Network Effects in the Rideshare Industry
“The Intentional Network Effects of Uber” (https://www.nfx.com/post/the-network-effects-map-nfx-case-study-uber/) outlines a case study about Uber and describes lessons that startups can learn by looking at the strengths and insecurities of Uber’s network effect along with the company’s defense strategies. The post starts by explaining how the strength of Uber’s large network is deceptive and not as strong as it looks. The essay includes a diagram explaining how an increased user base will benefit additional users, making the company more attractive to potential customers. The idea is that having a driver supply will reduce wait time and fares, increasing the number of users and, therefore, the potential earnings for drivers, which in turn increases the driver supply.
A previous blog post (https://blogs.cornell.edu/info2040/2019/12/04/the-benefit-of-network-effects-for-lyft-and-uber/) explains various types of network effects with Uber and other rideshares following a two-sided model. While the diagram included in the essay below seemingly follows a two-sided network effect (increasing the user base will increase the value of the product for other users), the cycle deviates in the step where a larger driving supply will lead to a lower wait time. If Uber had a true two-sided network effect, the cycle would keep repeating and the wait time would keep dropping but the cycle clearly has limitations (e.g. cannot drop below zero minutes). It turns out that not only is this instantaneous pick up physically impossible, but also a lower wait time is not always better since the extra time gives riders a chance to do last minute things before leaving. The system will reach an asymptote of sorts where large changes to the driver supply will lead to smaller and smaller changes in pick up time and user value.
With this being said, the essay continues on to describe why this asymptotic loop is weaker than the true two-sided network effect: they are vulnerable to new competitors in the market. The barrier to entry for the rideshare market would be matching the 4 or 5 minute wait time of current rideshare companies. Furthermore, it is not difficult for a region to amass a large supply and it is not difficult for users to be able to switch from the Uber app to the Lyft app (or others) to find a better price or wait time. The mobility also involves the supply since drivers are allowed to drive for multiple companies.
The solution to these weaknesses lie in a reinforcement strategy where new products strengthen their network and ability to expand the reinforcement. Some of these strategies include rapid expansion of the network, embedding their product into other companies, language/name appeal, 2- or 3-sided markets (i.e. freights or Uber Eats), and data collection among others.
The essay demonstrates how sometimes a large network is not always enough, even in a company with as much notoriety as Uber. The case study provides insight on how physical limitations, barrier to entry, mobility of users/suppliers, and many more factors affect the model of the network effect. The defense strategies were also interesting to gauge how startups can use a moderate core network effect and build off of the core to create a much larger, more stable network rather than risking the thin margins and high sensitivity to competition seen in other sectors like the airline industry. While it is easy to understand that the more layers of defense there are the better, the essay is helpful in explaining how something like expanding the products or having an effective name is in strengthening the company. Overall, the article provides valuable insight into asymptotic network effects are affected by competitors and low barrier to entry but can be strengthened in a multitude of ways as long as a core network effect is present.