From our field study, we created a low fidelity prototype and then formulated our experiment goals for A/B testing. We focused on comparing two aspects between versions of the app:
Experiment goals
Which version better supports:
1) the perceived casualness that users felt, and
2) the awareness that users had of status updates for clothing transactions?
We then created two medium fidelity versions of the app, where we implemented the workflow of borrowing an item from a friend's closet.
Shared designs between the 2 versions
Where it differs
After a borrow request is sent, the designs begin to differ. The Chat version (version A) supports borrowing clothing and organizing a meetup with a chat-based interface, while the Activities version (version B) forgoes using a chat feature entirely. The activities version uses an entirely separate component for handling transactions/status updates.
Borrow request pending

Chat version

Activities version
Setting up a meeting to handoff the piece of clothing

Chat version

Activities version
Requesting to extend the borrowing time of a clothing piece

Chat version

Activities version