Ok, Web advertising is starting to drive me insane. Back when I worked at Amazon a few years ago, there was a version of the search results page that put sponsored links into the middle of the results. This basically means that you could search for something like "nail polish" and see a list of 5 or so results and then a blue box with links like "Lots of Discount Nail Polish!" and "Nail Polish for Everyone." This box was followed by more search results. Because Amazon search results are so detailed, the sponsored links often (to my eye) seemed simpler and much more applicable to what I searched for than the long list of Amazon results. It was tempting to click on one of these sponsored links.

Unfortunately, I don't think it was totally clear to everyone viewing the search results that these were actually not links to products -- or even pages -- on Amazon. If you clicked "Nail Polish for Everyone" you ended up on another site entirely because this link was actually an ad. It seemed wacky to me to put these simplified ads in a place where they'd be the most attractive links on the page, because they were so simple and direct. After all, Amazon wants to sell you Nail Polish (and books, movies, t-shirts, and lawn mowers related to nail polish) -- so why make it so attractive and easy to zoom off to another site, no matter how much they got paid for each click on the ad? It made no sense to me.

Now I work for clients who are facing the same issue. Companies that build a Web strategy around advertising revenue have to figure out how to integrate ads into their sites. And they face a lot of issues, including:

• Everyone who has ever built a Web page has probably participated in training users to ignore the right-hand column. It's become a 'blind spot' because it so often contains ads. Case in point...have you even looked at the right column of this page yet? And when you do, surprise! It's...ads!
• Lots of users even ignore the left-hand column of pages--unless, at first glance, they find that it contains site navigation.
• Any ad that stretches across the width of any page will effectively cut the page off. If users see long ad (much like the 'something to add or update' bar below this text, only an ad...) they think the page is done. Most won't look below the ad.
  
• Many internet advertising consultants are advocating in-line advertising. The basic idea is that there are blind spots, so you have to put ads in the 'non-blind-spot' if you want anyone to see them. Essentially, this means you have to intersperse them with the content in the middle area of the page. Well, this is great--but isn't putting ads in a particular place the way we got these blind spots in the first place? Seems awfully circular to me.
  

Seems to me that Web advertising -- and users' opinions of ads on Web pages -- must be changing. Of course, I have zero data to back this up, which is why I'd LOVE to see some comments on this page. But things have changed a lot since the 'early days' of Web advertising, and they must be having some effect:

• Users are getting more and more sensitive to clutter.
  
• Advertising is getting more contextual and therefore is, in some ways, turning into a feature. Again, check out the google ads on this page--pretty relevant, huh? You might even call them...dare I say it...helpful? I'm certainly not going to say they are well-designed, but this will happen.
• There's way too much choice both between and within Web pages. People want to find what they are looking for, dammit. If site X looks promising and I click a link on it and suddenly I'm on site Y, that's not so helpful. I have to start all over again trying to find what I was looking for and/or do what I wanted to do. By the way, if you haven't yet, you MUST read The Paradox of Choice, by Barry Schultz. Another page on that coming soon.

So I don't really have an answer, but I must admit I'm totally frustrated when I talk to clients and they tell me "well, we have to put the ads here because our internet advertising consultants say that this is the most profitable spot. They have eyetracking data and every time we push back based on customer experience they haul out the eyetracking data." This drives me NUTS. But I also must admit that for some reason eyetracking data has always driven me nuts--I think it's because it looks at what you already designed and rates the relative value of each design element. Again, this seems like tail chasing. Excellent designs are holistic. And eyetracking can easily reflect what users are forced to do due to a flawed design. In other words, Sally might be looking at this area a lot because she can't figure out where else to look to find what she wants. This certainly doesn't mean that the area she's looking at is well designed. (so there. harumph.)

So anyways. What do you guys think? Anyone got interesting insights? Heard good new stuff? Do tell!!!