How Nielsen’s new ratings will play out

Effects of increasing its household sample and adding online viewing

By Diego Vasquez

December 2, 2015

   

Ratings for broadcast shows such as ‘Madam Secretary’ could see big bumps with the addition of online viewing.

One of the major criticisms of Nielsen is that the company moves very slowly with changes. But in the coming months there are going to be a lot of changes at the ratings measurement company. It’s bumping up its sample size of households, addressing one of the longstanding complaints from media buyers and planners. Nielsen will go from 25,920 to 40,330 households, which it claims will increase the reliability of ratings by nearly a third. Also next year, digital viewership from online sources such as desktops and tablets will be added to TV ratings. And Nielsen will begin tracking DVR playback beyond seven days. Each of these changes is sure to have an impact on how media is planned and bought. Networks are expected to make an even greater push to use seven-day DVR playback commercial ratings rather than the current three-day. And there could be some surprises relating to who’s actually watching which shows. Media people are generally excited about the changes and believe they’ll lead to more accurate ratings, at a time when Nielsen is feeling heat from the proposed merger of rival ratings firms Rentrak and comScore. John Morse, Ph.D., head of the full-service media marketing research company Byron Media, talks to Media Life about how the changes will resonate in media, whether buyers will ever pay for month-long DVR playback, and what measurements he’s still waiting to see.  

Nielsen will significantly increase its sample size come January. What type of impact could this potentially have on ratings?
The impact on ratings is unclear at this time.
Theoretically, the only changes should be greater statistical reliability and consistency. However, when samples change, there have typically been some companies that experience increases and others decreases. Supposedly, the weighting of the data to match the census on key variables should minimize this.

The greater reliability means that low-rated networks and programs are more likely to be reportable and therefore monetizable. Nielsen has calculated that this will increase reliability of audience data by 31 percent.

What do you think will be the effect of adding in digital viewership?
It is clear from numerous research sources that adding in the viewing on mobile, tablets, desktops will spike the ratings significantly. In fact, I expect the overall combined ratings to break records and document that Americans are putting in more digital video screen time than ever before.

What do all these changes mean for media buyers and planners? Do you think they will impact their decision making?
Until this audience data is publicly released through Nielsen’s new aggregated syndicated service in the middle of 2016, there will not be any changes in buying and planning.
However, everyone will be peeking at whatever data is available to begin to consider the impact on overall audiences.

In first quarter 2016 anticipation will build about the added audiences. Then will come the back and forth between buyers and sellers on pricing.
Will the audiences on newly measured devices be counted the same as traditional TV audiences? They should be, but there is likely to be some pushback and discounting.

One of the things being added in is DVR playback beyond seven days. Is that of any value to media people?
The value will vary by program genre.
Sports and news are mostly watched live or live plus same-day [DVR playback]. However, movies, documentaries and the top program series will get a small lift by adding in ratings beyond seven days.
Even these genres are mostly replayed in the first three days, so the value will not be huge but will add some revenue value. The question remains will the agencies accept anything beyond C7, now that the movement is under way to accept audience viewing beyond C3, the standard for the recent past?

Is there anything you would still like to see that Nielsen is not adding? Why?
Measuring OTT, especially Netflix, needs to be a priority.
With so many people dropping traditional multichannel subscriptions in favor of inexpensive internet-based ones, the audience migration needs to be fully measured. But currently Nielsen won’t measure sources without the cooperation of a service to code the programs for measurement.
Nielsen should find a way to apply its own code and measure these options–even at the risk of upsetting some services, which don’t want their numbers to be publicly reported.

Fox recently said it’s abandoning L+SD/overnight ratings. Do you think these ratings present an accurate picture of TV viewing in light of changing watching patterns?
We know that L+SD provides an incomplete measure of total audience. However, some programmers continue to run these numbers in order to get an early heads-up on how shows are doing. Sales, on the other hand, need to add as long a timeline as the market will bear.
There may be two accounting moments for settling up on audience delivery: L+7 and L+35.

Are you satisfied with Nielsen’s timeline for these changes–is it too late, at just the right time, or are they being rushed?
The industry is divided on almost everything that Nielsen is doing, with the exception of adding new national sample without increasing prices.
One faction wants to have more time to vet and critique any new data added into the Nielsen national sample. Others are anxious to get an expanded sample and to incorporate additional audience platforms.
However, there is no client consensus about what Nielsen should do. Most companies are wary but excited by the prospect that ratings will likely be higher. In part because of the competitive situation, Nielsen is feeling a lot of pressure to move forward with enhanced measurement.
I’m looking forward to it.

What should Nielsen be keeping an eye on for the future to ensure the ratings remain reflective of viewer preferences?
Given the proliferation of viewing options–the number of digital viewing channels on the internet–Nielsen needs to continue to expand its samples and find a way to add a representative sample of set-top box ratings for households.
There will need to have editing rules to make sure the household data is reliable. There are a lot of other issues related to using STB data, including cost, and Nielsen has been working on this for years.
But compare the statistical stability of Rentrak’s 18 million homes to Nielsen’s 40 thousand. Putting the STB data into Nielsen play would be a game-changer.