The Most Valuable Design Skill in 2027 Will Not Be Figma
When anyone can generate thirty interfaces, producing another interface becomes a less meaningful distinction. Knowing which one should exist becomes more valuable.
The most valuable design skill in 2027 will be judgment.
That is my bet. Figma will still matter. So will typography, interaction design, research, and the ability to make something feel considered. But as AI makes it easier to generate interfaces, the value of a designer will depend increasingly on the quality of the decisions behind the work.
When anyone can generate thirty interfaces, producing another interface becomes a less meaningful distinction. Knowing which one should exist becomes more valuable. Knowing whether an interface is needed at all becomes more valuable still.
I am excited by that shift. I am also concerned about what happens when we mistake the ability to produce something convincing for the ability to make something useful.
For designers, business leaders, and teams building with AI, the question is becoming more demanding: what deserves to be built, and what gives us confidence that we are right?
A surplus of options changes the job
Imagine a team exploring a new customer dashboard. Within an afternoon, they have multiple layouts, different navigation patterns, polished cards, and several versions of an onboarding flow. They can compare directions that previously would have required a much larger investment of time.
That is a real advantage. I want teams to use it.
But every option introduces another decision. Which information belongs on the dashboard? What does the customer need to do next? Does the proposed hierarchy reflect the customer’s priorities or the company’s reporting structure? Will the experience hold together once actual data replaces the placeholders?
Generating more screens does not settle those questions. Sometimes it obscures them. The team starts discussing which version feels most contemporary before agreeing on what the product should help someone accomplish.
A polished interface gives an idea weight. People react to what they can see. An assumption that felt tentative in a conversation can look settled when it appears inside a beautiful prototype.
Design judgment keeps that visual confidence from outrunning the evidence. It helps a team use abundant options without becoming distracted by them.
Judgment begins before the screen
The most consequential design decision often happens before anyone opens a design tool.
A business asks for a dashboard. A customer may need a notification. A manager asks for an AI assistant. An employee may need a clearer process and access to the right information. A team asks for more engagement. The user may want to finish the task and leave.
These requests are understandable. Businesses describe problems through the solutions they already recognize. A designer has to investigate the need underneath the request without dismissing the people making it.
That requires curiosity and tact. It also requires the confidence to slow down a conversation that is already moving toward production.
What is happening today? Who experiences the problem? How are they getting around it? What would improve if we solved it? What would remain difficult even if the proposed feature worked perfectly?
Those questions can change the entire direction of a project. They may reveal that the opportunity lives in the service, the content, the permissions, or the handoff between people.
AI can help explore possible solutions once the problem is understood. Human judgment helps determine whether we are exploring the right problem in the first place.
Taste needs a reason
I care deeply about taste. A well judged type scale, a clear hierarchy, and the right amount of restraint can transform how an experience feels. Visual quality affects whether people understand a product and whether they trust it enough to continue.
But taste becomes more useful when a designer can explain the decision.
Saying that one direction feels better can start a conversation. It cannot carry the entire argument. Why is the primary action easier to find? What does the language communicate? Where does the layout reduce confusion? How does the experience express the company’s character?
Consider two onboarding flows. One is visually striking and asks for substantial information upfront. The other is quieter and lets someone complete a useful action before requesting additional details.
Either could be appropriate. The decision depends on the service, the user’s expectations, and what information is actually required. A financial application and a community event tool may need very different introductions.
Judgment connects visual choices to those conditions. It gives taste something to work against and something to answer for.
Experienced designers should be able to make strong recommendations while staying willing to revise them. Confidence is useful when it moves a team forward. It becomes expensive when it makes a preference impossible to question.
Coherence becomes harder as output expands
A product can contain individually attractive screens and still feel confusing as a whole.
One screen calls an item a project. Another calls it a workspace. The same action appears in different places. A helpful tone becomes abrupt during an error. A new feature introduces a navigation pattern that makes sense locally but conflicts with everything around it.
Each decision may look reasonable in isolation. Together, they make the user do unnecessary interpretation.
As teams generate more work, someone has to protect the relationships between the parts. That means paying attention to language, behavior, content, accessibility, and the underlying model of how the product works.
Design systems can support this, but a component library cannot resolve every contradiction. Teams still need to decide when a pattern fits, when an exception is justified, and when the system itself should change.
This is where design leadership becomes especially valuable. A leader sees beyond the screen under review and asks what the decision does to the rest of the experience.
Coherence saves people from relearning the product every time they move through it. It also gives teams a shared foundation for making the next decision.
A working prototype is a hypothesis
I love how quickly we can move from an idea to something people can use. A working prototype changes the quality of a conversation. Someone can attempt a task, encounter a problem, and show us something we would have missed in a presentation.
The danger is treating that prototype as proof before anyone has tested the assumptions inside it.
A flow can work technically while solving a problem nobody considers important. An AI response can sound helpful while leaving the user uncertain about what to do. A demo can feel effortless because its data and conditions were carefully arranged.
Judgment means keeping track of what we know and what we are inferring.
Before testing, the team should be able to state the belief it is examining. Perhaps customers can complete a request without assistance. Perhaps staff can identify the next action from the information provided. Perhaps a shorter flow reduces confusion without removing necessary context.
Then observe behavior. Where do people hesitate? What do they misunderstand? What do they ignore? Can they recover when something goes wrong?
Speed becomes valuable when it helps us learn earlier. If it only helps us produce a more convincing version of an untested idea, we have accelerated the wrong thing.
The best decision may be removal
One of the hardest skills in design is deciding that something should disappear.
Every feature has an advocate. Every section can be justified. Every additional field seems useful to somebody. Over time, a product accumulates decisions that were reasonable when they were made but no longer serve the experience.
AI can make adding another option feel almost effortless. The effort of generating it, however, is only part of its cost.
Someone must understand it. Someone must maintain it. Someone must explain it when it behaves unexpectedly. The feature may create new permissions, new dependencies, and new ways for the experience to fail.
A designer with good judgment considers those consequences before recommending an addition.
Could clearer language solve the issue? Could two steps become one? Could the product make a sensible default decision? Could a rarely used option move out of the primary path?
Removing something requires an argument just as adding it does. The goal is to preserve what people need and reduce what they have to carry mentally.
Restraint is especially valuable when production is cheap. A team needs the ability to make less work feel like meaningful progress.
Judgment is accountable to real people
An interface changes what people notice, what they understand, and what they are able to do.
That gives design decisions consequences beyond appearance. A difficult cancellation flow may improve a short term metric while damaging trust. An inaccessible control may prevent someone from completing an essential task. An automated recommendation may sound more certain than the information behind it warrants.
Good judgment considers those consequences alongside business goals.
Who could struggle with this decision? What happens if the system is wrong? Can the person understand the choice being presented? Is there a practical way to recover or reach a human?
These questions belong inside the work, especially when a product uses AI. The team has to decide how uncertainty appears, when confirmation is appropriate, and which actions require a person’s involvement.
The right answer depends on context. Suggesting a headline carries different consequences from changing a customer’s account or influencing an important decision.
A thoughtful designer makes those differences visible. That helps the business build something people can use with appropriate confidence, and it helps the team understand the responsibility attached to its choices.
Leaders need to make judgment teachable
Judgment can sound like an individual gift: something an experienced creative director possesses and everyone else waits to receive.
That is a weak foundation for a team.
Design leaders should make their reasoning visible. When reviewing work, explain the concern, connect it to the user’s task, and identify what evidence would help resolve it. Give people criteria they can apply before the next review.
A team might agree that the primary task must be obvious, that important choices need clear consequences, and that new patterns should have a reason for departing from existing ones.
Those principles become useful through repeated application. Discuss where they help and where they conflict. Invite designers to explain their recommendations and challenge the assumptions underneath them.
Junior designers still need opportunities to practice craft. They also need opportunities to observe research, understand constraints, and see what happens after their work ships.
If AI absorbs parts of production, leaders must be deliberate about preserving those learning experiences. People develop judgment by making decisions, receiving feedback, and connecting their choices to outcomes.
The goal is a team that can reason well together, with enough shared understanding to move quickly and enough openness to change direction.
What I am betting on for 2027
At Artifact Digital, I want the promise of AI to show up in the quality of the work and the usefulness of what we build. Faster exploration should give us more room to understand the business, involve users, and improve the experience.
My bet for 2027 is that designers who combine craft with defensible decisions will become more valuable. They will help teams identify worthwhile problems, choose among abundant possibilities, and recognize when an attractive direction is leading nowhere.
For a business hiring a design partner, this changes the questions worth asking. Ask how the team chooses a direction. Ask how it tests an idea and handles conflicting feedback. Ask what it has removed from a proposed solution and why. A portfolio can show beautiful finished work, but the conversation should reveal the thinking that made it possible. You are trusting that team with decisions that affect your customers and your operations. Its ability to explain those decisions should matter as much as its ability to present them attractively.
Tools will keep changing. The responsibility to decide what deserves to exist will remain.
When thirty interfaces are available, the valuable contribution is knowing which one earns its place in someone’s life, why it belongs there, and what we still need to learn before we ship it.