One of the things I love about Bank Holidays is that I get to spend a whole day reading.   I’m halfway through this book and there’s already interesting stuff in it.

One of the ideas I  particularly like is that of ‘pure procedural justice’, where a process inevitably leads to the desired result (in this case, a ‘fair’ one), and where the desired result can be projected beforehand according to criteria that are independent of the process applied.

Pure procedural justice is rare, unless you are dealing with a simple outcome, such as dividing a cake up equally, but it does seem to be a useful way of approaching process design:

• What outcome do I want?
• How could I define it before I run any process designed to achieve it?
• How can I design a process so that I will inevitably achieve the outcome I want?
• How could I measure the outcome independently of the process?

For processes that involve human beings, part of the answer is to abstract the desired outcome.  Rather than trying to list out every possible acceptable outcome, instead you define the characteristics of a set of possible outcomes each of which would be acceptable, even though you have no way of knowing what they are, or which you will actually get as a result of any particular run-through of the process.

That’s what your Promise of Value is for, to help you define those characteristics.