Better practices in advertising can change a cost of doing business to wise investments in the business.
MANY VIEW ADVERTISING as a cost of doing business rather than wise investments in the business--a view that has created the roller-coaster rides of cutting advertising "spending" to meet quarterly profit objectives. This has most likely resulted from marketers not knowing, not being able to account for, not being able to improve, and/or not being able to achieve the return from investments in advertising consistent with their business objectives.
In an analysis of market mix modeling for 45 brands, Ephron and Pollak (2003) concluded that, on average, every advertising dollar spent returns just $.54 for consumer package goods and $.87 for non-consumer package goods.
While there are unique business objectives in each advertising situation, whether they be meeting specific growth or ROI targets, defending market position, or just keeping the distribution channels open, knowing the probable return or market impact of an advertising plan in advance of airing, with time to revise, and with the knowledge as to what and how to revise, are critical conditions for the achievement of consistent performance and the shifting of paradigms from a "cost of doing business" to "wise investments in the business."
In this paper, we will review specific knowledge about the television medium which provides insights into better advertising practices that, when adopted, can lead to more consistent and desirable contribution to the business enterprise:
I. Television advertising, today, may still be the most powerful element in the marketing mix.
II. Continuous airing produces more sales than flighting (with similar weight).
III. Airing advertisements--even those with modest impact--produces more sales than going dark.
IV. Ninety-four percent of all advertisements have a positive impact on sales.
V. Given these findings, it is no longer a matter of whether or not TV advertising is effective but whether it is effective enough to meet the specific business objectives.
VI. While never perfect, the knowledge, measurement, and models are available to account for advertising's impact after the fact, and to forecast the expected contribution of the plan for the next business quarter--before going to air and with time to adjust the plan.
VII. When there are indications that the advertising plan will not meet the business objectives, just a "couple of points" improvement will often make the difference.
VIII. Improvement of a "couple of points" can be achieved through several proven better advertising practices.
IX. Desktop tools for 24/7 decision support, facilitate the adoption of these better practices during the advertising-development and airing cycles.
[FIGURE 1 OMITTED]
In a nutshell, advertising intelligence, measurement, knowledge, models, and "best practice" desktop tools can empower brand and agency teams to know, account for, improve, and achieve a return from investments in TV advertising, consistent with their quarter-to-quarter business objectives. (See Appendix A for definitions of intelligence, measurement, knowledge, models, tools, and best practice as suggested by the Global Research Leaders Summit.)
The body of relevant knowledge on these subjects (how advertising works, what differentiates advertisements with more or less impact, how advertising can be improved, etc.) would be limited if we depended solely on the collective learning from the multitude of one-off studies conducted in the academic or business environments. On the other hand, with sound measurement housed in holistically integrated databases, along with continually funded basic-research activity (or what has often been termed "research on research"), the body of knowledge grows geometrically, adding to our "profound understanding ... of the business process or human behavior." The following learning comes from such an integrated database (see Appendix B) and is based on more than three decades of basic-research activity conducted by The ARS Group.
[FIGURE 2 OMITTED]
I. Television advertising, today, may still be the most powerful element in the marketing mix. The correlations of .71 and .72 (seen in Figures 1 and 2, respectively) indicate that the persuasive power of television commercials accounts for over 50 percent of the variation in market-share changes across brands and categories over a 4- to 12-week period of time (see Appendix C for more information about the ARS Persuasion metric). To our knowledge, no other single variable in the marketing mix has been shown to have such a strong relationship to market changes over a business quarter. (Note that this statement refers to changes in market share over a 4- to 12-week time period. One might expect greater or lesser contribution from other elements in the marketing mix over a shorter or longer period of time, and/or when viewing the business results on the basis of something other than change.)
Although there would be no market impact at all without airing the advertisements, the relationship between advertising weight alone and market response (r = .25) is not as strong as the relationship between advertising "quality" and market response. This finding is not surprising, as it converges with previous ARS Group studies (Blair, 1987, 1993) and other respected industry sources. For instance, Information Resources, Inc., found "no apparent relationship between the size of the weight increase and sales success" (Lodish, 1991). John Philip Jones's short-term advertising strength (STAS) analyses have suggested that "heavy advertising weight is not necessary to generate a positive STAS differential" (Jones, 1995). Note that combining the two measurements (ARS Persuasion levels and gross rating points [GRPs]) into the Persuasion Points Delivered (PPD) metric explains more variance in market response (r = .75) than either variable alone. (This PPD measurement is described in detail in Appendix C.)
A 5-year published case study of Prego spaghetti sauce highlights the power of TV advertising, even when the competition has price, promotion, and mediaspending advantages (Table 1). Former Campbell vice president of marketing research Anthony J. Adams (1997, p. 84) stated, "Prego's advertising managed to overcome Ragu's heavier competitive spending, retailer support, and lower price." (Details of this case study and what happened to the Prego brand over the following 5 years can be found in Appendix D.)
II. Continuous airing produces more sales than flighting (with similar weight). Of the cases in the integrated database for which media data are available over a quarter, about 60 percent involve advertising that aired continuously across three 4-week periods. In the remaining 40 percent, the brand "went dark" after two periods of airing (flighted). A 0,1 dummy variable representing continuity versus flighting proved to be statistically significant at the 95 percent confidence level, with continuous airing yielding a greater impact on market share than flighting (Figure 3).
For example, at a PPD level of 2.5, continuous airing yields a share impact of 1.7 points while a flighting schedule--for the same level of persuasive power and the same number of GRPs--results in an impact of only 1.3 share points. Taking this finding to the business level, say for a brand in a $2-billion market, this difference would mean $1.7 million in retail sales, $1.2 million in net sales for the advertiser, and $1.0 million to bottom-line performance over a business quarter (assuming 75 percent of retail sales go to the advertiser, with 80 percent margin on incremental volume). Using the Ephron and Pollak approach for calculating advertising payback (Ephron and Pollak, 2003), this difference between flighting and continuous airing would result in an incremental advertising payback of $.26 on the dollar (assuming an advertising cost of $3.75 million for a quarter) and would improve the average payback from $.54 on the dollar to $0.80 for consumer packaged goods (CPG) advertisers, and from $.87 to $1.13 for non-CPG advertisers.
These findings are also supported by other industry investigations. In their Ad-Works2 study, Media Marketing Assessment, Inc., and Information Resources, Inc. (1999) concluded, "Continuity plans are more effective than flighted plans." John Philip Jones (2001, p. 88) explained it this way: "If two brands with the same budget, size, media, costs, and advertising elasticity choose to raise their GRP support by say 20 percent, we would see very different volume returns as a result of different patterns of continuity. With additional weeks but no change in the weekly concentration of GRPs, the extra budget would generate extra sales. But if weekly GRP levels are lifted drastically and weeks on-air not increased ... [the] overall television effectiveness would not be improved in line with the budget increase." From ADIMPACT analyses, Reichel (1994) concluded, "The logic of single source leads to the correct solution. Minimize flighting. Strive for continuity." Ephron (1995, p. 21) has also spoken out in favor of continuity: "Flighting is a competitive trap.... If the competition has figured this out and advertises continuously at moderate levels, they'll clean your clock."
[FIGURE 3 OMITTED]
As shown in Table 2, taking into account all three television advertising variables explored--the ARS Persuasion levels of the advertisements airing, GRPs, and flighting--the relationship to quarterly market results becomes even stronger (r = .79). This high correlation between the three advertising variables and market response once again demonstrates the power of television advertising in the marketing arena, explaining over 60 percent of the variation in market-share changes over a business quarter.
III. Airing advertisements--even those with modest impact--produces more sales than going dark. Current analyses suggest that airing advertisements with ARS Persuasion levels of even 2.0 or lower--but greater than zero--has impact in the marketplace versus not advertising; and airing zero-level ARS Persuasion advertising has the same impact as going dark (Table 3).
That is, not advertising--or airing a zero-level advertisement--results in a loss of 0.4 share points over the next business quarter versus no loss with an advertisement scoring 2.0 or less of a loss with an advertisement scoring 0.5 to 1.5 (in the average market).
This "no advertising equals zero-level ARS Persuasion advertising" finding is also supported by no-stimulus testing conducted in the ARS laboratory. In a nostimulus test, respondents are asked to choose a product in the brand's category before and after exposure to television material in which no advertising for that particular category is shown. In other words, there is no advertising stimulus between the two brand choice occasions. As indicated by the nearly perfect relationship shown in Figure 4 ([r.sup.2] = .99), postchoice tends to equal prechoice in no-stimulus tests. Thus a brand's "ARS Persuasion level" is zero in the absence of advertising.
IV. Ninety-four percent of all advertisements have a positive impact on sales. Because the analyses previously described have indicated that airing even low-scoring advertising produces more sales than not advertising--and that not advertising is the same as airing advertisements with zero-level ARS Persuasion results--we can look to the integrated database to determine the proportion of all advertisements tested that have scored above zero and, therefore, would have some impact on market performance. The ARS database distribution of the over 10,000 most recently tested advertisements reveals that nearly all (94 percent) score above zero and would positively impact the market, relative to not advertising (Table 4). This study may be the first in history to suggest that Wanamaker was wrong: only about 6 percent of advertising is wasted, not 50 percent!
V. Given these findings, it is no longer a matter of whether or not TV advertising is effective, but whether it is effective enough to meet the specific business objectives. As the methods of measuring market responses have become more precise over the years--from bimonthly store audits, to split-cable studies, to scanner sales--we have observed more precision in the relationship between those outcomes and the size of ARS Persuasion results. This precision allows for the determination of just how effective an advertisement or series of advertisements is--and whether the level of effectiveness is more than sufficient, or less than sufficient, given the brand's specific business objectives.
[FIGURE 4 OMITTED]
As shown in Table 5, advertisements scoring in the 2.0-2.9 range maintain brand share about half the time, and advertisements scoring in the 3.0-3.9 range maintain share in 80 percent of the cases. As can be seen in the last column of Table 5, the higher the ARS Persuasion level, the higher the average share change observed. Note that when advertising persuasiveness is more than sufficient, additional investment behind the advertising will continue to provide positive returns (see Shepard and Ashley, 2002). When it is less than sufficient, just "a couple of points" improvement will often make the difference (see Section VII).
VI. While never perfect, the knowledge, measurement, and models are available to account for advertising's impact after the fact, and to forecast the expected contribution of the plan for the next business quarter--before going to air and with time to adjust the plan. When one speaks of models in the marketing context, what first comes to mind is "marketing-mix modeling," a method of analyzing how the various elements in a brand's marketing mix have impacted its sales volume. While not perfect, these models have done a good job of accounting for the contribution of television advertising, as well as the other elements in the mix. There are, however, some limitations to using this approach for managing advertising investments: marketing-mix analyses require complex modeling and significant amounts of data, brand by brand, and country by country; it is expensive, particularly for small brands with small budgets; the analysis periods are aligned with advertising "spending" cycles, rather than with the "quarterly" business standard; and perhaps most notable is that while market mix modeling may provide an understanding of how marketing events have impacted sales in the past, they provide little knowledge or direction about what and how to improve in order to achieve a particular business objective in the future. This is of particular importance given the recent conclusion from the Ephron and Pollak (2003) market mix database analysis that, on average, advertising does not yield a positive return.
These limitations can be addressed, however, since the main components of the advertising plan (current selling power of the advertisements on hand, GRPs, and flighting) are quantifiable, and their impact can be projected before going to air and with time to adjust the plan.
VII. When there are indications that the advertising plan will not meet the business objectives, just a "couple of points" improvement will often make the difference. A 2-point difference in ARS Persuasion points relates to a 0.5-point share difference in the average market (4-week period after airing versus 4-week period before airing), as demonstrated by the slope of the best-fit line in Figure 1. Over a business quarter, a 2-point improvement is associated with an average share difference of 0.4 points (Figure 2, best-fit slope).
For the brand in the $2 billion category, this 2-point difference translates into a difference of $2.8 million in retail sales and an improvement in advertising payback of $.45 over a business quarter (again assuming 75 percent of retail sales go to the advertiser, with 80 percent margin on incremental sales, and a $3.75 million cost of advertising for the quarter). Thus, improving the effectiveness of the advertising just a "couple of points" can improve the advertising payback, moving from the Ephron and Pollak payback of $.54 for the average CPG brand to $0.99 in this scenario.
The significance of these findings can also be illustrated over a longer period of time, in an 11-year "best in class" study comparing Duracell and Eveready batteries (Figure 5). Both Duracell and Eveready grew during the decade under study, but Duracell managed to build the brand to the number-one position, consistently fueled by more persuasive advertising of "just over a point." (A more complete description of the Duracell case study and its aftermath can be found in Appendix D.)
[FIGURE 5 OMITTED]
VIII. Improvement of a "couple of points" can be achieved through several proven better advertising practices.
Focusing advertising development behind a strong selling, or value, proposition leads to an average improvement of 2.0 ARS Persuasion points. Twenty years ago Ogilvy (1983, p. 160) proposed that "the selection of the promise is the most valuable contribution that research can make to the advertising process." Our findings support Ogilvy's assertion, demonstrating that creatives are most likely to achieve business success when they have a strong proposition from which to work.
A selling/value proposition as tested by The ARS Group is a "bare bones" video without executional enhancements, such as on-camera presenters, sound effects, music, background sets, and visual memory devices. This format allows the measurement to isolate the strength of the basic promise, apart from the advertising execution.
As shown in Table 6, executing from a superior proposition results in superior advertising executions over two-thirds of the time. Conversely, there have been no instances in which superior advertising resulted from a below-average proposition.
To quote Raymond (1989): "Your commercial and your advertising can be no better than the copy strategy on which they are built.... It takes an awesomely poor execution to bury a shining strategy. But no amount of executional brilliance can create a marketing strategy that isn't there."
While differences in creative execution generate advertising with a wide range of sales effectiveness, on average they will tend toward an ARS Persuasion level similar to that of their underlying proposition. As shown in Figure 6, improving the strength of the underlying proposition just "a couple of points" tends to have a similar effect on the average of the creative executions.
A recent analysis quantified the average increase in ARS Persuasion levels when advertisers have employed the discipline of testing video propositions in the ARS lab. On average, brands conducting ARS Firstep tests increased the selling power of their subsequent advertisements 1.2 ARS Persuasion points (see Figure 7). When advertisers were able to identify a superior proposition, the benefits of Firstep testing were even greater, improving their subsequent advertising executions an average of 2.0 ARS Persuasion points.
Understanding and using validated strategic and executional content "drivers" leads to an average improvement of 2.3 ARS Persuasion points. Normally, when advertisements are revised or "tweaked" after copy testing, but prior to airing, the revised versions are not tested to see if there was any improvement in selling power. According to SmithKline Beecham's Dan Shirley (1999, p. 36), "To evaluate the improvement track record of diagnostics, it is necessary to examine the broader context of the process in which they are used. Whether due to their rush to get commercials on air or their optimistic assumption that change is synonymous with improvement, there is the temptation ... to air revised ads without retesting them."
In the same article, Shirley revealed the results of a due-diligence exercise examining the track record of using traditional "communications" tests/"diagnostics" and validated "drivers" for direction. The drivers are a handful of 150+ content elements that, on an individual basis, and together, have been shown to relate to higher ARS Persuasion results (see Appendix B for description). The exercise suggested that adding these drivers to guide pre-airing revisions is another proven means of achieving "a couple of points" improvement. On average, ARS Persuasion results are 2.3 points higher when the validated drivers are used to guide revisions (Table 7).
Note that the same drivers can be used at the storyboard stage to review execution alternatives prior to production. When used at this early stage and in conjunction with proposition testing, the likelihood of developing strong advertising increases to over 80 percent.
Accounting for the wearout factor when planning the number of poolouts to shoot can increase selling power dramatically. In the 1980s, The ARS Group found that as GRPs are spent behind an execution, its selling power decreases predictably, indicating the speed at which the advertising power is delivered to market (Blair, 1987). Likewise, share levels increase when persuasive executions are first aired but diminish quickly (Figure 8). On average, the largest increase occurs during the first period the advertisement is on air, when the execution's ARS Persuasion power is strongest. A smaller increase occurs as the advertisement continues to air, and small decreases are likely to occur as it airs further.
It also is important to note that executions wear out independently; that is, airing one execution would not be expected to affect the wearout of another, even if they are similar. For 57 cases involving executions that were part of a larger pool, The ARS Group's Basic Research Team calculated two wearout projections, one using execution-specific GRPs and another using total GRPs for the pool. The data show a much stronger relationship when execution-specific GRPs are used (r = +0.87 versus +0.68), supporting the hypothesis that executions wear out independently from their counterparts or poolouts (Figure 9).
These wearout findings suggest the importance of planning the number of poolouts needed before the actual production "shoot," and later using a refreshment schedule based on the unique sales effectiveness of the poolouts available to air. For example, airing two executions instead of one, given similar persuasiveness and media weight, would result in a difference of $3.1 million in retail sales and incremental advertising payback of $.50 (using the same assumptions as in previous examples).
Allocating more media weight behind effective 15-second advertisements can also improve selling power dramatically. Rising media costs are causing advertisers to take another look at the viability of 15-second advertisements. The shorter spots can be purchased for about half the cost of 30-second units, making them an attractive option for advertisers interested in stretching their media budget. However, the real opportunity afforded by 15-second advertisements comes from knowing and leveraging their individual selling power.
In a recent study exploring the effectiveness and use of 15-second advertisements (Rabuck, 2002), the author found that in the 288 pairs examined, 24 percent achieved ARS Persuasion levels half that (or lower) of their 30-second counterparts, and 27 percent achieved results the same as or higher than their 30-second counterparts. While those scoring 50 percent or below do not represent a business opportunity, those scoring at or above hold tremendous opportunities.
Table 8 shows that the brand's three 15-second "poolouts" score the same as their 30-second counterpart. Placing all the weight (double the GRPs) behind the 15-second advertisements would result in a volume impact of $12.6 million over the quarter versus $5.6 million with all the weight behind the 30-second commercial, a difference of $7.0 million in retail sales over the business quarter. This difference would add $5.2 million to the advertiser's top-line sales and $4.2 million to bottomline performance (again assuming 75 percent of retail sales go to the advertiser with 80 percent margin on incremental volume). And once again, this shift in advertising behavior would improve the advertising payback by $.70, turning the average payback of $.54 to a positive $1.24 for CPGs and from $.87 to $1.57 for non-CPGs.
These findings and their implications suggest the better practice of testing all the executions (and lengths) produced in order to know their individual selling power for allocating media dollars regardless of length.
IX. Desktop tools for 24/7 decision support, facilitate the adoption of these better practices during the advertising-development and airing cycles. It is one thing to establish the measurement standards and integrated databases, another to accumulate the knowledge that leads to better practice insight, and yet another to facilitate the adoption of these practices. For marketing and advertising professionals who want to know, account for, and improve the return from their advertising activities, creation of the right support tools can act as the catalyst for this transformation--tools that deliver the right information (lead to improved performance/ROI), to the right place (desktops of decision makers), at the right time (24/7), and in the right form (easy to adopt and act upon, indicate a clear course of action). Following are descriptions of some "Best Practice" tools, and Figure 10 shows where they might be used in the context of the advertisingdevelopment and airing cycles.
[FIGURE 8 OMITTED]
The desktop advertising planner incorporates the knowledge about how advertising quality, GRPs, and continuity are likely to impact the market over the next business quarter. This type of Best Practice tool can support advertising development and management during several stages of the cycle:
[FIGURE 9 OMITTED]
1. Early on, it takes into account the media budget and business objectives to determine how strong the selling/value proposition needs to be.
2. After the appropriate proposition has been identified, the tool uses the selling power of that proposition and planned media spending as inputs to determine the number of executions/poolouts needed to meet the business objectives.
3. After the executions/poolouts have been produced, the planner takes into account the selling power of each one to plan the optimal allocation of GRPs (accounting for 30-second/15-second cost differentials) and to determine when the advertisements should be refreshed with other executions/poolouts.
4. The tool also indicates when the optimal configuration of the plan components will not deliver enough power to achieve the objectives (with time to adjust the plan components before airing), or if there is more than enough to meet the plan (with time to invest additional media dollars against the business opportunity).
Selling value proposition testing provides predictive measurement feedback upstream in the advertising-development cycle to identify propositions strong enough to support advertising executions that will meet the business objectives. This tool uses the same measurement and medium used to evaluate TV advertisements, thus avoiding the form, measurement, and modeling gaps that often exist when moving from "concept to copy." Testing at this stage in the process offers an opportunity to take risks, learn, and identify the "big ideas," avoiding expense and frustration later in the advertising-development cycle. Or put another way, get the proposition right, at the "get go," and improve the selling power of the subsequent advertisements "a couple of points."
Storyboard evaluation helps brand and agency teams transition from a strong proposition to strong advertising executions. A validated strategic-and-executionalcontent profile and interactive consumer feedback are used when reviewing storyboards to ensure that they stay true to the chosen proposition, incorporate executional enhancements that support the proposition, and avoid executional overload that may detract from the power of the proposition.
Advertising testing of the final executions going to air ensures that the upstream process is working, identifies the selling power of each advertisement for more precise forecasting of the business outcomes, and for optimizing the media investment by allocating weight based on the business value of each execution.
Competitive intelligence delivers "real-time" global competitive intelligence to the advertiser's desktop providing global vision and enabling the tactical and strategic responses necessary to compete in today's fast-paced business environment. When coupled with measurement feedback, it also serves to facilitate "best in class" determinations and to identify and quantify competitive opportunities and/or threats.
[FIGURE 10 OMITTED]
SUMMARY AND IMPLICATIONS
We have reviewed specific learning, or knowledge, about the television medium which provides insight into better advertising practices; practices that when adopted, can lead to a more consistent contribution to the business enterprise, as well as improvement in performance--when necessary or desirable--and therefore alignment with the investment objectives:
I. Television advertising, today, may still be the most powerful element in the marketing mix. Be sure to use sound measurements and models capable of accurately forecasting what the business impact will be, and invest accordingly.
II. Continuous airing produces more sales than flighting (with similar weight). Plan for continuous airing versus flighting over the business quarter.
III. Airing advertisements--even those with modest impact--produces more sales than going dark. Consider the market declines associated with not advertising before deciding to go dark.
IV. Ninety-four percent of all advertisements have a positive impact on sales. Do not waste media dollars on the 6 percent of advertisements with zero persuasiveness.
V. Given these findings, it is no longer a matter of whether or not TV advertising is effective, but whether it is effective enough to meet the specific business objectives.
* Establish specific quarterly sales objectives as the cornerstone of the advertising plan (e.g., defend/maintain share, grow a certain amount, etc.).
* When exceptional advertising opportunities come along, invest more media dollars behand them. Do not leave money on the table.
VI. While never perfect, the knowledge, measurement, and models are available to account for advertising's impact after the fact, and to forecast the expected contribution of the plan for the next business quarter--before going to air and with time to adjust the plan. Use measurements and models that take into account the main components of the advertising plan (current selling power of the advertisements, planned GRPs, and flighting) and determine their effects before media spending has actually taken place.
VII. When there are indications that the advertising plan will not meet the business objectives, just a "couple of points" improvement will often make the difference. Be aware of the plan's odds of success early in the process, and be prepared to take the necessary actions to improve.
VIII. Improvement of a "couple of points" can be achieved through several proven better practices.
* Test upstream in the advertising-development cycle to find a selling/value proposition strong enough to support advertising executions that can meet the business objectives.
* Sort storyboards and make revisions to advertisements guided by "validated-to-improvement" feedback. Test final copy to be sure that any executional enhancements and revisions have had the desired effect, allocating media weight accordingly.
* Account for the wearout factor before going to the production "shoot," planning for the appropriate number of executions or poolouts needed to deliver the desired selling power.
* Do not sell short advertisements short; allocate media dollars relative to an advertisement's unique selling power, regardless of its length.
IX. Desktop tools for 24/7 decision support facilitate the adoption of better practices during the advertising-development and airing cycles.
* In order to achieve the specific business objectives, use a desktop advertising planner and sound measurement to determine how strong a proposition is needed, how many executions or poolouts to produce, when the executions should be refreshed, and how to allocate media dollars behind them.
* View competitive advertising activity, in real time, in order to assess, quantify, and respond to any "best in class" threats before incurring a prolonged negative impact to the business.
The business implications for adopting these types of empirically supported better practices are enormous. For instance, the better practice examples cited earlier in this article would add $1.7 million to $7.0 million in retail sales over a single business quarter and improve the average advertising track record from a payback of $.54 and $.87 on the dollar to $0.80 to $1.57 on the dollar (see Table 9). Projecting these results to a business enterprise of, for example, 10 brands and for a full fiscal year would return $48 to $208 million more in net sales, and $40 to $168 million more in operating income, from a similar level of advertising investment.
Furthermore, these levels of improved performance and return on investment are not merely hypothetical. Many practitioners have used these better practices in advertising and have experienced exceptional results. Some have even published their experiences (Adams, 1997; Bean, 1995; Conlin, 1994; Cox, 1995; Masterson, 1999; Mondello, 1996; Shepard and Ashley, 2002; Shirley, 1999). As former Senate Minority Leader Everett Dirksen once observed, "A billion here, a billion there, and pretty soon you're talking real money."
The intelligence, measurement, knowledge, models, and "best practice" tools discussed in this article are the types of "products" being developed by the 21st century business researchers who are determined to add quantifiable value to the business enterprise, and the fact-based support being used by the brand and agency teams that are determined to win in the marketplace, quarter-to-quarter and year-to-year. These teams will know, account for, improve, and achieve the return from investments in advertising consistent with their business objectives (whether those objectives are to defend/maintain the current market position, to grow by a predetermined amount, or to meet ROI goals). The quest to meet quarterly business objectives will change from the roller-coaster rides of advertising spending cuts for bottom-line performance to consistent contribution to both the top and bottom lines--or from a "cost of doing business to wise investments in the business."
In January 2001, the first Global Research Leaders Summit attracted 50 market-research industry leaders. Organized by ARF and ESOMAR, the purpose of the global group was to identify and address the challenges and opportunities facing the research industry in the 21st century. The following working definitions used in this article come from Empowering the Relaunch of Research (RELEAS, 2003) and Realizing the Vision for Market/Business Research (RELEAS, 2002):
Business Intelligence: actionable, reliable insight into the current business/market context
Measurement: metrics that reliably identify business opportunities or threats, given the current context and potential actions
Models: analytical techniques that represent the causal relationships among various conditions and actions taken to achieve specific business results, and accurately forecast the future outcomes of various potential actions and conditions
Knowledge: profound understanding that yields a clear prioritization of action; learning or principles that yield true predictions with unvarying uniformity
Tools: systems that place all of the above on the desktop for 24/7 decision support
Best Practice: a documented method of operating behavior that yields a higher level of performance (ROI) than other operating behaviors
The Integrated Database
The ARS Group's database includes ARS Persuasion and recall measurement facts from about 40,000 tests of television commercials and selling/value propositions and spans three decades, 6,000 brands, and nine countries.
For each advertisement tested since the mid 1980s, the database also contains the 150+ strategic and executional content elements that were identified for the Stewart and Furse study (1984), based on collaboration between the Marketing Science Institute (its advertiser and agency members), Vanderbilt University, and The ARS Group. These elements range from promises and appeals, to comparisons, to advertising tone or atmosphere, to structure and format including the timing of various elements.
The database also includes the market results for over 2,000 advertisements that subsequently aired as tested, with corresponding media weight for several hundred of the cases. The market results are those generated from new-product tracking studies, split-cable tests, Nielsen bimonthly audits, and the current standards of SCANTRACK, InfoScan, IMS prescription audits for pharmaceutical brands, and marketing mix modeling outcomes.
This database also houses over 3,000 re-tests generated by our ongoing quality programs to ensure measurement standards such as reliability, validity, relevance, calibration, and precision, as well as those conducted for continual assessment of the "wearin and wearout" phenomenon.
The Advertising Measurements
Stanford University Professor Michael L. Ray (1979) wrote: "It is clear that if measurement is disregarded in marketing research, the field will be slow to advance. What most observers do not recognize is that measurement development is not only a scientific requirement, but also a practical necessity." In the same Journal of Marketing Research editorial, he explains the necessary direction of that development: "Managerial objectives can be met only when measures are reliable (contain little irrelevant measurement error) and valid (measure just the constructs they are supposed to be measuring)."
Over the years, The ARS Group has regarded advertising measurement as the critical cornerstone for productive research related to advertising performance and improvement. Consequently, much of the group's activities have centered around ensuring high-measurement standards.
THE ARS PERSUASION METRIC
The ARS Persuasion metric (a behavioral measure of brand choice taken before and after advertising exposure) has consistently demonstrated levels of correspondence, or validity, in the .80-.90 range when compared to the actual market impact in advertising-only controlled or isolated environments, such as controlled tests markets split-cable tests and marketing mix analyses (The ARS Group, 1983, 2004; Blair, 1987; Blair and Rabuck, 1998). When compared to the market results from "uncontrolled" environments where variations in the other elements of the marketing mix also come into play, the level of correspondence has been in the .60-.70 range (The ARS Group, 2004; Ashley, 1998; Buzzell, 1964; Dodd, 1964; Kelly, 1964; Murphy, 1968).
Currently, there is a strong relationship between ARS Persuasion scores and subsequent "uncontrolled" market results (in the following 4-week period) (as was shown in Figure 1). This relationship continues when the analysis period is extended to quarterly market response, the time period aligned with business planning and accountability cycles (Figure 2). The metric is also calibrated across brands, categories, and countries; a "2" is a "2" and a "7" is a "7" in the United States, Mexico, Germany, etc., making it easy to understand and act upon for the global advertiser. Note that this relationship holds across the following types of brands, categories, and advertising: established and new brands; large and small brands; multiple-purchase categories; food, household, over-the-counter, and personal-care categories; direct-to-consumer pharmaceuticals; high-, average-, and low-scoring advertisements; 15-and 30-second advertisements; and rational and emotional advertisements (The ARS Group, 2004).
The high correlations between the ARS Persuasion scores of advertisements aired and subsequent market responses demonstrate both the relevant predictive validity of the measurement and its appropriateness for identifying business opportunities (and/or threats from competitors), given the current market context and potential (airing) actions.
As found in previous examinations, the measurement of related recall alone has a statistically significant relationship to market results (r = +.31). Measurements of related recall can add to the predictive power of persuasiveness measurements, or not, depending on the nature or precision of the persuasiveness methodology.
Related recall also accounts for 14 percent of the variance in ARS Persuasion outcomes (r = +.37), but it does not add any sales-predictive power beyond that of the ARS Persuasion measurement alone; in other words, the effects of attention and memorability that relate to sales are captured in the behavioral ARS Persuasion measurement. As such, the measurement of related recall is useful as a diagnostic for ARS Persuasion results, but not for explaining additional variance in market impact (The ARS Group, 1997c, 2004).
UNDERSTANDING THE PERSUASION POINTS DELIVERED METRIC
Previous studies (Blair, 1987; Blair and Rabuck, 1998) have shown that when persuasive advertising is aired, sales build quickly--with diminishing returns as the advertisement's selling power is delivered to market. Likewise, the advertisement's selling power declines or wears out in the process. This market build and the corresponding advertising wearout are both predictable occurrences, given the persuasive power of the advertisement at hand and the GRPs spent behind it (r = +.85). The Persuasion Points Delivered (PPD) metric combines ARS Persuasion power and GRPs, while taking into account this wearout phenomenon:
Market-share change = f (PPD)
PPD = f[ARS Persuasion level] X [(GRPs) (wearout factor)]
The Case Studies
PREGO SPAGHETTI SAUCE
In late 1987, sales for Prego were in a slump. The brand and agency had high hopes for a new advertisement that positioned Prego as tasting more like homemade because it is rich with herbs and spices. The advertisement was tested and attained an ARS Persuasion level of +5.8, meeting Campbell's airing standard of +4.0. Prego's market share increased nearly 2 points when the advertisement aired, but a wearout projection indicated that the execution would be worn down to the +4.0 guideline within a month at planned spending levels.
In an effort to differentiate the brand from its competitors, the company decided to change the Prego strategy. While retaining the concept that Prego tastes more like homemade, the new direction emphasized Prego's thickness. The first execution based on the new selling proposition achieved an ARS Persuasion score of +4.5. Believing that the strategy had higher potential, the advertisement was revised, and the new version achieved an ARS Persuasion score of +10.0. When the second advertisement went to air, Prego share increased 4.5 points.
Later, two more executions were produced using the same selling proposition and achieved scores of +6.0 and +10.9. The stronger of the two advertisements went to air and once again share increased. Adams and Blair (1992, p. 24) concluded:
The Campbell story is one of a brand group and agency which are advertising more confidently. They know how to develop and identify ads that sell; know how to leverage media dollars behind them; know when to refresh creative; know how to plan both sides of the advertising dimension simultaneously; and know how to achieve sales results with television advertising both short-term and over time. The implications from this story speak to the request for advertising accountability. Campbell and rsc are laying in the systems for improving advertising productivity and establishing advertising accountability.
Over the next 5 years, The ARS Group monitored all the major brands in the spaghetti-sauce category, gathering data on sales and market share, ARS Persuasion levels for the advertisements aired, and media spending. There was significant new-product activity, with Classico, Hunt's, Contadina, Healthy Choice, and Campbell's brands entering the category and gaining about 25 percent share of market. During the same time period, Ragu lost roughly 19 share points.
Prego continued to air strong advertisements based on the "thickness" strategy and gained share--despite the new-product activity--by having found a strong selling proposition and by continuing to refresh with persuasive executions. Over the 5-year period, Prego's advertising managed to overcome Ragu's heavier competitive spending, retailer support, and lower price (shown in Table 1). Campbell's Dick Nelson (The ARS Group, 1996, pp. 2-3) came to this conclusion:
What underlies this five-year-long success story? A fundamental change in the advertising strategy and research process. Prego is the only Campbell's brand in the last five years to:
* consistently stay with the same [productive] selling proposition
* [ARS] Persuasion test every poolout prior to airing
* establish [ARS] Persuasion hurdles and stick to them
* utilize outlook to create an awareness of when to refresh creative
At the end of this successful 5-year run, the head of the Prego marketing team was promoted to head a larger division of the company's brands, and new members of the Prego brand and agency team drifted away from the research process that had guided the success (The ARS Group, 1997b). Share declined, nearly reaching preprocess levels. The decade-long study suggests that when better practices for achieving consistent advertising success have been identified, they should be adopted at an enterprise level and in such a manner that they will survive and improve over time, despite the frequent changes in brand and agency personnel.
During this 11-year case study, Duracell took over and fortified the number-one spot, fueled by more persuasive advertising (an average ARS Persuasion level of +5.1 versus Eveready's lower persuasiveness of +3.9).
Duracell stuck with the same brand-differentiating benefit (lasts longer) and brand personality (high quality/trustworthy) over the entire 11-year period. They leveraged this consistent strategy and equity across time and the world, producing phenomenal marketing and business results. A Proprietary Brand Drivers reading indicated that their "long-lasting" benefit was played back at 2.5 times the average advertisement. Their brand personality, "high quality/trustworthy," was played back at five times the average advertisement, and their overall "equity" score was over three times the average.
Duracell's brand building and equity can also be seen in financial terms. By the end of the 11-year period, Duracell had moved from the number-two position in the category to number one, with a market share of 44 percent versus Energizer's 32 percent--and with a higher price point than Energizer ($1.02 versus $.86). At $609 million, its profits were more than double those of Energizer, and the market value of the Duracell Company was over 8 billion dollars when sold, nearly triple that of Energizer when sold.
At the end of this successful 11-year run, the Duracell brand was sold to The Gillette Company. The new members of the brand and agency team did not adopt--and may not have even known of--the measurement and research better practices that had supported Duracell's success. Subsequently the health of the brand eroded.
The outcomes of both the Prego and Duracell case studies emphasize the wisdom behind this statement by Stalk, Evans, and Shulman (1992, p. 60): "The building blocks of corporate strategy are not products and markets but business processes."
TABLE 1 Prego Case Study (5-Year Overview) Prego Ragu Total GRPs 15,034 [left arrow] 20,400 Average displays 22 [left arrow] 43 Average retailer advertisements 29 [left arrow] 37 Average selling price $1.80 [left arrow] $1.64 Total TV power (PRP delivery (1)) 679 [right arrow] 448 Sales gains (units) (2) +22% [right arrow] -19% (1) PRP is an acronym for Persuasive Rating Point. The PRP measurement, a composite of ARS Persuasion scores and GRPs, was a precursor to the PPD metric described in Appendix C. (2) Market-share increase versus base period (last 20 weeks of 1987). Source: Adams (1997). TABLE 2 ARS Persuasion Measure, GRPs, and Flighting: Relationship to Quarterly Market-Share Change (n = 62) Adjusted F r [r.sup.2] [r.sup.2] F Significance GRPs alone .25 .06 .05 4.08 95% ARS Persuasion metric alone .70* .49 .48 57.46 99%+ ARS Persuasion level and GRPs (PPD) .75 .56 .55 75.75 99%+ ARS Persuasion level, GRPs, and flighting .79 .62 .61 98.16 99%+ * Among subset of cases with media data. r = .72 among total data set. Source: The ARS Group (2004). TABLE 3 Airing Low-Scoring Advertisements Produces Greater Sales Effects Than Not Advertising Starting ARS Persuasion Score Quarterly Best-Fit Market-Share Change* 2.0 [+ or -]0.0 1.5 -0.1 1.0 -0.2 0.5 -0.3 0.0 -0.4 No advertising -0.4 * Based on average per-period GRPs across validity dataset. Source: The ARS Group (2004). TABLE 4 Airing Low-Scoring Advertisements Produces Greater Sales Effects Than Not Advertising Cumulative Percentage of Advertisements ... ARS Scoring Below Scoring Above Persuasion ARS Persuasion ARS Persuasion Level Level Level 12.0 92% 8% 9.0 86% 14% 7.0 77% 23% 4.0 54% 46% 3.0 42% 58% 2.0 28% 72% 1.0 15% 85% 0.0 6% 94% Source: The ARS Group (2004). TABLE 5 ARS Persuasion Scores to Market Results (Period Before Airing to Period After Airing, n = 332) Percent of Advertisements Achieving Share-Point ARS Persuasion Difference of: Average Share Ranges n 0.0+ 0.5+ 1.0+ 2.0+ Change Observed 12.0+ 29 100% 100% 94% 83% +5.4 9.0-11.9 31 100% 97% 72% 49% +2.2 7.0-8.9 35 100% 87% 56% 36% +1.6 4.0-6.9 86 80% 58% 33% 9% +0.8 3.0-3.9 37 80% 46% 26% 6% +0.5 2.0-2.9 49 53% 19% 6% 0% +0.0 <2.0 65 47% 12% 2% 0% -0.2 Source: The ARS Group (2004). TABLE 6 Superior Propositions Lead to Superior Selling Advertisements- and Vice Versa (ARS Persuasion Results)* Resulting Execution Basic Proposition Below Average Average Superior Below average 67% 33% 0% Average 22% 68% 11% Superior 0% 31% 69% Note: The selling/value propositions and executions in this database are determined to be "below average," "average," or "superior" based on their relationship to the Fair Share degree-of-difficulty norm (at the 90 percent confidence level). This benclunark takes into account category/brand loyalty, the number of brands competing in the category, and the advertised brand's share to determine the score expected, on average, given the current category and brand's environment. *Based on 62 brands and 192 selling proposition/advertising pairs. Source: The ARS Group (2001). Average ARS Persuasion Selling Level of Resulting Proposition Advertising Executions 12.0+ +16.3 9.0 to 11.9 +8.9 7.0 to 8.9 +7.1 4.0 to 6.9 +4.6 3.0 to 3.9 +3.7 2.0 to 2.9 +2.8 <2.0 +1.1 Source: The ARS Group (2001). Figure 6 Stronger Propositions Lead to Stronger Advertisements (ARS Persuasion Results) Note: Table made from bar graph. Average ARS Average ARS Persuasion Persuasion level level after Firstep before Firstep testing testing All cases +3.8 +5.0 Brands achieving average or better propositions +4.3 +5.5 Brands achieving superior propositions +5.3 +7.3 Source: The ARS Group (2001). Figure 7 "Before and After" Firstep Analysis (Average ARS Persuasion Level of Executions) Note: Table made from bar graph. TABLE 7 Validated Drivers Help Improve Selling Power "a Couple of Points" Percent of Average Improvement Advertisements with in ARS Persuasion Results Significantly Higher (Versus Using Only Test-Revise-Test Outcomes Traditional Diagnostics) 40% +2.3 Note: In this ongoing study, advertisements that have been tested then revised are tested again in revised form. The ARS Persuasion score of the revised version is compared to the score of the original. The driver and diagnostic input is considered to have improved the advertisement if the score of the revised advertisement is significantly higher than that of the original at the 90 percent confidence level. Sources: Shirley (1999) and The ARS Group (2000). TABLE 8 Using Strong 15-Second Advertisements Can Increase Selling Power Dramatically, Adding More Return to Both Top- and Bottom-Line Performance $ Volume PPD Impacted Plan Scenario (1) Delivery over Quarter (2) Difference Three 15-second 3.6 $12.6 million $7.0 million in advertisements retail sales over the quarter One 30-second 1.6 $5.6 million advertisement (1) All four advertisements achieved similar ARS Persuasion levels, and both scenarios invest $6 million in media over the quarter. (2) Category with $2 billion in annual retail sales. Source: Rabuck (2002). TABLE 9 Improving Advertising Payback with Better Practices (Examples) Payback Better Practice Better Practice Better Practice Improvement Payback (CPG) Payback (non-CPG) (Base level) (1) (n.a.) (2) ($0.54) ($0.87) Continuity vs. flighting +$0.26 $0.80 $1.13 Advertisements based on strong proposition +$0.45 $0.99 $1.32 Revisions based on validated drivers +$0.50 $1.04 $1.37 Account/plan for wearout when producing +$0.50 $1.04 $1.37 More weight behind strong 15-second advertisements +$0.70 $1.24 $1.57 Note: Assuming a $2 billion market. 75 percent of retail sales go to the advertiser, 80 percent margin on incremental volume, and an advertising cost of $3.75 million per quarter. (1) Ephron and Pollak (2003). (2) n.a. = not applicable.
ADAMS, ANTHONY J. "Advertising Research." In Dartnell's Advertising Manager's Handbook, Fourth Edition, David Bushko, ed. Chicago, IL: The Dartnell Corporation, 1997.
_____, and MARGARET H. BLAIR. "Persuasive Advertising and Sales Accountability: Past Experience and Forward Validation." Journal of Advertising Research 32, 2 (1992): 20-25.
THE ARS GROUP. Advertising Caused Awareness and Trial: ARS Predictive Validity for New Brand Advertising. Evansville, IN: The ARS Group, 1983.
_____. "Better Practices for Improving Advertising Performance." Better Practices in Advertising, Issue 1 (2002): 1-2.
_____. Brand Performance Before and After Firstep Testing. Evansville, IN: The ARS Group, 2001.
_____. Campbell's Prego: An Advertising Success Case. Evansville, IN: The ARS Group, 1996.
_____. The Effective Delivery of Advertising and Commercial Wearout. Evansville, IN: The ARS Group, 1997a.
_____. Lessons Learned: A Case Study. Evansville, IN: The ARS Group, 1997b.
_____. More About Recall. Evansville, IN: The ARS Group, 1997c.
_____. Summary of The ARS Group's Global Validation and Business Implications. Evansville, IN: The ARS Group, 2004.
_____. Update on ARS Persuasion Levels for Firstep Videos and Resulting Executions. Evansville, IN: The ARS Group, 2001.
_____. "Using ARS Validated Drivers Can Help Improve ARS Persuasion Scores 'a Couple of Points'!" Quality Measurement in Advertising 5, 1 (2000): 7.
ASHLEY, S. R. "How to Effectively Compete Against Private-Label Brands." Journal of Advertising Research 38, 1 (1998): 75-82.
BEAN, BILL. "Oscar Mayer Lunchables: The In-Market Effects of Advertising." In Transcript Proceedings of rsc T.Q. Advertising Success Forum I. Evansville, IN: The ARS Group, 1995.
BLAIR, MARGARET H. "An Empirical Investigation of Advertising Wearin and Wearout." Journal of Advertising Research 27, 6 (1987): 45-50.
_____. "Moving Beyond GRPs to PRPs: Another Major Challenge for the 21st Century." Presented at the Advertising Research Foundation Twelfth Annual Electronic Media Workshop, December 1993.
_____, and MICHAEL J. RABUCK. "Advertising Wearin and Wearout: Ten Years Later." Journal of Advertising Research 38, 5 (1998): 7-18.
_____, and HANS-WILLI SCHROIFF. "Advertising: Today's Sale or Brand Building for Tomorrow?" Presented at the Advertising Research Foundation Annual Convention, March 2000.
BUZZELL, R. D. "Predicting Short-Term Changes in Market Share as a Function of Advertising Strategy." Journal of Marketing Research 1, 3 (1964): 27-31.
CONLIN, RONALD P. "Goodyear Advertising Research: Past, Present, and Future." Journal of Advertising Research 34, 3 (1994): 7-10.
COX, DOUG. "The Citrucel Case Study." In Transcript Proceedings of rsc T. Q. Advertising Success Forum I. Evansville, IN: The ARS Group, 1995.
DODD, A. R., JR. "New Study Tells TV Advertisers How Advertising Builds Sales and Share of Market." Printers' Ink, May 8, 1964.
EPHRON, ERWIN. "More Weeks, Less Weight: The Shelf-Space Model of Advertising." Journal of Advertising Research 35, 3 (1995): 18-23.
_____, and GERRY POLLAK. "Finding the Other Half." Presented at ARF/ESOMAR Conference, June 2003.
JONES, JOHN PHILIP. "Does Advertising Produce Sales Today or Sales Tomorrow?" Journal of Marketing Communications 1, 1 (1995): 1-11.
_____. The Ultimate Secrets of Advertising. Thousand Oaks, CA: SAGE Publications, Inc., 2001.
KELLY, P. J. "The Schwerin Model: How You Can Use It to Build Your Share of Market." Printers' Ink, May 8, 1964.
LODISH, L. M. "Key Findings from the 'How Advertising Works' Study." In Transcript Proceedings of the ARF Marketplace Advertising Research Workshop. New York, NY: Advertising Research Foundation, 1991.
MASTERSON, PEGGY. "The Wearout Phenomenon." Marketing Research 11, 3 (1999): 26-31.
MEDIA MARKETING ASSESSMENT, INC., and INFORMATION RESOURCES, INC. AdWorks2, Report of presentation, 1999.
MONDELLO, MIKE. "Turning Research into Return-on-Investment." Journal of Advertising Research 36, 4 (1996): 2-6.
MURPHY, M. P. "Empirical Evidence of the Effect of Advertising on Sales." Speech presented in Toronto to the Professional Marketing Research Society, October 1968.
OGILVY, DAVID. Ogilvy on Advertising. New York, NY: Random House, 1983.
RABUCK, MICHAEL J. "Don't Sell Short Ads Short: Trends in TV Commercial Length in the U.S." Presented at the ARF Week of Workshops, October 2002.
RAY, MICHAEL L. "Introduction to the Special Section: Measurement and Marketing Research--Is the Flirtation Going to Lead to a Romance?" Journal of Marketing Research 16, 1 (1979): 1-5.
RAYMOND, MINER. Advertising That Sells: A Primer for Product Managers. Cincinnati, OH: Black Rose, 1989.
REICHEL, WALTER. "Beyond 'Effective Frequency': How to Maximize Marketplace Impact Using New Data-Based Approaches to Media Scheduling." In Transcript Proceedings of the ARF Effective Frequency Research Day. New York, NY: Advertising Research Foundation, 1994.
RELEAS. Empowering the Relaunch of Research. From the Global Research Online Workshop of RELEAS3, January 2003.
_____. "Realizing the Vision for Market/Business Research." Presentation distributed by RELEAS, 2002.
SHEPARD, BARRY, and SUSAN R. ASHLEY. "Developing and Managing Advertising with a More Positive Return on Investment." Presented at the ARF Annual Conference, April 2002.
SHIRLEY, DAN. "From Copy Testing and Diagnostics to Process-Driven Improvement: A Paradigm Shift in Advertising Development." ANA/The Advertiser, April/May 1999.
STALK, G., P. EVANS, and L. E. SHULMAN. "Competing on Capabilities: The New Rules of Corporate Strategy." Harvard Business Review, March/April 1992.
STEWART, DAVID W., and DAVID H. FURSE. "Analysis of the Impact of Executional Factors on Advertising Performance." Journal of Advertising Research 24, 6 (1984): 23-26.
MARGARET H. BLAIR
The ARS Group MHBlair@ars-group.com
ALLAN R. KUSE
The ARS Group ARKuse@ars-group.com
MARGARET (MEG) HENDERSON BLAIR, D.Sc., is president of rsc THE QUALITY MEASUREMENT COMPANY, a position she had held since 1974, and chair of The ARS Group. She began her market research career at Schwerin Research Corporation in 1963. After Schwerin was acquired by an Indiana firm in 1968, she developed the ARS advertising research methodology (1968-1971) and became executive vice president of the new division. In 1974, with the purchase of the ARS division, she cofounded and became president of research systems corporation, now known as rsc THE QUALITY MEASUREMENT COMPANY.
ALLAN R. KUSE is executive vice president of The ARS Group. Before joining The ARS Group, Allan was a research faculty member at the University of Colorado, Boulder, where he obtained his doctoral degree in quantitative psychology. He began his career in advertising research in 1980 as The ARS Group's manager of information services. Allan was named vice president in 1985 and executive vice president in 1992.
|Printer friendly Cite/link Email Feedback|
|Author:||Blair, Margaret H.; Kuse, Allan R.|
|Publication:||Journal of Advertising Research|
|Date:||Mar 1, 2004|
|Previous Article:||IMC as theory and as a poststructural set of practices and discourses: a continuously evolving paradigm shift.|
|Next Article:||Targeting women's clothing fashion opinion leaders in media planning: an application for magazines.|