Lyft new-city pricing: trading take rate for match rate on a scheduled airport route

You own pricing for Lyft's scheduled rides, and Lyft is entering a new city, for example Toledo, Ohio. The city has several routes, but for this exercise consider only one: trips between the airport and downtown, in either direction. Market and launch facts: - Riders in the city expect to pay about \$25 for this one-way trip, and drivers expect to earn about \$19 for it. - Lyft launched at exactly those numbers: the rider pays \$25 and the driver receives \$19, so Lyft keeps \$6 per completed ride. At this price, only about 60 of every 100 requested rides find a driver. - Riders may never be charged more than the \$25 prevailing fare. - For this exercise the rider fare stays at \$25, so driver pay per trip is the only lever. Driver pay and Lyft's take move dollar for dollar in opposite directions: raising driver pay by \$1 cuts the take by \$1, and lowering it by \$1 adds \$1 to the take. Driver economics: - Acquiring a new driver costs between \$400 and \$600, and the cost climbs as you recruit faster, because each acquisition channel has limited depth. - At the \$19 wage, drivers churn at 5% a month and each completes 100 rides a month. Rider economics: - Acquiring a new rider costs between \$10 and \$20, and this cost also rises with the pace of acquisition. - A rider requests one ride a month on average. - Riders who never experience a request that fails to find a driver churn at 10% a month. Riders who experience one or more such failures churn at 33% a month. One experiment: for a few weeks, Lyft cut its take on every ride from \$6 to \$3, and the share of requests that found a driver rose almost immediately from about 60% to about 93%. Define net revenue as what riders pay minus what Lyft pays drivers. How would you maximize Lyft's net revenue on this route over the next 12 months? Support your recommendation with the numbers. Your answer must end in a specific change to driver pay per trip on this route: whether to raise or lower it, and by how many dollars. ```hint Price the failed match Compare the options by what one ride request is worth across the whole year, not by the margin on a completed ride. A failed match costs more than the lost fare, because it changes how long that rider stays. ``` ### Clarifying Questions - When the take fell to \$3 in the experiment, did driver pay rise to \$22 with the fare held at \$25, or did the fare drop? - Does money spent acquiring riders and drivers count against net revenue, or is it a separate budget? - Must driver pay be one fixed amount for every ride, or can it vary by ride, for example extra pay on a scheduled ride that has not yet found a driver? - After a failed match, does a rider's higher churn last only for that month or for the rest of their time on Lyft? ### What a Strong Answer Covers - Unit economics per request at each take rate, separating the margin on a completed ride from expected revenue per request - The link from match rate to rider churn, carried through the 12-month horizon and weighed against rider acquisition cost - Driver-side levers beyond one uniform pay change, such as targeted pay on hard-to-fill scheduled rides and paced driver recruiting, with their risks - Pacing acquisition on both sides when acquisition cost rises with speed - A test plan and guardrail metrics that check whether the few-week experiment result lasts ### Follow-up Questions - Drivers start leaving rides at \$19 unclaimed so they can wait for higher pay. How would you detect this and respond? - If paying drivers \$22 also lowered their 5% monthly churn, how would your recommendation change? - How would your answer change if the goal were the next three years instead of the next 12 months?