About a month ago David Sacks posted this model of Uber.
Uber’s virtuous cycle. Geographic density is the new network effect. pic.twitter.com/NpmUnZgVfH
— David Sacks (@DavidSacks) June 7, 2014
It depicts the flywheel of Uber’s growing business, a marketplace with interesting supply- and demand-side dynamics (and these dynamics have precipitated much of the debate around Uber’s surge pricing tactics).
Using geographic density as a network effect is of course not entirely new. The early days of Facebook’s growth were driven by high density activity and word-of-mouth within select colleges. However Uber is one of a breed of new companies built specifically for mobile and the network and scale this platform affords.
I sit in the forward-looking camp, holding the opinion that Uber will be one of the biggest and most impactful companies of the coming decade. A ton of upside exists that perhaps justifies a potential $100B+ market cap. In fast moving tech-driven markets, however, nothing is guaranteed. How Uber realizes its potential will come down to positioning and, of course, execution.
I want to make some assertions from the position of devil’s advocate, rather than homer, for the purposes of this post though. There are some challenges that Uber will have to overcome.
1. At least one reason Uber has raised a whopping $1.5B is because it’s a low margin business. There are a variety of variable costs ranging from marketing to managing regulatory hurdles that necessitate such massive upfront investments. What revenues and scale and customer loyalty must Uber attain in order to think about winding down these types of expenses? When will that happen? These are metrics that hopefully become increasingly evident, tangible, and measurable as their markets mature.
2. Uber is a marketplace with a strong offline component. It will be Uber’s mission to continue to deliver a safe and trusted experience to all participants of the marketplace. As the number of drivers and passengers scale, this mission becomes increasingly challenging. A small number of bad actors or poor experiences can have a disproportionately negative impact on customer opinion. The two-way ratings system definitely helps Uber’s cause but it remains to be seen whether this is sufficient to prevent all unfavorable behavior. The good news is systems with such feedback loops tend to improve over time. Anand Iyer has a great post on this issue at First Round Review. Trust takes a long time to earn and little time to lose.
3. Marketplaces are all about liquidity. Simon Rothman, well-known expert on marketplaces, hints at the right question at the end of that post. Since Uber has taken a more centralized approach to structuring it’s marketplace, growth will take longer than would otherwise. What rate of growth is acceptable while maintaining a more central role in the marketplace? In a space with low barriers to entry, rapid expansion is necessary to win across global markets. Uber must balance restraint, to maintain a world-class user experience, and growth, to be the global market leader.
4. Which brings me to my next point: Does Uber need to take all markets outright in order to win? This is unlikely. The ride-sharing business, and the car-ownership industry, are big enough to allow multiple entrants prosper, at local, national and global scales. This again is related to the liquidity of the marketplace. People and automobiles are everywhere, implying ride-sharing services can spring up anywhere. Could there be hundreds of players in this space? We are already seeing some of this behavior in Europe and Asia, and it is somewhat akin to the flurry of activity we saw in the daily deals space. There too we saw the industry leader with low barriers to entry push the pedal on growth. The underlying economics are different in the ride-sharing space, but some similarities persist. This is not so much a challenge Uber faces, but it is definitely an aspect they must weigh when considering the total addressable market.
5. This is one of my biggest question marks for Uber and the virtuous flywheel – and it’s related to #4 and the risks of a low barrier to entry. One of Uber’s most important moats is and will need to be customer loyalty, driven by an amazing customer experience. How does a service like Uber build loyalty? For passengers, faster pickup times is just one dimension of customer experience. In what other ways will Uber be able to power the flywheel? Cheaper prices is another way. However, there are fixed bounds on how fast and how cheap a service can be. On what other dimensions will Uber optimize customer experience? On the supply side, there is a fixed upper bound on the number of rides a driver can provide per day. (This is the contrary to most online marketplaces where the upside for suppliers is nearly limitless. But actually my issue is not with this limited upside – other marketplaces such as AirBnB face similar constraints.) In the long run, Uber will need to find other ways to further differentiate its customer experience.
6. Lastly, as intimated above, some may rightly argue that Uber is, for better or worse, not a true marketplace. In a true marketplace, the suppliers set their own prices and define how the customer experiences the product (one manifestation of this is peer-to-peer transactions). This creates a network of competition and diversification of services, both of which drive customer engagement. This growing network attracts even more suppliers. And the wheel turns. Uber has chosen a more centralized approach, one that favors a more consistent experience across customers. Whether this will continue to work well, or how Uber will respond in the face of successful market-driven competitors, remains to be seen.
Uber requires enough demand so that while suppliers may not make much per unit service provided, overall they are able to pull in non-trivial income. Some early experiments have shown that this kind of elasticity to exist but it’s yet to be proven across many markets. Uber will need to continue to scale its coverage and density in order to continue extending low prices to its passengers.
The types of experiences and marketplaces the mobile device has enabled is incredible. It has exponentially increased the density of available market participants both physically and temporally.
I would love to exchange the costs associated with car ownership with the convenience of Uber. I think at the least I would end up with one car instead of two. Regardless of whether Uber “wins” or not, it is in the consumer’s interest that these kinds of choices exist, opening up opportunities and markets that did not exist previously.
Update: Amazing. No sooner had I posted this, upon checking Twitter I see that Bill Gurley has written a long (likely excellent) post, yet to read, on Uber’s total addressable market. He also uses the same David Sacks tweet.