Measuring Ad Effectiveness from Several Media

Effectiveness of Advertising
"Measuring Ad Effectiveness from Several Media"
Article No. 9 (by Hideaki Koizumi)

SPI held a seminar not too long ago about ad effectiveness for advertisers, and there was a question about how to measure the ad effectiveness of not the four major media (TV, NP, MG and RD) but other media such as Transit and Out Of Home (OOH). I had to admit that it is very hard to measure those media as there is not much data available for those media, especially OOH.

The problem with comparing ad effectiveness of several media is how to judge ads of different sizes and/or with different content? In other words, how do you establish a standard? Is it worth while comparing a 4c1p print ad with a 15" TV CM? Another issue is that each media has its own function different from the others. From the point of view of increasing awareness, TV CM are superior to other media. On the other hand, print media can communicate more of the product's features than other media. So, the solution lays with setting KPIs, which I explained in my previous articles. KPI stands for 'Key Performance Indicator'. We should determine the most effective KPIs that contribute to communications' final objective; sales growth, market share growth or affecting some brand equity index. It is not difficult to measure ad effectiveness of several media if KPIs are set. This is to say that we should analyze how much each KPI contributes to the final objective. For example, when tracking surveys on brand recall are conducted, we should analyze how much each medium was able to contribute to any increases in the figures. SPI uses a quantitative method to do this analysis. There is not enough space in this article to explain a very complex method for doing so. So, I'll introduce a not very complex method instead. Still, even this not very complex method requires data.

This method is based on the idea of 'Netapps' (net-ad-produced purchases) method which was proposed by Dr. Daniel Starch. First, KPIs should be determined by following the way in which I explained in my previous articles. KPIs such as 'purchase intent' and 'brand awareness from media' need to be measured with data. Then, each medium's expenditure is divided by the rate of 'purchase intent out of all brand awareness' by the number of total respondents. In other words, obtain a coefficient, which is calculated by first obtaining the number of 'persons who have purchase intent' from the 'persons who have brand awareness from media' and divide by the number of total respondents. Then, divide each medium's expenditure by this coefficient. This way, we can compare several media. Of course, in this KPI's case, this method might not be perfectly accurate due to some other issues beyond the ads themselves affecting purchase intent (e.g., PR, etc.). But we can still analyze using this method. And we have been able to get results that were quite pertinent in our experience. One thing we need to be careful of is the number of the respondents. For instance, in the case of awareness of radio ads, the ratio will probably be very low. So, we need to have quite a few respondents to be able to properly do the analysis.

Using this method, not only can we choose effective media by understanding each medium's contributions to the KPIs at the time, but with ongoing analysis it can help us identify problems such as communications problems or media planning problems if some medium's contribution is declining. Actually, many modeling methods and new, sophisticated techniques are in use at SPI. But we still think it is worth considering this method as well.

The next article is the final article in this series. So, I'll summarize what was covered in the entire series in it.

Please contact us with questions or for more detailed information.
spiindex@spi-consultants.net