What Reason Have We Given Our Audience to Care?

Continuing on the Jakob Nielsen soapbox I've been on for the past several days, I came across his latest article, where he reluctantly released data showing that displaying ads that look like editorial content actually produces results. He argues against this practice under the presumption that ethical companies win in the long run.

No doubt there are still some disciples of John Maynard Keynes who believe, "In the long run, we're all dead." Thus, they'll allow their short-run thinking to overrule ethical business practices. That's too bad, but when we are all aware of the practice, our collective knowledge should somewhat dilute its effectiveness.

The reason the method seems to work in the short run is due to a phenomena called "banner blindness," which is a description of users' tendency to completely ignore anything other than the content they came to read. According to Nielsen:

Scanning is more common than reading, but users will sometimes dig into an article if they really care about it.

He then includes this evidence for his claim, which represents three different contexts in relation to a task. In every case here, the display ads were completely ignored:

"Heatmaps from eyetracking studies: The areas where users looked the most are colored red; the yellow areas indicate fewer views, followed by the least-viewed blue areas. Gray areas didn't attract any fixations. Green boxes were drawn on top of the images after the study to highlight the advertisements."

How you organize your copy will depend on your intended audience, but I'm amazed by the number of companies that actually think their customers want to read their fluffy marketing-speak (What Steve Krug calls "Happy Talk"). For people who don't know who you are or what you do, coming to your site will  often result in scanning like that seen in the left-most portion of the image above. You have to be able to convince them -- quickly -- that they have come to the right place and that you can solve their particular problem.

If you can't, that's okay - you're not the right solution for them anyway. It's still best to be respectful of their time, too, since they may need your service (or know someone who does) anyway, and if your site is easy to scan and use, then they're more likely to consider you.

In light of the shift in advertising dollars being moved from offline media to online, advertisers might be troubled when they reflect on Nielsen's advice to resist the urge to advertise over a network, and instead contract according to the content. (It's important to note that the "shift" that is occurring isn't a zero-sum game; there is still a very large net increase in ad revenue).

It's absolutely true that the more customized the ad is for the content and the audience, the more effective it will be, I think Nielsen draws too hasty a conclusion from the data (and to be fair, perhaps because he never tested it). Just because someone doesn't focus on a display ad does not mean they haven't seen it, and according to a study performed in October of 2006, the effectiveness of contextual ads increases by 249% when run in conjunction with display ads.

I doubt Nielsen, if he saw that study, would conclude such a dramatic increase had nothing to do with the display ads. But what can we conclude from this?

Simply this: We have to work harder to give people what they want and need, and we cannot rest on our laurels just because we're getting more ad revenue to purchase advertising that makes use of the interruption model.

I believe Nielsen's suggestion would, indeed, be more effective, but I have trouble believing it is the most efficient way to do this. The ad networks must figure out away to allow advertisers to incorporate ads that make use of, if not more audience data, then at least more data about the content. Specifically, if you are, indeed, Time's "Person of the Year," don't force on users advertising that suggests the contrary.

The worse we do up front, the less effective even the relevant ads will be, because people will become accustomed to treating ads (in whatever form) as having nothing at all to do with them. As advertisers on the traditional side are learning, once that shift in thinking has taken place, it is very difficult to break through the clutter to get our audience to care. - Cam Beck

Comments

[Paul Herring]

Cam,
Neilsen argues that banner ads don't work based on information from eye tracking studies. He's only partially correct. Some banner ads don't work. However, EVERY banner ad is measurable and can be optimized to work based on goals. We can talk to our brethern in online media (even those that work at our companies) and they can show us first hand. If they didn't work, you wouldn't have seen the dramatic increase in online media we've seen in the last two years.

Now TV ads, well, that's another story.