Jon Bell

As part of an ongoing series of interviews with designers, I’ve talked to Jon Bell about his design process. Jon Bell worked as a developer for ten years before shifting his focus towards design. He currently works as an interaction designer for frog design. He writes a blog about design called design dare. You can find out more about him at

Lukas: Tell me about your design process. Where do you start, and what kinds of techniques do you use at which stage of the process?

Jon: I’ve wanted to be a professional interaction designer since attending art school in the 90’s, but I kept getting hired as a developer. I did a lot of freelance design on the side, and was always the front-end specialist on my dev teams, but it’s taken until recently to have the word "designer" on my business card. (At frog, no less! Hurray!)

Jon Bell
Picture (CC) Randy Stewart

As a result, I’m not sure what my process is yet. I’m absorbing all the great ideas around me, watching my co-workers carefully, trying to stay healthy and positive, and letting my designs organically develop through whatever approach intuitively feels right at the moment. It’s messy but I’m having a lot of fun.

When you start on a new project, what exactly do you do? I imagine that you often have a pretty good idea of who the intended audience is when you’re working for frog, so do you start with user research?

Frog has a great three-part approach: Discover, Design, and Deliver. In the discovery phase, we do heavy design research. Lately I’ve been doing a lot of contextual inquiry, which means going to a customer’s office (or home, or whatever you’re designing for) and getting a sense of what they do every day. I find it's an exercise in studied ignorance as I try really hard not to bias the data with my own opinions.

Once we get our data, synthesis is a lot of fun. I like to post pictures of everyone we talked to on the walls (my team has interviewed 67 people in 2010 alone!), with sticky notes that highlight significant themes. Then it's all about tons of discussion and sketching, with a blend of alone time and group brainstorms, all under the gaze of our client's customer photos to keep us focused.

We make near-comical use of sticky notes. They're everywhere, grouped into topics, stuck to our co-worker while he's practicing a presentation, stuck to the image of a customer who had a particularly resonant quote, and so on.

The backdrop for all this customer-specific data and resulting discussion is the rich domain expertise that frog has. We often see emails go by that say things like "Does anyone have experience with living in Africa?" or "Does anyone have a Palm pre I can try out?" Combine our research skills with the combined experience of all frogs and it's hard to go wrong.

So that’s the discovery phase. I guess once this starts to settle down, you move towards sketching, storyboards, and maybe prototyping? How do you do that, and how do you use your research to inform your designs?

At the end of discovery, we present our findings and our recommendations to the client, and that leads into my favorite part of the process.

We have a bunch of activities designed to rank ideas, suss out details frog might not know about, be as collaborative as possible. It's not just about the official sessions, either. It's about lunch, dinner, going to the bar afterwards, trading war stories, adding each other on Facebook, debating the end of Up In the Air, and so on.

It's an intense time of collaboration, and by the end of it, everyone usually respects and understands everyone else's point of view. This part is crucial, because we're never going to agree on everything, but if there's a framework of respect, we can get through any issue.

(In freelance, I approach things the same, but at a much smaller scale :)

Then we return home and try threading the necklace: we have a handful of recommendations that everyone is in agreement on, so we just need to thread them into a viable product and story using flows, storyboards, and wireframes.

There’s some debate about what’s best at this point - do you go for the "rabbit out of the hat" moment, by not showing the client anything until the next presentation, or do you show incremental progress which reduces risk but also takes some of the emotional punch out of the final presentation and opens you up to last minute changes?

Jon Bell
Picture (CC) Ronald Woan

I like to stay in close contact with the client. I like thinking that we’ve all heard the same things in the contextual inquiries, we all coalesced around the important features together in during the deep collaboration, and as the fidelity goes from napkin sketches to grow into whatever their final form will take, the client always feels involved and appreciated throughout the process.

Once the fidelity of your design starts to go up, do you test with actual users?

Not enough. When a client is nearing the end of the process, I think they’re just itchy to get the work out the door. (my current client is interested, though, so kudos to them!) I still try to scrape together as much data as I can, but its never as systematic or solid as the early user information gathering.

I’ve used Silverback to great effect, but their motto, "guerrilla usability testing" really sums it up. It’s been surprisingly hard for me to justify user testing later in the process, despite how important I think it is. This applies to frog work, freelance jobs, and even my own projects, where I get to play the role of "guy who thinks late-stage testing is too much of a hassle". That guy sucks.

It sounds like you’re trying out a lot of different techniques and approaches in your design process. Are there things you’ve tried that didn’t work out, or didn’t produce the results you expected?

For a lot of things, especially since I haven’t done this professionally for long, it feels more like steering a car. As I go off track, (and it happens a lot) I just adjust slightly for the next time. But two things jump out as obvious mistakes. They’re less tactical and more strategy blunders.

First, I need to stay motivated, which means setting small goals that I can achieve, which inspires me to tackle the next thing. I’m involved with a few projects right now (two are a personal projects, the other is part of a group) that feel stalled because they don’t have a modest enough goal for version 1. So it’s hard to keep them going, and I hate feeling like a great project is dying for no other reason than over-reach. Making everything your opus is a stupid but common mistake.

Second, being grumpy, unhealthy, and stubborn worked pretty well for me as a developer, but I can’t do it with creative design work. This means I have to eat better, exercise, go to bed really early, constantly remind myself how great things are, and expose myself to designs, people, and points of view I might not have bothered with before.

When I don’t, it shows much faster with much worse results than when I used to stay up all night banging out code and/or yelling at jerks on the internet. It took me a whole career change to fully appreciate it, but a positive attitude is vital for creating great work.

Words we should all live by. Well, thanks for talking to me, Jon!

To find out more about Jon Bell, go to or, or feel free to email him about anything you'd like more details about.

If you require a short url to link to this article, please use

designed for use cover

But wait, there's more!

Want to read more like this? Buy my book's second edition! Designed for Use: Create Usable Interfaces for Applications and the Web is now available DRM-free directly from The Pragmatic Programmers. Or you can get it on Amazon, where it's also available in Chinese and Japanese.