Are advertisers still in a state of denial about display advertising? In 1997 Jakob Nielsen used eye tracking to identify a characteristic of web use – people simply don’t look at ads. Fifteen years later and what Nielsen called banner blindness is still an everyday hurdle for advertisers. We had a chat with Nielsen about why display still aren’t fit for the web – and what advertisers and publishers can do to make audiences pay more attention.

TheMediaBriefing: Has the online ad industry moved on since you first wrote about banner blindness?

The ad industry is in denial about the fundamental underlying problem, that the web is an interactive medium, not a passive consumption medium. That is why search advertising has been so successful, because it works with the flow of users wanting to decide where they want to go and what they want to do. The display ad industry is still dominated by one-way communication.

Even if they make interactive ads, let’s say a little game, that creates pseudo interest, is still all about how to spread brand message. It’s not that people aren’t willing to buy on the internet. It is that people don’t want to be talked down to, to be led by the nose.  

So why is display advertising still such a big growth market, with the likes of Google putting a lot of effort into it?

It is a paradox. Google see these marketing managers partying like it is 1955 throwing money away and Google think they might as well throw that money at us. But I doubt deep down they believe they are delivering value for the money. 

What do you make of new ad formats that are emerging, that seem to be more effective?

Anytime something new is tried it has a temporary effect of enhanced clicks and enhanced user retention. But banner blindness is inherently a psychological effect called selective attention. There is so much stimulus in the environment that for humans to survive they can only pay attention to the important things. If you get a new stimulus it takes a while before that system builds up again to automatically filter it out.

That is why when a new format comes out you always see these initial press releases saying it has something like five times the click-through rate of normal banners. But you never see releases two years later saying the click-through rate has declined to the same level as other formats. 

What difference does good creative make? 

High quality creative makes a big difference, but at the same time, there is a tendency for it to be so creative you can’t tell what it actually is. We see a lot of user behaviour looking at part of an ad, perhaps because there is an unusual  graphic, and then looking away. They need to be more utilitarian than they are used to.

Video ads in certain genres can be made highly entertaining. So for instance people might want to watch a movie trailer – which is inherently visually appealing, good-looking actors, explosions and action etc. You see a lot of that on YouTube with pre-roll commercials. But even then you have the ability to skip the ads and many people will do so to see what they are looking for. It is still fighting against your own customers, which is not a long-term prescription for success. 

Even very impressive ads work less well in pure display (banner ads and MPUs etc), because even if you target well, banner blindness means people aren’t looking at it at all – it doesn’t really matter whether it is interesting or not. We see this with site features which people were looking for but didn’t see because they looked like an ad. They see its general shape and they ignore it.

What about more interactive formats such as buttons (like those developed by ad company Response)?

Buttons attract attention, and they work with the basic mechanism of the web – people are scanning the pages looking for what they can do. People are always on the lookout for the buy button, the add to cart button, the search button. But the second challenge will be to offer some sort of interaction that makes sense. They need to be more action oriented and relevant. 

What about mobile?

We are still at a phase where it is relatively new for people to do a lot on mobile phones, so banner blindness isn’t as strong yet. And because the screen is smaller you are more likely to see all that’s displayed. But we are starting to see people don’t actually look all over the screen. Banner blindness comes from something more deep and will therefore play out in any format. But it takes a while to evolve.

What can publishers do to make ads work better?

They could look at content in new ways. If you can integrate the advertisement well with the content, it becomes almost more like a customer service than advertising in a traditional sense. They should screen out advertisers who are not credible, which means not just taking any advertiser with money, but credible advertisers that have passed some level of screening.

An example would be news aggregators for certain industries. They have what they call sponsored postings. The people who pay for that promotion tend to be people of interest to that audience. It might not be at the most relevant, but it could be bumped up for a fee. 

But there are some brands and content that don’t work that way. There is little chance of designing a Coca-Cola ad to feel like a natural part of an article about starvation in Africa. 

Are there some brands that don’t have a place in display advertising?

There are a lot of product categories that are very generic. They have very big ad budgets, but they are trying to create an artificial distinction between the product by the way they advertise. That type of advertising is not very well suited for the internet. 

One thing they can do is sponsored content. Doing something purely entertainment, like the old soap opera idea, is that you have a set of YouTube videos – perhaps an exercise video could be sponsored by Diet Pepsi.