18: Torque is cheap

In the last email (subject: ​Serendippidus​), we looked at the foundations of work design. And we discussed how the workflow user or AI requires:

• Context

• Instruction

• Guardrails

• Blueprints & templates

Now, one objection people will have upon seeing this list is we can’t possibly design all tasks to this level of detail. It would take years!

And they’d be right.

But not all tasks are created equal. Meaning, the effective execution of some tasks within a workflow are more important than others.

For example, if our goal is to make a video helping salespeople to deliver better product demos, it’s probably more important to include the right content than having perfect lighting. But ‘write the script’ and ‘setup equipment’ are both tasks in our workflow. And if we look at them without that context, they could be perceived as having the same importance.

But the outcome of the workflow is more dependent on great content than great video production.

So, when considering to what extent we design work, we should triage tasks based on their importance.

And this brings us onto something I’ve been excited to discuss since we began our journey seventeen emails ago…

Until now, I’ve referred to the intentional improvement of work as ‘design’. But I believe we should be taking it one step further, and referring to this as ‘engineering’.

Because what we’re doing here isn’t just haphazardly tweaking things to see what works. We’re systematically improving performance.

We’re engineering performance.

Because if our workflows are documented, and if we know what outcomes we’re hoping for, and if we’re measuring those outcomes, then through a process of controlled experimentation, we should be able to identify how changing the design of a task impacts the outcome.

A bit like tuning up our car.

Now, I don't know much about cars. But having recently seen Brad Pitt star in a very cliched Formula One movie, I can safely say a team of mechanics can fiddle with the car and make it go faster. Or adjust it to suit the conditions of the race.

And so, we can apply the same logic to our workflows. What we’re calling tasks become levers that can be pulled to improve performance.

Thus, ‘workflow engineering’.

Which is exactly how we shall be referring to it from now on.

Yours,
- Ant