The Media sector has given us many challenges we have been able to solve
challenge
1

Archimedes: Perceived Effectiveness Index

Our objective was to develop a study to regularly assess the communication effectiveness of a group of major brands operating in Spain, similar to a "stock market" value index. To achieve this, an innovative methodology of longitudinal and transversal analysis of advertising effectiveness was implemented.

Solution:

Effectiveness was analyzed with mathematical and statistical models from three perspectives: RECALL, IMAGE and BUSINESS. The final index, called "3D Effectiveness Index", is the result of the combination of the three dimensions, and each month can vary between 0 (minimum effectiveness) and a value of 1 (maximum effectiveness).

A Web application was developed , as well as an application for tablets and mobiles, allowing us to know the ranking of the 20 brands measured every week, and to perform simulations varying the weights of the three axes and know the time evolution of the various indicators.

Actions carried out:

Data were collected through a weekly CAWI tracking methodology, aimed at a representative sample of the Spanish population of between 16 and 55 years old. Five hundred weekly interviews were conducted, for a total of about 26,000 annual interviews.

The questionnaire was comprised of a simple battery of questions: brand recall; when was the campaign seen; the campaign's impact on brand image and purchase intention; overall evaluation of the brand; user / former user / non user of the brand; socio-demographic data and media consumption.

Results:

Perceived Effectiveness Ranking

Simulation of the value of effectiveness on the basis of the weights assigned to each of the three axes (recall, image and business).

Return to
top
challenge
2

TV: audience analysis

A cable TV distribution company, asked us to construct an audience prediction model for its main channels.

Their objective: estimate the audience levels both on a 24-hour basis as well as in prime time, and replicate the analysis in three countries.

Solution:

As a first step towards the development of the models, we carried out a comprehensive data analysis, in each channel, to determine the type and quality of content broadcast (variety of movies and series, repeat of programs, etc.).

Although historical data was limited (annual data had to be modeled), we still had data from various channels. Taking into account these characteristics, we applied repeated-measures models, which allow for an integrated treatment of cross sectional (channels) and longitudinal (historic) analysis.

Actions carried out:

The 24H absolute rating was modeled, and from this we derived models to estimate the rating in prime time and the relative rating.

In the models we tested drivers such as:penetration by channel and cable TV, audience engagement for each channel and some variables derived from the contents.

Results:

We managed to find out what variables were affecting the audience of each channel and what was their importance. Thanks to this it was possible to compare the variation of the drivers between channels and countries.

With the models we were able to predict not only the audience of their main channels, but also estimate the audience level of their competitors with a minimum margin of error.

Audience analysis.

Audience analysis
return to
top
challenge
3

Online: Effectiveness of Native Advertising

There is a type of online advertising that matches and integrates with the website, its content being relevant to users, this is known as 'Native Advertising'.

We perform a measurement of recall, analyzing the elements that capture attention, and thereby create criteria for measuring the effectiveness of native advertising and the advantage offered over other types of advertising.

Solution:

To study the impact of this type of advertising, an analysis of the data obtained through a CAWI survey was performed. In it, different creativities of native advertising were shown, in different devices (Smartphone, PC and Tablet) and supports

Various profile types were detected, revealing their behaviors and attitudes when faced with these creativities, comparing them with other types of online advertising.

Actions carried out:

Through this study, we detected the extent to which this advertising is integrated in the content in which users are interested, by support and device.

Then, the perception of the campaign by the users was evaluated, what makes them increase / decrease the probability of clicking on a given ad, as well as what subsequent behaviors the campaign generates on users (brand search on the web, social media monitoring ...).

Results:

Understanding how native advertising and the subsequent behavior it generates is perceived, its results can be improved.

We can increase visits to our website by integrating advertising with content, gaining greater user interaction, if we get more attention and increase monitoring and comments on social networks, generating linkages offering confidence.

Effectiveness of Native Advertising

Effectiveness of Native Advertising
return to
top
challenge
4

Advertising: measuring the response received by a call center

One of our clients was interested in quantifying the response to advertising at the call center.

It was necessary to quantify the direct and indirect effects of online and offline media.

Solution:

We identified and measured the response generated by the different online and offline channels.

Actions carried out:

Econometric models were created explaining:

  • Direct effect of offline campaigns.
  • Direct effect of on line campaigns.
  • Indirect effects of on/offline campaigns.

Results:

We estimated different business scenarios with different communication alternatives, both offline and online, according to the learnings gained from the econometric models.

return to
top