NBOP National Board of Operations Professionals

Part one. The Task

The three variables

Repeatable, no customisation, no thought. All three, or you are automating something else.

0:49. Captions on by default. Everything said here is also written above, so nothing is only in the video.

Read the transcript

Before automating anything, three conditions, all required. Repeatable. The same input produces the same required output every time, not most of the time. Frequency is not repeatability: a task done a thousand times a day in a thousand slightly different ways is not repeatable, it is just common. No customisation. The task does not change based on who it is for. If a step is performed differently for the largest customer, that difference is judgement, and judgement is what you are proposing to remove. No thought. Nobody has to decide anything not fully specified by a rule you could write down. Failing any one produces the same outcome: confident, wrong output for the remainder. And exceptions cluster around size and complexity, which is where your important customers are.

Before automating anything, three conditions, all required.

Repeatable. The same input produces the same required output, every time, not most of the time. Frequency is not repeatability: a task done a thousand times a day in a thousand slightly different ways is not repeatable, it is just common.

No customisation. The task does not change based on who it is for. If a step is performed differently for the largest customer, that difference is judgement, and judgement is what you are proposing to remove.

No thought. Nobody has to decide anything that is not fully specified by a rule you could write down. This is the one that gets waved through, because the thought is frequently invisible to everybody except the person doing the work, who does not experience it as thinking.

The reason all three are required is that failing any one produces the same outcome: an automation that works for the majority of cases and produces confident, wrong output for the remainder. And the remainder is where your important customers usually are, because exceptions cluster around size and complexity.

The test is to have the person who currently does the task list every case in the last month where they did something different, and why. If the list is long, you are not looking at an automation candidate yet. You are looking at a process that has not been defined.

Two things worth doing, neither of which costs anything

Read the standard in full, free and in the open, before you pay anything or give anyone an address.