How we approach design sprints at Shopify
Over the past few months at Shopify I’ve been running weekly design sprints to kick off new projects and features. If you haven’t heard of a design sprint before, it’s a 3–5 day collaboration of designers, developers, UX researchers, and content strategists where we work together to build several high-fidelity prototypes.
We start every design sprint with a full-day “Understanding phase” which is largely guided by a product brief: a concise document that paints a clear picture of the problem space and what we’re trying to accomplish. During the Understanding phase we intentionally don’t go into details about how we’re going to solve a problem; the what and why is what’s critical at this point.
A well-made product brief could be handed to anyone in the company (or even a random person on the street, for that matter) and they would understand exactly what our goals are after reading through it. A few of the most important parts of the product brief are:
- What problems we’re solving: A concise overview of what we’re going to be focusing on for the week.
- Who we’re solving them for: Our Persona Cards come in very handy to describe the types of users we’re designing for and what their specific needs and expectations are.
- Data to inform our design decisions: Both qualitative and quantitative. Things like “What is a typical workflow for processing an order? How do merchants ship international orders?” or “What percentage of orders have a single line item? How many products only have a single variation?”
Our product briefs are a combined effort by a few individuals who have expansive knowledge of the problem at hand. This is typically one of our UX researchers (who are talking to our users and understanding their pain points on a daily basis), and a PM / product lead (who are generally responsible for long-term vision of the product). We also lean on our data team to pull in the important numbers and relevant information that we need to gain valuable insights.
Why design sprints?
Design Sprints are a fairly new approach to design for us at Shopify. We’re only a few months into this methodology, but the response and results have been quite positive so far. There are several reasons that I think design sprints are a very effective way to approach design:
- Generate unique ideas: By having a diverse group of people all in one room working on the same problem, it helps you gather ideas that you may have not initially thought of on your own. Having a casual conversation with a multi-disciplinary group makes you look at things from a fresh perspective.
- Move quickly: We start every design sprint with low fidelity sketches. Sketching exercises like crazy eights are great for quickly brainstorming different ideas. Over the course of a design sprint, we’ll bring our best ideas into high fidelity prototypes that we can put in front of our users.
- Build the right thing: Moving quickly is one thing, but moving quickly in the right direction is much more important. Working closely with our UX researchers ensures that what we’re making is valuable to our users.
- Validate decisions: Since we have several different ideas to put in front of our users, it allows us to create a short feedback loop and validate our hypotheses.
"Moving quickly is one thing, but moving quickly in the right direction is much more important."
"Having a casual conversation with a multi-disciplinary group makes you look at things from a fresh perspective."
Data-informed, not data-driven
Our philosophy and approach for every design sprint is to be data-informed, not data-driven. We try to surface every piece of information that will help paint a clear picture of the problem we’re trying to solve. We leverage all of the data we can to understand the core problem, but we don’t blindly build whatever the data may suggest.
"If I had asked people what they wanted, they would have said faster horses."
This famous quote by Henry Ford may be cliché, but it’s interesting to interpret it in the context of data-informed design. It can be boiled down into one unambiguous truth: people want to travel faster. Looking past the literal interpretation of what your users tell you can help you think outside the box, and translate that into user needs and desires that may not have been immediately obvious.
Data is an extremely valuable tool and it’s critical to the design process. Designing without data is like flying blind, but purely data-driven design is dangerous and can lead to unintentional and uninspired design. Testing 41 different shades of blue may increase your conversion rate slightly, but if your design is flawed to begin with it will never be able to reach it’s full potential. Relentless A/B testing can only take you so far. Maybe your Google Analytics numbers aren’t quite giving you the whole picture.
Sometimes you need to take a step back, gather the facts, and explore some different concepts from a fresh perspective.
"Let data inform your decisions, not drive them."
Leveraging data in our design sprints
We’ve set up our design sprints in a way that we have all the facts laid out on the table right from the get-go. This helps to paint a clear picture and frame the core problem, which empowers us as a team to solve these problems in unique and creative ways.
Data helps us truly understand the core problem, which allows us to move fast and build the right thing. Data is what provides direction to our design process; it’s the important difference between speed and velocity.
This post was originally published on Ryan's Medium profile.