Updated March 2026: Originally written in March 2023. I have updated the tools section and added notes about WCAG 2.2 and the INP metric change. The core arguments haven't changed.
I have been building websites professionally since 2011. For most of that time, "UX" wasn't a word anyone at the companies I worked for actually used. We called it "making sure people can find the damn button." The vocabulary was different, but the problem was always the same: if people can't figure out your website in the first few seconds, they leave.
The difference between a website that works and one that frustrates people is almost never about how it looks. It's about how it behaves.
What UX actually means (and what it doesn't)
UX design is not visual design. They overlap, but they're different disciplines. Visual design is about color, typography, spacing, imagery. UX design is about whether someone can complete the task they came to your site to do.
A beautiful website with confusing navigation has bad UX. An ugly website where you can book an appointment in three clicks has decent UX. Obviously you want both, but if I had to pick one, I'd pick the one that works.
I learned this the hard way at a hospital I worked for early in my career. We had a nicely designed website, but patients couldn't find the appointment booking page. It was buried three levels deep in a dropdown menu, behind a link labeled "Services." We moved it to a prominent button on the homepage and appointment bookings through the website went up noticeably within weeks. No redesign. No new features. Just moving a link.
Why it matters for your business
Here's where I'm going to be blunt. You can ignore UX if you have no competition. If you're the only hospital in town, people will endure your terrible website because they have no choice. But the moment there's an alternative, the one with the better experience wins.
Some things I've seen firsthand:
- Bounce rate drops when you simplify. On one project, we reduced a landing page from 5 sections to 2 and a clear CTA. Bounce rate dropped by about 15%. People weren't scrolling through the original page, they were just leaving.
- Form completions increase when you reduce fields. A contact form with 8 fields performs worse than one with 4. Every additional field is a decision point where someone might think "this isn't worth it."
- Page load time is UX. A page that takes 4 seconds to load has already delivered a bad experience before anyone sees a single pixel of your design. At Reliance, we treated performance as a UX issue, not just a technical one. Improving load times by even a second had measurable impact on engagement.
The tools I actually use
There are hundreds of UX tools. Here are the ones that have stuck in my daily workflow:
Research and analytics:
- Google Analytics for traffic patterns and user flows. Where do people come from? Where do they drop off? The Behavior Flow report answers questions that guessing never will.
- Hotjar for heatmaps and session recordings. Watching a real user struggle with your navigation is uncomfortable but incredibly informative.
Design and prototyping:
- Figma has basically won this category. I use it for wireframes, prototypes, and design handoffs. The collaborative editing alone makes it worth it.
- [2026 update] Figma has added Dev Mode and variables since this was first written. Dev Mode gives developers a better view of design specs, and variables enable design tokens that map directly to code. These have improved the designer-to-developer handoff significantly.
Testing:
- Chrome DevTools Lighthouse for performance and accessibility audits. Run it on every page before launch. Non-negotiable.
- Manual testing on real devices. I keep a couple of old phones around specifically for this. Emulators miss things. Real devices don't lie.
Accessibility is not optional
This is the part that gets skipped most often, and it shouldn't be. Accessibility (a11y) is UX for everyone, including people using screen readers, keyboard navigation, or high-contrast modes.
Some basics that should be in every project:
- Semantic HTML. Use
<nav>,<main>,<article>,<button>for what they're meant for. Don't use<div>with an onclick handler as a button. Screen readers can't parse your intent. - Color contrast ratios. WCAG requires at least 4.5:1 for normal text. Use a contrast checker. What looks "subtle and elegant" to you might be invisible to someone with low vision.
- Keyboard navigation. Can you tab through your entire site and use it without a mouse? If not, fix that.
- Alt text on images. Not "image1.jpg." Describe what the image shows and why it's there.
[2026 update] WCAG 2.2 became a W3C Recommendation in October 2023, adding new success criteria around focus appearance, dragging movements, and target sizes. If you were already following WCAG 2.1, the jump isn't huge, but the new focus appearance requirements (2.4.11 and 2.4.12) are worth reviewing. They formalize what good focus styles should look like, which is something a lot of sites still get wrong.
I have been treating accessibility as a non-negotiable on every project for years. It's not extra work. It's part of the work. And frankly, a site with good accessibility usually has better UX for everyone, not just people with disabilities.
Common UX mistakes I keep seeing
After reviewing dozens of websites (both my own older work and others'), these are the patterns that keep coming up:
- Autoplay videos. Nobody wants this. Nobody has ever wanted this.
- Pop-ups on page load. The user hasn't even read your headline yet and you're asking them to subscribe to a newsletter. Wait until they've shown some engagement.
- Infinite scroll with no way back. If I scroll down 50 items and accidentally tap "back," don't send me to the top. Give me pagination or at least persistent scroll position.
- Mystery navigation icons. A hamburger menu is fine. But if your only navigation is three dots, a gear icon, and a bell, you've replaced words with puzzles.
- Ignoring empty states. What does your dashboard look like when there's no data? If the answer is a blank screen, that's a UX failure. Tell people what to do next.
Responsive design is UX
I wrote a separate post on responsive design, but it's worth repeating here: if your site doesn't work on mobile, your UX is broken for the majority of your users. Mobile isn't a secondary experience. For most sites, it's the primary one.
[2026 update] The replacement of FID with INP (Interaction to Next Paint) as a Core Web Vital in March 2024 reinforces this. INP measures how responsive your site feels across all interactions, not just the first one. Laggy menus, slow form inputs, delayed button feedback: these all count now. Google is explicitly measuring the user experience of interaction, not just loading.
What I'd tell someone building their first website
Don't start with how it looks. Start with what it does. Write down the top three things a visitor needs to accomplish on your site. Then make those three things as easy as possible. Everything else is secondary.
Talk to real people. Show them your site and shut up. Don't explain where things are. Don't say "you just click here." Watch what they do. The gap between what you think is obvious and what actually is obvious will humble you every time.
And remember: every person who leaves your site confused is a person who could have become a customer, a reader, or a fan. UX isn't decoration. It's the reason people stay or leave.