Behavioral targeting (BT), with which advertisers direct particular commercial messages to specific individuals based on a range of observed online behaviors (including sites visited, content examined, and searches made), is quickly becoming a dominant form of Internet advertising. Even as the technique grows in popularity among advertisers (and BT is expected to reach $4.4 billion in U.S. ad expenditures by 2012, a nearly six-fold increase over the $775 million spent in 2008), it continues to evolve, becoming ever more targeted as it draws on and compiles a growing range of data points.^[1]

Advertisers know a good thing when they see it, to be sure. Xtract , for example, a self-proclaimed “innovator in social advertising intelligence,” views its behavioral targeting capabilities as “the black gold of the 21^st century.” The company has developed “a unique and patented set of products that based upon social interactions, behaviour data and other dataflow can create accurate and dynamic real-time customer profiles.”^[2]

With the spread of the Web to mobile implementations and the overall increase in online transactions, such profiles are growing increasingly detailed. Marketers now have access to more consumer behavior than ever before, all of it digital grist for the expanding BT mill. In a statement to marketers, Xtract promises that “with highly accurate, three-dimensional profiling, based on customers’ social, behavioural and demographic information, Xtract SAI (Social Advertising Intelligence) will revolutionize your marketing and multiply the results from your advertising campaigns.” Described as “a powerful solution for advertisers and marketers to target and manage personalized and automated advertising campaigns,” Xtract SAI is touted as “a complete solution that enables the automized application of analytical modules to harness the vast amount of customer information.” These modules permit “the processing of huge amounts of data as well as automating analytical tasks,” “a data repository, capable of handling billions of transactions” and “input output packages that enable data input from a wide array of data sources.”^[3]

The “Data Jackpot”

The language of the interactive advertising trade, full of “auomized modules” and “input output packages,” may be fuzzy, but the implications of advanced BT are clear:Marketers are now aggregating data from a variety of sources, online and off, to produce increasingly complex profiles of individual consumers. Such data, moreover, are continuously analyzed and assessed for insights into consumer tastes, preferences, and vulnerabilities.

The marketing messages that result from these analyses, carefully honed to match particular goods and services to specific individuals, can be created on the fly, in the course of a given consumer’s online travels. Sarah Keefe, vice president of marketing for Bango, a mobile Web strategist, explains how online marketers are now able to update profile-based mobile targeting in real time:

Marketers can… compile an accurate and rich understanding of their target consumer’s profile. With this data jackpot, marketers can target messages to the right audience in the right geographic location. Also, real time data allows campaigns to be tweaked and refined to ensure success and optimize the marketing investment…. It’s a brave new mobile marketing world out there and the wealth of data and analytics capabilities that are part of the new landscape eliminate the risk of jumping right in. Why wait?^[4]

Ringleader Digital (formerly MoPhap) isn’t waiting, having succeeded in bringing to the mobile platform a version of the PC-based cookie technology (which places small, updatable text files on a user’s computers for tracking and targeting purposes). “The challenge with behavioral targeting lies in reaching the same customer across different media,” the company explains. “With MoPhap’s mobile server-side cookie, we’re able to target by creating a comprehensive profile based upon a user’s browsing history over time. The technology derives contextual meaning from each page that the user visits. This combines to create a psychographic user profile. We’re also able to identify the specific mobile device….”^[5] MoPhap offers a simple example of how the mobile and online worlds are converging for the purposes of BT-based advertising:

A consumer clicks on a mobile Web banner, opts in for an SMS [Short Message Service] auto response, and later receives a text message that includes a coupon code that must be redeemed online. With the mobile cookie, we’re now able to tie the subscribers together online and in mobile…. [M]obile advertisers can plug into the auditing, tracking, and reporting solutions they’ve used online to track consumer behavior.^[6]

Taking Online Advertising to the Next Level: Predictive Targeting

Simple tracking and targeting techniques are now giving way to even more invasive artificial intelligence tools that allow marketers to anticipate consumer actions online and actually predict user behavior. Enpocket, for example, a leader in “intelligent mobile marketing” purchased in 2008 by Nokia, has developed a “personalization engine,” which it describes as “a system of analytical models that scores mobile users based on their past behavior,” enabling the company “to predict which products and services a customer might purchase next.”^[7]

Personifi, similarly, offers a “uniquely powerful ad optimization solution [that] observes all available behavioral, contextual, demographic and other data to determine the most effective ad for each impression,” according to the company website. “Unlike many optimization solutions, Personifi’s optimization incorporates predictive capabilities which leverage learning from earlier campaigns, eliminating the need to waste impressions at the outset of a campaign. By observing all available data (including behavioral, demographic, contextual, transactional, and more if available), Personifi allows mobile providers to gain a more complete understanding of users’ characteristics and behavior patterns.”^[8]

Acxiom and Acuity Mobile, meanwhile, melding the former’s vast data and analytical capabilities with Acuity’s location-based and “Spot Relevance” ad servers, offer BT services that “facilitate relevant content delivery to a specific person based on preferences, time, context and location,” ensuring “real-time, location-aware, user-targeted mobile marketing.”^[9] In addition to the information gleaned from the mobile device itself, Acuity’s eMAP (Embedded Mobile Advertising Platform) Preference Engine “can mine multiple data sources to access any consumer information available,” ensuring that all available data, even from “third party databases of consumer data” are utilized.^[10]

We are only at the beginning of such predictive technologies, but already analytics specialists such as Targusinfo are promising to take “online targeting to the next level.” The company’s latest offering, called AdAdviser,

relies on a far wider range of data points to anticipate future behaviors across a wide range of interests…. Where other targeting methods track a limited set of fleeting “in-market” behaviors, AdAdvisor provides a holistic view of the consumer by using segment scores to track hundreds of predictive persistent behaviors and provide offline data to corroborate online behaviors…. AdAdvisor is unique in its ability to provide insight into:

• What a consumer is likely to purchase, across a wide range of products and services

• A consumer’s likely ability to afford a purchase

• Household- and individual-level demographics, including age and gender

• What websites a consumer is likely to visit

With the largest repository of US offline consumer information, Targusinfo is uniquely positioned to take online targeting to the next level.^[11]

Other new interactive-advertising approaches, according to Targusinfo, include affinity targeting (which “examines similarities in what and how people are interacting with specific content areas…. to identify specific affinities such as product-category preferences and price sensitivity”) and interest-based targeting (in which advertisers partner “with data sellers to aggregate anonymous shopping and research behaviors across the Internet”).^[12] But predictive targeting, which moves “online targeting from one-dimensional to multi-dimensional and from reactive to proactive,” appears to have the most promise, according to Targusinfo.^[13] More to the point, predictive targeting draws on off-line data, not only for “demographics but [also for] family status, number of children, hobbies, home ownership, lifestyle patterns, hobbies and avocations, discretionary purchasing priorities, brand and product affinities, education, income and occupational status.”^[14]

From Prediction to Prescription

The advertising industry is notoriously secretive about its plans for the future, but it is possible—by subjecting clues gleaned from the trade press to the same inferential, anticipatory research that fuels sophisticated BT campaigns today—to forecast where interactive advertising is heading. Mobile advertisers, for example, are hard at work on predictive systems to complement the long-standing cookie technology on PCs. One potential new system, the Mobile Individualized Lifestyle Knowledgebase, reportedly uses precise, real-time audience segmentation systems to generate unique universal consumer codes (UCCs) that correspond to universal product codes (UPCs, the ubiquitous bar codes that are affixed to all manner of consumer goods). Sellers of hairdryers or garden hoses, for example, will be able to purchase access to consumers whose UCCs indicate a high probability of interest in, need for, or attraction to certain personal grooming or landscape architecture devices.

The technology is still being refined, of course, but once it matures, the lifestyle knowledgebase and its custom UCCs will work in very much the same anticipatory, predictive fashion as Targusinfo’s current lead-generation application. Deployed at call centers across the country, Targusinfo’s system “instantly scores incoming calls to prioritize and route them to the optimal agent and to create a customized experience for each caller. Incoming calls are instantly scored before agents pick up the phone, matching callers with” specific items that a company may have for sale.^[15]

Eventually, as consumer databases become more robust and online analytics become more refined, these predictive technologies will give way to prescriptive advertising, command-based consumer calls to action that many individuals will be powerless to resist. Drawing on advances in neuromarketing, artificial intelligence, and emerging cyber-psychological studies, the Continuously Updated and Personalized Consumer Action Knowledge Engine will comprise a veritable “Digital DNA” of consumer behavior—from the earliest inklings of desire (e.g., Web browsing and window shopping), through active research and price comparisons, all the way to the completion of transactions (including extended credit arrangements and the inevitable buyer’s remorse). The results will have profound implications for us all.

^[1] David Hallerman, “Behavioral Targeting: Marketing Trends,” eMarketer, June 2008, p. 1.

^[2] Xtract, “About Us,” http://www.xtract.com/about-us/ (viewed 18 June 2009).

^[3] Xtract, “Xtract Social Advertising Intelligence,” http://www.xtract.com/products/xsai/ (viewed 18 June 2009).

^[4] Sarah Keefe, “Mobile: A Brave New World for Advertisers,” in Mobile Marketer’s Classic Guide to Mobile Advertising, Aug. 2008, http://www.scribd.com/doc/8640034/A-Guide-to-Mobile-Advetising2008 (viewed 18 June 2009).

^[5] Phil Leggiere, “Targeting Multi-Modally,” Behavioral Insider, 26 Sept. 2007,http://www.mediapost.com/blogs/behavioral_insider/?p=194 (viewed 18 June 2009).

^[6] Leggiere, “Targeting Multi-Modally.”

^[7] Enpocket, “Advanced Profiling and Targeting,” http://www.enpocket.com/solutions/enpocket%20platform/advanced-profiling-and-targeting (viewed 1 July 2008).

^[8] Personifi, “Mobile Optimization,” http://personifi.com/mobile_optimization.html (viewed 18 June 2009).

^[9] Acuity Mobile, “Acxiom Corporation and Acuity Mobile Partner to Power Targeted Mobile Marketing Solution,” press release, 21 May 2007, http://www.acuitymobile.com/docs/Press05212007.php (viewed 18 June 2009).

^[10] Acuity Mobile, “About Acuity,” http://www.acuitymobile.com/company/technology.php (viewed 18 June 2009).

^[11] Targusinfo, “Taking Online Targeting to the Next Level,” Mar. 2009, pp. 1-2, http://marketing.targusinfo.com/AdAdvisorLearningCenter.html (registration required).

^[12] Targusinfo, “Taking Online Targeting to the Next Level,” p. 9.

^[13] Targusinfo, “Taking Online Targeting to the Next Level,” p. 9.

^[14] Targusinfo, “Taking Online Targeting to the Next Level,” p. 10.

^[15] Targusinfo, “Credit Card Issuer Increases Premium Card Sales with On-demand Analytics,” p. 2, http://www.targusinfo.com/documents/credit_card.pdf (viewed 18 June 2009).