December 2018 (Vol. 58, Issue 4)
What Do We Know about Digital Attribution?
Editor-in-Chief John Ford walks us through the special package on digital attribution, and he gives special thanks to Gian Fulgoni, who whose publication in this issue marks his 20th —and final —“Numbers, Please” column for the JAR. The column will continue in 2019 with new authors.
Why We Need Better Measures of Research Impact in Advertising: Considerations for Best Practices to Expand Research’s Reach
It’s standard practice in research journals for authors to cite previously published articles as a way of either judging or building on earlier research. But, how those citations are managed and eventually counted as having an impact on knowledge in advertising is inconsistent, writes University of Missouri professor Shelly Rodgers. Not all citations are equal, having different characteristics, and there is concern that “too strict a focus on simple citation counts can result in narrow views on quality and impact,” she adds, citing research from 2016. In advertising research, there’s the additional challenge of increased number of specialties. “The field of advertising is fragmenting into smaller specialties. This reflects what is happening in the industry, due in part to advances in technology and the rapid rate at which digital media are changing advertising. As more specialties pop up, how confident are we that we’ve captured enough of the ‘splintered’ pieces of advertising knowledge? Other points Rodgers makes:
- An estimated 32 percent of papers published in social science journals—which includes advertising journals—go uncited.
- “Although there may be some authors who feel that ‘good’ advertising research speaks for itself, there are plenty of examples of good papers that go uncited.”
- Rodgers calls for a broadening of views on quality and impact that would involve “a deeper understanding of what a citation means” in terms of its physical and structural characteristics, and the measures used to determine research’s impact.
WHAT WE KNOW ABOUT DIGITAL ATTRIBUTION
How Limited Data Access Constrains Marketing-Mix Analytical Efforts: Why Data Barriers Are Preventing Marketers from Optimizing Marketing Spend
Digital attribution may sound simple in that it involves isolating advertising tactics and assessing their impact on consumer decision making. But without sufficient data at the individual consumer level, advertisers are having trouble building “actionable marketing mix models for accurately measuring marketing ROI,” writes Gian M. Fulgoni. The comScore, Inc. cofounder and former CEO points to walled gardens—Amazon, Facebook, and Google, for example—for not openly sharing what they perceive to be proprietary data: vast data banks that can provide valuable insight on consumer behavior and media spending. This “can constrain the efforts of independent research companies to measure the financial return from an investment in advertising on walled gardens’ platforms,” he writes. Granted, there are issues associated with multiplatform video-content consumption, among them: challenges of analyzing single-source models across a number of different touchpoints, and the nature of data privacy and the potential for overregulation. In the face of all of these hurdles, “analysts can pursue two paths,” he suggests:
- “They can run marketing-mix models that exclude the causal components for which data are unavailable, or
- they can use causal data that’s aggregated across users. Neither approach is optimal.
Will the situation improve? “To a large degree,” Fulgoni concludes, “it depends on the extent to which the various digital-advertising platforms come to see a benefit in sharing their data with third parties.”
Why Companies Risk Losing Customers by Not Reciprocating on Shared Data: Rebuilding the Data-Sharing Economy in a Consumer-Driven World
Customer concern over the lack of control over the ownership of their data has generated new challenges for companies to create meaningful and easily understood data policies. The tightening of rules in Europe have set a precedent to put greater control in the hands of consumers, which puts additional pressure on companies to protect consumers. Natasha Hritzuk, vp for consumer insights at Turner cautions that companies are not moving quickly enough to proactively develop new data policies and to comfort consumers. And, given the industry-wide tardiness in addressing this issue, Turner undertook its own research, sampling more than 9,000 people, to understand what customers are willing to share, examine ways that it could communicate exactly how data is used along with benefits for the consumer, and discover what consumers want in terms of benefits that would allow an increase in data sharing. Hritzuk offered a number of findings from that research, including:
- Consumers are willing to share most of the of the key data points that are vital to media and entertainment companies …
- … except for when data gets personalized or extends to focus on their family or friends. “In the long term, companies will benefit if they take steps to engage consumers in an open dialogue around data transactions.”
- With regard to examining the benefits that would incentivize sharing, Turner found two core types of benefits that consumers likely would welcome: opportunities for deeper engagement and seamless advertising experiences. “The more practitioners understand how, when, and what advertising consumers prefer … the more they will be able to consistently deliver better advertisement experiences.”
- Many consumers are indeed willing to share their data if they can easily see some reciprocity: “Companies need to be much more explicit about the connection between shared data and the subsequent benefits powered by that data,” the study concludes.
Attribution Modeling in Digital Advertising: An Empirical Investigation of the Impact of Digital Sales Channels
How do different sales channels impact the consumer’s journey to purchase behavior? In this article, two University of Southampton researchers compare and contrast four different attribution models: last-click, time-decay, uniformly distributed, and position-based. Given that different online channels are involved at different stages of the consumer’s purchase journey, they write, it is important to examine the ability of these models to properly attribute credits to the various channels that have an impact upon the final purchase. In order to test the models, Tahir M. Nisar and Man Yeung (who also is a senior associate at Digital Lab in London), included close to 1 million transactions with a total revenue of more than $158.5 million, and an average order of $112.50. The number of customer journey lengths ranged from 1 to more than 5 steps. Testing the different models, the study found that:
- “The more rigorous variety of statistics-based attribution models is a preferable attribution strategy,” and
- such types of attribution models “allow one to provide more stable credit assignments to the digital channels in a purchase funnel.”
- One caveat: “It appears that the attribution models currently do not value fully social media, which often do not directly lead to purchase but can have a strong behavioral impact (e.g., by shaping the consideration set).”
- The authors suggest that the value of social media actually improves with the more highly sophisticated attribution models, but “the consumer behavior implications need to be accounted for fully in future research.”
Coalition Game Theory in Attribution Modeling: Measuring What Matters at Scale
In the age of ever more fragmented customer experiences across multiple screens, optimizing media investments increasingly relies on attribution models. A research trio at GroupM analyzed the drawbacks of existing models and examined some new modeling choices which they believe may provide useful results. “Predefined heuristic rules to distribute the credit among advertising inputs remain in widespread use: last-touch attribution, uniform attribution, and first-touch attribution,” data scientists Seyed Hanif Mahboobi (now at Amazon Web Services), and Mericcan Usta (GroupM), and Saeed R. Bagheri (now director of insights and analytics at Amazon Advertising), write. “The importance of considering the full path to conversion, however, calls for more sophisticated methods.”
Mahboobi explains: “Practitioners well know that there is a clear tradeoff between the quality and speed of an attribution model. (On the one hand), a misattributing but quick attribution model, such as one that attributes consumer actions to the last medium they interact with (last touch), allows publishers to ride freely on others’ efforts. On the other hand, dividing the aggregated lift based on their average marginal contributions (aka as Shapley value method) stands out as a reliable attribution model with a reputation across-industry verticals. However, it is a computational challenge to bring Shapley values to, say, a dashboard.”
The study demonstrates “scalable ways to approximate Shapley values so that actionable insights can be generated on accurate models:
- “Optimizing media spend in the world of multiple devices and fragmented customer experiences requires accurate attribution models, but these models are hard to evaluate.
- “Probabilistic models and logistic regression can approximate such high-fidelity attribution models at scale.”
By using logistic regression and game theory as appropriate algorithmic approaches to attribution modeling testing, the study tested various models, utilizing the data from a three-month digital-advertising campaign for a national retailer. Their findings included the discovery that
- with the new approach to attribution modeling, programmatic media generated progressively higher attribution with more algorithmic approaches (like game theory), while paid search was getting progressively lower attribution.
- “Programmatic media effectively can target consumers interested in the product, regardless of where they are in the funnel,” the paper concludes. “Paid search, conversely, is focused more toward the end of the funnel.”
Different approaches provide different results, and it is always wise to use a variety of approaches to ensure the quality of the input for planning purposes. The authors offer logistic regression and game theory as valid choices for attribution modeling since “both approaches provide concise models that measure the incremental value each advertisement exposure adds to the consumer’s purchase decision.”
OTHER FEATURE ARTICLES
The Risk of Omitting Warmth or Competence Information in Ads: Advertising Strategies for Hedonic and Utilitarian Brand Types
Research has shown that consumers perceive brands as they do people, on the basis of two dimensions: warmth and competence. Hedonic brands (e.g. amusement parks, food delivery) therefore try to create a pleasant experience for the customer, highlighting the warmth dimension, whereas utilitarian brands (e.g. financial services) emphasize the competence dimension. This study is the first to analyze the use of warmth and competence information by hedonic and utilitarian brands in the context of the theory known as the innuendo effect. “Research on the innuendo effect has shown that if a person is described positively in terms of only one dimension, people make negative inferences about the other dimension,” Christina Peter, researcher at LMU Munich and Milan Ponzi, consultant at LOGIT Management Consulting in Munich write. “The current study provides evidence that this effect also emerges in the advertising context—that is, when a brand is described only with one of the two dimensions.” The results, in fact, can inform a key effectiveness measure: attitude toward the brand.
Peter and Ponzi asked survey participants to respond to 3 groups of posters they had created that referred either to the type of brand, type of company, or the description of the brands as follows:
- Hedonic brand, utilitarian brand (differing posters associated each brand type with either warmth more strongly than competence messaging, or vice versa, respectively), or a brand for which consumers judged both dimensions as equally important;
- Amusement parks for which warmth was viewed as most important, insurance companies for which competence was viewed as most important, and food-delivery services judged for warmth and competence equally;
- Descriptions referring to the warmth dimension, competence dimension, or no particular dimension
The results of the experiment showed that “if a brand was described only on 1 of the 2 basic dimensions—warmth or competence—customers made negative inferences about the omitted dimension, which, in turn, lowered the customers’ attitude toward the brand.
- “This means that even though there was a positive effect on brand attitude via the dimension addressed in the advertisement, innuendo counteracted this effect, making the advertisement ineffective,” the authors write.
- “This was the case regardless of whether the more primary or secondary dimension of the product was omitted.”
The innuendo effect was moderated by product involvement (high involvement, for example, for insurance companies, versus low involvement for amusement parks and food-delivery). But this influence was weak, the authors note:
- “Highly involved customers identified innuendos more easily, which makes the effect even more noteworthy, considering that this was also the group most likely to buy the promoted products.”
- Contrary to the authors’ predictions, however: “the omission of a dimension did not trigger reactance,” which leads the authors to assume that “negative inference about the omitted dimension might be an unconscious process.”
The Effects of Signaling Monetary and Creative Effort in Ads: Advertising Effort Can Go a Long Way Influencing B2B Clients, Employees, and Investors
Research long has shown that putting extra effort into advertising, in the form of more expensive or more creative advertisements, has positive effects on consumer perceptions and evaluations.
We know that “advertiser effort signals brand ability and commitment, which leads to beneficial effects on consumers,” professors Micael Dahlen and Sara Rosengren (Stockholm School of Economics), and John Karsberg, (project manager at H&M Online) write. In this article, they explain how their own research builds on that knowledge by testing “whether effort (in terms of expense or creativity) yields similar effects in business-to-business (B2B), recruitment, and investor contexts.”
The authors point to the seminal article on the signal effects of advertising, “Advertising as Information” (Nelson, 1974), which “suggested that spending more money on larger print advertisements signals to consumers that the brand makes better products.” A large body of literature followed, finding similar signal effects on the basis of the advertiser effort conveyed by such investments, but all of this work focused on consumer responses to advertising.
In four experiments, Dahlen, Rosengren, and Karsberg “tested the notion that advertising signal effects apply to other forms of advertising and target groups as well.” They divided the experiments in two sets of testing:
- One set tested whether more expensive or more creative ads yield similarly positive effects in other customer segments, such as (B2B) and recruitment advertising.
- The other set tested whether and how extra effort also might affect other stakeholders, including employees and investors.
Results from the first two studies:
- Extended the body of knowledge empirically by “applying it in new contexts on new target groups”;
- Provided managerial takeaways that putting extra effort into advertisements goes a long way in business-to-business (B2B) and recruitment advertising. “By increasing the perceived expense or creativity in advertisements, the advertiser sends a message that is more forward-looking than the specific message about the advertised (B2B) product or job offer,” the authors write;
- Showed that “receivers are particularly sensitive to soft information about the brand that extends beyond what the brand explicitly has to say and what it has to offer right now,” they note. Thus, “how the advertising communicates, rather than what it communicates, becomes particularly important.
Findings from the last two studies:
- Extend the body of knowledge “theoretically by showing that consumer advertising can affect stakeholders such as employees and investors as well.”
- Inform research on “employer branding and on advertising’s financial effects by testing the effects on an individual receiver, rather than on a macrolevel.”
- Show that “employees and investors recognize the advertiser’s extra effort in terms of expense or creativity and that they presume that the targeted consumers will react more favorably because of this.”
- Demonstrate that this “presumed consumer influence in turn, has an impact on the individual’s own ratings of the advertiser as an employer or investment opportunity.”
Face Presence and Gaze Direction in Print Advertisements: How They Influence Consumer Responses—An Eye-Tracking Study
There has been extensive research on executional devices that influence a commercial’s effectiveness. Some of it has focused on the presence of a model in an advertisement, but there has been even less study of the influence of the model’s presence in an ad. At the same time, psychology studies have found that the face and the eyes more likely will capture attention, compared with other stimuli. These two streams of knowledge informed the collaboration of three marketing professors in France—Safaa Adil (L’École Superieure de Commerce et de Management – ESCEM), and Université de Rennes’s Sophie Lacoste-Badie and Olivier Droulers—in their eye-tracking study about the impact of face presence and the direction of the model’s gaze in print ads. “For product manufacturers and advertisers, attracting consumer attention in a cluttered advertising environment is essential,” the authors write. The findings of their work “show that face presence and gaze directed toward the product—versus no face and gaze toward the viewer—have a strong influence on attention to and memorization of advertisements.” Among the management takeaways:
- “Face presence increases attention paid to advertisement elements, including product and brand. The product receives even more attention when the model’s gaze direction is toward the product, versus toward the viewer.
- “Face presence in an advertisement with the model’s gaze directed toward the product is the best combination to improve product and brand memorization, attitude toward the brand, and purchase intention.
- “Objective attention measures using an eye-tracking device can benefit managers and should not be neglected. Regarding the effects of attention on memorization, such measures could be of great help in advertisement tests.”
More research is needed to examine, for example, “whether the influence of face presence and gaze direction is strengthened or diminished when the model in the ad is a celebrity or an unknown endorser,” the authors suggest. Moreover, given the simplicity of the ads created for the experiment, “a follow-up study measuring the effects for real advertisements and brands may be necessary, provided that both product and brand familiarity are measured beforehand.” And, future studies should “examine the influence of face presence and gaze direction on real purchases,” the authors conclude.
Measuring Audience Reach of Outdoor Advertisements: Using Bluetooth Technology to Validate Measurement
Good news for out-of-home advertisers from the Ehrenberg-Bass Institute for Marketing Science at the University of South Australia: A method using Bluetooth technology measures important details that up until recently were available only for television advertising. With it, market researchers can estimate the frequency of exposure and level of unduplicated reach for outdoor advertising, while media buyers and campaign managers can examine in finer detail their targeted audiences.
Specifically, with this method practitioners can
- compare locations for outdoor and ambient advertising more objectively, based on measured rates of duplication, and
- decide how long an ad should be run at the same site, based on the measured efficiency of each spot.
The study has far-reaching implications, according to this research team:
- “Bluetooth data collection can be used to approximate the total reach of an outdoor advertisement and could be extended to be measured across multiple sites to determine the reach for a whole campaign. This allows for a noninvasive, anonymous, and real-time data-collection approach.”
- “An advertiser now can determine the average number of exposures for each respondent and the proportion of repeated exposures for a particular advertisement over a period of time. Practitioners can make informed decisions about the optimum duration of a campaign.”
Previously, practitioners have had to rely on their own managerial judgment to do these types of assessments, rather than on empirical data, according to Ehrenberg-Bass researchers Bill Page, Zachary Anesbury, Sophia Moshakis, and Alicia Grasby. “Even if all spots are not measured in the way,” they write “a sample of passersby, such as that provided by this method, can be used to improve or augment existing measures.”
For media buyers and campaign managers, the new method allows them to examine in far greater detail than previously the audiences that their advertisements reach. The authors explain two key points:
- “A network of detectors would be able to measure the patterns of reach for an entire campaign.
- “Given the increasingly digital nature of outdoor signage and the changeable nature of the displays, Bluetooth data could be logged against the actual advertisement being displayed at the time, which would allow for even more fine-grained campaign analysis.”
Event-Marketing and Advertising Expenditures: The Differential Effects on Brand Value and Company Revenue
Event marketing has become increasingly prevalent in recent decades. Researchers Lei Liu (Central University of Finance and Economics, China) and Jin Zhang (Tsinghua University), and marketing professor Hean Tat Keh (Monash University) point to knowledge that reinforces that claim: Nearly all U.S. companies (96 percent) engage in event marketing as part of their marketing communications. Scholars have defined event marketing as a “promotional strategy in which a themed activity is developed for the purpose of creating experiences for consumers and promoting a product or service” (Belch and Belch, 2004). Consumers participate in such activities, including sporting events, concerts, fairs, or festivals.
Despite the body of knowledge in this area, “surprisingly little research has examined (empirically) the effects of event marketing at the company level,” Liu, Zhang, and Tat Keh write. So, the authors conducted a study comparing, as the headline states: the differential effects of event-marketing and advertising expenditures on brand value and company revenue. They used a longitudinal dataset for 74 real-estate companies in China over a 7-year period (2006-2013) to test their research framework. The data were obtained from China Index Academy, China’s largest professional real-estate research institute. Among their findings:
- “Spending related to both event marketing and advertising made a positive impact on company revenue and brand value.
- “As a brand aged, advertising expenditure continued to yield positive returns on brand value and company revenue.”
- By contrast, “event marketing had diminishing marginal returns on brand value and company revenue.”
The authors see further opportunity for research in this area, including:
- Comparing the effect of event marketing “against that of advertising in reaching potential consumers, given that previous research has shown that gaining more buyers of all types is critical for brand growth (Romaniuk and Sharp, 2015; Sharp, 2010).
- Further testing on the authors’ expectation that “event marketing is more effective at targeting specific segments, whereas advertising might be more effective at reaching a wider audience.”
How Claim Specificity Can Improve Claim Credibility in Green Advertising: Measures that Can Boost Outcomes from Environmental Product Claims
Companies that advertise products for their green qualities should pay attention to how their green messaging is being perceived. Vague claims such as “biodegradable” and “environmentally friendly” just don’t cut it for the increasingly skeptical and cynical consumer. That’s why Benjamin Ganz, marketing manager at the Swiss online retailer Digitec Galaxus AG and Anthony Grimes, researcher at University of Manchester, collaborated on research that sought evidence to support two related propositions by earlier research:
- Specific claims are more credible than vague claims (Davis, 1993)
- Skepticism toward social and environmental advertising claims effectively can be reduced, and thus the credibility of such claims increased, by the provision of specific (over abstract) information (Pomering and Johnson, 2009).
In their own work, Ganz and Grimes found evidence from testing consumers that “supports the conclusion that increasing the specificity of the claim is a means by which to increase green advertising effectiveness. … This effect was found to be robust across a range of product categories and was not moderated by the environmental relevance of the product.”
Among their takeaways:
- Practitioners can expect to gain significant improvements in the perceived credibility of their green claims—and thus the effectiveness of their advertising—by developing claims that provide precise and meaningful information.
- That information should address concerns raised by the U.S. Environmental Protection Agency (EPA), which advises consumers to check whether a term such as “recycled” applies to the product, the packaging or both; whether it applies to preconsumption and postconsumption waste; and how much of this is recycled, from where it is collected (U.S. EPA, 1992).
- “This study may have implications for the selection and integration of media channels. For example, green claims might be expected to be more credible when they are delivered via channels that best lend themselves to the provision of detailed and meaningful information, such as print or online.”
Choosing Imagery in Advertising Healthy Food to Children: Are Cartoons the Most Effective Visual Strategy?
Food advertising is widely believed to influence children’s food preferences, although the evidence available is not conclusive. This study examined the effects of visual communication on children’s healthy food choices. As one might expect, cartoons were the preferred medium by children who participated in the study. But what surprised researchers Maria Lagomarsino Université de Neuchâtel, Switzerland and L. Suzanne Suggs, professor of social marketing , Università della Svizzera italiana, was that the children’s preference for cartoons did not translate into their consumption intentions. Among their findings from responses by children in Switzerland 6-8 years of age:
- When marketing healthy foods (such as carrots, broccoli, and apples and other fruit) to children, “it may be more effective to use photos of healthy foods rather than cartoons or other animations.
- “The use of a mix of visualizations techniques may maximize the attention to and adoption of healthy food marketed to children.
- “The use of peer modeling to advertise healthy-food consumption does not seem to be an effective visual technique.”
This area merits further research, the authors suggest. Their own work was limited by the fact that they conducted the study in just one region, when “cultural differences elsewhere can influence children’s responses.” Moreover, the language used in examples to children were “not real-world promotional messaging, which limits the applicability of outcomes in a real-world setting.” Finally, the authors did not control for food preferences. The authors recommend that future researchers test “whether differences exist in children’s perceptions of an actual product in which text and images interact, and to expand this work with other forms of communication, such as video.” Researchers also could “examine age-related changes in visualization preferences, and determine whether food-visualization preferences are consistent with food-choice behavior outside the experimental setting.”
Coming in March 2019:
What We Know about TV in the Digital Age
As a preview to March’s special theme section on television in the digital age, John B. Ford reviews seminal work that JAR has published on this topic.