The social experiment
Role
Lead product designer
Platforms
Web
Timeline
2 weeks in March 2023
For the biz
To ignite fresh thinking, the CEO proposed doing 2-week “innovation sprints,” where teams pause their regular work to build something entirely new and push the boundaries. My team's theme was Social.
For the user
It’s very difficult to connect with friends, classmates, or teachers on Quizlet to share content and study together.
And so what?
My team designed and built a social feed for a school community to connect and discover content. It was highly ambitious – in just a few weeks, we shipped a prototype to 5 schools that showed promising metrics.
Two main user jobs
Studying has a large social component to it, and Quizlet users are largely finding ways to do this offline. How can we promote this activity on platform? We synthesized the research into two main user jobs:
1
Find quality study content
... so I can make sure I’m spending my time effectively, studying accurate and comprehensive material.
2
Stay motivated
... so I can keep making progress toward my learning goal, like acing my next exam.
We quickly converged around the opportunity we wanted to test for the 2-week sprint:
A social network centered around a school that helps students feel less alone and jump into studying quickly. We can do this by:
Matching people to content — Vetting content quality is a big task. Seeing content from your course, a classmate, or someone you trust goes a long way.
Matching people to people — Misery loves company; feeling like you're not alone helps accountability.
Give users a reason to come back — Social networks need people and people doing things to succeed
Make it fun — Because otherwise studying is too serious!
How can posts in the feed encourage content discovery?
Since content discovery is a key goal, many posts in the feed will feature flashcard sets. I explored different levels of prominence of how to display them in the feed. We ultimately went with Approach B, which strikes the right balance between showing off the person (social aspect) and the flashcards (study aspect).
Approach A — Simple and small footprint
Approach B — Show some of it ✅
Approach C — Show even more
For the emoji reaction buttons, I decided on a neutral color for the background. The colored versions added too much visual weight. I decided on pill-shaped buttons, because they’re easily recognizable from other apps. For the scope of the test, we started with 5 emojis.
How can we keep users coming back to the feed?
We took this opportunity to inject a little bit of fun into the experience.
Mood check-ins
In the spirit of sharing and feeling less alone,select your “mood” for the day and post to the feed. This is one small reason that could bring users back every day. It has another benefit of populating more posts in the feed, and other users can react to it.
This-or-that questions
Another fun thing that could keep users coming back: every day we would release a new This or That question, and you can only vote once. It’s a small nod to games like Wordle where there’s only one per day.
How should we onboard users into the experience, keeping privacy at the forefront?
Social feeds need people in order to become valuable, but we need to let users have control on how much they want to participate. We decided on having users opt-in to the experience upfront. When opted-in, their study behavior will be automatically posted to the feed on their behalf (with the option to delete). For the sake of the 2-week experiment, this would generate more posts and more data to learn from.
For the first iteration, we launched with an interruptive modal over the page. Users have to make a choice one way or another. It's not easy to miss, and it's a "simple" click of a button to opt-in. Though modals have high pattern blindness, we wanted to get as many people seeing it as possible.
However, right out of the gate, the opt-in rate on this modal was low: at 4%.
Users can be cagey around showing their real usernames or pictures on Quizlet, so I tried options that bring the choice to stay anonymous more upfront.
Re-launched with this version, and opt-in rate increased from 4% to 20%. Around 80% are still opting for their full name (it is the default option), but having the choice make users more confident.
The impact
59%
of users are engaging with the feed, with the average # of activities per day slightly increasing over 7d.
36%
of users answering 2 or more mood check-ins and polls. They were one of the most popular features!
~1%
Only 1% of users followed someone, pretty low. Future opportunity area.
While the feature is no longer live due to the high investment it would’ve needed to scale and maintain, it proved to the company that we could flex our muscle to rapidly innovate and build. Internally, we got a lot of feedback like this one from the director of finance:
“This beta is freaking awesome. This is far and away the best new product I’ve seen during my time here... singlehandedly turned my perspective on Quizlet around.”
