< Part 2: Technological building blocks
The trip economy has created an explosion of choices for consumers. But sometimes choice is a burden. Bundling, as Chris Dixon lucidly outlines, is helpful to both buyers and sellers because it gives buyers more of what they want at a price they're willing to pay thereby growing the overall market for sellers. Bundles can also be helpful in breaking some of the psychological barriers that make trips seem more expensive than ownership by rationalizing costs in a way that is easier to compare.
By atomizing each transaction, the trip economy unbundles each trip forcing the supply side to improve efficiency and allowing externalities to be directly priced into the trip cost. Bundling trips may not always maximize these efficiency gains, but transportation decision-making is driven by more than just efficiency considerations. Reality is messy and the most appealing value propositions will be the ones that fit best into the consumer's existing reality. In the medium term, consumers are likely to continue to own assets to complete a wide range of trips even as the trip economy grows to meet a growing proportion of existing and new mobility demand.
This section will outline some of ways the trip economy is evolving to bundle itself into broader digital ecosystems, all from the perspective on consumers (the first corner of the trip triangle outlined at the end of Part 1). The next section (Part 4) considers the supply side of the trip economy and how it is responding to growing trip demand. The final section (Part 5) covers the remaining corner of the trip triangle: policy implications of the trip economy and how these emerging marketplaces can constructively be supported by regulation.
When comparing car ownerships to the trip economy, it is worth considering the full spectrum of business model options.
On one end of the spectrum is ownership, where an asset is bought by someone and owned by them. It is therefore dedicated to the owner's use and they also carry the financial risk and responsibilities for its storage and maintenance. The cost of ownership is front-loaded and therefore the marginal cost of usage is low.
A lease is a financial product that allows someone access to a dedicated vehicle without owning it by paying monthly installments, generally with a multiyear commitment period. The asset is therefore used exclusively by the leasee and they are responsible for storage and maintenance, but the financial risk is carried by a third party, usually a bank. The cost of access is fixed and doesn't impact the marginal cost of usage.
Somewhere between ownership and leasing is a lease to own agreement, in which the person leasing a vehicle has the option to pay an additional fee in order to own the asset at the end of the lease period.
Vehicle subscriptions are similar to leases in that the asset is not owned and is paid for in regular installments. However, in contrast to a lease, rather than paying for a specific vehicle, a subscription gives access to a type of vehicle without requiring a specific commitment period and bundles in costs like insurance and maintenance.
A rental is when you pay to use a dedicated vehicle for a defined, generally short period of time. Rentals mostly substitute car ownership during periods away from home such as work travel and vacation. The asset is dedicated to the renter's use and insurance is important for covering financial risk. Costs are generally not bundled since extras (e.g. distance allowance, fuel top up, navigation unit, insurance, tolls) are where rental companies make margin in exchange for damaged customer trust. Car-sharing and peer-to-peer models like Turo are contiguous with the rental model.
The final model to consider is of course the trip model in which the consumer pays a specific price for each standalone trip. The trip model is fundamentally different in nature from the other models since each trip is a discrete transaction and the user only has access to that vehicle during the duration of that trip.
Dogs have owners, cats have staff. And so too with these two models for thinking about mobility.
DOGs are Dedicated or Owned Goods which means vehicles that are owned, leased, subscribed to or rented since they are dedicated to the use of a single individual. DOGs are loyal, waiting patiently for their owner when not in use and therefore generally reliable unless the specific asset fails. They tend to prefer the suburbs and the countryside, but in cities they have a tendency to pollute the sidewalk.
CATs are Common Asset Trips. These are all the kinds of trips that have a fleet of vehicles that support the service (including transit). CATs congregate most densely in cities, but are found almost everywhere when you know where to look. While most people will claim they prefer the loyalty of DOGs, they recognize the independence of CATs. CATs get around and don't have a particular loyalty, but are usually available just when you need them so long as you're willing to pay them something in exchange. They tend to be nimble and adaptable and also cleaner, faster and more efficient.
A person need not be exclusively a DOG or a CAT person.
Most people have both in their life: even people who own three cars tend to use Amazon to replace some of their retail trips. Given that there are around 126 million Prime subscribers and about 129 million households in the US, that doesn't leave much room for exceptions.
Cars for the most part serve as the default mode for everyday mobility, a solution for general trip needs. But the trip economy is increasingly addressing specialty mobility, particular types of trips such as e-commerce, food delivery or quick, short distance trips with particular requirements not met by everyday mobility solutions. In the process, it is chipping away at the decathlete value proposition of car ownership and building a firm foothold across certain trip types.
Given the ubiquity of Prime and Walmart's shift towards delivery, it's not surprising that e-commerce has already reached significant scale. In the US, e-commerce was already a $360 billion market by the start of 2020 with around 250 million users (or about 75% of Americans) although this only constitutes about a tenth of the global e-commerce market, which is dominated by the Asia-Pacific region.
There have been clear indications that e-commerce is replacing retail trips, even before the onset of the pandemic. According to the 2017 National Household Travel Survey, the average US household received 2.4 deliveries per month in 2009, but that number had more than doubled to 4.9 by 2017, more than one delivery per week on average. Meanwhile, outbound retail trips per person have declined from an average of 54 per month in 2001 to 48 per month in 2009 and down to 39 per month by 2017 according to the same survey.
Significantly fewer people use grocery delivery services than e-commerce, with the number around 92 million in the US going into 2020. The market size is around $58 billion (less than a sixth of all e-commerce sales and less than 5% of sales in a market more than a trillion dollars in size). Globally, the market share for online food and personal care purchases was only about 3% in 2019, but growing rapidly even before the pandemic.
In the US, food delivery had about 38 million users and did around $20 billion in sales in 2019. China however gives a sense of the potential for growth: the Chinese food delivery market in 2019 was five times larger than the US at close to $100 billion (603.5 billion RMB).
Meituan, which dominates food delivery in China with about two-thirds market share, had 600,000 delivery people (dressed in distinctive yellow uniforms, occasionally sporting cute kangaroo ears) serving 400 million customers a year in 2,800 cities completing 20 million per day in 2019. Strikingly, more than one in eight of food delivery customers in China place one or more orders a day while about 80% order at least once a week.
Ridehailing has become increasingly popular over time. Between 2015 and 2018, the percentage of Americans who had used a ridehailing service grew from 15% to 36% with more than 50% of Americans between the age of 18-29 having tried it according to a Pew survey.
Ridership unsurprisingly varies significantly depending on density: according to the same Pew survey, 45% of urban residents and 40% of suburban residents have used ridehailing services, but only 19% of Americans in rural areas had done so. Urban users are also much more likely to use the service every week, with one in five falling into this category (compared to a less than 5% of non-urban users).
Ridehailing largely subsumed existing taxi trips while growing the overall market. For instance, in New York the trip market more than doubled in size, with ridehailing dominating this enlarged market. This shift is also reflected in corporate spending with taxis almost completely replaced by ridehailing and car rental taking a significant hit.
However, though the companies offering ridehailing services have become very valuable and many Americans have used the service, ridehailing constitutes just a small fraction of overall trips in the US. Uber's reported 1.3 billion trips and 10.3 billion miles during 2018 suggest that ridehailing services covered an overall 0.7% share of trips and 0.8% share of miles that year (up from 0.5% and 0.6% respectively the year before). •••
The global market for ridehailing, about $75 billion in terms of revenue in 2019, is dominated by the US and China and the respective national champions, Uber and Didi Chuxing. Uber, which operates in 68 countries in addition to the US holds around 37% global market share, while Didi, headquartered in Beijing with operations in Brazil, Mexico, Japan and Australia in addition to China, holds a comparable 32%. However Didi dominates in terms of total trips: by September 2019, Didi was serving around 37 million daily trips in China alone, or about 10x as many as Uber was serving in the US and double Uber's global volume. Didi has around 400 million monthly active users across its platforms in China. Meanwhile in 2019 Grab, which is the largest player in Southeast Asia, was completing about 5 million rides per day, while India's largest player, Ola, was doing about 3.6 million daily rides.
According to NACTO, in 2019, 136 million trips were taken on shared bikes, e-bikes, and scooters in the US, 60% more than 2018. Of these trips, 86 million were taken on electric kick-scooters, a more than 100% increase compared with 2018. The number of cities with dockless scooters increased to 109, almost doubling from 2018. Given this rapid growth, McKinsey predicts that the shared micromobility market could grow to be a $200-300 billion market by 2030 in the US alone, and up to half a trillion dollars when Europe and China are included.
Yet the proportion of overall trips done on shared bikes and scooters remains tiny. A survey conducted by the Washington Post found that during 2019, one in six people in Washington DC had used micromobility services and 7% of people across the broader DC area. However, when compared with the approximately 47 billion annual trips in the US of one mile or less, all micromobility trips account for less than 0.3% of such trips and just 0.06% of all trips.
The growth of the trip economy is supported by a long-term tailwind of rising urbanization. Cities are larger marketplaces with greater density to support a wider variety of better-quality trip services. Though in developed countries many people have moved out of dense urban cores to suburbs, more and more people have moved to metropolitan regions which have the density required to support the trip economy. Meanwhile in developing countries, there is a mass migration to cities and the global urban population is projected to grow from 55% to 68% by 2050, with 90% of this increase expected to come from Asia and Africa.
As the pool of specialty trips grows, the relative value of the general decathlete DOG bundle shrinks, but car ownership will continue to serve many trips. Vehicles such as e-bikes and e-cargo bikes also offer an appealing option as owned general use vehicles for everyday mobility that can coexist alongside the trip economy. But the trip economy is growing quickly and the key to accelerating this shift is to more deeply embed trip economies into the digital lives of users. That's where bundles come in.
In the context of mobility, technology and beyond, it is important to remember that people are not rationally transactional. Products must not only meet a need, but also work around the quirks of human psychology. Our irrationality creates opportunities as much as it creates obstacles. Catering to it effectively takes a unique blend of creativity, adaptation and luck.
What is impressive about a good bundle is that it transforms reality: consumers are often happier for the bundle even if it encases an incongruity. Starbucks somehow established and globalized expensive coffee, getting people to pay a substantial margin on what was before an effectively commoditized product. Cities now have coffee shops selling expensive coffee everywhere, but coffee wasn't always something that made sense for consumers to pay extra for. What's even crazier is that even though a key part of why people pay a premium for expensive coffee is the in-store experience bundled with it, they're still willing to pay the same premium on packaged Starbucks coffee sold in a supermarket! Before Starbucks, Pepsi realized that consumers are more willing to pay a premium for soda than food, and so acquired chains of quick service restaurants as a channel to market for soda sales. It is weird to think of fast food as a loss leader to sell beverages, but soda makes a 90% margin and the low prices on food are what get customers through the door. The fast food and Starbucks empires have been built around the car, with drive-thru lanes to smooth the process of making these irrational purchases. The intersection between these things incidentally helps explain why America has a problem with obesity.
Trip pricing and its supply side efficiencies mean that mobility has a fundamentally better value proposition. These services are often fun, convenient and empowering, but habits are hard to change. Like a healthy diet, the trip economy needs to change perceptions in addition to underlying economics. Trip providers are therefore constantly playing a cat and mouse game to overcome consumer psychology in a rapidly evolving digital landscape.
A smartphone puts trips in your pocket. Every new service has an app and the smartphone is the ecosystem from which they grow. In the process, it combines specialty mobility value propositions into a bundle that can substitute for the everyday mobility value proposition of car ownership.
While the smartphone is home to all of these value propositions, the way in which things are connected matters. Consumers don't like to download new apps while others are already deeply rooted. This digital ecosystem is like a jungle: occasionally a falling kapok creates space for new trees to grow, but for the most part it is easier for new branches to extend from already established stems. Successful apps with clear value propositions are like tall trees and new offerings that are looking to grow quickly can sometimes leverage these existing platforms.
About 5.2 billion people or two-thirds of all people on earth now have mobile devices and spend more than six and a half hours a day using them on average. Time spent on mobile is dominated by social media, messaging, video and gaming.
Apps where people spend most of this time are subject to a kind of gravitational force: if you use an app regularly you'll likely feel comfortable using it for more things, so long as these things are adjacent enough to the main thing you use it for. This plays out in a number of ways when it comes to mobility apps, some of which act like trees and the remainder of which act more like branches.
The first rule of mobility app gravitational dynamics is that more trip types is better than less. Uber started with just Uber Black, but saw rapid uptake when it added UberX. Other ride types (like Uber Green) have been added over time, but more importantly, Uber Eats and the acquisition of Jump expanded Uber's app offerings beyond ridehailing into food delivery and micromobility. Uber has also incorporated public transit into its app and is now adding grocery and alcohol delivery with the acquisitions of Cornershop and Drizly. In the process, Uber is building an ecosystem of trip offerings rather than a tool for ridehailing alone. Uber has also leveraged its platform strength to aggregate other mobility apps into its ecosystem. For instance, in Paris you can find both Lime scooters and Cityscoot mopeds on the Uber app.
Lyft has followed a similar strategy, expanding into micromobility through the acquisition of Motivate (which operates bike sharing services like Citi Bike) combined with its own scooter offering. Lyft has also launched a car rental product. However, it has struggled during the covid pandemic relative to Uber because it lacks an offering akin to Uber Eats, which surged even as ridehailing collapsed. Lyft is now working to address this by launching a food delivery service.
Ridehailing isn't the only potential launching point for bundled offerings. Lime tried carsharing and Tier offers mopeds in some cities (it acquired Bosch's Coup subsidiary). Sixt, which faces competition from ridehailing products, has in turn launched an app in a variety of German cities that offers ridehailing, car sharing, car subscription and scooter aggregation in addition to car rental. Sixt, which has a relatively small US footprint, has also partnered with Lyft to support its car rental product.
In China, Didi has expanded into a much wider array of services than its US analogs. Beyond rides of various kinds (taxis, private cars, luxury vehicles, corporate services, pooled rides), Didi also offers bicycles, DiDi Bus, drivers that can be hired to drive your car, Hitch intercity carpool rides (which have faced significant issues) and DiDi Foodie food delivery. It's now moving aggressively into grocery delivery.
When a mobility app has sufficient gravity through utilization velocity, it can expand beyond mobility. Didi's offerings include financial services like vehicle insurance, loans and crowdfunded medical insurance.
This strategy was pioneered by Tencent and Alibaba. WeChat for instance is not only the default messaging platform in China, but also a payments wallet which supports a panoply of other services and applications. Sending money through red envelopes is an ancient Chinese custom that was leveraged to popularize a now ubiquitous payments wallet integrate into WeChat. Digital payments through e-commerce and messaging gained massive traction since much of the population was unbanked. A lot of Didi's early traction and its ability to outcompete Uber in China came through Tencent and Alibaba's app ecosystem (both companies were early investors), but it has internalized these lessons to build similar services into its own ecosystem.
In contrast to China where messaging led the way, Southeast Asian mobility services were the first digital platform through which many customers started paying for things. To get a shared motorbike in Jakarta (colloquially known as an ojek), most people are willing to go to the trouble of giving cash to the corner store in exchange for credit on their Gojek app that can be used to pay for rides. But once you have this balance of credit on the app, you can use it to do your shopping and have it delivered, send something through the courier service function, pay a bill along with a range of other things.
Online payments in Southeast Asia are projected to rise from $1.4 trillion in 2019 to 2.3 trillion in 2025, with digital payments expected to increase 5x, driving a significant portion of the growth. GoPay reported $12 billion in transaction volume in 2020 and both Gojek and archrival Grab are actively investing and acquiring to support their digital payments ecosystems.
The payments velocity through their apps allowed both Grab and Gojek to become popular payments wallets in addition to their broad mobility offerings. These wallets can be used in stores to purchase goods or to purchase mobile phone data and these actions are even gamified in the app experience as "missions". Financial engineering is used to deepen investment into the payments ecosystem, including by bundling various offerings into "promo packs".
Wallets also naturally feed into a retail experience in which you can pay through the app and fulfillment is carried out by the mobility network. Gojek includes an online store through which consumers can browse offerings which are presented with promotions and discounts.
The ecosystem even extends into games and digital fitness, hosted by partner services.
And messaging is also built into the platform, in a way that allows for payments and extensibility. In this sense, the evolutionary journey of these platforms is somewhat the opposite of WeChat's (ending rather than beginning with chat), but incorporates many of the same learnings.
Grab hosts a similar array of services with payments at the center of the ecosystem (there have been sustained rumors of merger discussions between Grab and Gojek). More exotic Grab offerings include hotel bookings, movie tickets, insurance and gift cards.
Grab is even expanding into healthcare, partnering with Ping An Good Doctor to allow appointment bookings and medicine delivery through the app.
This potential for mobility services to be the driver for the creation of massive new digital ecosystems is fascinating to track as it rapidly evolves. A key factor for the emergence of such superapps, beyond mobility serving as an entry point into digital payments, is that mobility services have relatively broad appeal in emerging markets given the low rates of vehicle ownership and the relatively low cost of labor, which makes delivery and ojek ridehailing relatively affordable. In other words, lower labor costs mean that CATs can recruit the staff that they need with relatively little everyday mobility demand side competition from DOGs. As autonomy, electrification and new vehicle form factors enable cheaper and broader mobility value propositions, some of the possibilities demonstrated by these superapps will likely be imported to more mature markets.
A core value of mobility apps, especially in emerging markets, is their ability to serve as a convenient mechanism for transacting. This potential to act as a payment wallet creates a foundation for financial engineering. Creative financing solutions have long been an important tool for selling vehicles and are also quite helpful for selling trips.
There are two variations of trip pricing being rolled out that seek to shift the value proposition closer to that of ownership by offering a variety of benefits in exchange for a larger commitment from customers.
In addition to creating their own internal financial incentives, mobility apps can link into the branches of broader payment ecosystems. As mentioned, this was a key aspect of Didi's growth, leveraging Alipay and WeChat as ways to pay in addition to demand sources.
The trip economy could also link to existing payments infrastructure through widely used public transit payment systems such as London's Oyster card or Hong Kong's Octopus card. This kind of integration would allow regulators to collect data and subsidize externalities, especially when trips link into the transit system (e.g. using a scooter to reach a subway station). Access bundles are already used as a way to pay for transit within a defined geography (think zone passes) or time period (e.g. day pass, transfers within a designated period) and offer a natural pathway to link trips with public transit.
Employers are another financial input into transportation decision-making. Companies have long subsidized the cost of getting their employees to and from work. This often comes in the form of a company car (which also covers the employee's needs beyond the office) and at the very least, free parking onsite at corporate locations. But parking is not free, especially in large cities, and companies with offices in high value real estate zones are starting to shift their transportation spending to other modes of travel. New forms of trips combined with payment bundles let companies reallocate transportation subsidies to other modes and they can do so more precisely given the nature of the trip economy.
As we've seen, mobility superapps integrate e-commerce offerings. But e-commerce apps are powerful ecosystems from which mobility naturally extends. They not only incorporate payments but are already a key part of the trip economy even if they haven't been thought of in this way. And the ways in which they've grown give insights for other parts of the trip economy.
Fulfillment is the part of e-commerce that makes it a trip, but consumers hate paying for shipping. The clearest analogy to the general expectation of free shipping is the idea of free parking for cars. Of course, neither is actually free, but consumers often need to believe it for the value proposition to make sense. And in both cases, significant scale and organization is needed to effectively hide these costs.
This is the power of Amazon Prime: it is the preeminent example of an access bundle in action. It helped Amazon avoid what Eugene Wei calls an "invisible asymptote" or "product-market unfit" namely the point at which growth slows because the value proposition stops appealing to enough customers since they hate paying for shipping to "literally an irrational degree". By effectively pre-paying for shipping, "on some orders, and for some customers, the financial trade may be a lossy one for the business, but on net, the dramatic shift in the demand curve is stunning and game-changing."
Originally Prime meant free two-day shipping, but that has now come down to one-day shipping in many places. Some SKU-level wizardry is required to keep shipping windows optimally short: Popular items are identified and qualified while on the backend, the corresponding products have to be warehoused and fed into a distribution network that is constantly expanding and improving.
Amazon has also begun tackling adjacencies to e-commerce using Prime delivery as a key element of its attack. Specifically Amazon offers free two-hour delivery on groceries and essentials. Amazon originally launched this service as Amazon Fresh, but accelerated the rollout through the acquisition of Whole Foods, which gave it a physical retail footprint which helps strengthen the backbone of its grocery fulfillment network as its "first and best customer" for moving such goods. This is important because the delivery windows of shipping groceries are tighter and the logistical needs (including keeping items cool) are different. The acquisition of PillPack also allowed Amazon to launch Amazon Pharmacy and expand into delivery of prescription medication. It isn't hard to imagine Amazon expanding into other mobility verticals such as hot food delivery or even passenger trips, leveraging Prime to strengthen the appeal.
Amazon's main retail competitor, Walmart, has been forced to compete directly with Prime's core delivery value proposition, recently launching Walmart+, a bundle that is slightly cheaper but also skinnier. Walmart is likely to bulk up this bundle over time, but for now it includes free delivery "as fast as same day" as well as cashierless mobile phone checkout in Walmart stores and 5¢/gallon discount on fuel at Walmart affiliate stations.
What makes the Amazon Prime bundle particularly challenging for Walmart and others to compete against is how it is combined with an attractive media offering (Prime Video access, Amazon Music streaming and free access to a range of Kindle e-books). These are elements of commerce that Amazon absorbed as they became digital. Their virtual nature makes them fit well into a subscription bundle since their marginal costs are close to zero, helping to counteract the ways in which shipping costs scale more linearly.
Overall, Prime shows how effectively bundles can shift behavior towards trip modes when enough of the right elements are combined. But it also demonstrates how naturally Amazon and other e-commerce players could extend their ecosystem into new types of trips.
Maps are a search engine for physical space and are the ultimate mobility lead generation for outbound trips. That is to say, almost every time you open a map, you're considering making a trip. This makes maps a powerful trunk that naturally branches into mobility apps. Smartphone maps started by offering turn-by-turn navigation, which is still a key value proposition. But over time they've added public transit and Google Maps in particular has worked to incorporate other kinds of trips, including ridehailing, micromobility and food delivery. In some cases, it is possible to complete the entire transaction without leaving the Google Maps app. Google has also created APIs and integrations that help to route and manage a driver workforce.
Maps are hard to compete with since they have powerful network effects: as a map accumulates points of interest (POIs), the value to consumers grows, meaning more people use the map and add POI information - either passively by highlighting them through search or actively by filling in missing information - which in turn improves the map.
Trip operators face a tricky dilemma in choosing whether to integrate with maps: significant additional demand is hard to turn down, but it comes with the risk of aggregation. Looking further into the future, Google and Apple have invested significantly in autonomous technology and between them they own the two most popular mapping platforms. The viable alternatives are limited to HERE (owned by a consortium of German carmakers), TomTom and Mapbox. Similar dynamics exist in China where Alibaba, Baidu and Tencent own the leading mapping platforms.
It's noteworthy that Snapchat has a built-in map (powered by Mapbox) which allows you to explore what people are posting in particular locations around the world. It's also interesting how many of these videos show commutes. This is not to say that social media and mobility are particularly well integrated - Snapchat only lets you explore the world digitally through this map. But it is interesting to consider how mobility could tie our digital and real-world social interactions together.
While maps are about organizing information geographically, the social graph creates a different geometry: what matters isn't how far things are away in terms of physical distance but rather what are the human connections between them and how hard it is to bridge that gap. This is particularly relevant for social events: Facebook is where most people plan get togethers, so it shouldn't be too hard to tie trips into this loop. It doesn't matter so much where you're going as who you're going to.
This could extend in fun ways: in the heady early days of ridehailing, you could call an Uber through Facebook Messenger and there was an app called Bar Roulette that hailed an Uber to drop you and your friends at a random nearby bar. You could even send Snaps using an exclusive Uber filter en route. Those initiatives didn't last, but Bird and Bumble recently announced a more sober (though brightly colored) collaboration using decorated Bird scooters (which can serve quite effectively as outdoor billboards) to advertise the Bumble app, starting in Tel Aviv.
Messaging is of course also closely tied to all of this. Though platforms like WeChat are tied closely to mobility and payments, in the US and Europe messaging has tended to stay more siloed. As Facebook consolidates WhatsApp, Instagram and Messenger into a single backend and builds payments targeted at emerging markets into this platform, it seems that tie ins to mobility are likely to follow.
As the world was forced to stop by the covid pandemic, every aspect of mobility was impacted.
The segment of trips most dramatically affected was transit. In virtually every major city, public transit saw dramatic and sustained declines of around 50%. Even in places like Taiwan that have been successful in keeping major outbreaks of covid at bay, transit was heavily impacted, with Taipei seeing initial declines in ridership of around 40% and a decline of 20% through the end of 2020 according to Moovit data.
As car trips declined globally, congestion in cities decreased dramatically according to data from TomTom. The impact on driving was not uniform: weekday commuting and retail trips cratered as people avoided shared spaces such as offices and stores, but weekend trips increased as people found ways to get out of the house. For instance, in the Netherlands (where TomTom is headquartered), across six major cities weekday trips were down by more than 30% on average, while weekend trips were up by an even greater proportion. According to a Euromonitor survey, although many of the shifts to virtual platforms are regarded as temporary, working from home is regarded as a more permanent shift.
Meanwhile, inbound delivery has significantly substituted for outbound trips. E-commerce doubled its share of the retail market in the US, achieving a decade's worth of growth in the span of the first three months of the pandemic according to analysis by McKinsey. BCG found that even in segments that have been relatively slow to embrace e-commerce such as food & beverage and pharmaceutical delivery, gains have been substantial and significantly accelerated growth is expected as the pandemic subsides.
Food delivery has also surged, with sales up about 5x compared to 2018 levels with DoorDash capturing the lion's share of the growth, now holding 50% of the overall market. DoorDash reported 85% of order value came from new customers to the service by Q3 2020 and that it fulfilled 236 million orders in that quarter alone, almost triple the number of scooter trips in all of 2019. Meituan, which has also been boosted by covid even though China returned to relative normalcy more quickly, is currently fulfilling around 25 million orders per day, or about 10 times more than DoorDash.
Uber's business mirrored these broader trends: its outbound mobility business (primarily ridehailing) was down by 45% in the first three quarters of 2020 compared to the same period a year before, but its delivery business was up 100%, becoming larger than mobility and making up most of the difference. Micromobility had a mixed experience, taking an initial hit at the start of the pandemic, but bouncing back relatively quickly as people sought out safer ways to get around than transit.
On the whole, people have dramatically changed their movement patters. While outbound trips are likely to rebound as the pandemic subsides, many changes, especially dematerialization of work trips and digitally enabled retail are likely to be sustained.
Digitization changes how we access goods and services, but it sometimes also changes their fundamental nature. For instance, Amazon used to deliver physical CDs, DVDs and books to customers and these long tail categories were part of the complexity that allowed it to build a strong position in e-commerce. However, over time, Amazon started offering virtual equivalents of these products (MP3s, movie downloads and kindle e-books). Now, these product categories have become almost completely digital and available primarily through subscriptions and in Amazon's case, are all bundled into Prime.
Something similar to what happened to these product categories is now happening to entire categories of activity. Most of the places we travel to and spend our time outside the home are being supplemented with virtual equivalents and this process accelerated dramatically in 2020.
Since the beginning of computing, software has been used to virtualize certain aspects of work. Yet the proportion of workers that worked remotely had only reached about 5.5% of the workforce by 2018 compared with around 3% in 2001 according the US Bureau of Transportation Statistics.
Obviously, reality has shifted dramatically. A May 2020 survey conducted by the Stanford Institute for Economic Policy Research found that about 42% of US workers were working from home while about 26% worked on their business premises (including essential workers such as the 1.2 million now working for Amazon in the US). The remainder weren't working, a consequence of the acute recession.
Meanwhile, tools to enable virtual work and conferencing boomed. One year old startup Hopin, which enables virtual events, has skyrocketed to a $2 billion valuation in about a year since its founding. Zoom, which captured the zeitgeist, grew faster than Google Meet, but Teams has grown fastest of all, leveraging Microsoft's massive corporate customer base and salesforce. Already by April, Teams was being used by 30% of Office 365 customers. Microsoft's Productivity and Business Processes division (which houses the Microsoft Office productivity suite in addition to Skype and Teams), with an annualized revenue run rate of more than $50B, dwarfs high-flying entrants like Zoom and Slack, and Teams may already be contributing $5-10 billion of that.
Classrooms, like offices, were suddenly virtualized in 2020 with classes shifting almost completely over to video conferencing tools in the span of days. But the groundwork was already being laid long before, especially in higher education. In the US, overall postsecondary education enrollment stood at 28.3 million in 2012 and declined by about 7% between then and 2018. However the proportion of distance enrollment grew by more than a third to reach about 35% of overall enrollment, according to the National Center for Education Statistics (NCES). Digital substitutes have been growing steadily over this period, especially what are known as massive online open courses (MOOCs) through online platforms such as Coursera, EdX and Udacity. These platforms have surged during the pandemic, with the largest provider, Coursera, nearly doubling its user base to reach close to 80 million registered learners. Over the same period, educational app downloads have ballooned 90%.
According to McKinsey, telemedicine was growing at twice the rate of the overall healthcare market prepandemic although it constituted just 0.5% of all medical spending in 2019. While 11% of people used telehealth before the pandemic, now 76% of people are interested in it going forward. Medical providers reported a 50-175x increase in telehealth visits during the pandemic and providers are also shifting to approve significantly more virtualized treatments. Though a significant portion of treatments require physical presence, McKinsey estimates that 20% of the entire medical market could be virtualized.
Beyond substituting traditional medical interventions, health digitization enabled through wearable products such as the Apple Watch allows data collection and preemptive interventions before conditions reach a level of severity that prompts a trip seeking medical assistance. On the back of this trend, a growing array of apps are addressing the medical market and adjacencies like sleep monitoring. The market for such medical apps has been growing strongly and saw a 65% increase in downloads at the onset of the pandemic.
On an average day, Americans spend about 76 minutes on social media. Facebook holds the largest share of this time, but Instagram and Snapchat are both expanding. Part of the reason people spend so much time in these apps is that they are so engaging: Instagram is a bit like a mall where you can discover beautiful things, share them with your friends and buy them, but with a lot less driving and walking and more algorithmic curation.
The sticky elements of social media are being combined with the digitization of other aspects of our lives, transforming them in the process. Let's look at fitness and gaming.
Just as virtualized music, books, movies and the physical places that used to enable their consumption have been replaced by streamlined digital substitutes, so too gyms are facing the risk of replacement through in-home fitness virtualization. Peloton is the most visible disrupter in this space and has seen its sales increase by a factor of ten since the beginning of 2018 and its annualized revenues balloon to close to $2 billion. What is most impressive is that total workouts have grown at twice the rate of subscriptions, with the average subscriber doing more than 24 workouts a month. This has been driven by an expanding array of workout options coupled with carefully curated community-based gamification.
Like in other categories, games are increasingly shifting to smartphones with mobile games generating about half of all gaming revenue ($86.3 billion in 2020) and accounting for about two thirds of all mobile app revenue. Around 1.75 billion people now play games on their phone, an increase of 46% year over year.
Many successful mobile games (think Candy Crush Saga or Coin Master) simply have narrow feedback loops and casino elements that make them addictive and good at making money. However, other games are creating vast and interconnected worlds which are engaging and immersive. Just as technological building blocks have been implemented to offer an increasing array of new trip marketplaces, so too tools like messaging and video communication are maturing to allow gaming platforms to be social in a meaningful sense. In the process, games like Fortnite and Roblox are becoming places to hang out with friends or meet new people. They are even a context that can absorb real world experience like music concerts with each game's particular qualities animating the show. Recently Fortnite hosted Travis Scott (generating a rumored $20 million) and Roblox hosted Lil Nas X (you can even buy a Lil Nas X Roblox game avatar). Minecraft concerts have been a thing for longer. These social experiences are supported by communication platforms like Discord and streaming services such as Twitch that have both grown rapidly alongside this new category of games.
Minecraft and Roblox both passed 100 million monthly active users by the beginning of 2020 while Among Us shot up from close to nothing to about half a billion users (including AOC and Ilhan Omar) in November 2020 to become the most popular game ever according Nielsen's SuperData. Neither Fortnite nor Pokémon Go disclose their monthly active users, but annualized revenues in 2020 were similar to Roblox at over $1 billion. Discord and Twitch also both passed 100 million users in 2020.
Overall, the evolution of technology is making it harder to draw strict borders around gameplay, social interaction and work, causing unexpected consequences along the way. Using Slack for work and Zoom to view a webinar are different social contexts to playing Roblox while chatting on Discord or watching Starcraft streamed on Twitch, but they rely on very similar technological underpinnings. It's not a big step from there for millennials in pajamas to use Discord, Reddit and Robinhood to trade up GameStop shares to short squeeze professional investors and send shockwaves through Wall Street.
Meanwhile social media has completely transformed the way in which politics operates and is a big part of what brought the US to a point of political crisis under Trump. More quietly, it may also be causing a dramatic increase in teenage suicide especially amongst girls.
Technology democratizes access in a way that makes some things better while making other things worse. These rapid shifts involve systemic risk and opportunity which is hard to quantify and predict, making for a bumpy ride.
Dematerialization has a complex relationship with the trip economy: in some cases, like with remote work, it can completely replace a trip with a substitute digital experience or workflow. This shift has been accelerated by the pandemic, forcing us to get used to such virtual places and also to appreciate their enduring advantages.
But dematerialization can also change the nature of trips or even create new kinds of trip demand and lead back into the trip economy. For instance, Amazon Prime Video leads more people to pay for Prime, which in turn allows them to justify ordering more through Amazon, stimulating trip demand even as some trips are dematerialized along the way. Or Instagram, which allows you to virtually visit beautiful locations also causes you to want to visit them in person. Social media, like television before it, is a channel for introducing people to new products so that they buy them, whether it's through an outbound or inbound trip. Meanwhile, most people now meet their romantic partner through a dating app, but those connections lead to real world meetings which require trips.
Virtual gaming ecosystems can have even more interesting interactions. Take Pokémon Go, which was the coolest thing in 2016, as an example. Though the hype has faded, it has certainly not disappeared, with revenues growing to over $1.2 billion in late 2020 (see chart above). The game has strong social elements and is best played in groups. What is interesting from a trips perspective is that it causes players to leave their homes and increases visits to businesses such as stores and restaurants that have gameplay elements like PokéStops attached to them.
Dematerialization also powers the trip economy: ridehailing is a virtualized version of flagging down a cab and has significantly expanded the size of the overall market for rides in the process. E-commerce dematerializes the retail experience but requires fulfillment through the trip economy. And when time that used to be spent waiting in line at a checkout counter is unlocked and new kind of trips can easily be requested in more efficient packaging, new demand can be created even as reasons for past trips become virtualized. The expansion of digital tools and payment technologies coupled with market innovation also means that retail as a category and transacting as a process is expanding to cover an increasing swath of other trip categories.
This complex interaction between trips and dematerialization plays out in a fascinating way within superapps since they power trips but also virtual experiences such as gaming. By removing barriers to demand generation while simultaneously dissolving old categories of trips, dematerialization acts not just as a branch or the stem of a tree but also like a wormhole embedded into its trunk, leading to other worlds and back somewhere else in the physical one. Over time, the trip economy and virtual worlds are increasingly being tied into an appealing set of interlinked digital bundles that can more effectively substitute the everyday mobility value proposition of car ownership (while in the process also shifting the nature of the world that it helps connect).
The trip economy has evoked significant skepticism. Mobility services - from ridehailing to micromobility to food delivery and beyond - are all tough businesses when it comes to unit economics. They consume an incredible amount of capital since they require large outlays to finance fleets and to incentivize shifts in consumer behavior to form new mobility habits.
Mentions of profitability in Uber earnings reports are like sunshine in London: on the horizon but constantly obscured by rain and clouds. During the first half of 2019, Bird lost $100 million on $15 million in revenue while Lime's valuation buckled 79% in early 2020 although now the company brightly claims it can be profitable in 2021.
Some go so far as to argue these companies are fundamentally unsustainable with irrational business models fed by our distorted zero interest rate environment. These reports of imminent death juxtaposed with rosy projections of profitability and investor excitement can be referred to as the Schrödinger's CAT Paradox of the trip economy.
As investor fervor has extended into public markets, the Schrödinger's CAT Paradox has taken a new eschatological form in which electric DOGs will transform into fleets of autonomous CATs riding on a prophetic ARK. It is a wonderful and mysterious world we live in.
Schrödinger's CATs are a consequence of an important difference between the digital world and the trip economy: digital experiences can be scaled massively since they have low marginal costs. In contrast, mobility services operate in the physical world: they experience incredible demand since they solve urgent consumer needs, but face steep hurdles in scaling supply while still operating in a cost-effective manner. These challenges are why bundling them into digital ecosystems is critical for them to take root, but also why rapid evolution of new form factors and broader supply chains is critical. We consider this next.
Share this chapter with others: