Current Issue:

Current Issue

On The Cover:
Cumulus Chairman Jeff Marcus

Click here to subscribe to Radio Ink.

Radio Ink Writers

How Arbitron Plans to Help Advertisers With R.O.I.


By Carol Hanley

Department store merchant John Wanamaker is famous for saying, "Half the money I spend on advertising is wasted; the trouble is I don't know which half." In today's economy that is no longer acceptable.

Return on investment (ROI) metrics have become a critical focus for every industry, including advertising. Before any business spends its precious capital on marketing, it's more important than ever to know that the money is being used effectively to drive bottom-line results.

Traditionally, one of the most difficult fields in which to measure ROI has been advertising. It is difficult, if not impossible, to pinpoint exactly what factors drive a consumer to make a purchase. However, the fact that it is difficult has not stopped advertisers and marketers from trying to make that connection.

Many use a system called marketing mix modeling that employs complex mathematical algorithms (the "models" in marketing mix modeling) that correlate sales data with marketing activities to help marketers understand which elements are driving sales and delivering the best ROI for their clients. We believe we can help advertisers draw a better correlation between radio advertising and sales results by providing more granular data for marketing mix models.

We are therefore proud to announce that we will be supplying new "model-ready" data for marketing mix models designed to help link radio advertising and marketing ROI more directly. We have worked in partnership with Dial-Global, Premiere Radio Networks, and Sequent Partners to define this new model-ready data, which combines Arbitron Portable People Meter™ (PPM®) audience data with Media Monitors® ad occurrence data.

How Marketing Mix Modeling Works
Marketing mix modelers look at an advertiser's marketing activity such as advertising in radio, TV, and magazines, or direct marketing such as email or coupons and correlate that activity with sales data during the time the campaign was running. The results provide insight into which media have the most positive impact on sales and the highest ROI. Modelers also provide measures of confidence about their estimates of a particular medium's impact.

Arbitron has traditionally supplied data for marketing mix modeling based on the Diary methodology. Now, with the planned rollout of the PPM ratings service in 48 markets completed, Arbitron, along with partners Dial-Global and Premiere, undertook an extensive study to determine how radio data can be enhanced to improve the modeling process. We spoke with modelers, advertisers, and agencies to identify best practices and designed new model-ready data. We then worked with three leading modelers who compared the results using these new data with what they had used previously. The
results were very encouraging.

Three Case Studies
We performed three case studies. While radio was already showing a positive ROI in these models, replacing Diary data with the new model-ready PPM/Media Monitors information resulted in better scores for radio ROI and improved estimates of confidence in the medium's impact on sales.

One of the studies, undertaken by the modeling company MediaBrands, looked at an entertainment venue that measures success by its ticket sales. It looked at venues located in four top 10 markets across a two-year period. When the new, more granular PPM/Media Monitors data were used instead of the Diary data, the model showed that radio's contribution to ticket sales was significantly higher than previously indicated, with some results showing increases in ROI as high as 63 percent, while also improving the
level of confidence measurement quite significantly.

A second study by the modeling company Millward Brown looked at the impact of the more granular data for a fast food chain that tracks attributes like awareness and consumer recognition. With the new data, radio's score for contributing to awareness for the chain went from 1.45 to 2.61 – an 80 percent increase. When it came to one of the brand's main objectives, customers saying the food "has great taste," radio's impact grew by 192 percent. Commenting on the results of the case study, Bill Pink, senior partner of client solutions at Millward Brown, said, "Our recommendation to capitalize further on radio as part of the mix would have remained, but we would have been on stronger footing to make the recommendation."

In the third study, which was based on a national brand of candy, the new model-ready data significantly increased the level of confidence in the model, raising it from a 1.69 to a 2.23 – well above the level many marketers use as a threshold for determining their trust in a particular medium's contributions.

Why It Works
We believe the more granular PPM/Media Monitors data can improve the model's estimates of radio's ROI and confidence scores because they enable the audience data to be more closely aligned with how advertisers track sales.

The information in a Diary-based ratings survey is gathered from a collection of 12 distinct weekly samples, and for this reason there is not a sufficient sample to break out estimates for a specific day or quarter-hour. Therefore, modelers have been using quarterly averages rather than specific day and quarter-hour ratings. These average data tend to hide actual variations in radio's weight and presence, which might be affecting sales.

Because the PPM ratings service uses a panel, the sample size for any particular day or quarter-hour comes from nearly all respondents, not just 1/12th of the survey as it does with the Diary. For these reasons, modelers are better able to match the audience for the specific days and quarter-hours that spots ran and correlate that with the advertiser's sales data. The results of these cases indicate that advertisers can get a better read on the impact of radio, which may have previously been diluted when using data that came from a 12-week average.

The Next Steps

Arbitron is moving forward with providing new model-ready PPM data to help demonstrate radio's ROI and improve advertiser confidence in the medium.

We are confident that modelers, advertisers, and radio all stand to benefit from having this model-ready radio data, which provides better insight into radio's contribution to sales. And that should help increase everyone's ROI.

Carol Hanley is the Executive Vice President, Chief Sales & Marketing Officer for Arbitron Inc., responsible for all U.S sales of Arbitron services to radio, agency and advertiser customers and one of Radio Ink's Most Influential Women in Radio (see page 34 of our current issue). She can be reached via e-mail at


(1/8/2014 7:53:07 AM)
JlcCzr I really liked your article. Cool.

- NY
(12/15/2013 7:25:58 AM)
H7j98z I really enjoy the blog post.Much thanks again. Fantastic.

- NY
(10/31/2013 2:33:54 AM)
TB9NTz I truly appreciate this article post. Will read on...

- NY
(9/13/2013 10:54:07 AM)
ANzPcn Im obliged for the article post.Really thank you! Will read on...

- NY
(9/11/2013 8:46:03 PM)
t17yzu I cannot thank you enough for the blog post.Thanks Again. Want more.

- NY

Add a Comment | View All Comments


Send This Story To A Friend