Lyft's lower-fare shared ride option: what riders will pay and how to test it

Lyft wants to introduce a new ride option called Shared Saving. Assume it lets riders share the car with other passengers headed in a similar direction in exchange for a lower fare than a standard private ride. As the product manager, work out what determines how much riders are willing to pay for it, and design the experiments you would run to check your assumptions. ### Clarifying Questions - What does a Shared Saving rider give up besides privacy: detours for other passengers, longer pickup waits, a walk to a pickup point? - What is Lyft trying to achieve with the launch: attract riders who find standard rides too expensive, carry more riders per driver hour, or earn more per trip? - Is the fare quoted upfront and fixed, or does it change depending on whether a match is found? - Is driver pay for shared trips part of the problem, or only the price riders see? ### Part 1 — What shapes riders' willingness to pay Identify the factors that determine how much riders would pay for a Shared Saving ride, and say which of them matter most. ```hint Count every cost Compare a shared trip with a private ride and with the rider's other ways of making the same journey, and list everything the rider gives up, not only money. ``` #### What This Part Should Cover - The non-price costs of sharing for the rider, set against the size of the discount - The reference points riders compare the shared fare against - How willingness to pay varies by rider segment and trip context, with the factors ranked ### Part 2 — Experiments to test the assumptions Design the experiments and analyses you would use to check the assumptions from Part 1 and to set the price. ```hint Decide what to randomize Sort your assumptions into those that existing trip data can answer and those that need Lyft to change the price or the matching, then think about what unit you can randomize when nearby riders are matched into the same cars. ``` #### What This Part Should Cover - Each key assumption stated as a testable hypothesis tied to a pricing or product decision - Experiment designs with a randomization unit suited to a pooled marketplace, plus cheaper methods where a live test is unnecessary - Primary metrics and guardrails for riders, drivers and Lyft's revenue ### What a Strong Answer Covers - A definition of the option and of Lyft's goal, settled before any factors are listed - Factors specific to sharing a ride, ranked rather than simply listed - The risk that the option pulls riders away from standard rides, and how it is measured - A decision rule that links results to the launch price, a product change, or not launching ### Follow-up Questions - If most Shared Saving riders turn out to be people who used to take standard rides, how would you decide whether the option is still worth keeping? - How should the discount differ between rush hour in a dense city and a late-night trip in the suburbs? - How would drivers' willingness to accept shared trips change your experiment design?