I talk to people a lot about interest and social graphs these days – it’s a topic that feels like it’s moving in from the fringes of social media towards the centre, as we begin to understand what the wider implications the vast amounts of data on users tastes, relationships and interactions will eventually have on the Web.
Whilst I’m a huge believer that the kind of profiling which will inevitably come from this could well lead to a more personalized and contextual Web experience for us all in the very near future, I’ve been troubled by some nagging doubts about it recently.
F**k it, we'll do it live!
Our biggest ever edition of TNW Conference is fast approaching! Join 10,000 tech leaders this May in Amsterdam.
Where to start? Well, I feel quite strongly that if you give someone an experience that’s relevant to their tastes and/or their current environment (such as location, weather, time of day) then you’ll have a far more receptive audience, whether you want to raise awareness of an issue, promote a brand or even just and sell them something. The potential for such experiences is huge but then again so are the potential mistakes digital creators could make when profiling or presenting personalized or contextual Web experience.
You are what you click?
Perhaps my biggest concern is that users will passively believe that they are who the web tells them they are and that the knowledge of this behaviour may be just a little too tempting for those out for easy wins. In my experience people are less discerning and more easily swayed when browsing and interacting with the Web, and while its perhaps a lack of faith on my part, it’s easy to see how that could be abused when you’re presenting something which is supposed to be a personalised experience.
However, whilst no doubt tempting ventures for some, on-going ‘personalized’ web experiences based on inaccurate or forcibly skewed data are likely doomed to fail anyway since they’d eventually create the kind of negative feedback loop where even the most passive user would feel that their Web experience isn’t actually that representative of their tastes. You only have to look at Klout to see how badly some platforms can get this wrong these days – I’d expect that kind of inaccuracy on a larger scale would lead to some fairly pointless and wasteful Web experiences.
In addition to this, the fact that, so far, much of the existing personalised web is built around ads (Facebook, Google and recently Twitter) or the opportunity to sell stuff (Amazon) should mean we need to venture forward carefully and consider what experiences users might actually want – rather than jump headfirst into commercially driven ones. Essentially, the need for understanding human behaviour when it comes to the web is greater than ever and we need to be thinking seriously about how we can connect with users on multiple levels.
My final concern is that although this future social Web could be a wonderful development for humanity, there seems to be very little promotion of its benefits right now, least of all at the points where much of the useful social data is being collected. For instance, I have yet to see a Facebook Open Graph app which clearly details the benefits of connecting with it, in fact it almost seems as if developers want us to click that tempting blue button without a real knowledge of what’s involved.
Open and balanced
Surely a more open and balanced approach is needed here. If users clearly understand the positives of sharing social data then it’s likely they will volunteer it and not contribute to a well publicised backlash against the organisations which are collecting it, as we’ve seen in the past regarding Facebook and Google. You only have to look back a few years at the disaster that was Facebook Beacon to see where this process can go seriously wrong. Let’s hope that developers will remember this lesson going forward and seriously question the assumption that the collection and sharing of personal data is great for everyone, regardless of whether they have willingly opted in or not.
Right now, we’re seeing the birth of a more personalized Web, and no doubt the first iterations aren’t going to get everything right. However, if we keep user experience at the core, and resist the urge to go for easy wins, then we can ensure this future Web is truly a meaningful and personal journey and not some skewed and likely irritating one built for the person marketers and developers want us to be, rather than our genuine selves.