Mistakes and Learnings Designing at an Early Stage Startup

Apr 15, 2025

4min read

Over the past 8 months, I had the privilege of being a designer at Salt + Light, an early-stage startup focused on building a digital small group app to combat loneliness. It was an ambitious mission, but our team was fully dedicated to making it happen.

That said, there were a lot of firsts for us. It was our founder’s first company, my first time working on a product without a clear market fit, and my first experience balancing the demands of a live product with designing new features. In hindsight, we made quite a few mistakes that could have been avoided. I wanted to share these experiences and some of the lessons I learned, so we can all steer clear of similar pitfalls in the future—and hopefully save on some sunk costs 🤞.


Mistake #1 : Building Without Understanding Users

This one stings because everyone in design knows..you have to understand your users. So I have to take credit for this one, because I should have pushed for this from the beginning.

Instead I trusted the conversations the founders had with target users in their personal circle and our team prioritized shipping an MVP to do more accurate testing. We thought users wouldn’t know or tell us what they needed (which is true) so we wanted to get something in front of them as fast as possible (a very high-cost decision unfit for our stage).

We rushed to ship an MVP, thinking we knew what our users needed, partly from casual conversations and partly from our own needs because many of us also fit in the target user category. We assumed small group leaders wanted more ongoing communication, but we later realized their challenges were too unique for tech to solve.

What Would Have Been Better: Engaging with users earlier and testing our ideas could have saved us a lot of time and resources.

Later in my time at S+L, we took a last bet by working with Amy Jo Kim, known for pioneering game thinking in product design. One huge takeaway from her is that you can test your ideas without actually writing any code or touching Figma, and you can refine these ideas so you have a stronger foundation when you start building.


Mistake #2: Designing Based on Assumptions, Not Behaviors

We built features assuming that people would just use them, without considering how the app fit into their daily lives and existing habits. The real user journey always happens outside the context of the app, so it’s crucial to understand how the product integrates into the lives of users and helps them grow.

What Would Have Been Better: Trying to mine for the habits and existing actions users are taking and thinking about how the app can insert itself within that flow. This also leads naturally to considering how a user interacts with the product over time, instead of only in one day. The real golden question to think about is, “how do we help the user achieve their goals over time?”


Mistake #3: Scaling Through Friends and Connections

Relying on friends to scale the product gave us skewed data because they were more supportive than objective in their feedback.

What Would Have Been Better: Reaching out to a broader audience of people we don’t know earlier on would have provided us with more accurate feedback and could have also saved us from accumulating more sunk cost.


Mistake #4: Asking the Wrong Questions in User Interviews

We had some user interviews with small group leaders about their challenges of leading and thoughts on the app but missed the bigger picture. Focusing on their existing behaviors would have given us better insights.

What Would Have Been Better: Conducting interviews with more open-ended questions and behavior focused questions could have given us richer insight into what would actually help them the most, OR it would have shown us that the problems small group leaders face is not one that can necessarily be solved digitally since the crux of small groups is the relationships you have with people there.

That’s all folks! All in all, it might have cost a lot to learn these lessons, but I’m thankful for the experience of getting to learn these myself firsthand instead of hearing it from someone else. Now I can take these learnings with me to any role in the future!

thanks for visiting!

thanks for visiting!