Insight/ Branding & Communications

Advertising Learns To Adapt

by Samuel Greengard
Contributing Writer

See More by this author >

Article Highlights:

  • The idea behind adaptive advertising is to replace spreadsheets and manual levers with a seamless process.
  • Adaptive advertising helps marketers identify where a person is in the buying cycle and deliver ads in the right format for the right device and screen.
  • Adaptive advertising now comprises about 10 percent of overall media budgets, but the figure is expected to rise during the coming months.

It’s no secret that consumers find themselves increasingly inundated by marketing and advertising messages. Industry experts say that the number of ads a person views in a typical day has swollen from around 500 in the 1970s to more than 5,000 today.

What’s more, catchy slogans and jingles that once appeared in nationally distributed print ads or television commercials have given way to a highly fragmented and fractured environment that spans multiple delivery channels, including digital media.

Navigating this environment is no simple task. CMOs and other marketing executives increasingly find themselves attempting to break through the clutter and provide the right ad in the right context. “There’s a growing recognition that ads must be customized, even personalized, and they must be contextually relevant,” said Andrew Frank, a vice president and distinguished analyst at Gartner Inc., in an interview with What’s more, the systems serving the ads must “learn and adapt to constantly changing conditions and information.”

Enter adaptive advertising. The concept aims to create a more dynamic and flexible approach to marketing. Although there’s no single and precise definition for this method, it centers on the idea of using numerous data points and existing demographic information to target consumers more narrowly and connect to their needs and interests. The variables can include factors as diverse as weather conditions, geographic location, time of day, changing market or pricing conditions, and click behavior. The technique also plugs in account information and past purchase histories, and uses modeling to view potential outcomes and optimize a strategy.

“Marketers are now able to use artificial intelligence to generate real-world results,” said Eric Porres, chief marketing officer for data analysis and digital marketing firm Rocket Fuel, in an interview with “Traditional advertising revolves around campaigns. Today, effective digital advertising is all about learning, adjusting to constantly changing conditions, and being always on. It’s about identifying moments of influence when a consumer might be most receptive to a given message.”

Beyond Banners
Throughout most of advertising history, people have made virtually all decisions about the creative direction of advertising. Surveys, focus groups, and customer feedback provided some ability to adjust and adapt campaigns, but changes typically took place over the course of weeks or months. Today it’s possible to modify advertising and adjust campaigns in real time to match specific criteria and changing conditions. Much of the process can take place automatically and without human involvement.

It’s no small change. The premise of adaptive advertising, Porres said, is that “humans are great at storytelling. They’re great at providing narratives. But they aren’t able to sort through data and reduce the inherent complexity of the marketplace and the world.” He said the idea behind adaptive advertising is to replace spreadsheets and manual levers with a seamless process that delivers relevant ads at a point where a person is receptive.

Gartner’s Frank said that adaptive advertising, while incorporating elements of contextual advertising and native advertising, goes far beyond these methods by continually changing and adapting to outside conditions. Data sets can range from a handful of variables to hundreds or thousands of different factors—not all weighted equally. Adaptive advertising also relies on algorithms to learn behavior and predict outcomes. It also helps marketers identify where a person is in the buying cycle and deliver ads in the right format for the right device and screen.

The concept is attracting a good deal of attention among major companies across a wide swath of industries. The list of firms includes Lufthansa, Choice Hotels International, Birds Eye, Epson, Bridgestone, Brooks, and the Denver Art Museum. The ability to apply predictive modeling, adaptive controls, and portfolio optimization techniques can change the stakes across various delivery channels and media. Not surprisingly, adaptive advertising is now finding its way into online banners, videos, mobile campaigns, and the social media space.

The challenge, of course, is finding the right array of variables and producing results that really work. As Porres put it: “Even minor missteps can produce huge errors and deliver ads that are entirely off base.” What’s more, said Ritu Trivedi, senior vice president of digital strategies and partnerships at global marketing agency MediaVest, “There are a lot of different things you need to be mindful of and a lot of data that you need to ingest about consumers so that you can be relevant. At this point, everyone is still learning how to plug in the right mix of variables to produce optimal results.”


RocketFuel uses an approach that’s heavily dependent on external feeds (including demographic data), a company’s internal Web metrics and CRM data, and overall data modeling and analytics methods. The marketing company studies data patterns over time, continually tweaks modeling methods, and adjusts the nature and frequency of ads based on all of the data. What’s more, using computer modeling and artificial intelligence, marketers are able to explore models and simulations and understand how they might play out over time.

As these systems—and marketers—get smarter about what works, they’re able to adapt dynamically. Justine McGrath, director of digital at New York City-based agency OptiMedia, said that the agency has plugged in purchase data, information about online searches for recipes, and other lifestyle variables in order to send consumers more relevant coupons and incentives for competing products—at a time they’re more likely to act. “By more precisely targeting consumers, we have achieved success rates that are approximately 125 percent higher than through the use of conventional methods,” she told

The Denver Art Museum is also sold on the concept. Cindy Sewell Hohman, marketing and promotions manager, began using adaptive advertising a couple of years ago to target specific demographic groups, including younger visitors. “Digital advertising can become overwhelming. There are lots of promises but not always great results,” she told An adaptive approach has helped the museum fine-tune messaging and deliver ads more effectively. “The system has learned over time who clicks through and what works,” Hohman added. The museum has witnessed as much as a five-fold increase in response rates, along with higher ticket sales, using the approach.

Connecting The Digital Dots
A growing number of marketing analytics firms and ad agencies are embracing adaptive advertising. In addition, some marketers are experimenting with cameras, sensors, and other monitoring devices in retail stores. This allows them to adapt messages based on the demographics of who’s there, or using facial recognition to serve up highly personalized ads. For example, in Europe, Amscreen has introduced a system that determines gender, age, date, time, and volume of viewers, and dynamically adjusts ads on in-store display screens.

Yet despite all of the advances, bottom-line results are far from guaranteed. For one thing, companies using adaptive advertising must adopt a more flexible approach to marketing, including the way they budget, experts said. A campaign might require anywhere from minor to radical change and, ultimately, lead to a different approach than a marketing team first envisioned. This, in turn, can drive changes in Web sites, branding messages, and, in the end, the metrics used to measure success. There’s also the nettlesome issue of privacy and how data is stored and used.

OptiMedia’s McGrath said that adaptive advertising now comprises about 10 percent of overall media budgets, but she expects the figure to rise during the coming months. Although there’s a highly automated component to adaptive advertising, “It’s not something you put on autopilot,” she said. “You have to constantly think about the creative side and how to adjust messaging for different situations, scenarios, and outcomes.” What’s more, it’s crucial to keep an eye on the changing nature of digital media, including social tools, in order to create a more complete value loop.

MediaVest’s Trivedi believes that success in adaptive advertising spins a tight orbit around understanding consumer experiences—and the human experience. “You have to understand how people behave and how the touch points translate into buying patterns,” she said. The challenge isn’t collecting data—it’s finding the right data and connecting all the dots. Among other things, this requires organizations to break down data silos. “The media shouldn’t just reside within an ad agency or in a database for a marketing department,” Trivedi said. “It’s the interplay of data that creates real-time marketing insights.”

Although there’s a growing focus on big data, adaptive advertising can also hinge on small data, Frank noted: “Sometimes relatively few factors are crucial to how people buy things.” But there’s also an issue of how effectively an advertiser can act on the data. “There’s a trade-off between the number of variables you use and how many variations on a template you create--and comparing all of this to a handcrafted creative approach,” he added. “A lot of time and money is required to produce a relatively small number of creative treatments customized for a specific audience.”

Frank said that while adaptive advertising is still in the early adoption phase—and it remains somewhat experimental—the underlying concept promises to revolutionize marketing in the months and years ahead. Meanwhile, RocketFuel’s Porres was predictably exuberant about the approach. “It provides CMOs with a research lab for testing theories, messages and attributes--and refining things in order to maximize the probability of success,” he said. “We’re able to capture moments of influence and enter the realm of one-to-one marketing.”