Process directs your attention to impact

A good process needs to let you use the same tools every time. If you’re messing around with the process itself or the tools you use every time, you’re not focusing on the deep, intellectual work.

An IA/UX practice has pace layers, just like (nearly) everything else, going from fast to slow:

  • Tools - Mural, Figma, Post-Its, Excel
  • Artifacts - Flow diagrams, personas, journey maps
  • Processes - Object/content modeling, tree testing
  • Work structures - Who’s on the feature team, how work gets planned and assigned
  • Major questions - What should we build? How should we build it?
  • Purpose - Business goals, user goals

Cyd Harrell proposed this about about the larger community of UX practice, but it’s also true at the individual project/feature/etc. level. Changing the tool you use to create part of a feature is easy. Changing the purpose of the feature without re-addressing the major questions and work structures? It never works.

Ideally, your process will reduce the amount of thought you spend on the upper layers (What tools do we use? Where’s the template? Which artifacts do we need?) to spend more energy on the slower-moving and more impactful lower layers (What is this thing supposed to do? How do I make it work?)

Just choose a tool and use it every time. Just put together some artifact templates and keep making them. Tinkering is appealing, I know, but tinkering stops you from thinking about the deeper layers. On the other hand, if your tools, artifacts, and processes are getting out of your way, you can think about the major questions of the thing you’re building.

Cf: Attention is the beginning of devotion


Cyd Harrell. “Pace Layers, but for UX Practices.” Twitter, 15 Nov. 2021,

tools come & go; they can be important & they can affect the deeper layers if they persist, but they are not the deeper layers. practitioners who understand the work at those layers will have more durable careers.

tool & process argument can also take up a lot of energy that could otherwise be spent improving the deeper layers of our little UX institution (they’re called pace layers because change at the deeper ones is slower & usually harder)