2003-01-23

Press, No. 1, January 2003
TV ratings measurement panel of AGB Polska has been expanded by 35% since the new year. The institute wants to overtake the competition – TNS OBOP, increasing data accuracy and allowing media planners to structure new target groups. Increasing the panel from 1231 to 1650 respondents will also allow delivery of representative ratings for specific programs and spots, for example among women with babies under three years. In the previous panel there were 121 such women, now that number will increase to 165. Are forty people such a big change? Some media houses believe that audience viewing data in specific target group are representative once the sample contains at least 150 respondents – says Tomasz Kolanowski, research manager at AGB Polska. Now that will be possible in several new groups. The number of people aged 16 to 49 (basic group used in the planning of most advertising campaigns) has increased from 1900 to 2600. Total number of panel participants now reaches 5100. Expansion of the panel had no impact on ratings of large stations and their programs. Nothing was supposed to change. The panel was created proportionally -says Edyta Łyszkowska, communications manager at AGB.

- Perhaps some changes will become apparent after some time. For example a large group of new panel participants may have similar preferences, maybe they really like one specific channel - says Marcin Klimczak, research manager at MediaCom Warszawa. Changes in audience measurement data will only be visible in the case of speciality, niche TV stations – the number of broadcasting segments with zero ratings will decline (since the panel contains more households), so ratings of these stations may increase slightly.

The decision to increase viewership sample was made on the basis of annual AGB - Establishment Survey, as well as forecasts of the Polish Main Statistical Office (GUS). These sources have shown that the number of people using cable TV is on the rise and is already greater than the number of TV viewers with traditional antennas. Expansion of the national sample took into consideration previous overrepresentation of cable TV owners and Warsaw residents (both groups were underrepresented in the existing national panel, which is why both were spun off as separate panels). Now each panel participant represents 8400 Poles, which is similar to France and Turkey. However in Germany national sample consists of five million people, i.e. three times as many as in Poland, with twice the population. – To make the panel representative for nearly all target groups in Poland, it would have to contain 2500-3000 households. However our market can’t afford that now and won’t be able to for a long time – says Adam Rybarczyk, key account manager at AGB Polska.

Nevertheless, new AGB panel will not solve the greatest problem of audience measurement –differences between audience viewing data collected by AGB and OBOP, which has been a growing irritant for the advertising market for several years now (differences in average viewership of selected stations, or ATV, reaches up to 35 minutes per day). These differences are due to the fact that AGB has been modifying its panel each year based on forecasts of changing structure of the Polish society, for example by increasing the number of small households. Since television is more popular in these households, AGB’s figures were higher than those of OBOP. In the last few years the second institute has been using data from the abbreviated census completed in 1995, arguing that audience viewing research cannot be based on forecasts. The dispute was supposed to end with last year’s general population census, but full results will only be published in the second quarter of this year. GUS has only published preliminary results, including the fact that Poland has 13.8 million households – AGB forecasted 13.5 million, while OBOP 12.5 million. – It turned out we were right, we based our research on a figure that was close to reality – says Tomasz Kolanowski. AGB does not think it will be forced to redesign its panel after the publication of detailed census data. Stanisław Bębenek, head of audience measurement department at TNS OBOP replies instantly: - The number of households has been growing faster than the number of actual houses and apartments for several years, producing artificially high audience measurement data. We are planning to switch to home/apartment-based definition of a household.

At this point OBOP’s panel consists of 1100 households. – Audience measurements are the present, we are working on solutions for the future, for example on PPM – says Bębenek. PPM is a measuring system based on a device worn by research participant all day round and receiving impulses broadcast by TV and radio programs. However, OBOP will have to convince the biggest TV and radio stations before starting to test this technology. – TVP [public television] wants to give it a try, other stations are saying No – declares Bębenek.

Expansion of AGB audience sample will certainly increase cost of market research. – However, 35% increase of the panel does not translate automatically into the same price hike for the customers – emphasizes Tomasz Kolanowski. New rates are negotiated individually with each customer.

Jarosław Murawski