Let's Not Build Design Portfolios for Algorithms – Build Them for Humans Who Hire

I rejected a designer last month whose portfolio had perfect case studies, consistent branding, and a clear CTA to "Request Full Deck."

It looked like every other portfolio I'd seen that week. Which was the problem.

The work was fine. The presentation was polished. But I couldn't tell how they thought, how they handled ambiguity, or what it would be like to work with them. Their design portfolio looked like a landing page optimized for conversion metrics.

Designers aren't products. And portfolios aren't sales funnels.

So why are we building them like we're running A/B tests?

Take this post we stumbled across the other day:

This advice spreads because it feels actionable. When you're stuck staring at a blank Figma file, anything that promises structure is seductive.

But this turns design portfolios into sales collateral. And if you're trying to show depth, complexity, and how you actually work – sales collateral doesn't cut it.

I've reviewed about 200 portfolios in the last year. The ones that get interviews aren't the ones following LinkedIn formulas. They're the ones where I can see how the designer thinks.

Let's take the "greatest hits" and play them out:

"Limit it to three projects"

Why? What if two are similar? What if the project you're proudest of isn't visual at all?

I once hired someone whose UX portfolio had six projects – three shipped, three that failed. The failed ones were more interesting. They showed constraint navigation, stakeholder disasters, and what they learned. That's what convinced me.

The number matters less than the depth.

"Use one accent color, max"

Sure, if you've only worked on one type of product. But if you've designed a healthcare dashboard, a fintech app, and an internal logistics tool – shouldn't each project feel different?

Visual consistency across your portfolio site? Fine. Visual consistency across radically different problem spaces? That's just aesthetic performance.

"Add a CTA to 'Request Full Deck'"

Are you trying to help someone hire you, or are you trying to generate leads?

A good design portfolio makes people want to talk to you. You don't need to trick them into doing it. And if you're positioning yourself as a "freelance UX designer", that CTA is doing different work than you think.

This kind of advice turns real thinking into a formula. And formulaic portfolios attract formulaic work.

When I'm reviewing UX designer portfolios, I'm not counting accent colors. I'm looking for signals:

1. Can this person navigate ambiguity?

Most briefs are vague. Most stakeholders contradict each other. Most constraints emerge halfway through.

Show me how a messy brief became a usable product. Show me the trade-offs. Show me what you had to kill.

2. Do they understand the difference between decoration and design?

I've seen portfolios full of beautiful screens that solved nothing. Pretty UI doesn't fix broken strategy.

One paragraph about why you made a decision is worth ten polished mockups.

3. Can I tell what it's like to work with them?

Your portfolio is the longest job interview you'll have. If it reads like corporate marketing, I assume you sound like corporate marketing.

If your portfolio voice is "leveraging synergistic design thinking to create holistic user experiences," I'm out.

4. Have they ever been wrong?

The best portfolios include one project that didn't work. Not because failure is noble, but because learning is.

I want to know what you'd do differently now. That tells me more than five case studies where everything went perfectly.

5. Do they write in their own voice?

Most portfolios read like they were written by the same person. Bland, cautious, over-explaining.

Just say what you did. Use "I" not "we" (unless you genuinely mean "we"). Be human.

Mistake 1: The Feature Wall
"I led user research, wireframing, prototyping, usability testing, design systems, interaction design, visual design…"

That's a skill tag, not a story. Everyone does these things. What did you learn?

Mistake 2: The Perfect Case Study
Problem → Research → Design → Success → 40% increase in engagement.

Real projects aren't that clean. Real projects have politics, budget cuts, scope changes, and compromises. Show me the mess.

Mistake 3: The Mystery Process
"I conducted extensive user research and created personas."

Cool. What did you learn? What surprised you? What did stakeholders reject?

Process without insight is just busywork documentation.

Mistake 4: The Algorithm-Optimized About Section
"Passionate product designer focused on creating delightful user experiences through human-centered design thinking."

If this describes you, it also describes 47,000 other designers. Don't build your portfolio for algorithms. Build it for the specific human who might hire you.

Mistake 5: The Password-Protected Everything
I get it – NDA. But if I have to email you for a password before seeing any work, I'm moving to the next candidate.

Show something. Even if it's a side project, a redesign concept, or work you did in education.

The same impulse behind algorithm-optimised portfolios: AI checkers that claim 100% on anything.

Forget the LinkedIn formulas. Here's what actually helps:

Start with context, not decoration
Don't open with your design process diagram. Open with: "I designed [product] for [company] because [specific problem]."

One sentence. Then explain what made it hard.

Show trade-offs
"We wanted to add X feature but the engineering team said it would take 8 weeks. So we did Y instead, which wasn't ideal but shipped in 2 weeks. Here's what we learned."

This is more valuable than showing me your final Figma file.

Include something that failed
One project where you learned more from failure than success. Explain what you'd do differently now.

If you can't think of anything, you're either lying or you haven't shipped enough work yet.

Write like you're explaining it to a colleague
Not like you're pitching to executives. Not like you're defending a thesis.

Like you're walking someone through a messy project over coffee.

Make it easy to skim, but reward reading
Hierarchy matters. Someone should be able to scan your portfolio in 90 seconds and get the gist. But if they read deeper, they should learn something.

The algorithm problem extends beyond portfolios: what AI model collapse means for design tools.

Include:

• Context: company, challenge, constraints

• Your actual role (not inflated, not humble-bragging)

• What you did, how you approached it, what you learned

• Trade-offs and compromises

• One project that didn't work

• Screens or artifacts only where they help the story

• Your voice (seriously, just sound like yourself)

Cut:

• Generic process diagrams everyone uses

• "I'm passionate about..." statements

• Stock photos of diverse teams collaborating

• Every project you've ever touched

• The CTA to "Request Full Deck"

• Anything that sounds like it came from a corporate design RFP

Optional (but good):

• Side projects or experiments

• Writing samples (if you care about UX writing or content design)

• Tools or systems you've built for yourself

• Unfinished work with commentary

Design portfolios aren't sales funnels. They're not glossy brochures. They're not content marketing.

They're personal. Complex. Imperfect. Just like real product design work.

If you're stuck and tempted to follow LinkedIn's "5-step plan to portfolio perfection," step back and ask: What kind of work do I want to be doing – and does my portfolio show how I'd actually do it?

If it does, don't worry about the accent colors.

If you want to see design made for humans rather than algorithms, here is what my UI/UX design actually looks like.