The Science of Product Design

Design is typically thought of as a creative pursuit, more art than business or engineering. These days, where design is testable, measurable, and verifiable, designers are being held accountable as problem solvers, rather than creatives.

Design is tough because everybody has an opinion—designer or not. It’s easy for design decisions to become bogged down in politics and power struggles, to become victim to over-the-shoulder pixel pushing and top-down interaction design. Your manager or another exec may not say it (or they may), but it’s often implicit: “I’ve been doing this longer and I know what’s best for the user.”

The truth is that no amount of experience will give you Jobsian insights. But, you can use science!

Since the 17th century, scientists have been employing the scientific method to systematically observe, measure, and experiment to hypothesize about the natural world. It turns out that nature isn’t magic or subject to the whims of the gods, but instead adheres to certain observable laws and patterns of behavior. But I don’t need to sell you on science, you’re reading this article on your computer via the internet and everything is amazing!

In other news, we also know that during the past 20 years, 62 of 100 VC funds failed to beat returns available from the public markets. Another popular statistic bandied about recently is that 95% of startups fail.

What gives?

I contend that it’s the “creative” legacy that’s holding back product designers from truly empathizing with people and creating products they’ll love.

“For most innovators the word “idea” conjures up an insight that immediately requires a plan to bring it to fruition. In contrast, a hypothesis means we have an educated guess that requires experimentation and data to validate or invalidate.”

— Steve Blank

I’m not the first to write about the scientific method in relation to startups and technology, but I’m writing now to advocate for standardizing principles of product design the way the principles of the scientific method were codified so many centuries ago.

The Scientific Method as an Ongoing Process

I first became aware of the job title “product designer” when Facebook started using it for their designers sometime in early 2012. There’s been plenty of debate over differentiating design titles, but I think the dust has settled over last couple of years. This newer title implies greater ownership over the business aspects of a company and overlaps with the responsibilities of the product manager. With the ability to have greater influence over product development, designers have the opportunity to become the stewards of great process.

When you think about designers as problem solvers rather than creatives, you begin to see parallels between the product development process and the scientific method.

Paul DeVay, in his article What is Digital Product Design?, shared a clear visualization of these parallels:

Scientific Method and the Design Process

Graphic Credit: Paul DeVay

Julie Zhuo visualized these parallels in an abstract, but fun way that points out that experienced designers have a more methodical approach to design decisions:

Whether you’re a 7-person startup with a seed round, have found product/market fit and employ 250, or run Facebook or Google, the same rules apply.

The process of a senior designer above is intentional. A senior designer starts with a hypothesis and sets out in multiple directions to see which solution shows positive results and then builds on that work. The junior designer follows the whims of their imagination and the push and pull of management and customer feedback, without incrementally building on previous learnings.

The process of a junior designer can also apply to people with senior titles or even those in management, and as such can reflect on the entire product team. The same goes for the process of a senior designer. Both reflect opposing philosophies about design—that design is either a creative and political process or that it’s methodical and incremental with a clear path to success.

The reality is that having a standardized process takes the magic out of product development and helps develop a culture of customer empathy. It empowers everyone to become better facilitators and collaborators, keeping each other accountable and working toward common goals, rather than engaging in contentious rhetoric where it’s just like, your opinion, man.

It’s fairly common for a team leader or founder to observe something about the world that suggests a market opportunity. “I have a game-changing, industry disrupting idea! It’s going to make us billions!” But as we’ve heard so often, ideas are worthless. So, why not pivot and have a hypothesis instead? The great thing about hypotheses is that they’re testable—you can prove value with less cost than just building the next big thing.

“My co-founder and I spent $1,000,000 on a web hosting company that never launched. We were perfectionists so we built the best thing we could without even understanding what our customers cared about.”

Hiten Shah

I’ve been at that company. I’ve spent six months making painstakingly pixel perfect mockups, over and over again incorporating feedback from managers and higher-ups, only to present that work to the primary investor and have the entire project shot down because he had a project he decided was more important. When your company doesn’t have a culture of process, design is subject to the whims of power.

When you do have a team that shares a common process—it’s amazing. Design then has the autonomy to validate ideas through observation, research, and testing, and make informed predictions that will increase the likelihood of success once your team actually builds a feature.

Product Development as an Ongoing Process

Graphic by Stewart McCoy

The graphic above is represents a standardized process I believe applies to most product teams. Teams can pick and choose their techniques (e.g. competitive analysis, a/b testing) and tools (e.g. UserTesting, Optimizely), but in general should be solving customer needs methodically and incrementally.

What’s great about the product development methodology pictured above is that it’s a virtuous cycle. More research leads to more insights that lead to more opportunities. Everytime you release a successful feature, users begin to act differently and then you have new observations that you can form hypotheses about and go research.

Science! It’s the nearest path to objectivity we have. Sure, we still use the scientific method to inform our subjective opinions and biased perspectives, but it’s a good way to ensure we’re taking an approach that allows for repeatable, testable, verifiable outcomes. In the context of product development, such a standardized approach seems far more preferable to producing personal and expensive works of art that our users may or may not find useful, usable, and valuable.

As product designers, it’s our job to evangelize the benefits of such a process, and in doing so we not only empower ourselves as facilitators and curators of the best ideas, but position our teammates to be most effective and influential within their own domain expertise.

Now go forth, and share the good news!

Jumpstarting a Design Community

Understand Your Compensation

Designer Monoculture

The State of Design Leadership

Interview with Michael Flarup: Co-Founder and Lead Designer at Robocat

The Importance of Design Conventions

Interview with Aaron Quint: CTO-turned-Cookbook Author