You have ideas about how the world works. Direct experience informs some of these ideas, but others are less tangible. For example, you believe the earth orbits the sun, even though you’ve never observed the phenomenon directly. (You’d need to stand very far away to see the earth going around the sun!) Still, you believe it — and with good reason. We have ample secondary evidence for this model. Moreover, it’s proven to be useful for making good decisions.
Models are very important. They affect how you think and act. They allow you to go about your life without having to figure out everything from first principles. If you understand the parts of a system and how they relate to each other, you can then act more skillfully when interacting with the system. That is, your choices lead to more predictable outcomes. If you push button A, then action X happens. After several times of X reliably happening after you’ve pushed A, you come to expect the A → X relationship. You’ve built a model in your mind of how the system works. A good model maps closely to reality, allowing us to make good predictions about the effects of our actions.
We take much of this for granted, but it’s worth emphasizing that models aren’t a fixed, solid fact about reality; they’re mental constructs. There was a time when most people believed the sun orbited the earth. This model was reasonable given the (relatively) crude observational capabilities at the time. What could be more obvious when every day you can see the sun rise in the east, arc across the sky, and then set in the west? The ground is fixed; the sun moves. It’s reasonable from that evidence to deduce that the sun orbits the earth. But it’s wrong. Careful observations with more sensitive instruments reveal a different, contradictory configuration. Which is to say, the tools and systems that mediate our perception of reality inform our models. Whether we’re discussing astronomy, business, or politics, good media lead to good models.
Models are the designer’s stock-in-trade. The people who use the things we design bring to the experience models of how they expect things will work. Some models are more sophisticated than others. Some are closer to reality than others. If users’ models correspond to the system’s actual configuration, users will have an easier time learning the system. On the other hand, if they don’t have a good model — which is often the case with more complex systems — then they will need help getting started.
Given my experience as a design consultant and teacher, I’ve come to believe models are the most crucial unacknowledged factor in successful design work. Models are crucial because, without a good model of how the system works, the designer is relegated to stumbling about, making only superficial interventions. They’re unacknowledged because they’re hard. They’re hard to talk about and they’re hard to work with. Models are abstractions, and most of our interactions — both in design and in day-to-day life — deal with more tangible stuff. So, we lack practice. But practice we must, if we’re to become good at modeling. And we must become good at modeling if we’re to do good work — especially with the complex systems we’re designing today, which inform so much of our perception of reality.