From SharePoint to Yammer: Designing in Redmond vs. Designing in San Francisco

Yong Rhee

I may not be the designer on the Yammer product team with the longest tenure here (that honor belongs to Jeffrey Erickson), but I’m definitely the one who’s been at Microsoft the longest. I started collaborating with my Yammer counterparts immediately following the acquisition last year, and was so attracted to the team and how they worked that I decided to say a hard farewell to the team I’d been with for over five years and join Yammer down in San Francisco. It’s probably one of the best decisions I’ve ever made, and not only because I left the Seattle rain. At Yammer, I’ve learned some completely new ways to design product. Like what, you ask?

Build. Measure. Learn.

Over my five years in SharePoint and Office, I shipped twice. Over the last year at Yammer, I’ve already shipped around eight projects.

There are definitely product teams at Microsoft that have been shipping faster in recent years, but I worked on a team where we shipped major releases every two to three years. This meant we’d spend a lot of time planning for the release, iterating designs over milestones internally and with beta users later on, and then shipping a well-packaged product to everyone around the world. Although this approach and experience taught me many things—like coming up with good design guidelines and shipping a very polished design—it didn’t give me much learning on how our actual customers were using the product based on the designs I worked on. We’d get some feedback from the press, and from the customers through various conferences and through user research, but it took another two to three years for us to react to them.

At Yammer, I’ve spent less time on polishing and finessing the build to make my design sexy, and more on learning from the analytics data and iterating on the designs to ship a better UX within a couple of weeks, if not days. This sometimes hurts my designer passion to create pixel-perfect delightful experiences, but the trade-off of experimenting, iterating, then shipping designs faster has me convinced so far.

Data, data, data.

When dealing with real estate, the saying goes it’s all about location, location, location. When it comes to learning what designs to build and ship at Yammer, it’s all about data, data, data.

On my previous teams in Redmond, we based most of our design decisions on designer intuition, backed up by user research and market research. This is definitely better than only relying on the design team’s instincts, but at the same time it felt too dependent on opinions and anecdotes. Market research can be too broad, user research is often brought in only after design is already underway, dogfooding internally can be biased: all these add up to doing a lot of research, but potentially missing the way real users behave.

At Yammer, we have the extra resource of an ever-so-helpful analytics team. They provide us with data and analysis of how our users use our product and where our designs are succeeding and failing. Before, shipping a design could come down to how well you presented and argued your case. Now it’s not about which design can be better sold to the rest of the team, but about what we’ve learned from testing a design and finding out how people used our product to inform which design is actually more successful in the wild.

No More PowerPoint.

Well, maybe not no more. But definitely way less.

In Redmond, I used to give at least a couple of presentations per week walking the audience through the goals, scope, scenarios, iterations, and proposal or latest plan of my designs. This was super helpful for me to thoroughly think through and document my design process and get everyone on the same page and even let my designs be distributed to a broader audience. However, this meant a lot of time was put into getting those presentations together and making sure they covered all the content for any audience to quickly understand the design problem.

In the past year at Yammer, I can’t remember a time in which I put together a PowerPoint presentation to go over all the aspects of my design process to sell someone on an idea. It all happens online through Yammer and in real life through lighter-weight meetings. My time is spent less on putting together a deck so that others can take it and read it on their laptops, and more on going through my design process and producing designs to ship in the upcoming couple of weeks or days.

Over the history of product and software development, there have been many different schools of thought on what the best methodology and process is. I’ve experienced two very different ways of building products by being in two different teams in two different locations in the same company. I’m certain that throughout my career, I’ll be experiencing many more ways of building great products and many great teams of people to do that with. Now, back to designing…

Yong Rhee is a UX Designer at Yammer. He came to San Francisco by way of Seattle, Virginia, London, and South Korea. Blackjack is his game.