Skateboard, Bike, Car

Building Products The MetaLab Way

Andrew Wilkinson
7 min readOct 23, 2016

“You gave me a PDF full of incoherent garbage about chatbots and wasted months of our time!” I was visiting a client, the CEO of a successful venture backed company, and he wasn’t happy. He had hired us to help his team think through the next version of their product, and instead of jumping into a design phase, like we usually do, we had spent months building out a strategy deck trying to rethink the future of his business.

He was frustrated. He had come to MetaLab because of our reputation for building amazing products, and there we were, months into the project, handing him a 50-page deck about chatbots, AI, and a bunch of buzzwordy bullshit.

I grimaced and told him he was right. We’d blown it. I was embarrassed, and for the first time in 10 years I did the only thing I could to turn things around: refund the entire project and work for free until we earned his trust again. Every digit hurt as I cut the check. $-1–0–0-,-0–0–0, a hundred thousand dollars. Fuck…

Growing Pains

My product design agency, MetaLab, has never had a complicated process. Over the past decade, we’ve earned a reputation for helping the world’s most successful companies build digital products. From working on the first version of Slack to helping redesign big sites like YouTube and, we’ve been lucky to work on some amazing projects for an agency of our size.

Last year, we were doing some hard thinking about our growth and we started to get insecure. We had grown from 25 people to almost 100 in a couple years, and we realized that we were now competing in the big leagues with agencies like HUGE and Critical Mass.

As we dug into what the competition was doing, we realized that we focused on just a few core services, namely product design and development, while the big guys seemed to do everything from advertising campaigns to public art installations. Along the way, we noticed that they all appeared to offer something called “strategy.” We blushed. What’s strategy, and why don’t we have it? Are big clients going to take us seriously? Are we missing an opportunity here?

Like many business blunders, we made the decision to start selling strategy as a reaction to what other people were doing. We hired some strategists who had worked at big agencies and started adding strategy phases to each new project. Our team would fly in, spend an afternoon with the client, and then spend weeks building a huge presentation deck that would tell them about the future of their business. The decks were all about big ideas: what they could do and how they could do it using new disruptive technology. How they could use chatbots and machine learning to rethink their industry — that kind of stuff.

“Consulting: The art of stealing your client’s pocket watch to tell them the time, then keeping the watch.”


Unfortunately, none of it was faithful to what we do. The best businesses do one thing well, and at MetaLab we build stuff. Strategy is all about ten years from now when we are good at now. They wanted mockups and prototypes, not 50-page strategy decks.

Our design team was pissed. What happened to just sitting down and talking to our clients? What happened to talking to users to learn about their frustrations? What happened to the simple process that had served us so well for over a decade, where we’d just sit with our clients, learn about their product and goals, then roll up our sleeves and get to work.

After our short lived experience doing high-concept strategy, I’ve realized that our simple process is one of the things that set MetaLab apart from other agencies. We don’t get bogged down preaching vague visions of what could be. We shut up and build stuff, get feedback, and iterate until both our team and the client love it. It’s nothing fancy, but it seems to result in some spectacular results.

Planning < Iteration

I’m just going to come out and say it: planning is usually just a fancy word for guessing what will happen in the future. Whether it’s a strategy deck, a super detailed product roadmap, or a business plan, you can’t define the perfect 100 steps that will lead you to success. Doing these exercises make you feel great, but they are often akin to reading tea leaves. Predicting the future is hard.

“There are two kinds of forecasters: those who don’t know, and those who don’t know they don’t know.”

John Kenneth Galbraith

Don’t get me wrong, every company needs some degree of long-term strategy, and there are people who are very good at it, but I think there’s a strong case to be made for rapid iteration over planning in most cases.

Take a step. Look around. Take another step. Look around. Ensure you’re still heading towards the mountain. Course correct. Take another step. Over the past ten years, we’ve climbed some pretty big mountains by working in this way.

Skateboard, Bike, Car

Our big strategy misfire sent us back to our roots. We’ve realized that we had had it right the first time, and we’ve spent the past few months documenting our process. We call it Skateboard, Bike, Car.

Why does it have this dumb name? Because building a product should be tangible, satisfying, and fun the whole way through. Call it lean. Call it agile. Whatever you call it, it’s about building stuff that you can play with and give feedback on along the way.

The world’s top companies have already figured this out. Many of them have implemented some form of agile process, and yet, for some reason, many agencies still use the top-down, waterfall style that was popular in the 90s.

Sprawling scope documents. Multi-month research and strategy phases. Big decks. Presentation after presentation. A lot of talking about what can be done, instead of what is today. Nothing real. Nothing to you can use or hold in your hand.

We’ve decided to go the other direction. We work in one week sprints. The first week is spent with the client, with our design team learning about their product and what they are trying to accomplish. From there on out, we kick off on Monday and deliver something real every Friday. At first, it might be wireframes. The next week, a mood board. The next, high fidelity mockups. A month in, an Invision prototype they can play with on their phones. We deliver Friday, get feedback, Monday, and just keep prototyping until we get it right. It’s dead simple, but it works.

We’ve also refocused our services. When you walk into a restaurant and see that they do Chinese, Pizza, and British pub faire, you can safely assume you’re in for a night at home with a bottle of Pepto Bismol. Excellence comes from specialization. We want to be the artisianal pasta place that has been hand-making for 35 years, has 5 menu items, and a line-up out the door. We don’t want to be everything to everyone, we want to be absurdly great at one thing. We don’t do high level strategy or consulting. We don’t do campaign work. We don’t build marketing micro-sites. We don’t do print. We don’t do social media. We just do product. All day, every day.

Sticking To The Knitting

So far, it’s going great. We’re doing the best work we’ve ever done. And I’ve instructed my team to commit me to a mental asylum if they’re ever asked to design a presentation deck longer than ten pages…

It’s not like we’ve completely rejected strategy as a whole. We still do it, but we keep it simple, iterative and focused 100% on the product and its users. 5,000 feet not 30,000 feet. We do quick phases at the beginning of projects that focus on answering questions about our client and their goals so that our team has a framework for decision-making, and then jump into rapid prototyping. It has helped us do better work while filtering out all the pie-in-the-sky stuff.

And remember that huge refund check? I’m happy to report that we earned back our client’s trust and won a new project with his company. Instead of doing an overblown strategy phase, we kept things simple this time around. After spending some time with his team to learn about their user pain points, and kicking the tires on some of their competitors, we put four of our best designers in a room and started sketching. Two weeks later, we delivered a prototype that knocked their socks off and won them back, and we’re currently heads down on an amazing new product that we can’t wait to share.

I’m going to count this one as MetaLab’s version of New Coke. Sometimes you just need to stick with what works.

Andrew Wilkinson is the founder of MetaLab and lives in Victoria, Canada. You should follow him on Twitter 🐤