Mediapulse measurement method still controversial
To test its own newly introduced TV metering system, Mediapulse conducted a watermarking test. According to Mediapulse, the test showed that usage recording was working for its intended purpose. However, this conclusion is being challenged.
On January 1, 2013, Mediapulse introduced a new system for determining TV usage, which uses audiomatching to identify stations based on the sound of the TV signal being used. However, the reliability of audiomatching has been called into question. Mediapulse and Kantar Media, together with their client Goldbach Media and with the support of a German broadcaster, then conducted a watermarking test. It was intended to prove the validity of audiomatching technology in direct comparison with watermarking technology.
"If we compare the results of the audio matching with those of the watermarking test now carried out, the differences in the use and allocation of the tested station are within half the confidence interval" explains Manuel Dähler, Director of Mediapulse. The previously defined quality criterion would thus be fulfilled.
In a statement from Goldbach Media, however, it says that further tests are necessary to make a definitive statement about the Audiomatching measurement method. It is confirmed that the results are mostly within the half confidence interval. However, the quality criterion could not be met in every case. In particular, on one of the days investigated, an excess of several hours was found during the prime time. The test had also shown that there were some misallocations in the country signals. Goldbach Media is therefore calling for a broader watermarking test.
