Crafting Delight

I recently heard Tobi Lütke, the CEO and co-founder of Shopify, say that "experience minus expectations, if that's a positive number, that is the amount of delight caused."

We all desire for the people using our apps to be delighted. We want them to have 'wow' moments. Moments where using our software puts a smile on their face. Moments where they realize their job just got a little easier. Moments of magic.

Often, we think the path to this is features, and we're not wrong to think this. I remember the first time I used Roam Research — the app that popularized back-linking in note taking. From the first backlink, I realized that I finally found a note taking app that works how I've always wanted. I started bringing together ideas from every book I read by topic, by author, by cross reference of other books. I'm not sure how I would've written papers in college without Roam. The feature of back-linking brought delight.

And yet I do know how I'd do college without Roam as I switched to Tana halfway through. They copied the same features, added a bit more, and improved on how the app looked and felt. So I switched. But even then, since I've graduated I've found myself leaving these back-linking apps all together. Every sermon, blog, or business idea I've written since has been in the simpler, cleaner Notion. Why? Because features are not a moat. Features matter, but they're the easiest thing to copy and to come up with.

The key to delight it not features, it's craft. It's design that keeps people in a state of flow. In fact, I'm most delighted by the apps that get out of the way, make me forget I'm using an app at all, and let me get my work done.

But workplace apps are hard. Most people didn't sign up for their role with the expectation that 90% of it would take place on a screen. So building tools people use at work — from issue trackers to church management software — is an uphill battle. People come in with loads of suspicion (or at least incredibly high expectations), so if your app isn't well designed in both function and form, people will loath it. The app will create more work, not less. And the work it creates won't be fulfilling work, it'll just be pointless waste.

So how do we make workplace apps that are loved like Linear rather than loathed like Jira?

01. FOLLOW COMMON UI/UX PATTERNS

Every app — from Facebook to Figma, from Spotify to Stripe — uses the same UI/UX pattern for navigation: a sidebar. Is this because a sidebar is always the best UX to get around? It is because it looks the nicest? I'm not sure. But what I do know is that most users know how to use it which often makes it the right choice for the job. If you want to maximize delight, keep the important things familiar.

While a designer might be proud of coming up with something different and unique, users just feel a higher level of friction and cognitive load.

A few years ago, I was driving a rental car to pick someone up from the airport. Once we got the car loaded, I shifted into reverse, put my foot on the gas, and to my horror started going forward, nearly hitting the parked car in front of me. Did this happen because I'm a klutz? No, it happened because some designer thought that that it would be a wonderful idea to put the reverse icon under the park icon but to get in reverse from park I have to move the gear shifter up. It's not just me; there's nearly 100k views on a video showing you how to use the gear shifter in this one car. Mitsubishi (and every other modern car maker these days) reinvented something that worked perfectly fine, and now you have to relearn how to drive in every car you get in.

That said, not every app in the sidebar screenshot above is widely considered delightful. Linear, Notion, Stripe, and Open AI are brands that are constantly praised for their design. Meanwhile Spotify is regularly mocked in design communities for its bloat, Facebook has four navigation paradigms on its main page, and Mailchimp makes it hard to get to basic settings.

Following common patterns does not guarantee delight, but not following it will always increase the frustration in experience.

02. FOCUS ON THE INVISIBLE

Linear is the master of the invisible. From a screenshot, it doesn't look far from every other app out there: there's a sidebar, a table, and some breadcrumbs. But then you use it. Immediately you're blown away by the speed. Everything loads instantly, you can navigate solely with your keyboard, and you can edit anything inline (no clunky modals or edit states). On top of that, nothing in the UI re-renders when you navigate, you never accidentally un-select items, and every option you need shows up in the UI right when you need it but doesn't clutter it up before.

Everything feels thoughtful, and yet it's almost entirely unseen. Rather, it's felt.

At the end of your first day in Linear, you can't quite say why, but you have this feeling of delight. You didn't hit any pain points, nothing worked counter to how you expected, and everything just made sense. It just works.

Vercel design engineer Rauno Freiberg recently released a course called Devouring Details. My biggest takeaway from it is the idea of "inferring intent." It's where you know what the user is trying to do and making that easier without them even asking. It's what makes software feel like magic. You see this in Vercel when you add environment variables. Instead of having to paste the name and then paste the value and then repeat for each variable, you can just paste your full .env file into the first input and it's automatically parsed.

One of Linear's best examples of this is how sub-menus work in their contextual menus. When they realized that the sub-menu would close as a user tried to move their mouse to it, they started tracking your mouse to produce a safe area making sure you didn't have to perfectly position your mouse just to update a label.

When you infer intent, when you focus on the little details, you create software magic. You show people that you care. And while they may never notice, they will feel it.

03. CARE ABOUT HOW IT LOOKS

Obviously the way an application looks matters. Half the reason I switched from Tana to Notion was a desire for the the cleanest interface possible. I wanted something that focused on beauty and used progressive disclosure for anything that typically clutters lesser interfaces.

Something as simple as the colors you use can completely change the feel of the app. The right sans serif can set you apart. Choosing to fade away elements while a person is typing and the mouse is not moving creates focus. Making sure text is readable in both light and dark mode matters. Aligned icons feel right even if no one ever notices.

Every visual detail matters, and it's our job as designers to care more than anyone else ever will.

04. KNOW WHO YOU'RE BUILDING FOR (AND WHAT YOU'RE BUILDING)

Notion doesn't have common text editor features immediately visible. As I type this, I don't see an option to underline, highlight, or strikethrough my text. I can't change the font size, the font style, or the font weight. And yet, in Figma, every possible option is in my face. Why? Despite both apps being able to format text, the former is made for capturing and working on ideas while the latter is made for design. In Figma, how the text looks is the whole point, so having the formatting features readily available is the expectation while in Notion it would be clutter.

In the same way, Facebook — a social app — makes it easy to get to my profile whereas Linear — an issue tracker — make my avatar a tiny icon in the corner.

What you're building determines if something is clutter or required.

05. GIVE USERS THE CHOICE (BUT HAVE OPINIONS)

Many companies don't provide their users with "advanced" settings because they believe most people won't use them, and yet everyone I know has tweaked the size of their lock screen clock in iOS 26. Everyone wants to customize and tweak software, especially the software that they're stuck in all day.

When we write off these features as being "advanced," we say less about the users and more about ourselves. Often what makes a setting feel too complicated for an average user is designers not being thoughtful enough in where they place the option and how they explain it.

That said, there's something to balance here. You want to give users the option to make your app their own, making room for the Ikea effect. But this is not an excuse for laziness. Good, strong opinions from designers still win over a million options. The art is in knowing which choices you give users and which ones need to be made at a product level.

05. AND YEAH, BUILD SOME GREAT FEATURES

Of course features matter and bring delight, but even then, the best ones, the ones that feel like magic, don't draw attention to themselves. Notion's AI search has changed how I do note taking. I no longer have to tag and organize everything. I simply ask. Yet when I do, I'm never thinking about how great of a chat interface it is, I'm just glad that I could find what I was looking for while staying in flow.

Features matter, but it's the experience of the features that matters more.

CONCLUSION

An average user can't articulate any of this, but they feel it all implicitly. They don't know why Facebook feels bloated it, they just know it does. They don't know why Notion feels polished, but that's how they'll describe it to everyone.

People feel the smallest decisions we make in our software. Every detail — visible and invisible — matters. It shows that the software was made by people who care about their users and care about the work they do. It shows it was made by people who want to make something wonderful, often just for the sake of putting something great out into the world.

So make software that makes people feel like you read their mind. Make software that makes people smile. Pay attention to the beauty of how something looks, feels, and works.

Ultimately, make software that makes people forget they're even using it, leaving them only with delight.