Now that I've got your attention, I must say that the headline is somewhat misleading. Having said that the news is still rather exciting.
Nielsen today announced a revamped version of NetView, incorporating a new meter, new methodology, hardware and processing improvements and most importantly an increased / (ie more accurate) sample panel.
Up until this point, the market have made do with digital media planning data that did not accurately reflect the true extent of online usage, due to a panel that was strongly weighted towards at home usage. As a result, we all suspected that reported figures were likely to be under cooked (given enormous office based use).
Well, we were right, Nielsen data released from July 2009 shows a 16% jump in 'unique audience online'. By Publisher/site there are some significant variances, however, all (we think) have grown.
Of the top 5, the biggest gains were achieved by Fairfax @ +49% and Yahoo7 / Sensis both @ +41%. I'm sure all will be out beating their drums in the coming week.
Despite their numbers increasing, however, we don't expect this news to drive major change in share by Publisher. Perhaps some more competition regarding home page take overs, but little beyond that. If we are paying on CPM/CPC/CPA, we are still buying impressions or actions, not "total" audience.
The bigger story for us is in the TOTAL numbers and beyond total unique audience, there is more news. And that is, that reported average time spent online figures have actually doubled. This is huge news for the industry, illustrating more accurately just how much time individuals are engaging with the medium.
To end, it's all good. The quality of our planning data is becoming much more accurate. And there's more good news. The next step for Nielsen will see a Hybrid of their MI cookie based system and the improved NetView panel data re calibrated. Once the Hybrid system is released we can expect mobile and 'other' (library, Uni, cafe etc) data to also be covered. Again reported numbers will grow.
Note - Planners be warned, historical monthly trend data will not be relevant and will show a jump between June and July 2009. If presenting this data, a separation needs to be identified and an explanation of the changing measureument system foot noted.