In 2015, I sat in a packed ballroom at SXSW for a talk called "New Users Matter Too! Designing Better Onboarding" by Krystal Higgins.
I showed up expecting some light UX inspiration - what I got was a total reframe of how I think about product design.
The Icebreaker That Wasn't
Krystal kicked things off by asking us to turn to the person next to us and greet them the way our apps greet new users.
No one moved. No one spoke. We all just sort of looked at each other.
Then she said:
"Most of you should have just turned and said 'App name. Login.'"
Brutal. And 100% true.
It hit me:
- How much value do we hide behind a registration wall?
- What could we let people do before they create an account?
- How might we let them see the magic first, and only ask for commitment once they want to stick around?
That's not just onboarding. That's generosity. That's how you build trust.
First-time UX isn't extra. It's essential.
Every time I work on a product now, I ask:
- What's the very first thing the user sees, and how does it make them feel?
- How do we introduce just enough - without overwhelming them?
- When should we guide them, and when should we get out of their way?
- If they come back after a week or a month, what will they find?
I've carried these questions into every role since:
- At GatsbyJS, improving CLI and UI entry points so devs could get to "first success" faster.
- At Kaiser Permanente, untangling patient-facing workflows so people could book appointments or find care without getting lost.
- At startups and nonprofits, where trust has to be earned in minutes, not months.
Here's what I know for sure:
- You can't build loyalty without clarity.
- People don't churn because they're lazy - they churn because they're lost.
- New users aren't a funnel metric. They're people taking a risk on you.
- Onboarding isn't a tour. It's a trust-building exercise.
Krystal Higgins, if you ever read this - thank you. Your 2015 talk was a product awakening for me, and your work still shapes how I coach, design, write, and lead.
If you're building something right now, ask yourself:
"What would this feel like if I'd never seen it before?"
That's where the real work begins.
Frequently asked questions
Related work
Related services
Frameworks for this topic
Read next
Most teams argue about research as a binary: do it or skip it. Both extremes lose. Months of upfront research goes stale before build starts, and no research turns designers into pixel machines. I spent four years at Pivotal coaching the narrow middle: just enough research, at the right moments, to go fast forever.
We know consent has to be freely given, reversible, informed, enthusiastic, and specific when it's about our bodies. Our data is a body too. Here's what happens when you run your product's data handling through the same five tests, and why a cross-tenant leak is a consent failure, not just a security bug.
Most teams aren't balanced. One discipline decides and the others execute. I spent four years at Pivotal coaching the alternative, and the difference between the two isn't org charts. It's who has voice, who has a vote, and whether the team can celebrate a failed experiment without flinching.
Want to work together?
I help teams ship better products. Let's talk about your situation.
Get in touch