Growth, Innovation + Investing
from A Round to Apple Inc.

We’ve been delivering authoritative market analysis and perspectives, week-in and week-out for more than a decade. Sign up so we can update you with new releases of A2Apple and our other daily and weekly publications — sent straight to your email.

And make sure to keep a close eye on our research and commentary arm, GSV Tomorrow, where you’ll find original GSV insights, plus curated content revolving around our key Growth Themes. Just complete this form and they’re yours to use and share.

GSV's weekly perspective on Growth, Innovation + Investing...
from A Round to Apple Inc.
Marketplaces
February 4, 2018

The Fast Mile

Save to PDF

Market Snapshot

Indices Week YTD

During the time of Christ, the human population on Earth was at a steady 200 million. In the 1800s, at the height of the industrial revolution, the human population passed the 1 billion mark. But today, about 200 years later, the population boomed to be over 7 billion. And by 2050, it’s expected to rise to 9.7 billion.

Movement is in our DNA. Until about 10,000 years ago — or 99% of human history — there were few, if any homes or villages. People were nomadic, chasing food and gentler climates. While we couldn’t change the weather, we learned how to domesticate plants and animals in what is now called the Neolithic Revolution. And when the food stopped moving, so did we.

In the next 10,000 years, historians might look back and name our era the “Metropolis Revolution.” Today, the United Nations estimates that four billion people, or 54% of the World’s population, live in cities. In the next 15 years, the Economist projects that urbanization will increase average city density by 30%. By 2050, the ranks of urban dwellers will swell by 2.5 billion to nearly two-thirds of Global population.

THE POPULATION OF GLOBAL CITIES IS SURGING
Total Global Population Living in Urban Areas

Source: The Economist, United Nations

Around 80 million people annually move from rural to urban areas, going off the farm and into the harm. As such, the number of megacities — cities with a population 10 million or greater — has doubled in the past two decades, from 14 in 1995 to 31 in 2017. It’s estimated that by 2100, over 80 cities across the World will have a population over 10 million.

SHANGHAI SKYLINE, 1990 VS. TODAY

The rise of Global urbanization, coupled with a corresponding increase in the number of vehicles on the road, has pushed city traffic to the limit. In Manhattan, during UN Week, driving in traffic from Plaza Hotel to the Midtown Tunnel — roughly a mile — takes longer than flying from New York’s LaGuardia to Boston.

In Mexico City, for example, the city with the most traffic congestion in the World, drivers spend two thirds of their time in the car in gridlock. And with a population of eight million, Mexico City isn’t even considered to be a “megacity” (10+ million population).

MOST CONGESTED CITIES IN THE WORLD
10 of the 25 Most Congested Cities Are in China

Source: TomTom Traffic Index, United Nations, U.S. Census, World Bank, GSV Asset Management

Americans spend over $2 trillion per year on car ownership — more than what we shell out for food. But shockingly, the 250 million cars in the United States spend 96% of the day parked. In other words, there are 240 million cars parked at all times. As Lyft co-founder John Zimmer has observed, BMW doesn’t make the “Ultimate Driving Machine” — it makes the “Ultimate Parking Machine”.

ULTIMATE PARKING MACHINE: AMERICAN CARS ARE PARKED 96% OF THE DAY
Percentage of Time U.S. Cars Are Used for Key Activities

Source: Dr. Stefan Heck (4/1 Presentation: Driving Growth with Big Ideas — Private Capital and Global Innovation), GSV Asset Management

But the millennial generation don’t look to cars as their sustainable source of transportation. Coupled with the rise of ride-sharing platforms, the demand for car ownership among Millennials has plummeted. According to Bloomberg industry research, last year, more automobiles were sold to people aged 75 and older than 18-24-year-olds.

As is often the case, the greatest problems create the greatest opportunities — the bigger the problem, the bigger the opportunity.

And thats where the “Fast Mile” comes in. Arising at the intersection of several megatrends — the sharing economy, smartphones, urbanization, sustainability, and on-demand services — the Fast Mile encompasses solutions that will add efficiency to the problem of last mile transportation in congested cities.

THE FAST MILE: COLLIDING MEGATRENDS

Source: GSV Asset Management

BIG WHEEL, BIG DEAL

Before the invention of the wheel, 99% of the human population lived, worked and died within a 5 mile radius of where they were born.

But the invention of the wheel around 3500 BC led to a new era of human mobility. Horse drawn carriages were adopted a main source of transportation and bikes — which were first invented in the 1810s — empowered the general population with the option of mobility.

Since then, the core “technology” behind the bicycle has remained largely unchanged. On the flip side, the problem with last mile transportation has only grown exponentially as the population of cities continue to rise. Humanity has and will continue to face the challenge with how to get atoms from Point A to Point B. And while we’re a way off from beaming to places through Star Trek’s Transporter, the rise of fast mile services like shared bikes provide an convenient, affordable and eco-friendly alternative until then.

Bikes were first set free in 1965, when an anarchist group in Amsterdam painted bikes white and left them unlocked for anyone to use. Flyers were posted all over the city stating that “the white bike is never locked.” But eventually they were. The initiative was shut down after most of the bikes were either stolen or badly damaged.

Modern bikeshare systems have been in existence in major metropolitan areas for the past decade. But they never took off, as precursor services lacked convenience, were location specific, and were expensive — usually required a safety deposit of $100 or more. Since 2010, traditional docked bike-sharing services only amassed 88 million trips in the United States.

But suddenly, less than one decade later, bike-sharing has exploded into major markets, with dockless shared bikes seemingly popping up everywhere. What’s changed since then?

GLOBAL BIKESHARING TIMELINE

Source: CityLab, GSV Asset Management

Until recently, they haven’t been “smart.” Think about what Uber did to the black car industry. Believe it or not, Carey Limousine was a public company and its claim to fame was providing access to a Global network of high-end taxis. It was a network. Carey didn’t own any cars. Sound familiar?

In a similar way, an emerging group of bike-sharing startups are achieving massive scale — and they’re backed by serious investors. These companies are capitalizing on apps that enable users to easily locate, unlock, use and return bikes through an app.

Traditional bike-sharing services allowed riders to pick up bikes from docks scattered around metropolitan areas. Riders had to ensure that they picked up and returned the bikes to specific locations, otherwise they would be surcharged for their rides. While this service was useful, it didn’t allow riders the full flexibility to use bikes to get from point A to point B.

OLD VS. NEW: DOCKED + DOCKLESS BIKE-SHARING

On the flip side, dockless bike-sharing allows riders to pick up and drop off bikes wherever they want, whenever they want. Riders are able to unlock bikes remotely using a mobile app, ride the bike to their destination, and then leave the bike there for the next rider to use. Unlike traditional bike-sharing services, dockless bike-sharing grants riders the full flexibility to use bikes at their convenience. 

THE GLOBAL LAND GRAB

China is home to 10 of the 25 most congested cities in the World. So it’s not a surprise that the two leading bike-sharing companies — Mobike and Ofo — were born in Beijing. They raised over $1.9 billion alone in 2017. Why? In two years alone, these companies have amassed over 100 million users who are completing 25 million rides per day.

What differs between the two services comes down to the experience. Mobikes are considered to be higher-end “vehicles,” costing $440 to build. They have built-in GPS trackers, look like they belong in a SoulCycle class, and cost two yuan an hour to rent.

THE RACE IS ON
The Dominant Bike-Sharing Leaders: Ofo (Yellow) and Mobike (Orange)

Source: The Wall Street Journal

Ofo’s bikes feel much cheaper, because, well, at $35 to build, they are. With no added bells and whistles, Ofo’s bikes rent for half the price.

While dockless bike-sharing has hit major Asian markets, in Europe and the United States, the movie is just beginning. Global and homegrown players are rushing to secure the prime position in these markets. Mobike (China), Ofo (China), LimeBike (US) and Spin (US) are leading the land grab.

LEADING BIKE SHARING STARTUPS

Source: CB Insights, Crunchbase, Company Disclosures, GSV Asset Management

California-based LimeBike’s bright green bikes are now populating the streets of the United States. Founded in 2017, the company officially launched on the University of North Carolina-Greenboro’s campus. The company now operates in 35 markets — ranging from large metropolitan cities like the Bay Area, Los Angeles, Seattle and Washington D.C. to college campuses. Late last year, LimeBike announced it surpassed one million rides, and it is poised to launch in Zurich and Frankfurt.

Interestingly enough, one of the biggest beneficiaries from the bike-sharing boom is the bike manufacturer. Battle FSD, the World’s largest manufacturer of bikes, is partnered with and supplies bikes to many of the new bike-sharing providers, including Ofo and LimeBike. The company currently produces over 26 million bikes annually — a number that is expected to significantly increase as the bike-sharing industry matures.

WHAT’S NEXT

In a few short years, bike-sharing has become the fastest growing sector of the Sharing economy. Venture funding towards bike-sharing startups exploded in 2017, largely driven by the megarounds raised by Ofo and Mobike. According to Pitchbook, $386 million of investment dollars went towards the bike-sharing sector in 2016. In 2017, that number ballooned to $2.8 billion.

GLOBAL BIKESHARING ANNUAL FINANCING TREND

Source: PitchBook

Low barriers of entry and quick payback allowed for these bike-sharing companies to scale rapidly. Much like the nascent ridesharing industry, the bike-sharing industry will begin to mature as companies hit market saturation and face competition and heightened regulation from local governments.

Ridesharing Meets Bike-sharing

The Cambrian explosion of bike-sharing has created two Unicorns in three years and the birth of no less than 40 bike-sharing companies. In China, more than 2 million bikes are available from 15 bike-sharing companies. In the US, metropolitans like DC and Seattle are seeing companies like Mobike, LimeBike, Ofo, Spin, and Jump compete for customers. And even Paris, arguably the birthplace of bike-sharing, has been China-fied, with four leading Chinese bike-sharing services taking over the City of Lights.

OFO PEDALING INTO PARIS

Source: Getty Images

Much like the early days of the ridesharing industry, the bike-sharing industry will eventually consolidate as major players acquire other services to gain market share in new regions. And it already begun with Chinese firms Youon Bike and Hellobike merging last fall to compete against Ofo and Mobike. Mobike and Ofo themselves have been rumored to begun talks of a merger, though the reports are untrue as both companies are intensely focused on building global market share.

Ridesharing companies are taking on bike-sharing, with almost all of the Global players expanding into the sector. Since inception in 2012, Go-Jek has been the leader in “motorbike sharing” and has grown to be a $3+ billion dollar company from a fleet of only 20 bikes. This January, Go-Jek announced a $1.2 billion investment from Google, Singapore’s Temasek, and China’s Meituan-Dianping.

In the past two months, Dubai’s Careem, India’s Ola, and Southeast Asia’s Grab added bike-sharing services to their platform. In January, on the heels of its acquisition of the struggling Bluegogo, China’s DiDi Chuxing revealed that they too will launch their own bike-sharing platform.

And this week, Uber announced Uber Bike, a pilot which will launch in San Francisco next week in partnership with JUMP. Uber customers will be able to book JUMP bikes through the Uber app.

E-bikes + E-scooters

Down the line, keep an eye out for the proliferation of other fast mile services, such as dockless electric bikes and scooters. In San Francisco, Jump Bikes is launching their fleet of electronic dockless bikes under an exclusive 18-month permit.

LIMEBIKE’S E-BIKE: LIME-E

Source: LimeBike

Motivate and LimeBike both announced the launch of their e-bikes earlier this year. Motivate will make their electric fleet available through the Ford GoBikes system starting in April. LimeBike will start rolling out their Lime-E fleet this month starting in Seattle and Miami.

EMERGING SHARED “E-BIKE” SERVICES

Source: CB Insights, Crunchbase, Company Disclosures, GSV Asset Management

And companies are going beyond electric bikes to provide other shared electric transportation services. Scoot is currently operating a fleet of shared scooters in San Francisco, allowing commuters to easily travel through the city in Cherry red Vespa-type scooters. Down the coast, Santa Monica-based Bird currently provides electric scooters in the Los Angeles region.

Regulation

Earlier this month, San Francisco signed an agreement to allow JUMP Bikes have exclusive rights to operate in the city, shutting out competitors LimeBike, Ford GoBike , Spin, Ofo and Mobike completely for 18 months. But that may be too little, too late, as San Franciscans have already taken notice — and reacted — to the influx of colorful bikes on their streets.

SAN FRANCIS-NO
Vandalized Shared Bikes in San Francisco Discarded in Streets, Lakes and Trees

Source: Mercury News, CBS

Before dockless bike-sharing services have gotten away with negotiating with and getting operating permits from local governments. Companies now are under heightened scrutiny as cities take notice. For example, Santa Monica’s Bird recently got hit with a lawsuit by the city of Santa Monica for their dockless scooters.

And in China — the “Kingdom of Bicycles” with over 20 million shared bikes — the problem is exponentially worse. Shares are often discarded, stolen or heavily vandalized on the streets. One bike-sharing startup Wukong reportedly lost 90% of its bikes in the first six months. Already, cities are beginning to crackdown on issue. In Shanghai, the city hauls away thousands of bikes at a time that clog the streets and sidewalks. And in Xiamen, a bike graveyard exists that is filled with yellow, orange and blue bikes.

XIAMEN’S BIKE GRAVEYARD

Source: The Guardian

And to make things worse, Chinese bike-sharing companies wage wars against each other, purposely sabotaging each other by displacing one another’s bikes. As written in the New York Times:

One morning recently in Shanghai, I caught a glimpse of some suspicious behavior. I was walking down a tranquil, tree-lined street when a muscular man lumbered past carrying two orange-and-silver Mobikes. As he swept by, a wheel touched the ground and set off an alarm, causing him to heave the bikes even higher in the air. The man was not a bike enthusiast, but he wasn’t a thief, either. As I watched him slip down a side alley and emerge moments later empty-handed, I realized that he was a foot soldier in the bike-sharing wars, dumping competitors’ bikes in hard-to-find places. Rounding the corner, I saw the result of his handiwork: a sea of bikes in almost every hue. Yet not a single orange-and-silver Mobike was in sight.

But what’s most interesting here — and arguably how bike-sharing companies attract so much venture funding — is the Big Data play. Cities and businesses are enticed to work with bike-sharing companies to unlock the valuable data that these companies glean from their customers.

MOBIKING IN BEIJING
Map Depicting Saturation of Mobike Trips in Beijing (on September 5th, 2017)

Source: The Economist

Already, Mobike shares its data with think-tanks, universities, research institutes, and the World Bank to support city-planning development efforts. Ultimately, daily transit data on the movement patterns of millions of citizens could result in the creation of powerful initiatives that cut emissions, reduce congestion, and increase quality of life.  

Bubblin'

by Luben Pampoulov

Unlock The Gates

The IPO market started with a bang this year. Some 20 IPOs already priced on the public market, with the average performance at a solid +18%. At such a pace, we could be heading for 250+ IPOs this year — but it is too early for such predictions.

Also very appealing are the type of IPOs currently in the backlog — namely Spotify and Dropbox. With a bang, the two companies were reported to have filed back-to-back in the first two weeks of January. The filings are confidential, hence not available to the public, but we know that both companies are fundamentally strong and have significant revenue and users scale.

On January 4th, Spotify announced it hit 70 million paying subscribers, which was up from 40 million in September 2016, and up from close to 50 million at the end of 2016. In its 2016 annual filing, Spotify reported €2.9 billion in revenue, with the big majority of that coming from subscriptions. In other words, there should be strong correlation between paid user growth and revenue growth, in our opinion.

Spotify’s margin structure is another key metric; in 2016, the reported gross margin was 15%, but this past year, Spotify re-negotiated contracts with all the major labels — Warner, Sony, Universal, and Merlin. Accordingly, we expect there is an improvement in the gross margin, not only due to top line growth, but also because of the new royalty deals.

It is interesting to look at Spotify’s public comps, and how they trade. Netflix, Roku, Sirius XM and Pandora are all very similar to Spotify in terms of business model. All of them provide on-demand entertainment, on a freemium subscription basis. But while Netflix, Roku and Sirius XM are all growth companies with improving fundamentals, Pandora is a struggling business with broken fundamentals — having been disrupted by Spotify, in our opinion.

Spotify Comps

Source: Yahoo Finance, Public Filings

Dropbox has not been as transparent on revealing its user numbers, and it last reported hitting the 500 million user mark in March of 2016. We expect Dropbox is now close to 700 million users, but we will have to wait and see in their S-1 filing once it is public. But the company did disclose some important and significant milestones last year; in January, CEO Drew Houston said they hit $1 billion in annual revenue run-rate — and were the fastest SaaS business to do so! And three months later, Houston also reported Dropbox was EBITDA profitable, which was another impressive achievement given most of its competitors were/are still in the red.

While the vast majority of Dropbox’s 600-700 million users are still using the platform for free, the paid conversion rate is increasing, driven by a strong push into businesses. The potent mix of a dynamic freemium engine that is feeding a lucrative enterprise business creates powerful advantages for Dropbox, in our opinion.

Also, an estimated ~90% of the company’s revenue is inbound and organic — meaning it is not driven by marketing spend. This contrasts with competitors such as Box, Atlassian or Workday who derive the majority of their revenue through their salesforce. If Dropbox manages to grow its top line at a strong 30%+ rate, then we expect it will trade at a premium multiple to most of its comps.

Dropbox Comps

Source: Yahoo Finance, Public Filings

Somewhat surprisingly, another big 2018 IPO candidate pushed its going public plans to 2019; For a long time, Airbnb has been rumored to do an IPO, and we had expected they would chose to do so in the coming months. Most recently, the company hit $1 billion in revenue in Q3’17, and was growing at an impressive +100% while being profitable for 17 months, according to sources.

But ongoing internal conflicts between CEO Brian Chesky and CFO Laurence Tosi led to Tosi’s departure last week, therefore also postponing the company’s IPO plans to next year: “I know people will ask what these changes mean for a potential IPO,” Chesky said. “Let me address this directly. We are not going public in 2018. Our primary focus is becoming a 21st-century company and advancing our mission. We’re working on getting ready to go public and we will make decisions about going public on our own timetable.”

Pioneer Notes

by Li Jiang

The Master Algorithm

How the Quest for the Ultimate Learning Machine Will Remake Our World

I wouldn’t do Professor Pedro Domingos justice by trying to describe his entire book in this blog post, but I did want to share one core thought as I reflect on his book.

Domingos’ core argument is that machine learning needs an unifying theorem, not unlike the Standard Model in physics or the Central Dogma in biology. He takes readers through a historical tour of artificial intelligence and machine learning and breaks down the five main schools of machine learning (below). But he argues that each has its limitations and the main goal for current researchers should be to discover/create “The Master Algorithm” that has the ability to learn any concept aka act as a general purpose learner.

As with any great book, it leads to more questions than answers. My main question, as applied to startups, is this:

What’s the speed at which machine learning is improving?

Why is this an important question?

For the past several decades, the category defining companies from Intel to to Apple to Google to Facebook have benefited from 2 core unifying theories of technology.

First, Moore’s Law created the underlying framework for the speed at which computing power increases (doubling every two years or so) that has directly enabled a generation of products. Products that were at first bulky and expensive, such as room-sized mainframes, were able to ride Moore’s Law and become smaller and cheap, leading to mass products like phones and smart watches.

Second, Metcalfe’s Law governed the value of a network of users (n(n − 1)/2) that has enabled a generation of Internet services to effectively serve the majority of the world’s Internet population. As more users join a network, their value grow exponentially while costs generally grow linearly. This incentivizes even more users and the flywheel is set in motion.

So now the question is…is there a third “Law” that governs the speed of improvement of machine learning.

In Lee Kai-Fu’s (李开复) commencement speech at Columbia, he gives hints at this.

In speaking about his investments in now publicly listed Meitu and two other AI investments, he notes that in all three cases, the AI technology underlying the startups went from essentially not useful to indispensable.

The three software companies I mentioned earlier, when they were first launched: often made people uglier, lost millions in bad loans, and thought I was some talk show celebrity. But given time and much more data, their self-learning made them dramatically better than people. Not only are they better, they don’t get tired nor emotional. They don’t go on strike, and they are infinitely scalable. With hardware, software, and networking costs coming down, all they cost is electricity.

In god we trust, all others bring data…

To put some data behind it, if we look at the ImageNet Challenge, AI image recognition technology has improved 10X from 2010 to 2016, catalyzed by the introduction of deep learning methods in 2012.

On the backs of this “Law” of machine learning improvement, we’ll see a Cambrian explosion of new products and services that fits Kai-Fu’s description of products that are at first flawed, but with time and data, become essential and, for all practical purposes, perfect.

Questions, not answers

The ImageNet Challenge and image recognition is just one application of AI so it doesn’t give us enough to say what the “Law” of machine learning improvement is. I can’t say AI is doubling in intelligence every 18 to 24 months or that AI gets exponentially better by a factor of n(n − 1)/2 with each data point.

But I do think a particular “Law” governing the rate at which AI is improving exists and I can’t wait for someone in the field to articulate (or solve) it.

Because understanding the speed at which artificial intelligence is getting more intelligent will allow us to understand the third major foundational wave, in addition to Moore’s and Metcalfe’s Law, that will bring us the dominant companies and the brilliant products of the Age of AI.

Market Update

Week ending February 4, 2018

World Indices




U.S. Indices Snapshot

Valuation P/E Est. P/E/G Price/Sales
LTM NTM Growth LTM NTM LTM NTM

Saving...