After writing The error in advertising I’ve been thinking about what the future holds for the humble web banner advertisement. The likes of Google AdWords and Facebook Advertising deliver ads to the user based on ‘interest’. i.e. adverts relating to their Google searches, or Facebook ‘liked’ pages etc. While sites like BSA aggregate signed up websites with ad content relating to the site they appear on.

These methods are all very well but as discussed in my last post the question remains – do users consciously engage and react to these banner adverts in the way the advertiser intends?

Adverts that learn

Remember that film 2001: A Space Odyssey? I won’t go into detail, but a spaceship is controlled by an ‘intelligent’ computer (HAL 9000), and with it’s calm human voice it interacts with the ships Astronauts in a very human and emotional way. (Later he gets a bit too big for his boots and things get messy).

What about if all online advertising was controlled by a central company/application, and this application was able to read data from users of Google, Facebook, Twitter and Yahoo accounts? (Lets leave privacy out of the equation for now). Adverts could then be supplied in a very personal and targeted way on any site, using the data it’s collected.

Here is some of the data that Facebook, Google, Twitter and Yahoo could tell us:

– Who we are in relationships with, married to, or related to (Facebook)
– Birthday dates of our friends and loved ones (Google Calendar and Facebook)
– Our Google search history (Google)
– Items we’ve purchased or added to baskets without checking out (Google Checkout)
– Sites we like (Facebook and Google Buzz)
– What we’ve watched on TV or at the Cinema (Twitter and Facebook)
– Books we read (Twitter and Facebook)
– Food we like (Twitter, Facebook and Google)
– Where we’ve visited or ‘checked in’ (Twitter via FourSquare or Gowalla and Facebook)
– Our current GPS location (Twitter via FourSquare or Gowalla, Facebook and Google Latitude)

As well as basic information like:

– Our name
– Where we live
– Age
– Gender

A single style

One of the main problems right now is that all banner adverts vary in appearance quite a lot. Banner adverts rarely consider their environment, or position on pages and are usually designed to stand out from any other content. I think that this is part of the problem as we are conditioned to ignore, block, and close anything that gets in our face. Especially if it’s not invited.

So, with our new centrally and universally controlled ‘learning’, emotionally connected banner adverts need a visual style that doesn’t follow the current trend for adverts. We need a look that people feel comfortable with and almost feel pleased to see. We want to generate the same feeling you get when you see you’ve received a text message from a loved one, but we don’t want to try and imitate any OS type alerts (‘you have a virus – click here to perform a scan…’ – you know the kind of thing). These are spammy and dirty. Perhaps simple white on black text would work?

Idea 1

The advert below has looked at the users Facebook friend, in this case his mum. Worked out the relationship to him, worked out her birthday, (and maybe a Google calendar entry). Then combined that with his mums Google web history data:

Idea 2

Simple information could be used from Google checkout history to supply adverts based on users past purchases:

Idea 3

How about pulling information in from the users Facebook Friends status updates:

Idea 4

Getting GPS coordinates from the users mobile device and supplying location based adverts:

In reality

If you’ve seen the movie 2001: A Space Odyssey you’ll remember that HAL got a bit too intelligent for his own good, and there is certainly room for some mistakes to be made using the imaginary ideas above. We’ve also completely disregarded privacy issues for the sake of the blog post and I’ve no doubt that some people would get freaked out by the type of adverts above. In reality though, many of the ideas above are actually happening but in individual ways. Logins and online accounts are already talking to each other and it’s something I feel will develop quickly in the future. Who knows how much, and in what way our data will be used.