We do not describe the world we see, we see the world we can describe.

— René Descartes

Let’s say you’re working on a project. The project launches, other people start using it. How do you know it’s good? 

This is an important question, but trickier than it appears at first. I divide it into three parts:

  • Someone is being cast as the arbiter. (“You.”)

  • A continuum is implied: from no-good to good.

  • An objective is specified: you want the thing to lean towards the “good” end of the continuum.

All three are up for examination. Let’s start with the arbiter of “good.” If the project is being developed within an organization, it’s likely that many parties will have stakes in its success. And even if you’re working alone, there will still be other parties involved. For example, there will be people interacting with your system. They, too, will have a say on whether it is good or not. “Good” will likely mean different things to different people.

There’s a point along the continuum at which the thing is obviously no-good; it’s not meeting anyone’s criteria of success. As it evolves and (ostensibly) improves, it moves closer to meeting more criteria. (In theory, this process ends with a “perfect” artifact. In practice, this is an impossibility; the context around the system continues to evolve as the project progresses, changing the criteria for success for at least some participants.) Eventually, the system moves far enough along this continuum to flip over into the “good” side. (We say it’s “good enough.”) In other words, it meets the success criteria of a large enough group of stakeholders.

Who defines the threshold where the artifact flips from no-good to good enough? How is this measured? This is the crux of the matter: What “good” means in a particular context is up for grabs. Often there will be one lead stakeholder who will be calling the shots. There is an expectation that this person (or team) will be the arbiter of “good”. But these stakeholders are often buffeted by political forces that influence their decisions one way or another.

Designers working on complex systems need to understand how success will be measured. The criteria stakeholders will use to evaluate success will have an important influence on the structure and form of the system. For example, if the organization defines success and an improvement of a Net Promoter Score (NPS), the system’s designers will be strongly incentivized to structure it in such a way that feeds that measure. (As Jared Spool has pointed out, this may not be a good idea in the case of NPS.) Stakeholders and clients using NPS as a filter will see the world through that lens. This can leave out important factors for success.

What’s good for an internal team may not be good for an organization as a whole. And — more importantly — what’s good for the organization may not be good for society as a whole. As designers of complex systems, we’re called to see beyond the world we (and our stakeholders) can describe and measure. Our vision needs to encompass a wider (and longer-term) perspective if we are to provide real, enduring value.