Tuesday, March 15, 2011

Looking back and moving forward. The things I learned while working for Opera.

For the past four and a half years, Opera Software has been at the center of my professional life. It's also been nearly a year since I stopped working for Opera full-time. In April 2010, I shifted from a full-time role at Opera to a part-time role (as an independent contractor/consultant), focusing on Opera's once-a-month State of the Mobile Web report.

Over the last year, I've also been pursuing personal projects, helping my wife launch a successful style blog, celebrating milestones like my daughter's first birthday and my son going to kindergarten, and getting fitter than I've ever been (which will be the topic of my next post).

I've been plenty busy and moving forward. Eventually, my contract work with Opera will come to an end, and I'll be sad to leave, but I'm proud of what I achieved with the company and thankful for the lessons I learned. In this post, I thought I would reflect a bit upon my years at Opera and discuss what I learned.

My first two years at Opera

Some of you already know the story of how I got my job at Opera. Long story short, I was supposed to be working on my doctoral dissertation (on otaku culture in America), but I found myself procrastinating instead, thinking heavily about how certain browsers became popular while others did not (independent of technological superiority on one side or another).

To address such questions, I wrote an article in early 2005--Better branding for Opera—and shared it with the My Opera user community. Shortly thereafter, I was contacted by Opera, and when I completed my degree in late 2006, I started to work for Opera (in San Diego, CA).

I started my professional journey with Opera as a member of the Desktop Marketing team. I joined the company to help Opera understand its users better--especially in the United States--in order to improve how we communicated with them. As part of that task, I had the wonderful privilege of interacting with a lot of Opera users (and non-users) both online and in person. I got to meet people at conferences such as SXSW and numerous Barcamps, helped to create Opera's outreach program (Choose Opera), and wrote articles for My Opera, Opera Watch, Opera Labs, and other Opera-related blogs.

Learning what people cared about regarding the Web and the browsers they used was a lot of fun, and I really loved being a bridge between Opera (the company) and its vibrant user community. Even though I came from an academic background, I learned a lot about internet marketing and online community building. In 2008, however, I moved from Marketing to Consumer Product Management and Developer Relations (later the Products group) at Opera, and it's there that I really stepped up my research efforts.

Product and user research at Opera

As I mentioned above, I didn't have a traditional marketing, business, HCI, or UX background. I was a social scientist, fresh from the world of academia where I used ethnography as the main tool to understand social phenomena (e.g. media fandom's use of technology). As part of the Marketing team, my initial concern at Opera was how to improve communication between the company and the users (and potential users), with the secondary task of compiling and making sense of user feedback.

In 2008 however, user feedback became my primary concern when I moved onto Opera's Product Management team. Suddenly, it became my job to understand how people use and respond to the Opera browser itself (and not just the marketing). Based on my research, I made recommendations to the designers and engineers working on the product.

This was both exciting and scary. I now had some say (albeit small) regarding the direction of a software product used by millions of people all over the world, and I had to make sure that everything I recommended was based on factual evidence and stayed in line with the ideals of the company, not just my own biases regarding what a browser should be.

What I learned

Here are some of the guiding principles I followed and some of the lessons I learned in my role as a user researcher for Opera:

1. For a product with so many diverse users, making the right decision regarding who to study is absolutely critical.

For many years, before I started working for the company, I was a die-hard fan of Opera--interacting regularly with other Opera fans and even Opera employees on the My Opera community. Like my fellow Opera fans, I had no shortage of things that I wished Opera could do, or things that I thought had to be changed. When I started to study Opera's user base (as a researcher), my first instinct was to look at the fan community I belonged to, but I also knew that I had to look much farther beyond that. After all, there is a larger world of Opera users who are not part of (or represented by) the fan community on My Opera, and an even larger population of non-Opera users who are completely different from the core community of Opera fans.

2. Browsers are used in many different ways by all kinds of people.

It's important not to overgeneralize users. When researching users of a multi-faceted product like a web browser, it pays to be particular and precise. Specifics count.

To be most effective, I needed to study users (and their behaviors ) across many different contexts. For example, I had to carefully decide who I would interview among all the different groups of people using the browser, and come up with recommendations that best served as many users and potential users as possible.

Even when conducting survey research, I had to make sure that the results I compiled were organized properly and representative of the groups I was interested in (existing Opera users, new Opera users, users in different countries, non-users, users who were uninstalling the browser, mobile users, etc).

3. It's not a good idea to assume that users, even the technical ones, are very knowledgeable about your product.

It's an understatement to say that Opera is "full-featured" (right out of the box, without having to download any extra extensions, widgets, etc.). For the most part, even "power users" are not completely well-versed in everything Opera can do. A person who knows how to use one advanced feature in Opera might be completely clueless regarding a different feature, even one that is considered "basic" by most people. In other words, almost all Opera users are like novices in one way or another. Even most "browser geeks" (like myself) are normal, everyday users most of the time, so it's not a bad idea to treat everyone as novices regarding most things in the browser.

4. Helping people learn about your product is difficult but worthwhile.

When I first started using Opera, I liked the fact that it held all sorts of secrets for geeks like myself to uncover and share with everyone. Eventually, however, I started to care more about the browser as something that could immediately help more people overall, not just hardcore browser tinkerers.

It's not that I wanted to see the browser simplified to the point that it contained nothing interesting to be discovered. Instead, I thought it was important that we should help all people who use Opera to learn about the things that make it unique. Instead of Opera being a browser strictly for "power users", I've come to believe that Opera should help out anyone who wants to become more effective at browsing the Web. In my mind, Opera should be the program for inquisitive (but not necessarily techie) people who enjoy exploring their technology and what it can do for them, beyond what they're used to.

In that sense, Opera might be considered "the browser that makes you smarter". In my opinion, Opera is the one browser that has the biggest potential to show everyday people (who aren't already browser geeks) all the amazing things you can do with a browser that will help you save time, organize your data, search for information, and communicate with others.

5. Mechanisms to educate people about your product should be subtle, elegant, non-intrusive, relevant, and fun.

There are many ways to help users understand your product, but not all of them are equally effective:

  • You can rely solely on documentation provided to users (e.g. help files and tutorials), but many people don't like looking for or reading directions, and there's an unfortunate temptation to blame users for not reading the manual (or not searching forums and FAQs for previously answered questions).

  • You can have things like the notorious Clippy, but that's the example everyone brings up when discussing intrusive help measures.

  • You can show "tips of the day" within your software, but there's no guarantee that any given tip will be relevant to the user who's reading it at the moment.

On the other hand, many agree that some of the best and most empowering products make sure that:

  • users understand the consequences of their actions (e.g. using clear visual cues to show cause and effect).

  • users are presented with information in a contextual way, such that relevant information unfolds in front of them based upon their actions (as opposed to appearing arbitrarily).

  • users are not afraid of (or otherwise prevented from) exploring the product. Trying out the capabilities of something you own, whether a car or a browser, should be fun and exciting. That's only possible if users can play around with the product without fear of accidentally breaking something that they don't know how to fix.

Parting gifts and fond farewells

Although I haven't been part of the team working on Opera Desktop for over a year now, Opera 11 came out in December (and Opera 11.10 beta debuts at SXSW this week). I've been thrilled to see that some of the research I did made an impact on Opera 11 (and hopefully future versions of the browser).

Specifically, I was involved in conducting research that informed Opera's decision to implement things like:
a) third-party extensions

b) a new way to manage and organize open tabs. (In my opinion, the specific implementation of that--which I had nothing to do with--is brilliant!)

c) visual mouse gestures (see video below). Reflecting upon the principles discussed in point #5 above, I daydreamed the idea and wrote up the initial set of requirements for visual mouse gestures while sitting all alone in Opera's now-closed San Diego office (US operations are now based in San Mateo). As such, I feel a little proud and somewhat nostalgic every time I use that feature.

With a renewed focus on usability and the overall user experience, along with a continued emphasis on pushing the envelope with cutting edge features, I think that Opera is evolving nicely. Before I was an Opera employee, I started as a fan of the browser, and when I finally hand in my keycard, I'll be leaving as an even bigger fan. If all goes well, I'll be using Opera until there's no more Web to browse.

I hope you've found this retrospective interesting, or maybe even useful. When I was in grad school--studying anime fans to earn my degree in Science and Technology Studies--I never imagined I'd have the privilege of working in such an exciting industry.

Post-Opera, I look forward to taking on fresh challenges and exploring new opportunities (either consulting or full-time), so if your company or small business is in need of user research insight, please don't hesitate to contact me. My CV can be found here (and my resume is available upon request): http://www.cjas.org/~leng/cv.htm
Finally, I just want to thank all the amazing members of the My Opera community and other Opera fans I've had the chance to meet and interact with over the years. Together, we were always part of something bigger than ourselves, and I hope to see you again soon!

PS: My heart goes out to my friends and colleagues in Japan, and anyone else affected by the tragedy.