Personally, I am not familiar with what Lyft shuttle is, so I am going to have to make some assumptions.

1. Lyft shuttle is a service provided by Lyft that allows users to request a shuttle pick-up instead of a standard car or carpool pick-up.
2. The incentive being that since you are sharing a ride with other people the trip will cost less but may take a while for you to be picked-up/dropped-off.
3. Lyft has its own fleet of small shuttles that it uses for this service.
4. Users can pre-schedule a shuttle trip so that Lyft can better optimize the route.
5. This service is only available in large metropolitan areas where demand for car service is high.

With those assumptions in mind, I would like to clarify success. What is the goal of Lyft shuttle? Based on the 2nd assumption I made I would like to focus on this service saving users money by allowing Lyft flexibility to schedule a more optimal route per ride request.

I would like to transition to how customers interact with the service. I can think of two main types of users.

Pre-planned trip – a user who knows they need to be somewhere and schedule their trip well in advance.

real time trip schedulers – a user who isn’t in a hurry and would just like a ride to their destination on the next available shuttle.

Some metrics to consider:

Shuttle request / total requests – How often do users utilize this service over other services provided by Lyft?

Successful or completed trips / requested trips – How many trips were successfully scheduled and completed?

Total users – how many total users have used Lyft shuttle at least once?

Total active users / time range – the number of active users per day, week, month can be a good indicator of the health of the service over time?

Returning users – how many customers have used the service more than once?

Average price / mile – what is the price compared to other services?

Referral codes / total rides – what percentage of users are using special referral codes?

I think to best target our goal of making this a cost-efficient service for flexible riders we should focus on shuttle requests per total requests, successful trip percentage, number of returning users and average cost per mile. I think these KPIs will inform us about how well this service is being adopted by users and what the cost to the user is compared to other options.