we had one but I wasn’t willing to speak officially until I was sure.

In response to ‘it go wrong’ and our $60m methode platform – unfortunately we did have some issues with video when moving some of our major mastheads onto methode. We would have preferred not to have any issues, but these things sometimes happen with massive projects. We have now resolved the issue and our video stream numbers are recovering, as I wrote above. You can rest assured the $60m is money incredibly well spent. if you’d like an overview of what methode is doing for us, I’d be delighted – just drop me a line at News.

In response to Ann – yes, the introduction of digital subscriptions has had some effect, as we always knew it would and indeed talked about when we very first announced digital subscriptions back at the first mumbrella 360 in June 2011.

You can switch autoplay off on our network with one click, without having to register. On any story with video, just switch autoplay to off and your office environment won’t be upset from there on 🙂

Cheers
Steve

Steve from News once more…. 🙂

‘Researcher’ – my apologies, I forgot to respond to your point about news.com.au. You are correct about news.com.au remaining very stable. However, I think you are mistaken in thinking that a decline in video would affect this overall number. Video is a comparatively small part of the overall pie and so has a negligible effect on the top line number.

Cheers
Steve

Steve – I think you have misunderstood my point, about overall audience and video streaming audience. The overall audience for the news sites and video audience for content on the same sites are both large and should be relatively stable. One (the video streaming) is a relatively large subset of the total unique audience. On this basis it would seem surprising/interesting (see above) for the total audience to remain so stable whilst the audience for a significant subset should be decimated. This is true of the increase too.

Consistency, replicability and stability as much as accuracy are cornerstones of currency measurement. I wait to see what has happened in this instance.