Words: 5114
Branding 101: An Overview of Branding and Brand Measurement for Online Marketers Report Contact: Molly Hislop Director of Research & Development, Dynamic Logic 800-245-2455 molly@dynamiclogic. com April 2001 Dynamic Logic’s Branding 101: An Overview of Branding and Brand Measurement for Online Marketers Table of Contents INTRODUCTION3 WHAT IS BRANDING? 6 MEASURING BRANDING OFFLINE9 STANDARD PRACTICES IN TELEVISION ADVERTISING9 Case Study: Kraft Uses Copy Testing for DiGiorno Rising Crust Pizza9 STANDARD PRACTICES IN PRINT ADVERTISING10
Case Study: Advertising Works in Print11 TRACKING STUDIES12 Case Study: Avis ??? Rent-A-Car12 MEASURING BRANDING ONLINE14 ONLINE COPY TESTS14 Case Study: Findings From Ipsos-ASI Interactive14 IN MARKET MEASURES15 Case Study: Travelocity16 COMPARING ONLINE AND OFFLINE BRANDING18 TABLE: COMPARING MEDIA EFFECTIVENESS IN BRANDING AND DIRECT MARKETING18 MSDW OBSERVATIONS ON MEDIA EFFECTIVENESS18 Brand Recall18 Generating Product Interest19 Generating Brand Awareness19 Cost Effectiveness19 SUMMARY20 GLOSSARY OF TERMS21 REFERENCES22 Dynamic Logic’s Branding 101:
An Overview of Branding and Brand Measurement for Online Marketers Introduction Professionals involved in online media are met with increasing scrutiny surrounding the medium’s value. Online media is not alone, however at their inception, all forms of media have been pushed to show value. Through that pressure have come some innovative and widely accepted approaches to measure advertising effectiveness. One simple way to measure the effectiveness of advertising, as stated by the head of online advertising at a large packaged goods company is to “stop advertising and see what happens. While that comment was made humorously, it underscores the notion that we know advertising works – it’s just a challenge to know precisely how. Since the stop-and-see-what-happens approach is not feasible, smart marketers need to devise ways to measure the value of their advertising on an ongoing basis. Of course, the measurement tactics and metrics chosen must meet the clearly defined objectives of the campaign: branding, direct response, or both. When looking at why ad budgets are spent in TV, Print, and Radio, over half of it (59%) is for branding, according to the Direct Marketing Association.
How much online advertising is geared towards branding is a topic of some debate: AdRelevance says approximately 63% whereas Nielsen/NetRatings says 11%. Either way, branding campaigns are clearly a part of traditional ad spending and online marketers need to understand how to include brand metrics in their measurement approach. Some online marketers hear they need to measure branding and think: ‘oh no, here is more data we have to deal with. ‘ While the data-obsession can get overwhelming, the answer is not to turn away from the important data, but perhaps be more judicious about what data-point one seeks.
Smart marketers find the right balance, gathering select data points without trying to measure every single thing and getting frustrated. Like a mosaic or the famous point-style head-shots used in The Wall Street Journal, a few key data-points can paint a clear picture. Online effectiveness tracking has historically been framed in terms of ad impressions, click-through rate, and the resultant direct-response action (registration, purchase, etc. ). These metrics, while important, only tell a small part of the effectiveness story.
Getting people to react immediately to advertising, while possible, may not be feasible or appropriate for many advertisers. Imagine the challenge for the brand manager of Soap X who is thinking about running an online advertising campaign; the campaign should be measured by what are realistic outcomes. Outcome A is best measured by behavioral metrics such as click-through and post-click conversion analysis. These are behavioral in nature and measure the actions as driven by the computer/mouse.
Outcome B is best measured by branding metrics such brand awareness, message association, brand favorability, and purchase intent. These are attitudinal in nature and measure people’s minds through their attitudes. Widely used in the offline world in terms of effectiveness tracking, branding metrics can only be gathered through consumer interviews. This paper begins with a definition of branding and brand equity and then outlines the various tactics marketing researchers have used to measure branding both offline and online.
Case studies with actual data are included wherever possible. Note this paper investigates various methodologies for measuring ad effects on brand equity measures. As such, the methodologies reviewed focus on attitudinal/survey based measures and not behavioral measures linking ad exposure to sales. The issue of linking up advertising and sales will be explored in a future paper. What is Branding? Before a discussion on research techniques for measuring ad effects on branding can commence, the term brand itself must be defined.
A brand is a distinguishing name or symbol designed to: ??Identify to origins of a good or service ??Differentiate those goods or services from those of the competition ??Protect the consumer and producer from competitors who would attempt to provide products that appear to be identical Branding is the process of creating an association between a symbol/object/emotion/perception and a product/company with the goal of driving loyalty and creating differentiation. For example, through product packaging and advertising, Coca Cola has created an association between many different objects and its brands.
The hourglass shaped bottle, the red and white colors, and even the font of its logo together make Coca Cola distinctive from competitors. The product’s past ability to satisfy consumer needs and its widespread distribution play an important role in consumption, however constant advertising has been successful in reinforcing past positive associations between the brand and consumers (Assael, 1998). In addition to recognition, branding may consist of building emotional responses (Volvo with a feeling of safety) or cultural responses (Mountain Dew with youth).
As consumers are bombarded with a variety of products to meet the same need, branding provides a way for consumers to reduce their decision making to consider only those products that they feel are relevant to them or that have met their needs acceptably in the past. There is no question that a strong brand is an important corporate asset. Brand equity cannot be measured in dollars and cents but rather it is a direct result of how consumers value a brand based on their experiences and perceptions (Spaeth, 1993).
It is these experiences and perceptions that permit the brand to earn greater volume or margins than it could without the brand name. There are many steps involved with building a brand’s equity including; brand awareness (unaided/aided), brand attributes, message association, brand favorability, brand preference, and ultimately brand loyalty. Each has an important role in moving a consumer towards a purchase and should be understood in terms of their specific function. Branding MetricQuestion it answersStage of brand building
Brand awareness (unaided)Is the brand “top of mind” for the consumer? Early Brand awareness (aided)Is the brand something that the consumer recognizes when presented with the name? Early Brand attributesIs the brand cool? Hip? Intelligent? Good value? Stable? Innovative? etc. Middle Message associationDoes the brand offer a specific value proposition to the consumer? Middle Brand favorabilityIs the brand well-respected and appreciated beyond being known and even used? Late Brand preferenceWhere does the brand stand when consumers are asked to choose among a competitive set?
Late Brand loyaltyIs the brand strong enough to keep consumers coming back for more? After conversion Most, if not all of these metrics are derived through attitudinal research. While loyalty can be measured behaviorally, behavioral metrics alone cannot help marketers understand why consumers act the way they do. Having the answer to the ‘why’ question, or what Duboff and Spaeth call ‘true commitment,’ will identify the reasons for loyalty and as a result help marketers identify the source of brand profitability (Duboff and Spaeth, 2000).
So how do you measure ad effects on brand equity components such as brand awareness, brand familiarity, brand favorability, brand image, and brand loyalty? By definition, these metrics are cognitive in nature and cannot be inferred from consumer behavior. As a result, marketing researchers must derive the branding value of advertising through interviewing consumers. While methodologies vary for different media, most involve the widely practiced exposed and control methodology.
In this design, the impact of an advertisement is isolated from other potential variables through placing a sample of consumers into one of two groups; exposed to an ad or not exposed to an ad. Both take a survey with questions regarding the test brand and differences between the two samples are tested for statistical significance and attributed to ad exposure. Measuring Branding Offline Standard Practices in Television Advertising In the world of television advertising, one widely accepted approach to measuring the branding value of a television ad is through copy testing.
Copy testing involves sampling a group of consumers to participate in a research study and splitting the group in half. While viewing a program, half of the group is exposed to an advertisement (exposed) while the remaining half sees an alternative ad (control). After viewing a program, both groups answer a brief questionnaire with questions concerning a particular category as well as their preferences and attitudes for various brands in that category. If the test brand scores higher among the exposed group than the control group, the difference is attributed to the commercial.
Case Study: Kraft Uses Copy Testing for DiGiorno Rising Crust Pizza The extent to which communication objectives are met will determine if the commercial is ready to air, requires modifications, or if it is back to the drawing board. The March 1999 edition of American Demographics outlines how copy testing was used as part of a research program commissioned by Kraft to change the consumer perception that frozen pizza tastes like cardboard (Lach, 1999). Based on extensive research probing why consumers eat pizza, especially frozen pizza, Kraft produced a number of TV spots.
It then commissioned Millward Brown to copy test each 30-second television ads and found the following: ??64 percent of respondents recalled the spots’ main message whereas the average commercial tested scored about 24%. ??The ads generated strong brand identification with 53% recalling the DiGiorno name. According to Nielsen data, Kraft did notice a steady rise in sales for its DiGiorno brand since its launch in 1996. Additionally, awareness for the brand has also increased from 23% during the launch year to 77% in 1998.
While powerful in determining a commercial’s potential, copy testing is not the appropriate tactic to measure the impact of a campaign after it is running. Most TV advertisers who want to measure in market performance use a metric entitled Day After Recall. Day After Recall tests typically occur 24 hours after a commercial has aired in a certain market. Since it is not possible to understand who in a given market was actually exposed to a TV ad, consumers in the market are called at random and asked to recall if they have seen any advertisements by companies in a particular category.
They are then asked to recall anything about the commercial; what was said, what was shown, and what the main idea was. The “day after recall measure” is the percent of those in the commercial audience who were watching the show before and after the commercial was shown and remember something specific about it; such as the sales message, the story line, the plot, or some visual or audio element. To avoid biasing the sample, advertisers interested in brand measurement can ask a battery of questions on the test brand before recall measures are gathered and use the recall information to plit the sample into exposed and unexposed groups. They can then look to see if there are any differences between those who recall seeing an ad and those who could not recall seeing an ad. Standard Practices in Print Advertising The Starch Ad Readership Program measures the readership of advertisements in magazines and newspapers. Through one on one personal interviews, Starch uses a recognition model to assess a particular ad’s effectiveness on four degrees of reading: ??Nonreader: A person who does not remember having seen the advertisement in the issue. Noted: A person who remembers seeing the advertisement in the issue. ??Associated: A person who not only “noted” the advertisement, but who also saw or read some or part of it and clearly indicated the brand or advertiser. ??Read Most: A person who read 50% or more of the written material in the ad. Cost data from newspaper and magazine space can be used to calculate a “readers per dollar metric. ” Each year, Starch measures over 25,000 ads in over 400 magazine issues. On the most basic level, clients get raw readership scores — the percent of readers who saw the ad and read the copy.
Then the data are put into a context: The ad is ranked not only against other ads in the issue, but also against other ads in its product category over the past two years. These norms help clients to judge the performance of their ads over time and against their competition. Case Study: Advertising Works in Print During the 1960s, Alfred Politz conducted three classic studies to show the effectiveness of print advertising. Each study took place in a controlled setting whereby specially prepared magazines were left with consumers.
These subjects were asked to review a magazine for editorial content and were unknowingly exposed to varying numbers of test ads in these magazines. Later, consumers were asked to rate the advertised brands on a number of dimensions (Kim, 1992). In all three studies Politz found: ??Brand familiarity increased with number of exposures ??Claim familiarity and belief increased with the number of exposures ??Purchase intent increased with the number of exposures Some critics of the study cite the potential bias that results when experiments are conducted in a controlled setting.
Since the subject knows he/she is participating in an experiment, they may spend more time than they normally would interact with the stimulus. In association with Seagram and Sons, Time, Inc attempted to replicate Politz’s findings outside of a controlled setting. Ads were rotated for eight Seagram products in issues of Time and Sports Illustrated. Weekly questionnaires sent to subscribers in both test markets supported the Politz findings. Overall, increases were noted in use and purchase for the brands advertised (Kim, 1992). Tracking Studies
Oftentimes a campaign utilizes various media and it becomes increasingly hard to attribute advertising effects to any one media using an in- market test of advertising effectiveness. It is not uncommon for advertisers to monitor campaign impact through a tracking study whereby periodic sampling of a target audience provides a time trend of branding measures. Of particular interest is how the campaign, as opposed to a specific advertisement, is affecting the brand. In this type of study, a baseline measure of brand metrics is gathered prior to a campaign going live (pre) and then again as the campaign is running (post).
Any information gathered in the post wave is compared to the baseline levels seen in the pre wave. While this methodology is common, it cannot attribute lifts to any one media or advertising execution and it does not control for the influences of non-advertising related variables. Case Study: Avis ??? Rent-A-Car In responding to a request from Avis management, the marketing research team at Avis worked with Bates USA, its advertising agency, and Lieberman Research to demonstrate that Avis advertising had a positive effect in building brand equity and ultimately share of market for Avis (Georgiou and Miller, 1997).
Avis turned to a tracking study conducted over a 10-year period among car renters. Variables measured over this time period included brand and advertising awareness, communications, brand image and share of market. Avis et al first looked at the relationship between changes in advertising spending and various measures of brand equity and share of market. Avis found that there is a strong relationship between the amount of money spent on advertising and its share of market. Additionally, Avis also found that there was a significant positive relationship between advertising spending and measures of company awareness and advertising awareness.
Additionally, spending was also correlated with advertising communications (theme/slogan recall) and to a lesser extent company image (Georgiou and Miller, 1997). Once these relationships were identified, Avis worked with Lisette Berry & Associates, a statistical consulting firm, to develop a model that describes how variations in advertising spending influence share of market. This model was based on a regression analysis designed to identify the mechanisms in which advertising has its influence (Georgiou and Miller, 1997).
Based on this model, Avis found: ??Changes in advertising spending have a greater effect on company awareness, advertising awareness and advertising communications than on company image. ??The two measures most highly correlated with share of market are company awareness and company image. These insights led Avis to conclude that changes in company awareness (particularly unaided awareness) and imagery are more likely to be translated into improved market share. Being able to identify the predictors of market share enabled Avis to focus its marketing strategies and advertising tactics on these two important variables.
Measuring Branding Online All of the methodologies we have reviewed are the result of demand for accountability in advertising expenditures. Over time, these research tactics have become standard in measuring the branding impact of offline advertising. As Internet advertising is also under the scrutiny of accountability, it is important for online advertisers to understand how their online efforts are impacting measures of brand equity. Online Copy Tests It is possible to replicate a copy-test methodology on the Internet. In this practice, users are recruited from a site or database and asked to take a brief survey.
Users are randomly assigned to either a control group or an exposed group. Exposed group respondents see a page with a client’s banner on it while control group respondents see an alternative ad. The user then answers some questions about the test brand and differences between exposed and control groups are measured and attributed to the ad exposure. Millward Brown Interactive is largely credited for bringing the copy-test approach online and showing that online advertising is effective regardless of click-through (Briggs and Hollis, 1997).
This service helps online advertisers reduce the risk that banners in their campaign will have no effect. Dynamic Logic’s online copy test product is called AdIndex Lab. Case Study: Findings From Ipsos-ASI Interactive Ipsos ASI was an early mover in measuring the branding effectiveness of online advertising. Since 1997 the company has been measuring Internet advertising based on the premise that, as in traditional advertising, the effectiveness of online advertising is based on a consumer’s ability to recall an ad and absorb and decode a message before behavior and attitudes can change.
Ipsos’ Internet Advertising Consumer Model sheds some light on the value of different advertising formats, sponsorships, and broadband advertising (Flores, 2000). All findings are based on online copy tests. ??Larger ad units are more noticed and communicate more however, if ads are too disruptive to the consumer experience (slow download), the consumer may take away a negative perception of the ad, brand, or even site itself. ??Online sponsorships can positively impact advertising awareness of a brand. ??Rich media increases an ads ability to be recalled but does not impact ad likeability.
Like the Politz print effectiveness findings, there is always the possibility of a bias in any research conducted in a controlled environment. Because of this fear, the industry has pushed for in market tests to measure the branding effectiveness of online advertising (recognizing of course that online copy tests remain an important tool for pre-testing a campaign’s effectiveness). In Market Measures There are certain challenges faced in the online world in terms of measuring advertising impact on brand equity when the ads are in-market.
Unlike television or print advertising, Internet advertising is served dynamically so that two people visiting the same site may never see the same ads. Because of this, recruiting people who were exposed to a specific ad is a challenge. While ad exposures can be tracked by third-party ad serving databases; most ad servers do not have the ability to link exposure information to the survey data information required to derive brand equity measures. Marketers can modify a tracking study to measure the impact of an online campaign as it is running in market.
Like the offline version, online tracking studies recruit respondents in waves, often from the same sites the ads are running on. Surveying typically happens before a campaign is launched and then again in subsequent waves after the campaign has launched. While recruiting and interviewing are conducted online, oftentimes exposure information cannot be linked to survey data. As a result, it is impossible to attribute a change in brand equity measures to advertising exposure. Through the use of cookies, it is possible to match survey data with ad exposure data. Dynamic Logic’s AdIndex product does just that.
AdIndex measures consumer perceptions and attitudes toward an advertiser’s brand by capturing consumer opinions through an online survey while the online campaign is running in market. To isolate the impact of advertising exposure on consumer attitudes, two groups of online consumers are sampled at the same time from the same Web sites on which a test campaign is running. As the only difference between the groups is the presence of the advertising, any attitudinal differences between the two groups can be attributed to the exposure to the specific online ad campaign.
AdIndex captures frequency data so that advertisers can understand how varying levels of exposure to advertising impacts branding. Furthermore, AdIndex works with any ad-server and can measure a campaign’s performance across multiple sites so that advertisers can understand how the different media properties on a buy impact branding. Case Study: Travelocity 24/7 Media commissioned Dynamic Logic to research the branding effectiveness of online advertising for its Travelocity client. Results show that Travelocity has significantly increased awareness of their brand through the use of online advertising.
Using AdIndex to measure the branding value of their campaign, Travelocity found that their banner campaign significantly lifted aided awareness of the Travelocity brand by 16%. Aided awareness refers to the percentage of respondents who indicated that they are aware of Travelocity when presented with a list of travel service providers. This greatly exceeds the average lift in awareness of all campaigns measured using AdIndex, which is 6%. The results vary by frequency level – the more times a person saw the banners, the greater the impact in awareness lift.
Among those people who were exposed to the banners four or more times, the lift in awareness of Travelocity was 44%. While Travelocity’s overall goal is to sell seats, the company recognizes that there is a way to measure how successful its online advertising is beyond counting sales. By quantifying a significant lift in brand awareness, Dynamic Logic was able to quantify that more consumers may place Travelocity in their consideration set when they need to make travel arrangements in the future. Comparing Online and Offline Branding
Recently, Morgan Stanley Dean Witter released a report entitled Correction: Does Internet Advertising Work? Yes, But… Based on data provided by Dynamic Logic and others, MSDW eventually concluded that Internet advertising banners are a cost-effective branding tool when compared to TV and print. “The Internet performs well across branding measures, but we believe it is better for those who already have a brand than those who are trying to develop one. ” “At today’s prices, we believe banners are cost-effective in generating brand recall and brand interest…but they are only a moderately effective direct marketing tool. Table: Comparing Media Effectiveness in Branding and Direct Marketing TVMagazinesNewspaperBannerse-Mail CPM – Cost per 1000 impressions$16. 00$6. 00$19. 00$3. 50$20. 00 Indexed to Internet457%171%543%100%571% Generating Brand Awareness36%29%N/A14%N/A Indexed to Internet257%207%N/A100%N/A Cost Effectiveness56%121%N/A100%N/A Brand Recall Ability17%26%23%27%N/A Indexed to Internet63%96%85%100%N/A Cost Effectiveness14%56%16%100%N/A Generating Brand/Product Interest46%44%N/A44%N/A Indexed to Internet105%100%N/A100%N/A Cost Effectiveness23%58%N/A100%N/A Source: Morgan Stanley Dean Witter, 2001
MSDW Observations on Media Effectiveness Brand Recall According to the data gathered by MSDW, the Internet leads magazines, newspapers, and television in brand recall. After seeing an ad on the Internet, consumers showed a 27% greater ability to recall a brand than before. This compares to 26% for magazines, 23% for newspapers, and 17% for television. Generating Product Interest Consumers were 44% more interested in learning more about a product after seeing a banner ad than before. This number is similar for magazines but slightly lower than television, whose comparable figure is 46%.
Generating Brand Awareness The Internet has been shown to raise awareness of a brand by an average of 14%, lower than both television and magazines with 36% and 29% respectively. Cost Effectiveness MSDW calculated the effective CPM for the Internet to be $3. 50. At this price, banners remain less cost effective in generating brand awareness than magazines but more than television. In terms of brand recall, the Internet is more cost-effective than television, magazines, and newspapers and in terms of generating product interest; the Internet’s $3. 0 CPM ranks it as the most cost effective media. Summary Online marketers have traditionally looked at behavioral responses to measure the effectiveness of the medium. As click-through rates have been declining, critics have been reporting that online advertising is ineffective. However, looking only at behavioral responses to online advertising ignores the fact that online advertising can impact a brand’s equity. Measuring the branding effectiveness of advertising has been done in the offline world for over half a century.
Many of the tactics marketers have used to measure advertising’s impact on branding have been outlined in this paper. These traditional metrics can be adapted to online advertising effectiveness studies to give advertisers a more valid assessment of their online advertising efforts. Finally, online advertising has been shown to be a cost effective branding vehicle especially when it comes to generating brand recall and brand interest. Conversely, banner advertising has been shown to be only moderately effective in terms of generating direct response. Glossary of Terms
TERMDEFINITION AdIndexIn market measurement of online advertising performance on branding variables. Isolates online advertising’s impact on brand equity metrics and can be repeated over time for more longitudinal measures. AdIndex also allows advertisers to understand how frequency and site selection impacts branding. AdIndex LabProduct offered by Dynamic Logic designed to let advertisers pre-test their online creative to reduce the risk that their ads will have no effect. Aided AwarenessRefers to the percentage of respondents who indicated that they are aware of an advertiser.
BrandDistinguishing name or symbol designed to identify the origins of a product or service, differentiate the product or service from the competition, and protect the consumer and producer from competitors who would attempt to provide similar products. Brand EquityThe value of a brand as derived from consumer attitudes, behaviors, awareness, and perceptions. BrandingThe process of creating an association between an object or feeling and a product/company. Copy TestingTesting the branding value of a television commercial in a controlled environment.
Copy testing is done before commercial airs on television. Control GroupThe segment of respondents who are recruited to provide a baseline measurement of branding levels – these respondents are either NOT exposed to the test brand advertisement or they exposed to a “control” or “placebo” ad Day After RecallPhone based methodology to understand how memorable a commercial is. The “day after recall measure” is the percent of those in the commercial audience who were watching the show before and after the commercial was shown, who remembered something specific about it.
Exposed and ControlResearch methodology designed to isolate the impact of a stimulus (advertisement) on a response (awareness, purchase intent, etc. ). Exposed GroupThe segment of respondents who are recruited to provide an ad-exposed measurement of branding levels – the data from these respondents is compared to control group FrequencyThe number of times a consumer is exposed to an advertisement or message. Online Copy TestTest/Control based methodology designed to measure the branding potential of an online ad.
Requires respondents to view a Web page with either a client’s banner (exposed) or an unrelated banner (control). Scores on branding metrics are calculated for each group and differences are attributed to the ad. Online Tracking StudyMeasures branding in waves using the Internet for recruitment and survey hosting. Does not allow advertisers to link exposure to attitudes. Starch Ad Readership ProgramUses one on one interviews to determine a print advertisement’s effectiveness in terms of readership. Ads are compared to other ads in an issue as well as other ads from competitors.
Statistical SignificanceA finding (for example the observed difference between the means of two random samples) is described as statistically significant, when it can be demonstrated that the probability of obtaining such a difference by chance only, is relatively low. It is customary to describe one’s finding as statistically significant, when the obtained result is among those that would occur no more than 5 out of 100 times when the only factors operating are the chance variations that occur whenever random samples are drawn. Test BrandRefers to the brand being measured.
Tracking StudyLongitudinal research program whereby respondents are recruited in waves to take a branding survey. The result is a time trend of branding metrics. Tracking studies are meant to give an idea how a campaign is performing rather than any individual advertisement in that campaign. References Assael, Henry. Consumer Behavior and Marketing Action. 6th Ed. South Western College Publishing, (1998): 104. Briggs, Rex and Hollis, Nigel. “Advertising on the Web: Is There Response Before Click-Through? ” Journal of Advertising Research, 37, 2 (1997): 33-45.
Duboff, Robert and Spaeth, Jim. Market Research Matters: Tools and Techniques for Aligning Your Business; John Wiley & Sons, Inc. 2000. Flores, Laurent. “Internet Advertising Effectiveness: What Did We Learn and Where are We Going? ” Worldwide Advertising Conference, Rio de Janerio, November 2000. Georgiou, Paul and Miller, Stephen. “10 Years of Tracking in the Rent-A-Car Business. ” Advertising Research Foundation, 1997. Hoffman, Howard S. “Statistics Explained. ” The Animated Software Company, 2001. Kim, Peter. “Does Advertising Work: A Review of the Evidence. The Journal of Consumer Marketing, 9, 4 (1992): 5-19. Lach, Jennifer. “Upper Crust: Kraft Foods Pours the Sauce, Sprinkles the Cheese, and Delivers Results with a Tastier Frozen Pizza. American Demographics, March 1999. Russel, Michael J, Keith, Robert J, Feuer, Rod N, Meeker, Mary and Mahaney, Mark. “Does Internet Advertising Work? Yes, But…” Morgan Stanley Dean Witter Internet Direct Marketing and Advertising Services, February 22, 2001. Spaeth, Jim. “Brand Equity and Advertising; Lessons from Jimi Hendrix. ” Advertising Research Foundation, 1993.