A landmark collaboration evaluates which metrics can help guide advertisers and planners to more impactful use of the magazine medium

The pressure to increase the return on marketing investments has never been greater. The media landscape is as challenging as it is fragmented. The traditional reliance on large-scale TV campaigns is giving way to nontraditional initiatives that promise to break through the clutter of advertising and promotional messages bombarding consumers. Integrated, multitiered media campaigns now compete for the attention of hard-to-define, hard-to-reach, and hard-to-motivate audiences. This environment presents advertisers, media planners, and publishers with a shared need: Improve the effectiveness of magazine advertising.

This research—which was presented in October at the 2005 Worldwide Readership Research Symposium—identifies and prioritizes the drivers of print advertising effectiveness. We focus on a pool of currently used metrics instead of introducing new measurement concepts. In essence, our work audits the predictive power of the common currency of print media planning.

Traditional approaches to planning magazine ad campaigns

When media planners evaluate alternative magazines for an ad campaign, they have at their disposal a broad range of tools to determine the strength of one magazine versus another in meeting the plan's objectives. Each tool or metric provides a plausible measurement of whether an ad in one magazine will have more impact than in another. At the heart of this process is the search for a greater return on the advertiser's investment; to the extent that metrics can connect advertisers with target audiences, they become indispensable. But planners draw on an odd mix of subjective judgment and opaque algorithms to tell them which of the many readily available metrics are t he best predictors of ad effectiveness (often operationalized as "ad recall"). Are planners focused on the right metrics to predict success? Are the new planning tools genuinely helpful, or do they simply provide false comfort without any predictive power?

IN TODAY'S COMPLEX ENVIRONMENT, WHICH METRICS ARE THE BEST PREDICTORS OF AD RECALL?

Today's planning metrics range from a straight count of "eyeballs" to the increasingly popular measure of "engaged" readers. Over the past five years, increasing attention has been paid to the role that "involvement" or "engagement" with a medium plays in generating advertising recall or attention— especially in the magazine publishing community, but more recently also in broadcast TV and other media. Studies have been conducted by publishers (Magazine Involvement Alliance), agencies (Starcom), and the Magazine Publishers' Association, among others, focusing on the value and role that metrics like involvement and engagement play in enhancing advertising effectiveness.

But engagement is only one aspect of a complete and realistic understanding of the factors driving magazine ad recall. Planners also evaluate editorial environment and positioning. In fact, there is a variet y of tools available to determine the ideal media plan, each with its own pros and cons. But little work has been conducted on the predictive power of these tools. To this end, a collaboration was under taken between Meredith Corporation (a leading magazine publisher); Foote, Cone & Belding (a leading advertising agency); and Knowledge Networks (a leading research consulting firm) to prioritize the generally available print planning resources and metrics with respect to their potential to predict advertising effectiveness—which, in this case, is defined as ad recall.

The new approach to predicting ad effectiveness

This work is a first step in identifying which metrics are most predictive of ad recall (quantified as the percent of variance explained by the statistical model developed). The researchers selected one branded advertising campaign within each of three product categories for this pilot research: a widely recognized consumer packaged good, a prescription pharmaceutical drug (promoted directly to consumers), and an apparel brand. All three campaigns—for leading brands within their respective categories—ran in the U.S. in the spring of 2005.

A large, random sample of more than 1,500 consumers, recruited from Knowledge Networks' representative online consumer panel, was screened for readership of magazine issues in which the three test ad campaigns appeared. Those magazine readers exposed to the campaigns then answered a series of questions probing their involvement with the magazines using MRI's standard qualitative metrics, including

• frequency of reading
• source of copy
• place of reading
• time spent reading
• actions taken after reading the magazine
• interest in advertising
• overall rating of the magazine

Other characteristics of the magazine included in the analysis were

• audited circulation
• ad/editorial ratio (more than 50 percent advertising)
• positioning
• editorial environment

The study also captured the impact of some variables that influence advertising effectiveness but are usually beyond a print planner's immediate control. These included category participation, strength of copy, and potential frequency of exposure to the campaign.

All of these measures served as inputs to a logit model that explained the degree to which they predict reader recall of the advertisements (measured in a variety of unaided and fully aided questions). As expected, there are differences in the influence and predictive power of various metrics by product category and by magazine genre. These findings point the way toward a new, richer approach to magazine ad effectiveness—one that, with further study, could result in an improvement in the magazine selection process and could suggest new strategies for maximizing the efficiency of the magazine planning and buying process.

Ad quality and reader involvement are the strongest predictors of recall

In total, 26.6 percent of the variance in ad recall is explained in this model—a substantial improvement over prior studies that attempted to explain the predictors of advertising effectiveness in terms of only one or two variables. (By comparison, only 12 percent of TV ad recall was explained by executional factors, such as humor, in a study conducted by Stewart and Furse.) Ad quality (10.6 percent) and magazine involvement (7.9 percent) emerge as the two strongest predictors of ad effectiveness; overall, they explain 18.5 percent of the total variance, marginalizing the effect of other metrics like category participation and positioning.

WE FIND THAT A "ONE SIZE FITS ALL" APPROACH TO PLANNING

This research also indicates that the factors associated with recall differ from product category to product category— suggesting that a "one size fits all" approach to planning decisions does not work.

• Ad quality is the strongest predictor of ad effectiveness in the CPG category, explaining 13.4 percent variance— compared to 9.5 percent in the apparel category and 5.5 percent in pharma.
• Involvement is most influential in the pharmaceutical category and least influential in CPG.

See Chart 1.

These differences—though they may be partly attributable to the individual characteristics of each campaign tested— suggest that

• creative may be less of a determinant of the effectiveness of pharmaceutical advertising, given the strict regulatory limits around DTC advertising (and the stricter enforcement of these limits in 2005)
• a reader's involvement with the magazines takes on greater importance in the pharmaceutical category

The CPG ad tested was more visually arresting, which may account for the higher percent of variance explained by ad quality in this category. The apparel category example tracks most closely with the overall model.

The influence of ad quality is not uniform

Ad quality is the single most important contributor to advertising effectiveness in two of the three product categories and three of the four magazine genres tested. Ad quality is defined across five dimensions that are rated by respondents. These five dimensions—chosen because they are most used in copy-testing conducted by advertisers—are then distilled into two factors. The most influential is the factor that encompasses communication/believability/ persuasiveness. This suggests that a print ad's ability to convey the core message in a persuasive and credible manner is even more important than its ability to capture attention and be liked.

See Chart 2.

An ad's ability to command attention and appeal only emerges as a highly influential driver of ad effectiveness in the CPG category. As described earlier, one possible explanation is that the CPG ad is eye-catching, whereas the apparel ads use traditional visuals and the pharma ad's creative potential is constrained by strict regulatory guidelines.

The influence of ad quality also fluctuates across magazine genres, exerting a greater influence in Health magazines than other genres like Parenting, Home Service, and Women's Interest magazines. Interestingly, the "attention" and "appeal" aspects of ad quality are very important in predicting recall of ads that appeared in Health magazines, whereas the "communication/ believability/persuasion" factor exerts the greatest influence in Parenting magazines and is least influential in Health magazines. However, only the CPG and pharma ad executions ran in Health magazines.

Five Key Drivers of Creative Success in Magazine Ads Appealing Attention-getting Communicates the Main Message Believable Persuasive

Involvement proves more complex

Similar differences are also observed around the influence of the readership involvement metrics. These factors account for 7.9 percent of total variance across all ad campaigns tested, yet this increases to 17.6 percent in the pharma category and declines to 4 percent in the CPG category. In fact, this illuminates a key finding for planners—involvement may not be a key driver in all product categories.

MANY CONVENTIONAL METRICS ARE NOT SIGNIFICANTLY RELATED TO AD RECALL.

Additionally, there are significant differences in the influence of the components of involvement across categories and magazine genres. For example, "any action taken" (i.e., clipped an article, ad, or coupon; sent for product information) is most predictive of ad effectiveness within the context of Parenting magazines, whereas newsstand acquisition is more predictive of effectiveness of ads appearing in Health magazines. The importance of actions taken reinforces the media planner's intuitive beliefs; however, this is not a significant predictor of ad effectiveness in the pharmaceutical category and in three of the four magazine genres tested (although this may be driven by constrained sample sizes within some of magazine genres).

Magazine likability is highly influential in the pharma model. The counterintuitive importance of readership "away from home only" and newsstand purchase in this category are also noted. While one could speculate that this is the result of readership in physicians' offices, there is nothing in the study to suggest that people are more likely to read pharmaceutical ads there. This finding lends credibility to the importance of total audience, versus a focus on subscribers or primary audience only.

Conventional media planning metrics are often unimportant

Many of the conventional media planning metrics fail to be significant predictors of ad recall in any of the product categories or magazine genres tested; metrics include:

• "Far forward, right hand page"
• Read three or four out of four issues
• Primary versus pass-along readership
• Percent of pages opened
• Interest in advertising
• Special position (other than back cover, not measured)
• Total pages in the magazine

See Table 1.

AD CAMPAIGNS IN DIFFERENT PRODUCT CATEGORIES MAY CALL FOR DIFFERENT PLANNING TOOLS. ACKNOWLEDGEMENTS

Characteristics of the publication and ad placement (the other magazine characteristics) account for 4.8 percent of variance across all three ad campaigns. This is driven by the importance of a high ad/edit ratio (more than 50 percent advertising) to the apparel ad campaign (2.2 percent of variance). Intuitively, apparel advertising works best in magazines where advertising is an important component of the magazine's character, although interest in advertising fails to be a significant driver of recall.

Although magazine exposure frequency accounts for less than 5 percent of variance, it is a significant variable in all three product categories. This finding is consistent with the intuitive belief that frequent exposure enhances the likelihood of recall.

Purchase of the product category accounts for up to 4 percent of the variance in ad recall. This suggests media planners should continue the common practice of targeting print advertising to product users. Counterintuitively, future product purchase is not a significant variable in predicting the effectiveness of the pharma ad. "Intend to purchase" explains some of the variance in the CPG and apparel categories, but not in pharma, perhaps because intention is driven by circumstances such as an unexpected diagnosis or a physician-driven decision.

These findings should be refined and corroborated by additional studies utilizing multiple brand campaigns within categories. These findings may also be idiosyncratic to the campaigns tested. Nevertheless, the findings suggest that a planner's time would be better spent using metrics that are more predictive of ad effectiveness than many of the conventional metrics that prove to be less influential.

Refocus on the metrics that truly affect recall

This research provides some valuable insights for the advertiser and researcher communities — among them are the following:

• Ad effectiveness is influenced by actions of the agency (creative), the client (product user satisfaction), and the consumer (level of category interest/involvement)
• Negotiated ad placement is less important in some magazine genres and/or product categories than in others
• There is a discernible difference in the value of some magazine measurement tools over others
• Different ad campaigns in different product categories may call for the use of different planning tools

These findings also provide a number of new directions and thoughts for planners, publishers, and researchers; collectively, these groups need to

• Increase the focus on understanding the consumer dynamics of individual ad campaigns and product categories to determine which metrics will most enhance ad recall
• Increase the focus on the quality of print ad creative and the importance of testing copy to ensure it meets the campaign objectives
• Reduce the focus on analyzing metrics that have little to no impact on ad recall in particular ad categories or genres but potentially can be time-consuming for media planners

We believe that this work should be considered a pilot test for further studies involving multiple brands across multiple product categories. Ongoing research should focus on

• consistency of metrics that truly affect recall across campaigns within a product category
• the role of magazine genre vs. product category
• the relationship between ad recall and the metrics planners use to select magazines, in addition to other measures of advertising effectiveness

This work is a first step toward greater accountability and efficiency in planning magazine ad campaigns. It quantifies the differences in the predictive power of planning metrics across product categories and magazine genres. However, the primary value is in the revelation that all planning tools are not created equal. Our study provides a valuable first step toward greater accountability and efficiency in planning magazine ad campaigns.

Readers can contact the authors for more information or for a copy of the full paper.

Britta C. Ware is Executive Director, Corporate Research, at Meredith Corporation. Contact her at britta.ware@meredith.com.

Roger B. Baron is Senior Vice President, Media Research Director, at Foote, Cone & Belding. Contact him at rbaron@fcb.com.

Justin Edge is Senior Vice President and Senior Managing Director at Knowledge Networks. Contact him at jedge@knowledgenetworks.com.

Acknowledgements
A large team of professionals supported the authors of this research. Special thanks to the three anonymous advertisers that agreed to submit their campaigns to this analysis. We would also like to thank the team at FCB, including Scot Thompson, Melissa Handley, and Maggie Connors; Jim Collins of Scarborough Research; Jill D'Ambrosio of JRD Consulting Services; and the research team at Knowledge Networks.

Send this article