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.
General Commentary
January 22, 2017

Cracking The Code

Save to PDF

Market Snapshot

Indices Week YTD

Interest in gene exploration dates back nearly 200 years to Gregor Mendel, the father of modern genetics. But it wasn’t until the mid 1980s that technology and Moore’s Law made it possible to use software to sequence the over three billion genetic bases that make up the human body.

Unprecedented in terms of international collaboration and public-private partnerships, the Human Genome Project was launched in 1990 with the goal of mapping the entire human genome in what seemed like an unrealistic 15 years.

In April 2003, it was announced that the entire human genome had been mapped at a cost of $3.8 billion. Beyond scientific gains, the ROI was staggering. The Human Genome Project generated an economic impact of $796 billion from 1990 to 2010 alone.

Moore’s Law and the Megatrend of “Software Eating the World” are having a revolutionary impact on the cost for an individual to receive their own genetic map. In 2008, it cost $1 million dollars for a person to create a personalized gene map. By 2011, it was $100,000. In 2014, it cost $1,000 — less than a chest X-Ray — and today you can get a personal genetic report for less than $300.

Rapid Decline of Cost to Sequence a Human Genome
Cost to Sequence a Human Genome Outpaces Moore’s Law, 2001-2014 (Log Scale)

Source: National Human Genome Research Institute, GSV Asset Management

Today, your mechanic knows more about your car than your doctor knows about your body. When you bring your car into the shop, the first step is usually to plug it into a computer that scans performance data and runs a battery of tests. In a World where everyone can map their personal genome, healthcare gets a whole lot more personal, efficient, and effective.


Anne Wojcicki co-founded 23andMe in 2006, on the heels of the Human Genome Project, after a decade on Wall Street as a health care analyst convinced her that the way we treat illness and create new drugs is fundamentally broken.

A chance encounter with Markus Stoffel, a leading molecular biologist at ETH Zurich, left her convinced that the key to creating a new health and medicine paradigm was to aggregate the World’s genetic data and discern patterns to prevent and combat diseases like Parkinson’s and Alzheimer’s. Big Data meets Big Pharma.

23andMe launched with a mail-order test kit that generated a detailed genetic report for $1,000. A simple saliva sample produced insights across nearly 200 categories — including risk factors for inherited diseases like cystic fibrosis, genetic traits like lactose intolerance, and various details about your genealogical history. By 2012, 23andMe had driven the price down to $99.

In 2013, Uncle Sam dealt 23andMe a potential death blow. The Food and Drug Administration ordered the company to stop marketing its flagship tests, deeming them unregulated medical devices. Regulators argued that consumers could misinterpret this health data, which had not been clinically validated, and take action based on a “false positive” result.

Pioneers: 23andMe
Anne Wojcicki, Co-Founder + CEO, 23andMe (Left) Interviewed by Katy Steinmetz, San Francisco Bureau Chief, Time Magazine, at the 2015 GSV Pioneer Summit

Source: GSV Asset Management

While many would have wilted, Anne Wojcicki played offense. With its popular product sidelined, 23andMe partnered with pharmaceutical companies like Genentech and Pfizer, trading access to its DNA database in exchange for upfront payments and a cut of revenue from new drugs developed using it. At the same time, the company worked with the FDA to get an approved genetic test back on the market.

In late 2015, with the required Federal approvals in hand, 23andMe relaunched its genetic reporting service targeting four categories: ancestry, wellness, traits, and carrier status. Today, the company has over one million customers. Backed by NEA, Google Ventures, Fidelity, Genentech, and Illumina, 23andMe is valued at $1.1 billion.

Emerging Personalized Genetics + Health Testing Companies

Source: Fusion, CrunchBase, GSV Asset Management

The fact that everybody can and should have their own genetic profile has profound implications to the future of medicine and human life. This isn’t science fiction.

Instead of the normal reactive treatments patients receive under “modern medicine,” genetic mapping unlocks the potential for personalized prescriptions that could mean taking a new or modified drug, changing your diet or fitness routine, or any other combination of treatments that are tailored to your specific needs.

Today, over half of 23andMe customers learn something from their test results that is “medically meaningful” — from a disease risk to a dietary trait. Understanding risks and health conditions enables people to take preventative measures, and as tests continue to improve, health and wellness service providers will be better equipped to help people avoid or stave off risk.

Color Genomics, which is backed by Khosla Ventures, Formation 8, and Emerson Collective, sells a $250 kit that lets people find out if they are at risk for certain hereditary cancers. Then it connects them with genetic counselors to analyze the test results and discuss prevention, monitoring, and treatment options. As CEO Othman Laraki observed in a recent interview with Recode, “We have a torrent of data in our bodies… and today, we make use of almost none of it.” That’s changing, fast.

For $250, Color Genomics enables people to test for hereditary cancers and discuss prevention options with a certified genetic counselor.

Source: Color Genomics

Helix, which launched in 2015 with $100 million from backers including Warburg Pincus and Illumina, aims to create an “app store” for personalized health apps, powered by your genetic map.

As reported by MIT Technology Review, the company’s idea is to collect a spit sample from any new user, sequence and analyze their genes, and then digitize the findings so they can be accessed by third-party software app developers. Helix calls the idea, “sequence once, query often.”

Sequoia-backed Assurex Health, which was acquired by Myriad Genetics in 2016, uses DNA gathered with a simple cheek swab to analyze a patient’s genes and provide individualized information to help healthcare providers select medications that better match their patients’ genes. Initial clinical studies have shown that when care providers use Assurex to guide personalized treatment decisions, patients are up to twice as likely to respond to the selected medication.


Interestingly, 23andMe may have provided the clearest window to the future when it was on death’s doorstep, courtesy of a crackdown by the FDA. As we described above, the company survived by selling access to its data set to Big Pharma companies.

Why were they such eager buyers?

23andMe has assembled the World’s largest database of genetics and phenotypic data. Mining this data can help drug developers unlock insights about how various treatments impact people across genetic profiles. As pharmaceutical R&D increasingly integrates massive genetic datasets, it will enable the development of tailored drugs that are more effective for sub-populations, or that minimize the risks of adverse reactions.

A benign example is Benadryl, which makes some children sleepy and others hyperactive. In the future, parents will be able to pick the Benadryl “flavor” that doesn’t make their kids bounce off the wall.

Today, 23andMe has contracted with over 13 drug companies to access its data. Genentech, alone, anted up $10 million to study the genes of people with Parkinson’s disease. Incidentally, in 2015, 23andMe hired hired Genentech’s former research chief, Richard Scheller, to lead an in-house therapeutics research strategy. The company, other words, could be moving up the value chain. Look out Big Pharma.

A wave of emerging startups are attacking the opportunity to apply big data analytics to pharmaceutical development. Zephyr Health, for example, enables Life Sciences companies to extract insights from disparate global health data to streamline the drug development process (which typically takes 10 years and costs $4 billion) and better align therapies to emerging areas of patient need.


Source: CrunchBase, Company Disclosures, GSV Asset Management

But Big Data is not just about drug development. Founded in 2013, Khosla Ventures-backed Lumiata is creating the health industry’s “medical graph” to help medical practitioners more quickly identify patient needs and risk factors. The company combs oceans of data — from medical textbooks to scientific journal articles and public data sets — to map how patients and illnesses are connected.

Increasingly, leading technology companies are turning their attention to genetics, using Big Data to make healthcare smarter. IBM, for example, has been on a healthcare and genetics buying spree. In 2015 it acquired Explorys, a company spun out of the Cleveland Clinic with access to the anonymized health records of over 50 million people, and Phytel, which has access to population health management data. Big Blue could be sitting on one of the largest, most comprehensive health and genetic databases in the World.


Source: MIT Technology Review, Fusion, GSV Asset Management

Apple is collaborating with U.S. researchers to catalyze the creation of apps that will collect and apply user DNA to provide health insights and medical services. The apps will be based on Apple’s ResearchKit software platform, which launched in 2015.

Alphabet (Google) launched Google Genomics for researchers to store and share genomic data, with the aim to accelerate key research breakthroughs and organize the World’s genomic data.


Tailoring treatments to align with our personal genetic profile is one frontier. Another frontier that is quickly moving from science fiction to science reality is modifying and enlisting genes themselves in the treatment of various diseases. Here, the story is all about “CRISPR.”

CRISPR — which stands for Clustered Regularly Interspaced Short Palindromic Repeats — is a series of gene repeats that acts as an immunity system set up in the genetic code of our cells. When combined with an enzyme called Cas9, which acts as a type of genetic scissors, the CRISPR/Cas9 system is able to target, snip out and replace undesirable genetic code in our cells.

In November 2016, a group of Chinese scientists were the first to use CRISPR technology to genetically edit cells in patients to fight off a form of lung cancer. The World is waiting to see the results, but the China’s first-to-the-clinic victory may fuel positive rivalries in the around the World. As the leading cancer immunotherapy expert Dr. Carl June observed in a recent interview with Fortune, “I think this is going to trigger Sputnik 2.0.”


Source: CrunchBase, Company Disclosures, GSV Asset Management

The CRISPR technology, though a massive breakthrough in the field of genetic engineering, is still an imperfect system. When Cap9 cuts genes at the wrong locations, it can actually cause cancer. Despite the difficulties, Editas, a publicly-traded genome editing shop, has proposed running a CRISPR trial in 2017 to treat genes causing blindness in humans. Stanford University has similar plans to use CRISPR to repair genes causing sickle cell anemia.

As part of a $250 million study, another group of U.S. scientists will use CRISPR to treat lung cancer. Funded by Sean Parker’s new cancer institute, this study will take place at the University of Pennsylvania. It has been approved by the NIH (National Institute of Health) and is awaiting FDA approval.

We expect to see a wave of emerging startups focused on bioengineering as pioneering labs break through the initial regulatory hurdles.


Beyond the science of personalized genetic testing, pharmaceutical development, and, bioengineering, companies that provide or facilitate efficient, on-demand health and wellness services are beginning to gain broad traction.

One Medical, for example, has created a 21st-century doctor’s office focused on patient convenience and on-demand services. Customers pay a yearly fee on top of their normal insurance copays to get perks from the provider direct digital access to doctors, online scheduling and prescription renewals, same day appointments, and treatment recommendations for common medical issues through a mobile app.

To date, the company has raised $182 million from a syndicate of investors, including Benchmark, Google Ventures, and JPMorgan.

Companies like Clover Health are changing the paradigm by using data and mobile technology to reduce costs, improve outcomes, and deliver a superior customer experience. Backed by Sequoia and First Round Capital, Clover’s initial goal is upend the Medicare health insurance market, which is dominated by sprawling incumbents like UnitedHealth and Cigna, which have a combined market value over $150 billion.

Pioneers: Clover Health
Kris Gale, Co-Founder, Clover Health (Right) Interviewed by Phineas Barnes, Partner, First Round Capital (Left) at the 2015 GSV Pioneer Summit

Source: GSV Asset Management

To do this, Clover collects a range of patient data like lab test results, radiology results, and routine checkup absenteeism to get an overall profile of a person’s health and risk profile. It then uses software models to automatically identify issues — like patients not regularly taking a prescription — and intervenes early with nurse practitioners and social workers that are equipped for home visits.

As Clover co-founder Kris Gale observed in a recent interview with TechCrunch, “If we know something is on a 30-day refill and we haven’t seen a claim in 35 days, we know they aren’t taking it regularly. We can reach out and intervene. This is info that’s available to us because we’re the payer… The doctor doesn’t know if they’re getting that filled unless they’re asking them regularly. That’s part of the data advantage.”

The net result is lower costs for both Clover and its customers, not to mention better health outcomes. In 2015, for example, seniors using Clover Health had 50% fewer hospital admissions and 34% fewer hospital readmissions than the average group of Medicare patients in the New Jersey areas it serves.

Smart Services: Insurance + Health Benefits

Source: CrunchBase, GSV Asset Management

Personalized, On-Demand Health Services Companies

Source: CrunchBase, Company Disclosures, GSV Asset Management

Looking forward, as personalized genetic mapping, pharmaceuticals, and bioengineering applications are increasingly commercialized, we see next-generation health services and insurance companies as a key catalyst to drive adoption.


by Luben Pampoulov

Biking Boom

It’s been seven years since Uber started the ride-sharing economy boom. During that period, the taxi industry was put upside down, for very obvious reasons, and several multi-billion dollar companies emerged; Uber, Didi, Lyft, Grab, Ola to name a few. (Disclosure: GSV owns shares in Lyft).

While Uber has been the dominant force until now, it is running into very strong competition. On one hand, China’s Didi is beefing up its international position, as shown by its recent $100 million investment in Brazil’s ride sharing platform 99. Didi also has strategic partnerships and investments in Lyft and Grab, and it famously bought Uber’s Chinese business last year.

On the other hand, Lyft has been winning ground against Uber domestically. Over the last 2-3 years, Uber’s lead on Lyft has shrunk from about 4x to about ~2x larger. Looking at the recent leaked financials for both Lyft and Uber — their respective growth rates confirm the trend.

So while the ride sharing economy is still growing at over 100%, it is now shifting towards an Oligopoly, with only a few major players enjoying the global market share.

Meanwhile, a new ride-sharing economy is booming, and specifically so in China; Bicycle ride-sharing services have attracted significant attention lately. In late September, Haidan-based Ofo raised $130 million at a $500 million valuation from Didi, TPG, Ctrip and CITIC PE. And earlier this month, Shanghai-based Mobike announced it raised $250 million from Tencent, and from existing investors Hillhouse Capital, Sequoia and Warburg Pincus.

Famous for its orange wheeled bikes, Mobike allows users to unlock, rent and pay for bikes through the app. Its bikes are considered higher-end “vehicles” in comparison to Ofo’s bikes. They cost $440 to build, have built-in GPS trackers, and look like they might be from the local SoulCycle… Mobikes rent for about 2 yuan per hour ($0.30).

The Mobike

In comparison, Ofo bikes are much cheaper — both price-wise and visually. The manufacturing cost is only about $35, and they rent for 1 yuan per hour. Ofo certainly has the much better margins at this point, and according to its 25-year old founder Dai Wei, “it is making money.” Similarly, Ofo riders rent, unlock and pay the bikes via the Ofo app.

Ofo (Yellow) and Mobike (Orange)

Currently, Mobike says it has more than 100,000 bikes in Shanghai alone, and it operates across 9 major cities in China. It is also expanding internationally and is expected to launch in Singapore in the coming months.

Back in October, Ofo said it had 85,000 bikes across several Chinese cities, with most pick up locations being near University campuses.

There is also a third bike sharing service called Bluegogo. It operates mainly in Shenzhen, and uses a slightly different subscription model; user pay an annual $99 fee, and then pay $0.99 per ride.

Based on the recent funding activity, it seems that Mobike is the best positioned player. Mobike is clearly going for a big marketshare in and outside of China, and will have enough money to spend to pursue that strategy in the near future. Its higher-end, unique looking bikes might turn out to be a strong advantage for the company’s branding efforts.

Pioneer Notes

by Li Jiang

10 Most Important People in Artificial Intelligence in 2017

“As soon as it works, no one calls it AI anymore.”
— John McCarthy

John McCarthy coined the term Artificial Intelligence in 1955.

Since then, the AI industry at large has seen dramatic ups and downs — progress and promise mixed with disappointment and disillusion. But now with the convergence of Megatrends on massive data, lightning fast processing speeds, and renewed competitive fever from the American MAFIA (Microsoft, Alphabet, Facebook, IBM, Amazon), AI is poised to cause disruption on a scale that could surpass the Internet itself.

As we prepare for a wave of AI first companies (@sundarpichai) and AI natives (Ryan Hoover), every person in the innovation economy will need to understand how AI will (or will not) change their industry and their lives.

Source: Waitbutwhy

So start following the work done by these 10 industry titans and, as a bonus, 10 rising stars you may not have heard of today, but will soon:

Part 1: Industry Titans

These titans shape the conversation and have the most ability to move the entire AI industry. In alphabetical order [0]:

Amit Singhal (@theamitsinghal)

In his new seat as the head of Uber’s engineering team, Amit will be driving forward autonomous vehicle breakthrough across the Uber fleet.

Andrew Ng (@AndrewYNg)

As one of the creators of the Google Brain and now Chief Scientist at Baidu, the “Google of China”, Andrew shapes the AI conversation in the two largest economies in the world. It doesn’t hurt that he’s the co-founder of Coursera and has 100K followers on Twitter. (Disclosure: GSV owns shares in Coursera)

Elon Musk (@elonmusk)

Because. While not a researcher or the head of an AI company, when Elon speaks on AI, people listen. He does chair OpenAI…more on them later.

Jeff Dean

As the creator of MapReduce, Google Brain, TensorFlow, Jeff is at the heart of some of Google’s most important projects and will continue to make waves at the company who first coined the “AI-first” strategy.

Jeff Dean

Ginny Rometty

IBM has been working on Watson since Ken Jennings’ 74 winning show run on Jeopardy in 2004. Now IBM has positioned its entire future on Watson.

Greg Brockman

When Elon and Sam Altman chairs OpenAI, Greg is its CTO and engine for discovery on a daily basis. OpenAI is an unified attempt to make AI technology accessible rather than closed to a particular company or nation, for the benefit of all of humanity, rather than a select slice of humanity.

Martin Ford

The author of Rise of the Robots, Martin shapes the cultural conversations around AI and robots. He’ll have many more best-selling books to come.

p.s. Come see Martin Ford, Charles Jolley, Riva Tez, and Nikhil Budama speak at the Collider event next Tuesday.

Ray Kurzweil

Perhaps the real granddaddy of the group, Ray Kurzweil has been making accurate predictions for the computing industry for decades. He created the concept of the Singularity, predicting that in the next few decades we’ll reach a point where computing power will be so overwhelming that it will eclipse any reason human attempt to process the speed of innovation. 

Sebastian Thrun

While he is one of the leading minds on self-driving cars, Sebastian also found time to start Udacity. What’s with AI professors from Stanford starting massive online learning companies? Maybe they realized before anyone that the disruption in AI will require society to re-educate now more so than ever.

Sebastian Thurn

Yann LeCun (@ylecun)

Yann heads up Facebook’s AI group in New York. Need I say more? I will. Facebook’s reach and its inherent nature will mean that its progress in AI will be incredibly personal and intimate for billions of people. Yann, with great power, comes great responsibilities.

Market Update

Week ending January 22, 2017

World Indices

U.S. Indices Snapshot

Valuation P/E Est. P/E/G Price/Sales