Know Thine Customer

Part of my job is to research new methods of online interactivity, and I have recently been looking further into this area. I am trying to come up with streamlined systems that might produce a high degree of interactivity with the user when embedded in online content.

Targetted marketing is no new idea. That any product should be tailored to the market it is designed for is an obvious fact. But it was only in the 20th century that large scale market research became the vogue. Since the pioneering work of characters such as Herta Herzog, any kind of consumer article tries to second gues the perceptions of what it guesses are its audience and tries to satisfy them as well as possible.

The area of interest these days is the way in which websites that we visit can tell a large amount about the visitor simply by analysing their online movements. Amazon.com, for instance, has progressively improved its systems for giving customers recommendations. This has been a bit of a headache for them, but through improving their data-crunching skills, analysing what each user buys, looks at, searches for, it comes out with a recommendation that is supposed to be very closely tailored to its customers.

Yet amazon has barely scratched the surface of what is possible, and ehre we get into territory that is easily scandalous. Websites automatically save all manner of information about the user, and this can easily look like a form of espionage. Several big marketing, web analytics, and content management companies are investigating the possibility of bringing the user targeted content that is dynamically linked with their own behaviour. As they dance around the web, the web dances with them. So far Amazon, though proud of its achievements, has been treading on its users' toes.

For instance, were my website to be a bookshop, if a visitor arrives having come from a gardening website, I might display a range of popular gardening books prominently on the page. However, if the user is surfing using firefox and linux, I might guess that they would be more technologically-minded and less amenable to being baby-sat, and display the site in a more directory-like way. If the user is using windows and internet explorer, however, the opposite is a good guess. Furthermore, from the kind of network or connection that the access is routed through we might guess about whether the user is at work or not - perhaps displaying a less playful scheme to those whose colleagues/bosses might be peering over the shoulder. Location, too, has long been exploited by advert companies, but can do much more than simply suggesting local businesses. Much can be guessed about tastes from the location, and much more can be guessed from the synthesis of location, software, patterns of use, IP address all combined.

With scandal after scandal rolling over our tv screens/RSS feeds about government losses of private data, it might be surmised that the British public are neurotic about information, and any form of data collection is therefore suspicious. But if we come to terms with the fact that this information is collected anyway, we might enter an age of interactive surfing willingly, appreciating the intelligence of websites. How we respond to electronically generated content, and what psychological mechanisms are thereby triggered, is an issue that is really interesting to follow, and a battleground for the marketeers of the future.

I find these things a bit of a headache to work with, and many of the firms researching these things hold their cards close to their chest. Its fascinating though. I feel like I am involved in cutting edge stuff, even though our company is really quite small.

Otherwise, life is good here in Hoxton.