There's something I see designers do often, and it's terrible for their design process. They sit down to start working through a problem, and before they've even figured out what they're supposed to be solving, they're already making something polished. Formatted. Labeled. Something that looks like it was meant to be presented to someone.
It wasn't, though. It was just supposed to help them think. This is the deliverable trap, and a lot of us fall into it without even noticing.
Not Everything You Make Needs to Go Somewhere
We talk a lot in UX about deliverables. These are the artifacts we hand off to stakeholders, developers, clients, or whoever else is waiting on our work. Deliverables matter. A well-crafted research report, a thoroughly annotated set of designs, a clear task flow — these are things other people need to understand your work and act on it. They take time and care to produce, and that's appropriate.
But there's another category of artifact that we make constantly and rarely talk about. I'm calling them design disposables.
Design disposables are the things you create not to deliver to anyone, but to help yourself think. They exist to get what's in your head out into the world so you can look at it, poke at it, and figure out if it's any good. The moment they've served that purpose, they've done their job. Most of them get thrown away. That's not failure. That's the point.
Early concept sketches? Disposables. Sticky notes on a whiteboard during a brainstorm? Disposables. That page in your notebook where you mapped out a rough research plan before you knew what you were trying to learn? Disposable. Even a quick prototype can be a disposable, if you built it specifically to figure out what you should be building.
Why Disposables Matter (and Why We Avoid Making Them)
The reason disposables are so valuable is also the reason designers often resist making them: they're allowed to be wrong.
When you sit down to work through a fuzzy problem, a research question you haven't fully formed, a design direction you haven't committed to, you need to be able to put something imperfect on paper without it meaning anything yet. You need the freedom to sketch a flow that you might throw out entirely, or write a research plan you'll gut and rewrite before you ever share it.
The trouble is, a lot of UX professionals have been trained (by feedback, by culture, by perfectionism) to treat every artifact as something that will be seen and judged. So instead of roughing something out quickly to test whether it's a good idea, they spend hours making it presentable. And by the time they're done, they're so attached to it that they can't see its flaws anymore.
This is the sunk-cost fallacy applied to design artifacts, and it slows you down in a very specific way. It makes the exploration phase of design feel expensive, which means you do less of it.
Here's the thing: exploration is supposed to be cheap.
Sometimes You Have to Make the Thing to Know If It's the Right Thing
One of the sneakiest forms of design disposable is a high-fidelity artifact you're treating as low-stakes. Sometimes a quick prototype, a draft set of designs, or even a nearly complete task flow is the only way to figure out whether you're going in the right direction. You truly cannot evaluate the idea without seeing it.
That's fine. That's a legitimate use of your time. But you have to go into it knowing that you're making a disposable, a thing whose purpose is to teach you something and then potentially get thrown away. If you don't make that decision consciously, you'll treat it like a deliverable instead. You'll present it, defend it, and ship something you should have discarded.
The difference between a disposable and a deliverable isn't always the artifact itself. It's the intention behind it.
AI Makes This Easier and Harder at the Same Time
There's one more wrinkle — how many of us are making things now.
AI tools have gotten good at speeding up UX-artifact creation. You can generate a rough research plan, a first-pass task flow, or a set of early concepts in a fraction of the time it used to take. From a disposables perspective, that sounds great. Faster to make means easier to throw away, which means less attachment, which means more freedom to explore. But there's a catch.
When you make something yourself, even something rough and disposable, you learn by making it. You discover which parts of a research plan are hard to write because the question isn't well-formed yet. You notice which step in a task flow feels awkward because the underlying logic doesn't hold up. The friction of creation is where a lot of the thinking happens.
When an AI generates the artifact for you, you can skip straight to having an output without going through the process of making it, and if you're not careful, you can also skip the learning part. You end up with something that looks like a disposable but functions more like a placeholder. It gives you the feeling of progress without the understanding that should come with it.
By all means, use AI to speed up your process. Generate that first draft, spin up that rough prototype, get something on the page faster than you could alone. Just make sure you're still doing the work of evaluating it critically, questioning it, and pushing back on the parts that aren't right. The point of a disposable was never just to have an artifact. It was to think through a problem. Don't outsource the thinking.
A Quick Gut Check Before You Start
Before you start making anything, ask yourself: Who is this for?
If the answer is "for me, so I can think through this problem," you're making a disposable. Keep it rough. Keep it fast. Give yourself permission to be wrong. The goal is to get clarity, not to produce something impressive.
If the answer is "for someone else, so they can understand my work," you're making a deliverable. Take the time to do it well. Format it. Annotate it. Make sure it communicates clearly to someone who wasn't in your head when you made it.
Both kinds of artifacts are valuable. Mixing them up is where designers get into trouble.
If you want to go deeper on this, specifically on how to tell the difference between when you need a disposable versus a deliverable, and how to use both more effectively throughout your design and research process, I created a new self-paced course. We get into the practical details of how to make these decisions in the moment, so you can spend your time on the right artifacts at the right stages of your work.
Because the goal was never to make beautiful deliverables at every step of the process. The goal was always to design something great.
