Online question-and-answer platforms allow consumers to learn various knowledge from crowd wisdom. Such platforms’ performance critically depends on both quantity and variety of knowledge contents contributed by the crowd. This paper studies how early-stage knowledge production outcomes influence the future crowd’s knowledge production behavior. Using a novel data set from one of the largest question-and-answer platforms, we construct measures of knowledge variety using an unsupervised learning method. We find early knowledge content has substantial effects on the quantity and variety of the knowledge content the future crowd produces on the knowledge-sharing platform. Specifically, we document that (1) longer early knowledge content decreases the quantity of future knowledge contents but increases the variety; (2) a higher number of upvotes of early knowledge content leads to more diversified future knowledge contents but does not affect the quantity. Moreover, we find that whether the early knowledge producer is an expert moderates the interrelationship between early knowledge content and future knowledge content under the same question on the platform. We discuss the implications for the question-and-answer platform’s interventions to trigger high volume and diversified knowledge content.
Consumption is usually associated with time spending. Sometimes the length of time itself is part of the products, for example, the length of consulting services or time duration of transportation. In other scenarios, there is a minimum requirement of time, or waiting time, that customers have to spend during their consumption process. Examples include the long queue outside an amusement park, shipping time after placing an order on Amazon, and waiting time when hailing for a ride on the road. Marketers have started to offer various types of time-related service packages correspondingly, such as premium pass, 1-day delivery, and priority order. In this paper, we use anonymized and normalized browsing and order data from Didi—an online ride-hailing company, and estimate the value of waiting time in riding scenarios. We find that the value of saving one standardized unit of waiting time is equivalent to lowering 0.97 standardized unit of price for an average passenger. As a reference, upgrading to a higher level of car type is equivalent to lowering prices by only 0.4 standardized units. The marginal value is decreasing when the time horizon is short, but there exists a turning point at about 1.5 standardized time units. When the waiting time is less than 1.5 units, the marginal utility is decreasing; whereas when it is longer than 1.5 units, passengers exhibit an increasing marginal value in waiting time. Furthermore, we find a larger turning point in peak hour orders (relative to off-peak hour orders), and in weekday orders (relative to weekend orders). Our study provides new insights on understanding the value of time and is also practically relevant for managerial decisions on time-related product design and pricing.
See the online version on SSRN: The Value of Time: A Study of Pricing Strategy on A Ride-Sharing Platform