Editor’s note: Stephen J. Hellebusch is president of Hellebusch Research and Consulting, Inc., Cincinnati.
The intent of this article is to help those who evaluate advertising via tracking studies to think about the practice and how it might best be done. Advertising tracking studies are a common phenomenon in the research world, and there are many ways to conduct them and many ways to look at the results. They are sometimes referred to as usage and awareness (U&A) studies, awareness and usage studies (A&Us), or awareness, trial, and usage studies (ATUs). Sometimes they are continuous, and sometimes they are conducted at points in time around the advertising events, or just at selected times for monitoring.
One thing all of these types of studies have in common is an attempt to measure the effects of advertising activity. They always measure changes in brand awareness. Sometimes they also measure changes in claimed brand usage, advertising awareness for brands, and brand imagery. Some marketing researchers and marketers are comfortable looking at changes in the total awareness measures and some have a preference for unaided measures. Those preferences can drive the method chosen to conduct the research and the way the data are tabulated.
A few definitions are necessary. If awareness information is obtained from respondents prior to the mention of any brand names, it is unaided awareness. An example would be “What brands of ready-to-eat cereal can you think of? What others? What others?” If the method is telephone or in-person interviewing, the question is asked until the respondent indicates he or she cannot think of any other brands. In online interviewing, it is an open-ended question. The questioning elicits brand names the consumer can recall. They were in the consumer’s head, and, if not top-of-mind (first brand recalled, technically), at least not too far from it.
If awareness is obtained by asking questions that include the brand name, it is aided awareness. An example would be “Have you ever heard of Grape Nuts?” This demonstrates whether or not the consumer can recognize the brand name, a far easier task than recalling it. It was in the consumer’s mind but may not have been anywhere near the top. Since people make mistakes, it may not even have really been in the mind. I can answer “yes” to “Have you ever heard of Johnson’s Corn Flakes?” when, in truth, I have never heard of the product.
In telephone interviewing, all unaided questions are asked first, before any brand names are mentioned: What brands are you aware of? What brands have you seen advertising for? What brands have you used?
The interviewer will have a list of brands in front of them and check the ones mentioned. If a respondent mentions a brand “unaidedly,” the respondent is not asked the aided question for that brand. Consequently, to get total brand awareness, unaided and aided are added together. Normally only two of these three related measures are reported - e.g., total brand awareness and unaided brand awareness, since total = unaided + aided.
In online interviewing, it is more difficult to handle the unaided questions. As with mail interviewing (the other main self-administered method), it is nearly impossible to skip asking the aided question when the respondent has mentioned the brand “unaidedly,” because the unaided question is an open-ended question, and there is no way to know ahead of time what each respondent will say. When unaided and aided are asked in online interviewing, the aided is asked with an “irritation reducer” - “You may have typed in some of these brands before, but have you ever heard of __________?” The brands are then listed. Basically, this procedure makes “aided” equal to total brand awareness, and no adding is needed to obtain the percent aware of a brand; total = aided.
Thus, while you can have both types of awareness with any method of data collection, telephone or in-person make unaided easier and cleaner to gather and online and mail are more suited to aided awareness.
Which is the best type of awareness to examine in looking for the effects of advertising and promotional activity? The practical problem one faces when focusing on total brand awareness is that people are very good at recognition. If a brand is well-known (Kellogg’s Corn Flakes) it may have near 100 percent total awareness (unaided + aided) all the time. It is impossible to detect changes if that is the case, and unaided awareness may be a much more useful measure. Additionally, if consumers are “unaidedly” aware of a brand, one can be certain that they have the brand in mind, while recognizing an established, well-known brand may be meaningless in terms of assessing the brand’s current marketing activities.
On the other hand, if a brand is relatively unknown, the unaided measure may take years to change, while the total awareness increases markedly over time. Since recognition is easier than recall, it grows faster. In this case, the only way to note success in the promotional efforts is via the total awareness measures, and, while few recall the brand name, the fact that higher and higher percentages recognize it means that the efforts are having a positive effect.
When trying to select the proper measure, one other helpful tip is to make certain that the way the data are processed accurately reflect total unaided brand awareness. Often programmers who have no marketing research knowledge are creating data tables that show the results for each question, and nothing else. To get at total unaided brand awareness, some additional programming must occur.
Suppose, via telephone, a respondent is asked, “What brands of ready-to-eat cereal can you think of? What others?” The respondent mentions many brands, but not Frosted Mini-Wheats. The next question is “What brands have you seen advertised recently? What others?” When asked that question, the respondent says, “Frosted Mini-Wheats.” At this point in the interviewing, no brands have been mentioned, so this person is aware of Frosted Mini-Wheats on an unaided basis. If the data tables show just the responses to each question, the unaided awEditor’s note: Stephen J. Hellebusch is president of Hellebusch Research and Consulting, Inc., Cincinnati . He can be reached at 800-871-6922 or at info@hellrc.com.areness of Frosted Mini-Wheats will be underestimated, since this person was not aware at the unaided brand awareness question.
It is necessary to either work with knowledgeable programmers or specify in advance that a total unaided brand awareness table be created, showing the number of respondents who mentioned a brand anywhere in the unaided questioning. This number will often be higher than the number seen on the unaided brand awareness table that is generated by just reporting the responses to one question.
Think and discuss
In attempting to measure the impact of promotional spending, it is always wise to think and discuss. It is hoped that this small effort contributes to that greater one.