The Journey

Phase 1

My journey at Pop started as an intern in the summer of 2020. My task for the summer was to prototype a new way to onboard users through their account creation. What was wrong with the current process? How might we encourage more engagement with the onboarding process?

Example screens of the original onboarding.

Hypotheses

What could be possible reasons behind the onboarding's poor numbers?
  • Users are getting bogged down by the number of steps and screens.
  • Users are being given non-optional open entry questions that require time to fill out.
  • The onboarding process doesn't have much of a human touch, it is reserved and distant.

Design + Rationale

The design I decided to pursue and develop was inspired by the chatbot interfaces found in apps to aid the user in collecting information while creating the illusion of conversation. My major inspiration was the app Flo, a menstruation tracker, that has a chatbot feature to converse with the user and gather data about their monthly period. I've always found it to be unique from other apps and I find the usefulness of the app to be increased due to the simulated conversation.

I conducted interviews with users in our target demographic and questioned them about their experiences with chatbots, their preferred apps with an easy onboarding, and problems they encountered that they wish they could change in onboarding across apps. With this information I was able to expand on my idea.

While chatbot-based features are not present in other social apps, I was tasked by Pop to "think outside the box" and come up with a different solution to the time and completion problems. And so, I spent a few weeks refining this design, thinking of the visuals, the copy (tone), and the kind of information we were collecting.

Sample screens of the first iteration of Onboarding (Summer 2020).

Impact

This version of the chatbot performed poorly and users were taking even longer or leaving the app.

Opportunities for Growth

I created a product that did not provide the correct solution since my design process was focused on the idea first, not the user needs.

Challenges

I was a new design intern and the UX method was unfamiliar to me. This was also the first UX team in the startup and there was not a lot of design guidance.

Moving Forward

This experience taught me not to dive too deep into high-fidelity design early on and not become attached to my designs.

Phase 2 (Dec-Jan 2021)

In the second iteration of onboarding, I tackled the problems identified in usability tests to improve on the issues of the first version of our chatbot-based process.

I identified issues with interaction, notably how a different tutorial feature (separate) negatively affected this feature.

Changes made included: removing this tutorial, modifying the copy to shorten it and match the company tone, and allowing a certain question to be undone. As a result, completion rates recovered what had been lost and gained 5% more.

Sample screens of the second iteration of Onboarding (fall 2020).

Impact

Completion rates: Recovered what was lost + gain of 5%, resting around 80%

Opportunities for Growth

Failed to ship on time, as there were staff changes and ongoing development projects that took precedence.

Challenges

I had been made a part time Product Designer and a co-lead of the UX team during the fall, and it was an intense amount of pressure and work while I was still in school for my senior year.

Moving Forward

While this iteration was focused on fixing issues with the past design, during this time I came to terms with the decision that an intercom-based onboarding was not the best method for a social app, based on competitive analysis. This taught me to hold myself accountable for my designs and acknowledge when it's time to move on.

Phase 3 (Feb-Mar 2021)

In the third iteration of onboarding, it was time for a reality check. Yes, the numbers for the chatbot-based onboarding were better than before, but was that good enough? 80% completion rates were not satisfactory for the product manager. So, for this phase, I took a closer look at the research from other apps and conversations with users to choose the next best path forward, even if it meant undoing the work from previous app versions.

Breakdown of Hinge's Onboarding to identify strengths and weaknesses.
Sample screens of the most recent iteration of Onboarding.

Impact

Completion rates: 85% (5% increase)

Opportunities for Growth

In this iteration I felt more confident in my designs and my process. I still felt that more user testing would have made us more confident in our designs before shipping.

Challenges

The company had undergone new restructuring and began operating in cross-functional teams. This was a challenge to operate in at first, but proved to be most effective for creativity and pace of development.

Moving Forward

I left the company shortly after the launch, but was aware the new flow would be compared to the previous and the team would keep the solution with better performance.