A SUMMARY OF 55 IN-MARKET EXPERIMENTAL ESTIMATES OF THE LONG-TERM EFFECT OF TV-ADVERTISING

This article reports on the first in-market experimentally based measu rement of long-term TV advertising effects for a representative cross- section of 55 tests for established consumer packaged goods. The typic al BehaviorScan(R) weight test has a test group exposed to a heavier T V advertising weight than a matched control group for a one-year perio d, after which the experiment stops. For this analysis, the two groups were tracked and analyzed for another two years, when the only differ ence between the two groups was the TV advertising treatment during th e first year. The analysis shows that when TV weight increases had a s ignificant impact during the year of the weight increases, during the following two years, on average, the sales impact of the first year is approximately double and, on average, it was from an increasing in bu ying rate in the test group. The analysis also shows that if TV weight increases had no significant impact during the first year, on average , they had no impact in the two following years.