design process, explained
why the “HCI way” is failing designers and what “design process” means in 2026
This is the second piece of the explained series I’m writing. you can check my first piece about founding designer here:
now to the new article:
I’ve been thinking about the phrase “design process” a lot lately, and why the “HCI way” is failing designers in practice. I keep hearing the same story from portfolio critiques, interviews, and friends who graduated from great HCI programs: the methods are useful, but the promise doesn’t match the job.
You know the diagram:
Research → Personas → Ideation → Wireframes → Testing → Hi‑fi → Ship.
And I get why people love it. It’s clean. It’s comforting. It makes design feel predictable.
But the longer I work in industry, the more I think that diagram is being taught wrong. School gives you real methods, then sells them like a linear recipe - when in real life they’re a toolbox. And if you internalize the recipe, you start to feel like you’re “doing it wrong” the moment reality gets messy.
The tension: methods are real, the recipe is not
I don’t want to do the edgy “school is useless” take. I actually think design education is valuable.
It gives you the tools. It shows you what exists:
interview techniques
usability testing
heuristic evaluations
journey maps
personas
competitive analysis
prototyping
experimentation frameworks
design principles and conventions
Even if you never use all of them, it matters that you know they exist. You can’t pull a tool you don’t know is in the drawer.
So that’s part one of the solution:
Part 1: You need the tools
You need the vocabulary, the methods, the fundamentals. You need to understand why conventions exist. You need to be able to look at a problem and go, “Oh, this is the kind of situation where talking to users could actually change the decision,” or “This is the kind of problem where I need to prototype because nobody can reason about it in words.”
Design education helps with that. It builds your toolbox.
But there’s a second part that matters just as much, and it’s the part I think school is way worse at teaching:
Part 2: You need to know when and why to use each tool
And the word I keep coming back to for that is intentionality.
Intentionality is the invisible layer under the steps
It’s the part where you’re clear about:
Why you’re doing something
What you’re trying to learn
What decision it unlocks
What risk you’re trying to reduce
What would change your mind
Intentionality is basically the difference between doing research because “process says research comes first” and “we’re about to commit engineering time and we’re missing a key piece of truth”
Same method. Totally different energy.
And honestly, without intentionality, the toolbox doesn’t work. You just end up making artifacts because artifacts are what you’re supposed to make.
The contrast: the diagram vs. how work actually shows up
Traditional HCI teaching loves the clean storyline. It’s neat. It’s presentable. It’s portfolio-friendly. But real work, at least the real work I’ve experienced - usually starts as confusion.
It starts as messy prompts like:
“We need onboarding.”
“Users don’t get it.”
“Sales is losing deals because of X.”
“This feature is table stakes now.”
“Retention dipped, can you look into it?”
None of that is “step one: research.” It’s just… a foggy situation someone wants you to un-fog.
And this is where intentionality changes everything.
Because when you start in confusion, “process” isn’t about producing the right artifacts in the right order.
It’s about moving from fog → clarity.
When you start in confusion, the real step one is questions
If confusion is the starting point, then step one isn’t personas.
Step one is questions.
Not performative questions. Not “for the workshop.”
The slightly annoying questions that force the situation to become real:
Why are we doing this now?
What problem are we actually solving?
Who is this for, specifically?
What does success mean?
If we ship nothing, what breaks?
What are we assuming that might be wrong?
What decision are we trying to make by Friday?
This is what I mean by intentionality: you’re not just “following process.” You’re trying to unlock the next decision.
And then, and only then, you choose a tool.
Sometimes the right tool is interviews. Sometimes it’s competitive teardown. Sometimes it’s analytics. Sometimes it’s prototyping. Sometimes it’s just writing down the decision tree and getting alignment.
The point is: intentionality is the bridge between “I know methods exist” and “I know which one to use right now.”
What intentionality looks like in different orgs
Once I frame it this way, the differences between agency/startup/big tech stop feeling like “they don’t follow process” and start feeling like: they’re using the same toolbox for different reasons.
Agency: intentionality is narrower (because upstream decisions are pre-made)
In an agency, the “why” is often already decided.1
The client or founder gives you a prompt, the constraints, the timeline, sometimes even the opinionated direction. You’re not there to question the existence of the project. You’re there to execute.
So intentionality becomes narrower and sharper:
How do we turn this brief into something coherent quickly?
How do we make it feel high quality?
How do we translate someone’s vision into something real?
The toolbox is still there, but you’re mostly living downstream - craft, speed, clarity, presentation.
And honestly, it can be an amazing craft bootcamp.
Startups: intentionality is about reducing guessworks under uncertainty
Startups are where the fog is thickest.
Prompts are vague. Context is missing. Everyone is moving fast. Everyone is guessing. And the company can’t afford long ceremonies that produce pretty decks but don’t change outcomes.
So intentionality in a startup starts with questions and rapid validation:
What are we actually trying to prove?
What’s the smallest thing we can ship to learn something real?
What would “good” look like in user behavior?
The process becomes less about formal steps and more about reducing guessworks.
Opinions are cheap, prototypes are semi-cheap, but real user behavior is the expensive truth needed. So a lot of “research” in startups becomes things like:
playing with competitors deeply (not just screenshots)
reading user complaints and tickets
watching funnels and retention
shipping and measuring
iterating fast without getting emotionally attached
The goal isn’t to do research “properly.” The goal is to make the next decision with less fantasy.
Big tech: intentionality is de-risking at scale (methods as insurance)
Big tech is basically the same toolbox, but used with a different intention.
At scale, mistakes are expensive. A tiny UI change can move real money. It can create PR issues. It can break trust. It can confuse millions of people.
So intentionality becomes: de-risk.
How do we validate without blowing up the product?
What’s the smallest rollout we can do?
What do we need to be confident enough to ship?
This is why big tech can spend weeks (or months) on changes that feel “small.” The process is not discovery for discovery’s sake. It’s often insurance.
Same methods. Different reason.
Intentionality on a project: the onboarding example
Onboarding is the example I keep coming back to because it looks simple on the surface, but it’s secretly a perfect “intentionality test.”
The prompt always sounds like:
“We need onboarding.”
If you treat process like a recipe, you’ll jump to the standard onboarding artifacts and flows, because you’ve seen them a thousand times.
But if you apply intentionality first, you start by asking:
Why do we need onboarding right now?
What problem are we seeing - activation, retention, confusion, trust?
What does success look like? (Completion rate? Activation metric? Fewer support tickets?)
What do we want the experience to feel like?
What’s the first impression after the marketing site - do we match that promise?
And those questions unlock a very different space.
Because “onboarding” isn’t just “screens you tap through.” It can be:
a typical wizard (safe, familiar, efficient)
a guided tour (more hand-holding)
progressive disclosure as you explore (less friction, more autonomy)
a chat-like guide or assistant (more human, more conversational)
an avatar / character (brand personality)
a map or “mission” structure (game-y, directional)
something playful (if your brand can support it)
All of these can technically satisfy “we need onboarding.”
But they communicate completely different things about the product: how confident it is, how serious it is, how human it is, how it sees the user.
Same prompt. Different questions. Different intention. Different outcome.
That’s intentionality in action: you’re not choosing a pattern because it’s the next step in a process - you’re choosing it because it’s the right lever for what you’re trying to accomplish.
Where I’m landing (for now)
I think “design process” is not the steps.
To me, the real structure is:
You learn the tools (education matters).
You build intentionality (why am I doing this, what decision does it unlock?).
Then the toolbox actually becomes useful.
And maybe that’s the whole point:
The process isn’t there to constrain you. It’s there to help you stay honest - about what you know, what you don’t, and what you’re actually trying to decide.
I’m sure I’ll refine this take later. But right now, this version feels true enough to write down.2
footnote:
my take on design process at agencies is based on a combination of my personal experience running a design studio and conversations with friends who worked at design agency or studios. if you have different takes, feel free to disagree and i would love to hear new perspectives!
Thanks to all the people who made this post possible by sharing your insights and stories with me <3