Defence Against the Dark Arts

Following @TimHarford 's illuminating advice on the dark art of ‘drip pricing’ (Financial Times, 21 Aug 2010) and my post on badly designed websites, @gagan_s comments "Sometimes bad websites, phone-trees and policies have a dark purpose".

The word "dark" can mean "hidden", "obscure", "mysterious", "secret", "unconscious"; or it can mean "devious", "evil", "malicious", "treacherous".  Harford's use of the term "Dark Art" in relation to drip pricing clearly denotes a practice that is not just unclear and confusing but also morally questionable.

In terms of the dark purposes of behavioural economics, a website designed to exploit drip pricing needs to be just complicated enough that when the customer reaches the webpage that demands an additional payment for paying by credit card, the customers have already wasted so much time that it isn't worth starting again with a competing website, so most of them grumble but pay.

IT professionals who understand commercial software and/or website design also grumble about the incompetence of the developers of the website. That just goes to show how naive most IT professionals are, if they imagine that all companies genuinely want (and are willing to pay for) websites that are simple and easy to use. Or they may feel morally outraged that companies are not willing to invest the time and money to do things properly.

As regular readers of this blog will know, our starting point is Stafford Beer's maxim The Purpose Of a System Is What It Does (POSIWID). According to this principle, there is no such thing as an unnecessarily complicated website: the complication emerges from some conscious or unconscious dark purpose. In the case of drip pricing, the purpose is to chisel a few more dollars from weary and/or confused customers. This purpose may exist even if the managers of the company aren't aware of it; they may intend (one day) to simplify their pricing scheme, but they feel no pressing need to do so; meanwhile, the company continues to profit from complexity.