An Interview with Victor Hsu
Published: February 24, 2014
Since 2003, Axure has been one of the go-to tools for prototyping designs—not only for UX designers, but for business analysts, product managers, and others, too. Victor Hsu and Martin Smith founded the company Axure in May 2002, in Berkeley, California. Before founding Axure, Victor was an electrical engineer turned software developer turned product manager, while Martin was an economist and self-taught hacker. Working together at an Internet startup and trying to build software from PowerPoint and Visio wireframes and Word specifications, they decided that there had to be a better way, and Axure was born.
As a longtime fan of Axure, I was delighted when Victor Hsu agreed to be interviewed for UXmatters.
Peter: Victor, thanks for your time today! How did you move from working in a startup to deciding to make solving the prototyping problem your obsession?
Victor: When we started Axure, we were focused on solving the communication problem that existed between product managers and developers. I was a project manager at the time and Martin was a developer, so we thought in terms of time, budget, features, and the frustrations of having to rework code. But what made prototyping an obsession was getting to know our customers. They didn't want just to ship software on time. They wanted to build applications that gave users an exceptional experience and delivered a ton of value to them. When we saw how prototyping could help them do that, we were hooked.
Peter: You must get a lot of feedback from Axure users. How do you manage that feedback and decide what new features to develop or features to refine?
Victor: We get a lot of direct feedback, and we also actively look for feedback that customers are sharing with each other. We enter just about every idea into an issue-tracking system, and it’s a regular part of our day to openly discuss new requests and get to an understanding of users’ core issues. After each release, the requests naturally revolve around certain areas, and those surface in the next release. We also like to organize requests by how issues currently make customers feel and how they could feel. Some categories have been offended, annoyed, and wowed. We like to go after features at the extreme ends of that spectrum.
Peter: Having a target audience of UX designers must make the user experience of Axure itself a pretty tough design challenge! What process do you go through to make sure that you’re happy with the design?
Victor: It is a challenge, but it’s also a blessing if you have a little humility and a really thick skin. The design of Axure is in constant iteration, and our customers are a huge part of that process. Internally, we have a very open design process. From early concepts through each design iteration, we’re constantly getting internal feedback, and no one’s pulling any punches. If a design makes it through our team, I can feel pretty confident that it’s a solid starting point. From there, we listen carefully to customers and keep iterating.
Peter: How is the Axure team organized?
Victor: All but one of our team is colocated in our San Diego office. We’re organized into four teams: development, marketing, customer support, and operations. Each team has a manager to make sure things are running smoothly, but otherwise, it’s a flat structure. We try to make sure everyone is doing what they’re best at and that everyone is doing something essential.
Peter: One of the most impressive things about Axure is the community that has developed around it. There are forums and AxureWorld, and some books have been written about Axure. How do you nurture and work with that community?
Victor: We participate in the discussions and try to help when customers ask questions. We try to bring some attention to the events, books, sites, and libraries that they share. But we try not to interfere too much. I think the best things that we can do are to continue making Axure better for the community and run a business that we can all be proud of. Beyond that, it’s really the community and its leaders who have created an environment that makes others want to contribute.
Peter: Your licensing model means that, recently, users have received free updates. What led you to that model?
Victor: Versions 6.5 and 7 were both free updates for all customers. Prior to that, there was an optional, yearly renewal fee that users paid to continue getting updates. The main reason for the change was that renewing a license was a terrible experience for customers. We sold a lot of renewals, but I don't think anyone enjoyed going through the process. It felt really good to make that change. It’s possible that a future update will have a charge, but hopefully, not until the process is much, much easier.
Peter: Axure users range from those using just the basic functionality to power users. Before the responsive features in the latest version became available, people were coming up with various hacks to get some measure of responsiveness into their prototypes. How do you balance usability against the needs of power users?
Victor: I like features to be consistent and flexible. Interactions are a good example of this. Once you learn to create a basic link, learning to respond to mouse movements is fairly simple because it’s consistent. This is great both for getting started and when you need to go beyond the basics only once in a while. At the same time, the event-driven model is flexible enough so power users can build things like the pre–version 7 “crazy flicker” hack for responsiveness.
This level of simplicity isn’t always possible, so advanced features are sometimes a little harder to get to, but hopefully they’re worth the extra click. If you take a look at adaptive views in version 7, there’s just a single button in the user interface to start adding views. It’s a huge feature under the covers, but if you don’t need it, it stays out of your way.
Peter: What challenges have you faced during the eleven years you’ve been working on Axure?
Victor: One of the tougher business decisions came after releasing version 1. After spending an unpaid year developing version 1 based on an embedded browser, we sold only a couple of handfuls of licenses. We almost called it quits, but decided to keep going and started over with a custom diagramming environment.
From the development side, the first Mac version was easily one of the toughest releases. We combined a C# application with Mono and a custom UI framework that we developed internally. There weren’t a lot of resources to learn from, and I think we did some things that just hadn’t been done before.
Because it’s fresh in my mind, I think version 7’s adaptive views stand out. We decided to address responsiveness fairly late in the project. We knew that adding the feature would dramatically extend the project timeline and that it would be difficult to get right. So far, it looks like it was the right decision.
Peter: Finally, we’ve seen a lot of changes in the last eleven years, as you’ve developed Axure to better address the challenges of changing Web technologies—most recently responsive design. What does the future hold for Axure? Could we see product teams using Axure to generate production code?
Victor: We’re going to continue building software to support UX professionals and others who are moving the software industry forward. There are both some small steps and some big steps that we think we can take to help move prototyping forward. We’re already seeing some production sites being built with Axure—in addition to axure.com—and this has a lot of room to grow.
Peter: Thank you for your time, Victor, and for all of the effort that you’ve put into Axure over the past eleven years!