So you want to write design principles

Originally posted on Medium.

You want to write some design principles? Cool.

You’ve never done this before? That’s ok. Going from zero to principles can be a little daunting, but you’ll be fine. You’ve got this. I’ll help.

Start with why.

What are these design principles for?

Maybe you’re trying to establish a guiding vision. Maybe you’re trying to get your team aligned. Maybe you just need to bring clarity to a confusing space. Figure out the point you’re trying to make, and back it up in your principles.

Is your audience internal or external? Do they have an established point of view? How technical are they? How much design experience do they have? Tailor your language to your audience. If everyone is coming from the same background as you, build on top of that shared knowledge. If not, make sure to explain any fundamentals that won’t be obvious.

It’s not about you.

Although you’ve been tasked with writing design principles, you aren’t responsible for generating them by yourself. Your job is to record insights, not to author them.

Build a team.

Start with the people you’ve already got — the people you work with most closely in this area. Aim for a mix of perspectives and experience levels. Get PMs and designers in the room, and make sure to include both junior and senior people.

Keep the size of the group manageable. You need to be able to have real conversations with one another, so stick to single digits.

Include stakeholders in the conversation. If other groups are affected by your work (or you’d like them to be), make sure to keep an open communication channel. You shouldn’t invite all of them to your discussions, but you can’t expect to throw principles over the wall with no preamble and have them follow everything blindly.

You’re the moderator.

Your most important job is to make sure that everyone’s voice is heard. You’ve got a group of people from all sorts of backgrounds. Some of them may report to other people in the room. They may come from entirely different teams, with all of the history that comes with that.

Your job is to create a space where those dynamics can be shelved, and everyone can speak up as an equal. Make it clear to any leads or managers in the room that they are there as individual contributors, not as leaders.

If someone isn’t speaking up, figure out why. Maybe they are just introverted. This isn’t a problem — being in the meeting is valuable for them, but they may not have much to say until hours or days later. Create a way that they can add their thoughts in their own time (for example, tell them explicitly that they can email you or talk to you one-on-one after meetings). Maybe they aren’t comfortable because of the other people in the room. This is a problem, one that you need to fix as quickly as possible. Some things that you can do to help:

  1. Be assertive. If your establish control over the flow of conversation, you can more effectively create space for other people to speak.
  2. Don’t let anyone interrupt or railroad other people. If you see someone failing to get their opinion in, pause the conversation and ask them to share their thoughts.
  3. Listen actively. When someone speaks, especially if they’ve been quiet, pay attention, and reiterate their points to them and the group. Put your opinions aside and focus on making sure you understand what they’re trying to communicate and not on whether or not you agree with them (other people will take care of that for you).
  4. Provide positive reinforcement. Figure out what’s useful about their comments, and call that out. For example, if someone says something controversial, it’s an interesting point that merits further discussion. If they say something obvious, it’s an important fundamental to keep in mind.
  5. Watch your body language. Make sure you’re redirecting your gaze periodically to look at everyone in the room, and be aware of your blind spots. Don’t just point your body at the most senior people in the room; show equal deference to the junior folks as well. If you’ve done a good job establishing yourself as moderator, everyone else will cue off of how you treat people. Use that wisely!

The first meeting lays the groundwork.

Your first meeting is crucial for setting the tone of the entire process. Luckily, everything that would normally make a meeting hard is valuable here!

Your job in this meeting is to figure out what all of the important topics are. This usually goes one of two ways: statements that everyone agrees with, and statements that cause a lot of tension.

Find what you disagree on.

An easy way to identify meaningful topics is to tell everyone to come to the first meeting with their 5 most important design principles written down. Be clear that they should work alone on this.

Once you’ve got everyone in the same room, have them write their principles on note cards or post-its. Take all of the cards from everyone, to separate ideas from people. You may need an individual to take ownership of a statement long enough to explain it, but focus on separating statements from the people who wrote them (and their relationships and reputations) as much as possible.

With the group steering, affinity diagram the principles that you’ve got. Statements that agree go together. Statements that disagree or have tension also go together. Let the room talk about each one, but keep it brief. If tensions aren’t resolved quickly, star that section of the map and table it for future discussion.

Reiterate to the room that these areas are valuable, because it means that this is a place where having a design principle is meaningful. Use neutral, nonjudgemental language (for example, I like to refer to things as “open questions”).

If any other interesting questions come up during discussion, write them down, ideally where everyone can see them.

A word about definitions.

Sometimes disagreements arise because everyone means something slightly different when they talk about the design space. If you notice this happening, or someone asks something like “what do we mean when we say ___?”, take some time to define confusing or controversial terms. I promise, it will make life easier.

Keep track of your disagreements.

Record all of the groups you ended up with in your affinity diagramming, and make sure that you know which areas had strong agreement (i.e. almost everyone wrote a principle about the area) and disagreement (two or more people had conflicting principles). This is your roadmap for all future discussions.

Answer your open questions.

Now all you have to do is get the group to resolve all of the disagreements you’ve highlighted. Piece of cake!

  1. All that moderation stuff you were doing earlier? Keep doing it.
  2. Tackle one issue at a time. While it can be hard to keep the group on track, staying focused on one issue at a time helps prevent drawn out conflicts and epic arguments. Write down anything insightful that comes up in tangential conversations as a future discussion point, and then reset the group.
  3. Tease out the fundamental question before struggling with answers. It’s easy to pick one side or the other based on the opinions that highlighted the disagreement, but it’s not useful.Figure out what the root of the tension is and resolve that. For example, if the disagreement is on whether or not touching in the middle of a block of text should highlight a word or move the cursor there, the real question may be “Is touch forediting existing content or creating new content?”
  4. If someone is having an epiphany, give them space to think. Don’t rush forward without them, and encourage the room to listen quietly while they sort out their thoughts. It may feel inefficient, but disrupting their train of thought before they get to an answer is a far bigger waste of time than pausing for a minute or two.
  5. Write down insights. If you’re doing a good job getting at the heart of your open questions, answering them will lead to important insights about the design space. These insights are what you’re trying to find, so when they come out you should immediately record them.

Now you can write your principles.

Good job! You’re finally in a place where drafting a set of principles makes sense. This can be something that you do alone or with your group. Go with what makes sense to you.

Remember your roadmap? Go back to it, and find all of the areas you noted before. Meaningful design principles will most likely come from the two areas we talked about before: strong agreement, and tensions that led to open questions.

Areas with strong agreement are probably so fundamental to the area that they warrant a principle, but if they’re obvious to your intended audience you may be able to get away with skipping them (even so, they can be valuable to reiterate). Write down drafts of however many principles this leads you to.

Once you’ve finished those, go through your list of open questions. Which ones had interesting answers? What insights did you identify along the way? Pull out everything meaningful, especially things that weren’t obvious, from here and draft principles around them.

Now you’ve got a list of draft design principles. High five! Put them in a rough prioritization order (don’t spend more than 5 minutes here) and share them with your group.

Design is iterative, and principles are too.

You’re not quite done (sorry). Get your group together and go through all of your principles. Does everyone agree with them as they are written? Does anything seem off about them? Is anything important being left out? Make sure that you’re still creating an environment where everyone feels safe voicing their concerns.

Keep rewriting them until the whole group feels good about every principle.

Once you’ve got internal agreement, share your principles with close partners and stakeholders. See how they interpret the principles you’ve written and what concerns or disagreements they have. This will show where you can improve your wording to be clearer, and highlights any mistaken assumptions you may have made about your audience’s understanding of the space.

It’s also possible that a thoughtful and well-informed partner will bring up something that your team has overlooked. Don’t panic! You know how to handle this. Get your group back together and tackle this new area until you’ve gotten it sorted.

You’re finally done!

Congratulations, you have a set of design principles! Use them well.

Seven months of Microsoft

This is a response to the open letter to PM recruiters that Ellen Chisa wrote, which is in turn a response to a tweet by Jon Bell.

I’ve been a PM at Microsoft for about 7 months now. In that time, I’ve gotten solicited by quite a few recruiters aiming to poach me. The interesting thing is that they don’t seem to care about my current job as anything more than a sign that I am, in fact, hireable. I’ve gotten comments on my UX background, and my essays, and my Github profile, and my philanthropic work, but not a single recruiter has asked me “So, what do you do at Microsoft?”

It’s a shame, because my experience is that Microsoft PMs do a lot of great things that would be valuable anywhere.

Caveat: These observations are based on my personal experience working in the Windows Shell. They cannot, and should not, be extended verbatim to every organization in the company. However, I believe they’re still a good representation of organizational culture at Microsoft.


In my day-to-day I work closely with 5 designers, 1 user researcher (soon to be 2), and 2 other PMs. My team, which is part of the Windows Shell, also works closely with our sister team on the Windows Core, so we regularly meet with 2-3 PMs from that team.

I’m involved in a cross-group initiative, which includes meeting weekly with PMs from half a dozen teams.

We’re also trying to help app teams do great things with our features, so my teammates and I have been reaching out to people from various app teams. We’re actively talking to about half a dozen of those too.

I also own a feature that spans multiple areas in Windows, which means working closely with PMs from at least two other Shell teams.

Not to mention daily interactions with the rest of my team – even though we’re focused on different areas, we regularly chat in the hall, eat lunch together, and find time to bounce ideas off of one another.

In a normal week, I’m working with upwards of 30 people. And we’re in planning right now, so I’m really only interacting with Design and PM. That number is only going to grow as we start working with Developers and Quality folks (formerly known as Testers).


I haven’t met a single person at Microsoft who isn’t willing to help others. Even when old conflicts resurface, individual people are willing to help individual people. Playing well together and teamwork are highly valued here, and it’s reflected in our new employee assessment format (which specifically calls out helping others get work done).

The most common sentiment expressed in our many cross-group collaborations is “Tell us what you need to be great, and we’ll see what we can do to help.”


PMs aren’t designers, but we work closely with them and we all care about great design. Furthermore, every PM I’ve worked with feels a responsibility to champion end users and deliver great experiences. People come to the company because it’s opportunity to touch hundreds of millions of people. We have to make hard decisions along the way and things don’t always work out the way we expected, but don’t doubt for a second that Microsoft employees don’t want to build things that make the world a better place.

Just about any idea you can imagine to make a product better, someone at the company is thinking about. Hopefully they’ll get to build it! But if they can’t, it’s because shipping a product to a billion people is hard. Maybe the onscreen keyboard doesn’t have as many cool accelerators as you’d like because someone spent the last release making it support Buginese (and several other languages that previously didn’t work with computers).


The engineering culture here is unquestionably data driven. User research, usability studies, flighting and telemetry, and competitive analysis are all part of day-to-day life. If you can’t back up what you’re saying, people aren’t going to take you seriously.

Product vision

It’s easy to assume that employees at such a massive company just show up and build what they’re told to build. I’m sure there are places where that is fairly accurate, but it’s far from true in my organization. We get direction from leadership, and we the areas we own are scoped, but PMs are given autonomy and expected to use it to think strategically.

At the end of the day, features have to add value. How will this feature appeal to people? How will it move product off the shelves? How does it affect our app ecosystem? Our device ecosystem?

The best PMs are the ones who pitch ideas with answers to those questions. Want to get a proposal through? Explain how what you’re working on helps grow the business.


At the end of the day, a PM’s job is to make things happen. It’s the primary thing that we’re assessed on. A good PM anywhere makes sure that the right thing ships, in the right condition, at the right time. Because so many people have to work in tandem, execution is rigorous. PMs have to deliver.

Welcome to the colorblind future

Originally posted on Medium.

A stranger stops my mother as she carries me across the UW campus.

“Did you get the baby from Korea?”

“No… I made it myself.”

When my mother picks me up from daycare, the other kids tell me that I’m adopted. I try to argue with them, but they won’t listen to me.

My sixth grade classmates switch to Chinese whenever I walk near them. I know that they speak English more fluently; this is just to drive the point home.

Although he was born and raised in the Chicago suburbs, my father started school without knowing a word of English. Scarred by years of intense bullying, he never taught me Cantonese. 11 years old and desperate to fit in, I’m oblivious to the irony of resenting him for it.

“Mixed children are always the most beautiful.”

I’m not sure why my guidance counselor is telling me this, or what it has to do with approving my class schedule.

My mother and I are in line at the grocery store, stealing some time together while I’m home from college. The Asian woman behind us accosts my mother, offended that she is wearing Chinese jade. My mother explains that it was a gift from my grandmother, a family heirloom. The woman doesn’t seem to care.

I’m on vacation with my family, enjoying a brief respite from grad school. The clerk at the ice cream shop tries to take my father’s order with the Asian 20-somethings standing next us, skipping past me, my mother, and my brother.

Eventually she realizes her mistake and blushes.

She doesn’t apologize.