In today’s confusing world of proliferating media choices, a primary role of advertising pre-testing research is to bring clarity to the problem of how to divide up your precious advertising dollar. TV… Print… Radio… Outdoor… Direct Response… The Internet. How do you decide in advance which of these media will work hardest for you? In theory, conducting research on the creative elements of your integrated campaign idea should be able to help. But does it?
The problem is that, over time, many ad professionals have developed very different mental models of how to think about different media. Tactically, one might think of television as an awareness-generating medium while print is an educational medium, outdoor or radio are media that keep your brand “top-of-mind”, while direct mail is highly targeted and new media such as the internet are interactive.
As a result, researchers who specialize in testing the different forms of advertising operate in silos of research, with measurement constructs that rarely intersect those used by other researchers. For instance, how would you compare a banner ad’s click-through rate to a print ad’s eye-tracking data to a TV commercial’s branded attention score?
With all the different research that is used by different media to prove the advantages of using their channel to reach the consumer, it is easy to lose sight of the basic fact that all advertising is, in the end, supposed to be a “sales” medium. In the end, all ads have to be consumed by one and the same mind—your target customer’s.
By looking at the problem of advertising effectiveness in all its forms with a coherent, overlapping set of mental models, we can improve the way we organize and communicate our research data. This is useful because it makes it easy for ad managers to do comparisons of performance metrics across media platforms. It is a way to fit together all the different pieces of research onto a decision-maker’s spreadsheet.
With that in mind, I would like to show you how you can start with one pre-testing model of how advertising works for a particular medium and, by systematically changing only one or two variables at a time, use it to create corresponding models for other media. (See figure 1.)
1. The TV Model
Since it is still the major line item in the advertising budget, we start with television advertising. The Ameritest® model of TV advertising is a well-known heuristic or teaching framework for testing television creative. Without going into the details of particular testing procedures which vary in practice from one research supplier to another, we can still use the model in a general way to identify a set of theoretical constructs that almost all research suppliers can agree are important to measure.
Our conceptual model says that for a TV commercial to be effective it must accomplish three things: (1) it must win the fight for the attention of the consumer in a medium filled with the clutter of competing advertising; (2) it must be well branded, so that the consumer properly attributes the advertising message or experience to the sponsor’s product or service and not to someone else or to the category in general; and (3) it must motivate the consumer. That is, it must move the mind of the consumer somehow closer to the sale, either by persuading the consumer of the value of the sponsor’s offering or by creating a either a preference, consideration or simply an emotional bond with the brand.
At the report card level, these are the three things that almost most researchers can agree are important predictors of advertising performance. As a result, virtually all the major pre-testing suppliers provide some kind of score for each of these three dimensions of performance as a way of evaluating TV commercials.
The second level of the model describes categories of diagnostic information that can provide useful insights to help advertisers understand the performance scores their commercial is getting. This is important for the purpose of optimizing commercial executions. For example, figuring out ways to re-edit average performers so that they will work harder before you start spending large sums of money on very expensive media time.
Looking from left to right, we see that attention-getting power is primarily a function of the advertising execution. In particular, executions can be conceptualized in terms of their form and content. By content we are asking whether or not the creative idea is fresh and different, funny, enjoyable or involving in such a way that it provides a reward to the consumer for the time spent attending to your advertising. By form we mean that the commercial is a well-edited, or well-formed, piece of film, one that grabs the viewer’s attention early on and leads it from moment-to-moment, focusing attention on the important ideas and images that you are trying to get across. The Flow of Attention, therefore, is a construct that reminds us that television is a dynamic experience. To fully understand television commercials we must understand how the consumer cognitively processes moving pictures.
Branding is linked both to the creative execution and to the brand’s strategy, something we call “focus and fit”. A well-branded piece of film tightly focuses the attention of the consumer on the identity of the brand somewhere in the ad. Although depending on the creative concept, it might be at the beginning, somewhere in the middle, or not until the end of the ad. A well-branded commercial also fits the brand like a custom-tailored suit, so that ideally no competing brand could be substituted without seeming out of place in terms of emotional character, positioning or values.
The right side of the model explains motivation. To be motivating, the execution must be grounded in a good strategy, with a compelling promise being made to the consumer. Motivation always has a rational and an emotional component, although the balance of the two may vary from ad to ad. The key to motivation is to communicate a relevant idea in a dramatic way. The Flow of Emotion construct reminds us that, like attention, emotion is dynamic. It flows and changes from the beginning to the end of the film. Motivating creative is designed to organize the emotions of the consumer along the way for particular dramatic effects that express something relevant about the brand experience.
It should be noted that this same model can be used for designing research for a radio commercial. Attention, branding and motivation are still the primary measures of performance and can be measured with identical research questions. The only differences are in the diagnostics. Being an auditory rather than a visual medium, the dynamic measures of cognitive processing and emotional response—the Flow of Attention measures—would be handled differently with radio and TV commercials (See figure 2.).
From a pre-testing standpoint, print and television advertising are a lot more alike than most marketers realize. Both have to compete for attention against competing ads in highly cluttered media environments. To be successful at this, both have to provide the reader (or viewer) with a reward for the time they are asked to spend with them, by providing creative content which is fresh and different, involving, or entertaining in some way. Both kinds of ads have to be well-branded in order to be effective. And both are expected to motivate a sale by communicating relevant ideas in an emotionally engaging context.
In terms of performance, therefore, both print and television have the same criteria of success. It is only in terms of diagnostics—that is, getting inside the ads to understand how they work, that researchers must modify their measurement tool kit.
There is a common misconception that print is a static medium while television is a dynamic medium because it uses moving pictures. This represents a media-centric rather than a consumer-centric point of view. To be sure, the printed page does not physically move, but the mind of the reader moves through the page. Indeed, the flow of the consumer’s attention through all forms of media is dynamic, because whatever the channel of communication, the mind of the consumer is actively “shopping the information in the ad,” looking for ideas and images and emotions that are relevant and meaningful.
So, how is print really different from television?
First of all, with television, the linear sequence in which ideas and images are exposed to the viewer are predetermined by the advertiser: image B always follows image A, and image C always follows image B, and so on. This is not the case with print. Like a manager of a retail store, a creative director can use the layout of the print ad to try to shape the path the consumers mind travels on as she looks over the ad. But the consumer has a way of looking at things differently than we might expect.
As a diagnostic construct, the Flow of Attention deals with the sorting and scanning functions of the right brain, the perceptual basis of how we perceive the world visually. With a TV commercial, it measures how selective attention varies over time, as the viewer sorts through the series of the images from the beginning to the end of the film; with a print ad, it measures how attention varies over space, as a reader scans the layout of the ad.. (See figure 2 for an example of the difference.)
The other difference is that television comes in fixed units of time: 30 seconds, 15 seconds, or perhaps some other quantum unit, which is obviously not the case with print.
To transform the TV ad model into a model for testing print advertising, we need to add one new variable that needs to be measured: the amount of time the reader is willing to spend with the ad, or stickiness. Importantly, the model makes no a priori assumptions about the amount of reading time it takes for a print ad to work; stickiness is only a diagnostic variable, not an end in itself.
Finally, the testing of outdoor and newspaper ads does not require a different ad model. While the visual clutter of a street scene or of a pile of newspaper inserts may be different from that of a magazine, the attention-getting power of the advertising is still a primary performance variable. Communication is a driver of motivation for both types of ads, though, of course, the communication objectives of outdoor must be more telegraphic in comparison to newspaper. In fact, all of the theoretical constructs of the print model are still relevant for these other forms of print. It is only the logistical details of research testing procedures that need to be adjusted.
3. Direct Response
By reputation, direct response is the most accountable of the advertising forms because you can accurately measure the response rate after an ad has run. By comparing the response rates for different executions it is relatively straightforward to create a closed feedback loop for continuously improving direct response campaigns over time. For that reason, unlike TV or print advertisers, direct response marketers have historically made little use of pre-testing research.
However, this has begun to change. More sophisticated advertisers have realized the value of obtaining diagnostic research in advance, for the purpose of optimizing executions before the budget is spent. Research has shown it is important to not only know which of two mailings has the greater selling power, but to know the reasons why.
In addition, pre-testing research can provide value by helping the advertising manager synchronize the creative elements of direct response with the other elements of an integrated ad campaign.
Direct response, whether it be television or print based, is different from the more general type of advertising in one important respect. It is expected to trigger a behavioral response from the consumer right away (or at least within a very short period of time) rather than be stored in the consumer’s memory in order to enhance sales of the brand at some point in the future.
Indeed, the consumer’s behavioral “response” may be engaged from the very beginning of the interaction with one type of direct response, direct mail. When you sort through your mail, what is it that causes you to open one unsolicited piece of mail, while you toss others unopened into the trash? Is it the gold lettering, the high quality of the paper? Is it the personalized address? Are you more, or less, likely to open an envelope addressed to “resident”? Or is it the brand logo on the envelope? Are you more likely to open a piece you receive from American Express than one you receive from the Acme Loan Company? In any event, attention-getting power and branding are still relevant constructs of interest to the direct response researcher.
But from a pre-testing standpoint the essential difference that describes direct response advertising can be captured by our basic ad model by adding a qualifying variable to Motivation: the call to action. This variable defines the specific response that the advertising is designed to evoke. For example, calling an 800 number which is usually a higher level of commitment on the part of the consumer than is elicited by most advertising in general.
To achieve this commitment direct response is usually very promotional in its communications, e.g. “We are making a special offer!” In addition, direct response ads are frequently designed to define a time period for action, e.g. “The special offer ends Sunday!” The purpose of these messaging strategies is to impact consumer emotions in a particular way by creating a sense of urgency which drives the call to action. Of course, more sophisticated users of direct response can deploy emotion in negative as well as positive ways to create the sense of urgency. One way is to raise the level of concern about some problem, e.g. “Do you have these symptoms….?” which creates stressful emotional anxiety that is resolved when the consumer responds by taking the prescribed action.
4. Web Advertising
Web advertising, with all the degrees of creative freedom the Internet provides, is chameleon-like in that it is able to emulate any of the others. Like TV, Web ads use moving images and visual novelty to attract attention. Like print, Web ads try to engage consumer interest while consumers are reading. Like direct response, Web ads call for an immediate action—just click on the image!
A web ad is a barker who stands by the entrance to the carnival tent and solicits customers with loud, colorful sales talk. The first click-through is actually part of the attention-getting function of advertising. It is the first hesitant decision the consumer makes to engage with the advertiser, which is really no different than the decision to open unsolicited mail.
Once the consumer clicks through the door of the ad to the advertiser’s video or website, the consumer finds they are in another ad or a virtual store. Here, the friendly, sticky salesman must keep communicating to keep the consumer engaged, inviting the consumer to browse through information until they find something they like and motivates them to go to the virtual check-out counter to click again, on “purchase now.”
With the Internet, the boundary between advertising and in-store or point-of-purchase communication has dissolved. As a result, pre-testing research can be defined narrowly. And it can focus on the attention-getting function of the first click, which is analogous to the outdated practice of only using recall scores to evaluate TV commercials. Or it can be defined broadly. The ad model we show here provides a road-map of all the things the researcher needs to take into account in order to plot the path of the consumer’s mind all the way through to the second click.
To describe that path from a diagnostic standpoint, the internet ad researcher may need to use all the tools in the toolkits developed for other media. figure 2 shows you some parallel techniques we have developed to measure how the attention of the consumer flows: first, through moving images, second, through graphic layouts, and third, through lines of copy. What these three techniques have in common is an appreciation for the cognitive processes of the mind. In particular, the consumer is not—and never has been—a passive receiver of brand communications, but is always actively engaged in Googling the ideas and information in advertising.
There is an old saying in the packaged-goods business that “the package is the last ad the consumer sees before buying the product.” Indeed, most ad researchers seem to consider packaging research to be almost a separate, highly specialized discipline. Yet all of the constructs we have been talking about apply equally well to effective package design as they do other forms of advertising.
In terms of risk, getting the package wrong can be extremely costly. The reason for this speaks to the essence of what is special about the package-as-advertising. Unlike other forms of advertising which may simply create an enhanced intention to purchase the brand, the package must actually close the sale!
From a communication standpoint, a package needs to systematically remove all the barriers to purchase that make the consumer hesitate. It must provide a point-by-point argument about why it’s better than the competitor’s product that is sitting beside it on the shelf. And it must do this in the most emotionally appealing way in order to drive home the impulse to purchase.
Because of the media constraints of other forms of advertising such TV,—i.e., what you can reasonably hope to accomplish in 30 seconds—a tight focus on one clear idea may be the best practice . But when the consumer is holding the package in her hands, trying to decide between your brand and a competitor, it’s best to cover your bets. The package needs to communicate all the advantages of the product.
The need for multiple-points of communication does not mean package designs should be cluttered or confusing. Quite the opposite. The information on a package should be easy to find. That means the order in which the consumer “sees” the information on the package should match the order of importance in order to help make the sale, with the most important information standing out first in the consumer’s perceptions.
Breadth and Depth of Campaign Engagement
The traditional way of making media allocation decisions is to trade off reach versus frequency in the media buy. This is still a useful approach to thinking about programming advertising exposures. It also provides us with a useful conceptual framework for thinking about how consumers are engaged by different creative executions across an integrated advertising campaign.
Even though the different elements in a campaign are designed to work together, that does not mean that all the creative executions will work equally well. Consider the scenario shown in figure 3 where the motivation scores (using standard top box purchase-intent scores) for creative elements from two alternate campaigns are being compared. In this hypothetical example, the TV commercial from Campaign 1 is significantly more motivating than the TV from Campaign 2; on the other hand, the print and outdoor from Campaign 2 is more motivating than the creative from Campaign 1. Which campaign would you choose?
A key question to ask when testing creative elements across media platforms is whether or not the different pieces of creative are motivating to the same consumer segments. In this example, if we were to ask the question, “How many consumers were motivated by at least one execution in the campaign?”, this is the intersection of the sets of consumers motivated by different elements in the campaign. We see that Campaign 1 has more “breadth” of appeal, 85% to 72%. What if, on the other hand, we were to ask the question, “On average, how many different executions impacts a given target consumer in a motivating way?” This is the intersection of the sets of consumers motivated by different elements of the campaign. Now we see that Campaign 2 wins, with an average of 1.6 versus 1.2, of what we call “motivational touchpoints” or motivating ads per consumer. So we can say that Campaign 2 has greater “depth” of appeal.
Just as media buyers multiply reach times frequency to get a measure of the net effectiveness of a media buy, we can do the same thing here. By multiplying breadth of appeal times the depth of appeal we can say that Campaign 2, winning 115 to 96, is the more impact-full campaign when viewed as a whole.
This example makes several important points about campaign testing. First, it points out the importance of using standard measures of key performance across different media, so that the relative impact of the creative can be compared. This is one reason why the adoption of an integrated set of media models is so useful. Second, it underscores the importance of diagnostics. In this case, either campaign choice is like a wobbly stool—one of the media legs supporting each of the campaign ideas was sub-optimal. From a diagnostic standpoint, the question for Campaign 1 is, “Can I fix the print and outdoor to make it work harder in support of the TV?” The question for Campaign 2 is, “Can I re-edit the TV film to make it work harder with the print and outdoor idea?” Hence, we see the importance of diagnostic measures for different media creative elements.
Finally, it reminds us that in this new age of exploding media choices there is the fundamental need to view advertising holistically. And if this is to happen, advertising research needs an integrated testing approach.
Kastenholz, J., “How We Do It”, Harvard Business Review, pp16, April 2005
Young, Charles E., The Advertising Research Handbook, Seattle, WA, Ideas in Flight, 2005.