Using Lean Thinking to Become a Better Service Designer
Two UXers share what they learned from NYC Service Design Collective’s recent meetup.

On October 29th, the NYC Service Design Collective meetup held a workshop titled “Using Lean Thinking to Become a Better Service Designer” at Bionic’s beautiful office near Union Square. Having toyed with the idea of conducting Lean experiments in our day-to-day work, we were excited to learn more about the subject and further our education as service design practitioners.
The session was led by Akanksa Chaubal, a business designer at Bionic. She explained that her day-to-day work involves conceiving and testing startup ideas for Fortune 100 companies. Efficiently validating ideas early is at the core of her work.
We’re both UX practitioners by day. While we’ve heard about the power of Lean UX, it’s been difficult to put the theory into practice. This workshop was extremely helpful in that regard: it provided us a foundational understanding of the lean experimentation, a safe environment to generate ideas, and forum for some solid feedback.
Here are some of our biggest takeaways from participating in this workshop:
- Takeaway #1: Lean experiments are learning-centric, while product development is delivery-centric. A Learning-centric approach is advantageous when working in a startup environment. Traditional service design methods fall short in uncertain environments where you have to learn quickly, iterate, and pivot as efficiently as possible. Lean experimentation emphasizes focused learning. Each experiment is designed to help you figure out if you should double down and invest more resources, or pivot and try something else.
Say you’re trying to determine whether or not people in your neighborhood are interested in starting a community garden co-op where people volunteer their time in exchange for produce. Instead of figuring out where you’d get fertilizer and shovels from, you can first run a lean experiment to see if people are interested in signing up to volunteer their time or not. If people are interested, you can confidently take the next step towards building. If not, reconsider ways to provide more value to potential members.
- Takeaway #2: It’s important to understand what assumptions you’re already making in this space. In an uncertain environment, many things we take for granted could actually be assumptions-based, not fact-based. What are we already assuming about the users? About the environment? About the setup of the service itself? Writing it all down made us realize the sheer number of assumptions we had in this space.
In our community garden co-op example, we assume that potential users are from our local neighborhood, that they don’t have challenges with mobility, and that they would rather invest their precious time for a ripe tomato than go to the store and pay for it. These are all important assumptions, but where do we start with testing?
- Takeaway #3: To be successful, figure out the single most important assumption you need to test. You may find that there are so many different things you wish to learn that it’s hard to narrow down to just one variable to focus on. The advice we got was to focus on what was the riskiest. The part that absolutely needs to be clarified to move forward– that is the first variable to focus on.
Some variables we might consider: How many hours should members volunteer? What do members want to plant? How much does each member get at harvest time? These are all important variables, but does anyone actually want to participate in the first place? Without a list of potential users who have signaled their interest by signing up, nothing else matters. Thus, the single variable to focus on first is: will anyone sign up?
- Takeaway #4: Good experiments are small enough in scope that they can be reliably proven or disproven. This is critical for deciding what to do next, especially in the early stages of validation. As a rule of thumb, it’s usually a good sign if users are willing to give you their time, money or data. If not, it might be time to revise your approach or pivot.
In our example, we’re trying to figure out if people are interested in becoming members at our community garden. We’ll determine this by pitching the idea to people in the community. If our hypothesis is true, people will sign up and share their name and email address. If our hypothesis is false, people won’t sign up. Pretty simple right?
- Takeaway #5: Lean experiments are deceptively easy to set up and hard to execute on. The process of learning anything is vastly accelerated when we’re thrown in and forced to come up with something. There seemed to be a million assumptions we wanted to call out, a thousand variables we could pursue, and jumble of hypotheses to test. We had no idea lean experiments required so much thinking and deliberating until we gathered in our groups and hashed it out.
Are we trying to produce more vegetables? Or get people to eat more vegetables? What about just charging people for vegetables, how does that fit in? Where exactly should we go to find our potential garden members? We had a really hard time zeroing in on our first experiment, but it helped to frame things around the question of desirability.
- Takeaway #6: Lean experimentation is a never-ending process that feeds your product’s growth and development. It’s certainly not feasible to run experiments on every last assumption you have about your product, but it’s important to keep asking yourself, “what’s the most significant assumption we have?” Use this to drive new features and get the desired outcome, with as little waste as possible. The rule of thumb was this: the fidelity of our experiment should increase as we do more experiments.
Assuming we did get people to sign up for our community garden idea, we could move on to the next experiment and flesh out more details of our proposed service. That might involve testing options for how much volunteer time users need to give to get access to the garden, or testing the real engagement of people who signed up– will they show up and get their hands dirty? There are countless ways to approach this- planning out the sequence of experiments you’ll follow on your way to success is tricky.
Now that we’ve shared our thoughts we’re curious, what are some things you’ve learned about lean experimentation, either through a workshop, reading, or through using it on a project? We’d love to hear about your tips & tricks in the comments.
About the authors: Kathy Liu and Craig Stover are both members of the UX team at CBRE Build. Kathy is a UX researcher and Craig is a UX designer.
Disclaimer: The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of our current employer.
