Point of View
With people spending so much time in the digital world, marketers need to be just as effective there as they are in traditional media. But so far, many marketers have yet to see a good return on their digital investments. For all the successes, there are many more failures. Most video does not go viral. Few fan pages garner many fans, and those that do are often simply collecting "likes," not building brands. And according to our research, many digital ads are ineffective, ignored or, worse, have a negative effect.
One problem is that marketers often rely on faulty assumptions about people's online behavior. Too much digital marketing is informed by naïve beliefs such as these:
Before marketers throw money behind ill-conceived digital efforts, they need to understand all the ways people behave and respond in the digital space and set clear and realistic goals for what they hope to accomplish.
The digital realm is not one monolithic universe; rather, it consists of a wide variety of distinct domains.
This seems obvious, but the implications are huge because consumers approach each domain differently.
Someone who is searching for the best price for a specific product has a different mindset than someone who is hanging out on Facebook.
But at least two commonalities do exist across the various digital domains. First, in almost every digital environment, users find a rich and complex array of options for becoming informed or finding diversion. And second, whether they are seeking facts or amusement, users are in control, choosing their own routes through the online territory according to their moods and objectives.
Therefore, the potential to irritate people with online marketing must be taken seriously. A user who is looking for information is on a mission and will not appreciate an ad that gets between him and his objective. People who are checking out content for fun may be slightly more tolerant of ads, but even so, our research shows that of all the ads that move purchase intent in a noticeable way, 27 percent have a negative effect.
The implications of this are frightening: could a lot of online advertising actually be damaging brands?
Whether they are purpose-driven or kicking back, people make quick, often split-second decisions about where to click in the digital environment. They are not necessarily going to follow a predictable or linear path. And while online users enjoy being in charge of their experience, being constantly in control is demanding.
To cope with the demands of the very choiceful environment found online, both our instinctive and our deliberative decision-making systems are forced to work overtime. Content is quickly judged as being relevant or irrelevant. Work conducted by Carleton University in Ottawa demonstrated that it takes one-twentieth of a second for people to make a decision about the visual appeal of a web page, less time than it takes to form a conscious thought.
We can assume that people make the same instinctive judgments about online ads. Our eyetracking database suggests that, on average, only 41 percent of people let their eyes rest on an online ad for as long as three-tenths of a second. When people do look at ads, they look for an average of just 1.3 seconds—not a long time, but long enough to make an instinctive judgment. Video ads get more attention online than other ad types; they are looked at by 70 percent of online users. But even with video, people make quick decisions—the overall video dwell time is just 1.8 seconds.
The web has been sold through the numbers, the eyeballs, the clicks, the case studies, and other research data. It is often called the world's most measured medium. But we have reason to wonder if all of the measurements are truly meaningful. Ted McConnell, Executive Vice President, Digital, for the Advertising Research Foundation, reported on an experiment in which he and some associates ran blank ads in different sizes and colors. The average click-through rate across half a million impressions served was 0.08 percent. (For a brand campaign, this would be good; for a direct-response campaign, so-so.) But when asked why they clicked on blank boxes, only half of the "clickers" said they were curious to see what might be behind the blank box. The other half admitted to a mistaken click.
So if half the people who click on an ad do so by mistake, what does that imply about the value of online advertising? As Ted said, "At a minimum, the data suggest that if you think a click-through rate of 0.04 percent is an indication of anything in particular, you might be stone-cold wrong." ROI calculations based on click-through start to look even more dubious.
Fortunately, we know that people don't have to click on ads to be influenced by them. While digital marketing can create valuable interactions between brands and consumers, these don't have to occur for a digital campaign to be successful. It's great when people do want to engage with your digital content, but the people who will play your online game, download your app, or create and submit a video number in the thousands at best. You need to plan for that level of engagement while also making sure that the rest of your exposures—which may number in the millions— are not wasted.
The good news is that digital advertising can build brands the same way traditional advertising does—by creating and reinforcing brand associations that strengthen a person's predisposition to buy the brand. This has been obvious right from the beginning of our online ad testing in 1996. Indeed, if we look at more recent data from Dynamic Logic, we see that increases in purchase intent are 14 times higher than suggested by click-through alone. This is also evidenced by the fact that people who are merely exposed (i.e., those who don't click) generate a significant portion of the sales effect of a piece of online communication.
Our understanding of the digital environment and the ways people behave there leads us to make some simple recommendations on how to succeed in the digital space.
Online communication needs to connect instantly and make a great first impression on the many who see it, not just the few who may engage with it. Make it simple for people to judge whether an online ad is relevant to them or not. Don't make them think; enable them to simply respond.
A TV ad has the luxury of time to develop its story and integrate the brand; online doesn't. If a brand is not front-and-center throughout the execution, exposures will be wasted. Even the location of the brand in the ad is important. We tested a rich media banner that had action moving from left to right while the brand was static on the left-hand side. Thus viewers' eyes were consistently being pulled away from the ad. In this execution, the brand would get more attention if it were placed on the right-hand side.
The mindset of online users will vary with context. You need to understand the motivations of online users in the various spaces where you want to play. For example, people are not on Facebook to be unpaid promoters of your brand. Some of them become fans just because they want to get coupons and offers, while others "like" your brand because they believe that signals something positive about them. They are using your brand for their purposes, not yours. So think about what they want, and deliver it to them.
Context and mindset also come into play in determining how ads might best be deployed, as when deciding whether a video should be placed in-stream or as an on-click execution. Table 1 shows AdIndex data on the average response to these two types of ads. (The numbers represent the difference between those who saw the test ad and those who saw a control ad.)
In-stream advertising is over twice as memorable as click-to-play, but does not produce the same gains in purchase intent. This points up something about how users process these different viewing experiences. The in-stream ad makes an impression at the time it is viewed (because all events make an impression on us, automatically, at the time they occur). But people don't pause to reflect on the message because, as soon as it finishes, they become engaged with the content they really wanted to see.
However, when the click-to-play ad is finished, nothing happens until the viewer makes another click; thus the viewer does have a chance to process the message he or she just saw. On-click ads, therefore, are more effective at delivering explicit messages. In-stream executions deliver implicit impressions, much like many commercials seen during TV viewing.
Behavioral metrics should not be relied upon on their own to optimize spend. Click-through, search, and liking depend as much on things that happen elsewhere as on the immediate online circumstances. Set expectations about what your digital marketing is meant to achieve, and judge effectiveness against those goals.
The digital space, in all its diversity, represents a huge opportunity with the power to engage people in new ways. But we need to remember that even activity that is "free" has both risk and opportunity cost. The risk is that it will be negatively received and undermine brand allegiance, and the opportunity cost is in the allocation of precious resources to activities that may be ineffective or even largely unnoticed.