Juicer program is about letting user make money by charging our scooter. Since we launched the MVP on March 3th, we are still iterating on weekly basis. After two month, 1,200 scooters around the country harvested by 160 juicers per night averagely, that is approximately 70% of our total scooter fleet.
I was asked to have a minimum viable product (MVP) design during a weekend. Therefore, I skipped the data gathering, user research and analysis, and directly jump into design thinking.The first stage of the design thinking process is to gain an empathy understanding of the problem I’m trying to solve, and the user needs. In this case, how can a user collecting, charging and releasing our scooter?
After define the core problem, I generated some ideas for the solution. How can I test it out in such short amount of time? Competitive Analysis. Bird is the only company has the similar program, and they are only in LA at that time, so I cannot try the entire flow. Then, I watched Uber Driver and Doordash Driver flow on YouTube, since the flow has some parts in common. Below is the Juicer V1.0 flow I created during one weekend:
New Onboarding Flow
The all-time activated Juicer conversion rate is only 6.42%. There is a 50% drop-off from signup to add payment. To increase the conversion rate, we add a Juicer onboarding flow including in-app tutorial, add contact info and payment info. Below is the new onboarding flow:
New Serve Flow
Problem: How to improve Juicer work quality (KPI). Solution: Increase serve Lime-S efficiency. Before, juicer has to go to “My Limes” screen, tap on “Lock & Serve” button to scan only one scooter, wait for couple seconds, confirm harvest, back to “My Limes” screen and repeat until all the scooters served.
To simplify the serving process, we introduced batch serve feature in Juicer V1.6, so that they can scan multiple scooters and serve them all at one time. Below is the batch serve user flow: