Oct
31
2011

Personalization: trick or treat?

When a friend drops by unexpectedly with a birthday gift, it cheers us up.  But what if a total stranger were to walk up at the train station and wish you a happy birthday?  That’s rather creepy.

Halloween – when we acknowledge and mock our most visceral fears – is an appropriate day to introspect about personalization and privacy.  The increasing amount and integration of personal data enables both fiendishly clever personalization and ghoulishly undesirable loss of privacy.

For example, the Gilt Groupe, an online, members only buying service makes every visit to their site unique and personalized based on the tastes and preferences of the member.  This is made possible by the fact that the site requires registration, and by sophisticated analytics used to present only the most relevant choices to each visitor.

On the other hand, consider the targeted advertising in Google’s Gmail service.  Though mostly innocuous, there are times, after noticing an advertisement clearly triggered by the content of my private communications, I am reminded that the price of “free” is a loss of privacy.

Even if we look past specific privacy concerns, personalization may have the unintended consequence of limiting our view of the web as noted by Eli Pariser, MoveOn.org executive, in his book, The Filter Bubble.  Nonetheless, personalization will continue to increase as consumers ignore “one size fits all” advertising.

Amazon.com was the pioneer in friendly, helpful recommendations.  Even though they use purchase history to predict and direct our interests, it never feels intrusive.  Why?  What makes personalization a trick versus a treat?

Personalization is the behavior of a service changing based on personal information.  Here are three rules for “good behaviors” that will banish those eerie feelings of being followed:

  • Relationships – personalization doesn’t work well with strangers.  However, if a site or service establishes a relationship in the form of registration, or an Opt-In, then it is perceived positively.
  • Transparency – like Amazon, make it clear why I am seeing the specific action or recommendation.  Don’t be mysterious.
  • Consumer Control – allow me to modify my “presumed” attributes.  Maybe things have changed, or maybe someone else last used my computer, or maybe I want to buy for someone else.

If personalization is done right, more people will be willing to take off the mask of anonymity and share a bit of themselves in order to get a more relevant, valuable online experience.


Comment are closed.