Jobs To Be Done

by Nick de Voil 5. January 2017 18:12

I've been reading Alan Klement's book "When Coffee and Kale Compete" about his JTBD thory.  You can download it for free at http://www.whencoffeeandkalecompete.com/ .

In my opinion, it's a really helpful book and it goes some way towards relieving some of the problems I feel are inherent both in standard UX-style and also traditional BA-style analysis of people's needs and motivations. I think it will help us to come up with a better synthesis of the two. There's a lot of wisdom in there.  For example, he points out what's wrong with longitudinal studies (but also clarifies when he thinks they're appropriate) and and also (one of my pet hates) the Five Whys. And this on personas was music to my ears:

Personas include data such as race, age, and gender; however, these data represent only the natural, common variation among the people who use the product. But common variation doesn't help you understand... For example, a persona may describe a customer who likes to use the product on weekends. Now, is that important to the design, or is it a distraction? Is it real, or was it fabricated to "bring the user to life"? How many customers said they use it on weekends? One? Ten? One hundred? When invalid data are co mingled with valid data, how can you tell the difference? Personas do not distinguish variation due to either common or special causes. The layman who does not understand statistics will believe any variation within a system is due to special cause.

Have a read and tell me what you think.