Designing While Shipping

My latest talk. (Art by my friend @bysusanlin)

A lot of designers are good at pixels but struggle to ship something they’re proud of. This is because the skills it takes to be a designer are different skills from the ones that make you great at releasing a product. I assume you’re already familiar with design, so today I’d like to talk about the finer points of shipping as part of a team.

Milton Glaser has this great talk called Ten Things I Have Learned. He’s been around for a very long time, so his ten lessons are pretty great. One in particular is wonderfully true but misunderstood: “Some people are toxic, avoid them.”

The funny thing about this line is that everyone agrees at first. You immediately think of a troublesome co-worker, or manager, or someone you just don’t get along with. You think “Ain’t that the truth. There are definitely toxic people among us.”

But this isn’t what Glaser is saying. He’s not saying there are good people and evil people. Because if so, things would be so much easier!

See, the evil people would stand out, which would make my talk today pretty straight-forward. STEP ONE:

Find the evil person or persons in your group. STEP TWO:

Everybody run! Escape the evil people! STEP THREE:

Ship amazing designs that everyone loves because you got rid of the toxic people, yayyyyyyyyy!

But, of course, we know this isn’t true. This isn’t how things work, and this was not what Milton Glaser was saying. He’s not saying there are good and evil people, he’s saying that some relationships are toxic. And I bet you’re not evil, but we’ve all experienced good and bad sides of relationships.

For example, when you first start dating someone, everything’s wonderful. You’re great, they’re great, and the relationship is great. But then let’s say you have a falling out, and it gets, well, toxic. It doesn’t mean you’re toxic, or they’re toxic, it’s just that the relationship didn’t work out. It happens.

With that in mind, I’d like to share ten things I’ve learned, roughly based on Milton Glaser’s talk. First, I’m going to share five things about “opening.”

Ah, opening! The beginning of a project. The “blue sky” phase. It’s a time of high hopes and expectations. Lots of brainstorming happens here and for most people this is the part that’s the most fun. But there’s also some anxiety involved. Who’s running things? What’s going on? How can we get started? Did we already start? I’m excited but a little nervous!

I find a good way through this anxiety is something I call McDonald’s Theory. When no one knows where to go to lunch, I say “How about McDonalds?” and once there’s an idea to discuss, it’s easy to come up with other ideas. Better ones.

So I use this all the time on the whiteboard. I’m not coy about it — I very clearly say “this is probably a dumb idea but I’m just trying to get us thinking. What if we …” and then I draw something, anything, on the board.

With something to react to, with the ice broken, it’s easy to provide feedback. Someone will. Whatever they say, write it on the board too. This will trigger another person. Write that as well. Maybe just hand off the marker and sit down. You’ve done your part.

The key is not to think your first idea has to be good.

In fact, it’s often better if it’s bad. Because the real goal is getting people thinking and talking. The great ideas will come in due time.

I have another technique that’s pretty similar, called “If you had a magic wand”. It’s helpful in the process to remind everyone not to artificially restrict thinking. So in practically every meeting, I say “pretend we had infinite resources and infinite time, how would we do this?”

And then of course once you understand the magic wand scenario, you have to stage it out, shoehorn it into a schedule, etc. But it’s important that those considerations come second, after you determine what perfection looks like.

As well as this works with features, I find it’s absolutely vital to do the same thing with people. I learned this technique from my friend Jake Zukowski at frog design. At the beginning of a project, he makes a coffee date with each person he’ll be working with. Even if he already knows them.

He’s there to ask them what each person thinks success looks like. “Describe how you like working with designers. Do you like meetings? Do you hate getting long emails? Do you like scrums? If you could paint a picture of this project going perfectly, what would that look like?”

It’s such a great trick — I’ve never met someone without an opinion. If you ask, and you’re willing to listen, you will become so much more effective during the project. It really is like magic.

If you do this from day one, you can “Take Them With You”. This slide shows five people hiking up a mountain together, all on the same team.

But unfortunately, in our industry, it’s really common to say “I’m the designer, that’s the PM, and those are three devs”. It’s true that everyone’s roles are different, but it’s really important to genuinely believe, deep in your heart, that you’re just “five builders”.

I don’t mean that in a cheap semantics way, or in an “everyone gets a trophy for participation!” way. I mean really, sincerely meaning it deep in your heart. It means you invite design research at the beginning of the process, and don’t treat them as an afterthought. It means inviting developers to design brainstorms and crits, and accepting that PM and dev have great design ideas too. It means not practicing high priest design, where the design team walls themselves off because CAPITAL D DESIGN OMG.

You are all building the product, and great ideas come from everywhere. But only if you’re actually blending. That means eating lunch together, going for drinks together, and respecting that everyone has something to contribute.

On a well-blended team, design is able to earn more influence and wield it appropriately. I call this next technique “Invade the Vocabulary”, and I got it from my friend Mike Kruzeniski.

He got the official bug tracker for his project to add a field for “Soul Features”. It was for protecting and championing certain design priorities as the “soul” of the product, like animations not being choppy, or the logo not being upside down, or using the right branding colors, type, etc.

And how did he do this? By respectful conversation with the other stakeolders. He didn’t declare it by fiat, he had discussions about how it’s hard for design to get “priority 1" bugs in the system, and how even egregious brand violations too often fell off the radar.

And he realized he had succeeded not when he found himself saying “that’s a soul feature” but when the other stakeolders, such as dev leads, would say it for him. I love this technique, and I always think about it. Design needs to find ways to speak the language of the company or it won’t succeed.

(Although maybe “soul” is a bit too new-age. He still can’t belive that terminology worked for him.)

Finally, let’s talk about using the right fidelity. This is an image of two people talking over a sketch, and it points to one of the most important things I’ve learned about shipping: a sketch opens a conversation, whereas high fidelity can often close it down or cause it to rathole.

This is a technique I learned from Bill Buxton, and it’s what he talks about in his book. Stay in sketch mode. Much longer than you think. Longer than you’re probably used to.

Design is problem solving, and as soon as you go into Photoshop you’re slowing yourself down and walling yourself off. In the beginning of the process, when alignment and problem solving are paramount, you need to stay in sketch mode as long as possible.

Ok! So those were five items on opening. Now let’s talk about …


Closing is tough. The blue sky is gone, probably replaced with thunderclouds, fear, and anxiety. The release is coming into focus and it’s not looking good. Worse, there are way too many open issues and some of the most important things haven’t gotten closed yet.

So here are five techniques I’ve learned that help with this critical phase of the project. It all starts with …

Using the right fidelity! We talked about this on the previous slide, when I said a sketch opens a conversation and high fidelity can close it down. Now the two people aren’t just talking about the sketch of a building, they’re looking at the result of the design thinking.

And that’s what is desperately needed from design. Not more re-hashing sketches at the whiteboard, or layers and layers of explorations in PSDs that you hide/show while a bunch of people disinterestedly check their email. No. You’ll never land anything that way.

At this point of the process you need to go high fidelity. Motion studies. Prototypes. Actual code. A pitch deck that you’ve actually practiced in front of a mirror multple times.

This is where designers get themselves in trouble. If one end of the spectrum is sketching and the other end of the spectrum is the Steve Jobs polished pitch, too many designers live in limbo between the two. Tons of artboards and layers, but no clear story or point of view. No pitch.

You should be sketching way more than you think. You need to stay in that phase through all problem solving. But when it comes to land, you need to up your game or the design principles will not be heard.

But the easiest way to drown out your voice is hating on other teams. We’ve all been there — we love our team but we’re struggling to work with some other team. It’s human nature to express frustration, to vent, to say things like “ugh I hate the orange team why can’t they be more like the green team”. But you have to learn to bite your tongue.

Shipping a product is really hard and it requires each team to treat each other well. Design can’t do it all alone, neither can engineering, neither can marketing. The best results come from teams that are able to “reach across the aisle”, as they say in politics. If you can’t, it will affect your product’s design for the worse.

So let’s say you’re all on the same page, you respect each other, and you’re in sync. That doesn’t mean you’re always going to agree, but it sets you up to disagree in a productive way.

I always try to remind myself that there is no such thing as a perfect design. There are only tradeoffs. For example, a tugboat is great at pulling things and bad at speed. So which is it, good or bad? It depends on the task it’s trying to achieve.

Design tradeoffs are the same — the person that disagrees with you isn’t stupid, or inexperienced, or toxic. They’re simply prioritizing X higher than Y, and you’re doing the reverse. Figure out what your teammates are prioritizing, compare it to what you’re prioritizing, compare it to the business objectives, and you can often find a way forward.

Which is not to say that everything’s 50/50 and everything’s going to give you an equal result, so you might as well flip a coin. You should have a point of view and you should be pushing it forward. So what do you do when you can’t agree? Push it until it breaks.

This is a technique I was taught about breaking stalemates. Assume that both sides have already explained why they like their approach more. Assume that both sides have framed the tradeoffs, compared it to business objectives, trust and respect each other, but the way forward still isn’t agreed on.

What you do is print out all the implications of either approach. If you do option 1, it means this and this and this. Put everything up on a board and try to sell it. Then do the same for option 2. Then collect everyone together to discuss as a group.

One of two things will happen. Either it’s obvious to the group which way to go (even if it’s not your favorite), or the team needs to discuss priorities further. Both are good outcomes, and this exercise can help make either direction a lot clearer.

Finally, let’s talk about “braiding time”. We spend tons of time in meetings, and we know we get more work done when we’re focused on our own. I’ll start with the obvious first: you should be balancing between the two.

And now for something provocative: balance means saying no more often. Software builders get hit from all sides: scrum leads to team meeting leads to firedrill then crit then lunch at your desk then three hours of meeting then another 900 crits scrums 1:1s checkins executive reviews offsites post-mortems drinks with the team it’s madness.

I carve out multiple three hour chunks every single week. I block off whole days. When I see a 30 minute gap between two meetings, I fill it so no one else can grab it. Surprise: the more I do it, the better I am at my job.

When I have time to think, to really chew on a problem, my contributions to meetings get exponentially better. Which means my solo time is extremely valuable, which leads to better meetings, which leads to better solo time. Repeat, repeat, repeat.

Can you imagine Hemingway trying to write a book in-between status meetings? Fight back. Braid your time or you won’t be able to think and design at the level you’re being asked to.

So those were ten things I’ve learned.

And a funny thing happened when I wrote all these items out in a list like this. I saw two themes. First, courage. It takes courage to throw out bad ideas like McDonald’s Theory, or decline a meeting you know will not be the best use of your time.

And of course the industry celebrates courage. It’s in all our books, movies, tv shows, the people we call heroes. Courage gets a lot of press.

But the more I looked at this list, the more I saw humility. The humility to not be the smartest person in the room. To realize that everyone has design ideas to share. To truly let yourself hear what people are saying to you, even when it’s uncomfortable. To act, to feel, to truly believe that everyone you work with has something to teach you.

But the industry doesn’t talk enough about humility, and it’s holding us back. If we keep emulating Mad Men or Game of Thrones, we’ll wonder why our resulting product designs aren’t as good as they can be. What we need to do is take a good, hard look at “rockstar culture”.

As an aside, I don’t mind the word “rockstar”. You can say ninja, or thought leader, or pro, or champ, or whatever. They’re all just ways of saying that someone is good at their job, so if we want a catchy label, that’s fine by me.

But I do have a bone to pick with the attitude behind it. We all say that people are our most important asset, but that’s only half the story. It focuses way too much on the person at the expense of team harmony. “Ok, fine,” we say. “The team is the most important asset in a company. Is that a better way to put it?”

Well, no. It’s closer, but not quite there. See, talking about team still leads us to hire “rockstars”. Just, ya know, four of them. Boom, a talented team! But there’s a lot more nuance than that.

Take the Beatles. They were successful early on, they had an amazing run, and then they broke up. So what happened? What changed? Did John Lennon become less talented? Did they need more rock stars? Of course not. The real secret to their success and their failure can be found in simple math.

For any group of four people, there are six distinct relationships. In this case, Ringo to John, George to Paul, and so on. Six relationships. When the Beatles started, their relationships were strong and drove a lot of great creative results. By the end, when they broke up, what happened? Simple, the relationships broke down.

So it’s not just about people, or teams, it’s about the relationships they’re able to cultivate. And rockstars, or people with a rockstar mentality, aren’t very good at building relationships. Or practicing humility. Meaning they’re a pain to work with.

I didn’t come up with this insight myself. I had a friend and soccer teammate in high school named Bo. One day we were walking to the soccer field and he said “Jon, I know how we could become undefeated.”

“Oh? I’m all ears,” I said, since our team up until then was, for lack of a better word, defeated.

“Yup, here’s what you do … you clone me eleven times.”

I laughed.

“I admire your confidence!” I said, “but what an arrogant thing to say!”

“No no no,” he explained. “It’s not that I’m so great on my own. I mean, I can pass, shoot, run, whatever. We all can. But think of the mind meld! Think of the no-look passes, or how everyone would be in exactly the right place all the time! Think of the ease of communication! The teamwork!”

He was right. I thought about our existing team, full of talented soccer players. “Rockstars”, even. But on offense, everyone was trying to get all the glory. No one passed. No one meshed. No one set up the goal with an assist. It was just a bunch of selfish people trying to make a name for themselves at the expense of the team.

Sounds like a lot of the tech sector, doesn’t it?

Now think about your current team. Maybe you work with 5 people, maybe you work with 10. But for this example let’s just compare it to a soccer team and say you have 11 co-workers plus you for 12.

Remember that calculation from before, where the four members of the Beatles needed to cultivate six relationships? The math gets even more interesting with 12 people.

How many distinct connections exist between 12 people?


Which is kind of extraordinary. This right here, all these dots? The health of each of these dots directly influences the quality of the team’s work, and the end result it’s able to deliver. Whether soccer or software.

Of course we all want “talent” on our team. But each time you hire you’re not just taking on one person’s ability to work. You’re also adopting the three, or eleven, or fifty relationships this new worker initiates and maintains, for better or worse.

I don’t care how good the person is in Photoshop, if no one wants to work with them, the ROI just isn’t there. With too many rockstar attitudes on your team, you will not thrive. You might not even survive.

Which is why it’s important that we zoom out a little bit in this industry. We all want to make great products. We all want to change the world. And we believe the way to get there is to hire great people. I agree.

But let’s start by defining great people as people with humility, grace, kindness, good humor and the ability to make people around them better.

In my experience, these people don’t get on magazine covers, or become millionaires, or get a gazillion people following them on Twitter. So hiring them doesn’t necessarily send shockwaves through the industry. They’re not necessarily bringing a “wow” factor just by being on your team.

But you know that quote from Steve Jobs, how “design isn’t just how it looks, it’s also how it makes you feel and how it works?”

I think it’s time we apply that same insight to our own teams. The people we choose to surround ourselves with. The work cultures we call out as exemplary and want to work in. It’s not just how the worker looks, it’s how the worker works. If the experience of working with them is bad, they’re a bad person to have on your team.

Remember that Milton Glaser quote I started with? “Some people are toxic, avoid them”? Remember how I joked that it’s not as simple as “spot the toxic person, run, and everything works out?”

Well, sometimes you come across a self-described rockstar who has an attitude to match. They think they’re above working with others. They clearly think they’re doing the team a favor just by being there. They’re not good at listening. And they’re long on ego and short on humility.

Which leads to the most important thing I’ve learned in all my years doing this. The number one thing to keep an eye on if you want to ship something great: when you meet someone who can’t build relationships, and doesn’t seem inclined to try, call a spade a spade: they’re toxic.

And you shouldn’t just avoid them. You should run.