Workshop Overview

In this workshop, Matt spoke to our members about what really goes into UX design, beyond the usual sticky notes and user personas. He shared practical truths about working in UX - from how to be persuasive to what actually matters when applying for jobs.

Key Insights

1. Being persuasive is more important than being right

Design isn't a debate to win—it's a solution to sell. If the best idea doesn't ship, it wasn't persuasive enough.

How to be a persuasive designer:

• Design in high fidelity
• Build a prototype
• Let the decision maker play with it

(Arnell Group famously did this and landed a $1 million project.)

2. Execution matters more than strategy

Plans and diagrams don't make impact—released work does. Slide decks won't improve a product unless they become something real.

Less of this:

"Here's our UX strategy deck."

More of this:

"Here's the prototype in your hand."

3. Impact matters more than intention

Designing for impact sometimes means skipping the sticky notes.

Adam Fisher-Cox redesigned bad signage at JFK on a whim. No formal research, no brief. Just a designer solving a problem he noticed in the wild. He posted it online. The people in charge saw it. They hired him.

Solve the problem you see. Sometimes that's enough.

4. Hiring managers will only spend 30 seconds with your portfolio

And that's being generous.

Tips:

• Don't password-protect your website
• Put your work on the homepage
• Make it obvious what sets you apart

Include 3 case studies:

1. Show depth
2. Show breadth
3. Add a third so it doesn't feel incomplete (no one will read it)

5. You should be really, really good at Figma

The tools matter. Whether it's Figma or whatever's next, mastery of your design tool makes the difference between ideas and persuasion.

“When artists get together they talk about where you can buy cheap turpentine.” – Picasso

6. It isn't real until it's code

Design stops being theoretical once it's implemented.

• Get technical constraints up front
• Learn how to vibe code
• Use code-adjacent tools

The closer you are to the medium, the better your work will hold up.

7. Most user research is unnecessary

Research is often idealized, but if it never informs what ships, it's not useful.

37signals once reduced chargebacks by 30% by changing their billing descriptor. No AB test, no survey—just a domain name and a redirect.

A good idea and a bias toward action often beat a long research cycle.