TV ratings in Italy with SuperPanel: Interview of Paolo Duranti, Auditel

World TV updates - October 2017

Starting from the 31st of July 2017, AUDITEL has been publishing TV ratings collected through the so-called SuperPanel. Could you please explain what does this SuperPanel consist in?

SuperPanel represents the innovative evolution of the TV rating measurement in Italy. The high and growing TV offer fragmentation boosted by the fast development of the digital Channels, requires a larger sample, able to better grab the audiences of very small channels (new digital/thematic channels and local stations as well). The original sample of 5.700 Households has been expanded to 16.000 Households: the SuperPanel sample is now the largest panel in Europe in absolute value, and the largest in the World compared to the population.

With the aim to give to the reader some details about such big innovation, we are pleased to share some information about the new sample structure and the way it works.

In the original panel (5.700 HH) the families are equipped with a people meter which is able to detect the channel viewed by the member/s of the family, and who is in front of the screen. The main technology used by meters is called “audio-matching” and allows the detection of channels viewed even for short periods of time (something similar to the well known Shazam). The remote control (push bottom) is used to identify who (which member of the family, and/or guest) is watching the TV.

Such methodology is very accurate and largely used internationally but comes with a very high cost; so, in order to conciliate the need to expand significantly the sample size without increasing proportionally the costs, Auditel decided to equip the additional families (about 10.000 HH) with a set meter: this device is able to detect the channel viewed by the family using the “audio-matching” technology but does not collect any information on who is in front of the TV. The advantage of this approach is that 2/3 of the SuperPanel HH do not require any collaboration by the panel members and need less maintenance.

To compensate for the missing information about who is in front of the TV, Auditel developed a very sophisticated and unique model for estimating the individual viewing also in the family equipped with set meter. The strong robustness of the statistical algorithm is even empowered by the fact that it has to estimate just 1 of the 8 variables used for measuring the individual viewing. It took nearly 15 months for developing such very advanced and unmatched methodology which has been tested for 8 months before its official release to the market.

Hundreds of micro/macro tests have been run by the Auditel’s Technical Committee on this proprietary model which is now up and running since 31st of July and it provides more accurate figures.

What are the main benefits and findings resulting from the implementation of this SuperPanel?

As a general comment, we can say that the main benefits of SuperPanel are more addressed to channels with little audiences (the most difficult ones to be represented by a panel).

We can briefly summarize here which are the main advantages:

  • More stable data: the weighting factor of each individual has been reduced of 2/3 in average. Thanks to this, Superpanel data show more stable trends with less picks (up and down)
     
  • Less “zero minutes” phenomenon: the audience curves of small channels, often show a significant presence of minutes in which there are no viewers in front of the screen. This fact is depressing the audience. With SuperPanel, the presence of “zero minutes” has been reduced of more than 60%. This gives a significant contribution to a better representation of small audiences that, in average, results more accurate than before.
     
  • New Reach calculation: SuperPanel has introduced a new formula for the Reach calculation which shows 2 main advantages:
     
    • The New Reach calculation reduces the negative effects of the (physiological and unavoidable) sample rotation: so, new Reach estimates are more accurate.
       
    • The new calculation represents a Standard Golden Rule: all users must adopt the new standard with the relevant benefit to introduce more transparency into the market place.

What is the next big step forward for Auditel?

Looking forward to future developments, Auditel is happy to communicate that a new project for the digital devices audience measurement already started in 2017. The number of screens where people could watch videos has increased and needs to be reported by measuring the digital content consumption.

Though is a little premature to go into thin details of such initiative, we are pleased to give to the reader just a couple of key facts:

  • The architecture of the project:
    the digital measurement requires Census data (collected with the cooperation of the broadcasters) and Panel data. Census data gives the overall “size” of the viewing, Panel data give the profile of the individual viewer. One of the most critical facts in the digital device measurement is the high level of cooperation requested to the individuals in the sample (for measuring all their personal devices). Auditel exploited its SuperPanel large sample for building a Single Source sub sample in it, where to measure simultaneously both traditional and digital screens. This methodology is fundamental to properly profile the viewers and to de-duplicate the viewing of the same content done simultaneously on more devices (i.e. more accurate Reach)
     
  • The technology adopted:
    Auditel has already selected the technology to be adopted in its Panel. A router installed into the house allows to track all viewings done on digital devices connected to it. This includes out of home consumption (measured once the device is back home).

We have monitored over the last years the fast evolution of the technologies for measuring the digital viewing and we are confident that their current level of accuracy is enough for guaranteeing the quality standards required by the market. Auditel aims to release the first data on digital viewing by 2018. Sometimes not to be the first mover bring valuable benefits and helps to be up-to-date and to design very solid and reliable projects!

Paolo Duranti is Commercial Director at Auditel Italy with the responsibility to design and launch a new go to market approach. He spent about 30 years working in Nielsen and GFK dealing with Consumer and Media research; lastly, he managed the Nielsen Media business in Southern Europe and South Africa, consolidating and expanding the research activities introducing new services.

Confidence interval calculus

Taille de l'échantillon ou d'une cible dans l'échantillon

n =

Proportion observée dans l'échantillon ou sur une cible dans l'échantillon

p =

%

Attention : ne s'applique qu'à une proportion. Le Taux Moyen est une moyenne de proportions et la Part d'audience un rapport de proportions.
Cet outil est donné à titre indicatif. Il ne saurait pouvoir s'appliquer sans autres précautions à des fins professionnelles.

Test of significance of the differences between two proportions

Permet d'évaluer si la différence entre 2 proportions est significative au seuil de 95%

Proportion

Taille de l'échantillon

Échantillon 1

%

Échantillon 2

%

Attention : ne s'applique qu'à une proportion. Le Taux Moyen est une moyenne de proportions et la Part d'audience un rapport de proportions.
Cet outil est donné à titre indicatif. Il ne saurait pouvoir s'appliquer sans autres précautions à des fins professionnelles.

Solutions associées
En complément