Does Behavioral Targeting Work?

One-to-one marketing… personalized content and information… targeted advertising… All of these ideas have been the big promise of the web since it became a marketing vehicle six years ago. While marketers have promised this type of targeting capability and precision, the problem is that behavioral targeting doesn’t always mean greater results for the marketer. So why has this kind of targeting failed?

The Problem of Accuracy

One way targeted marketing can fail is when incorrect inferences are made. This can happen easily when marketers try to target individuals by psychographic or behavioral data.

The underlying principle is illustrated in this example: I am a web surfer, and I visit an outdoor travel web site. I surf around a few pages in the site and then leave. The marketer then infers from this bit of information that I’m interested in outdoor topics and products, and subsequently begins pushing me offers of this type.

This is usually accomplished via an advertising or marketing network that tracks surfer activities and puts the information in a database. The problem with this approach is that the inferences drawn are often incorrect. Many people surf on a lot of different sites, spending a decent amount of time there. They look at all kinds of things, but this doesn’t mean they are necessarily interested in the general theme of the web site.

For example, I visited a site this morning looking for a writer I thought I might have gone to high school with. I spent about fifteen minutes on the site until I figured out the writer and my former classmate are not one and the same. The site was about fashion, and anyone who knows me knows this would not be a good topic to target me on. (!)

Even targeting on purchases isn’t a reliable indicator because people buy merchandise for others. This is the underlying reason this type of targeting has failed: The inferences driving the targeting are just plain wrong.

Legal and PR Issues

Another problem with delivering on the targeted marketing promise is the fact that many people object to being targeted. This resistance has been expressed in lawsuits and public campaigns to inhibit such database targeting.

Most people don’t want to think of themselves as marketing targets. They don’t want their perceived or actual privacy invaded. Companies that finally got their technical ducks in a row so that they could deliver targeted marketing messages have found that a large group of people want to stop their efforts. This has caused the movement to take a step backward and makes it necessary for these marketers to re-evaluate.

It’s All About Permission

To really make the targeting capabilities of the Internet work, you need to utilize permission-based efforts. In this format, people have freely provided their demographic information and have agreed to be targeted. This type of permission and in-depth information must then be exchanged for value to the individual. People need to feel that they get extra value by allowing marketers to target them. This value can be in the form of important information, rewards, or even payment.

The technology is available to achieve detailed targeting, but it will only be successful if the audience is willing. Value-added, permission-based targeting is the bridge to this capability, and the successful marketer will take advantage of these opportunities.

Related reading