Elon Musk’s Las Vegas Loop might only carry a fraction of the passengers it promised

In pandemic-free years, America’s biggest trade show, CES, attracts more than 170,000 attendees, bringing traffic that jams surrounding roads day and night. To help absorb at least some of the congestion, the Las Vegas Convention Center (LVCC) last year planned a people-mover to serve an expanded campus. The LVCC wanted transit that could move up to 4,400 attendees every hour between exhibition halls and parking lots.

It considered traditional light rail that could shuttle hundreds of attendees per train, but settled on an underground system from Elon Musk’s The Boring Company (TBC) instead — largely because Musk’s bid was tens of millions of dollars cheaper. The LVCC Loop would transport attendees through two 0.8-mile underground tunnels in Tesla vehicles, four or five at a time. 

But planning files reviewed by TechCrunch seem to show that the Loop system will not be able to move anywhere near the number of people LVCC wants, and that TBC agreed to.

Fire regulations peg the occupant capacity in the load and unload zones of one of the Loop’s three stations at just 800 passengers an hour. If the other stations have similar limitations, the system might only be able to transport 1,200 people an hour — around a quarter of its promised capacity. 

If TBC misses its performance target by such a margin, Musk’s company will not receive more than $13 million of its construction budget — and will face millions more in penalty charges once the system becomes operational. 

Neither TBC nor LVCVA responded to multiple requests for comment. 

Fire regulations limit the load/unload zone near the cars to 800 people per hour. Credit: TBC/Clark County

The LVCC always realized that it was taking a gamble on the Loop. Although Musk built a short demonstration tunnel near Los Angeles, this would be the first public system with real customers and service requirements. An analysis by Las Vegas Mayor Carolyn Goodman in May 2019 concluded that TBC’s unproven system presented a high risk for the LVCC’s parent body, the Las Vegas Convention and Visitor’s Authority (LVCVA).

So when the LCVCA wrote its contract with The Boring Company, it did its best to incentivize Musk to deliver on his promises. The contract would be for a fixed price, and TBC would have to hit specific milestones to receive all of its payments. When the bare tunnels are completed, which could happen any day now, TBC will have earned just over 30% of the total. The next big milestone is the completion of the entire working system, which would result in a pay-out of over $10 million. 

That was scheduled to have happened by October 1, so that the system would be ready for the next CES show in January. Although CES 2021 has now gone virtual and there is less time pressure on Musk to deliver, he presumably still wants to get paid. 

In a tweet this week, Elon Musk wrote that the system would be open in “maybe a month or so. Some finishing touches need to be done on the stations.”

After another milestone for the completion of a test period and safety report, the system’s final three milestones relate to how many passengers it can carry. If the Loop can demonstrate moving 2,200 passengers an hour, TBC will get $4.4 million, then the same payment again for hitting 3,300, and the same again for 4,400 passengers an hour. Together, these capacity payments represent 30% of the fixed price contract. 

Even if TBC achieved those numbers during testing, the LVCVA was worried that it might not be able to maintain them once the system was operational, so it inserted yet another requirement: “[TBC] acknowledges liquidated damages are applicable for [TBC’s] failure to provide System Capacity for Full Facility Trade Show Events.” 

For each large trade show that TBC fails to transport an average capacity of 3,960 passengers per hour for 13 hours, it will have to pay LVCVA $300,000 in damages. If TBC keeps falling short, it keeps paying, up to a maximum of $4.5 million. 

So what is stopping TBC from transporting as many people as both it and the LVCC wants? There are national fire safety rules for underground transit systems that specify alarms, sprinklers, emergency exits and a maximum occupant load, to avoid overcrowding in the event of a fire.

Building plans submitted by The Boring Company include a fire code analysis for one of the Loop’s above-ground stations: 

Image source: The Boring Company/Clark County NV

The above screenshot from the plans notes that the area where passengers get into and out of the Tesla cars has a peak occupancy load of 100 people every 7.5 minutes, equivalent to 800 passengers an hour. Even if the other stations had higher limits, this would limit the system’s hourly capacity to about 1,200 people. 

“That sounds correct,” says Glenn Corbett, a professor of security, fire and emergency management at the John Jay College of Criminal Justice in New York. “But if that’s the bottleneck, the question from a safety standpoint is, what controls that [800 per hour]? Is it just pure honesty and people following the rules, or is there a mechanical thing that keeps them out?” 

The plans do not show any turnstiles or barriers to limit entry.

Even without the safety restrictions, the Loop may struggle to hit its capacity goals. Each of the 10 bays at the Loop’s stations must handle hundreds of passengers an hour, corresponding to perhaps 100 or more arrivals and departures, depending on how many people each car is carrying. That leaves little time to load and unload people and luggage, let alone make the 0.8-mile journey and occasionally recharge. 

[gallery ids="2061530,2061529,2061685,2061528"]

Although TBC’s Loop website says that the system will use autonomous vehicles, a TBC executive told a planning committee last year that the cars would have human drivers “for additional safety.” TBC had proposed developing a larger capacity autonomous shuttle for the Loop, capable of carrying up to 16 people. The latest plans all show traditional sedans, however, and another Musk tweet this week admitted: “We simplified this a lot. It’s basically just Teslas in tunnels at this point.”

The most recent documents filed by TBC also show changes to the Loop’s original design.

Gone are striking curved roofs, with both aboveground stations now having flat photovoltaic canopies to help charge the Tesla vehicles. These terminal stations each have a single Supercharger station, and a “showpiece sculpture” consisting of a concrete segment similar to those used in the tunnels below.

The central, subterranean station has a large, open platform, and also houses the electrical, fire safety and IT equipment. Each station will have bays for 10 Tesla vehicles to load and unload passengers. 

Even before the first Loop is operating, TBC is planning two more Loop tunnels nearby, connecting the LVCC to the Wynn Encore and Resorts World casinos.

The tunnel to the Encore is long enough that safety regulations require an emergency exit about halfway along. Plans indicate an emergency egress shaft and a small hatch, but it is unclear whether passengers escaping a fire or breakdown would be expected to climb stairs or even a ladder.

Loop extension egress. Image source: The Boring Company/Clark County NV

TBC last year suggested an emergency ladder for its proposed Loop between Baltimore and Washington, D.C., a system that Corbett called “the definition of insanity,” as it did not account for passengers with limited mobility. That project is now on pause

TBC’s stated aim is to expand the LVCC Loop from a local people mover to a Vegas-wide transit system serving the Strip, the airport and eventually extending all the way to Los Angeles. If the company struggles to deliver capacity — and revenue — from its small-scale Convention Center system, the future of those ambitions could be in doubt. 

Enabling humanoid robot movement with imitation learning and mimicking of animal behaviors

Over the past two decades, humanoid robots have greatly improved their ability to perform functions like grasping objects and using computer vision to detect things since Honda’s release of the ASIMO robot in 2000. Despite these improvements, their ability to walk, jump and perform other complex legged motions as fluidly as humans has continued to be a challenge for roboticists.

In recent years, new advances in robot learning and design are using data and insights from animal behavior to enable legged robots to move in much more human-like ways. 

Researchers from Google and UC Berkeley published work earlier this year that showed a robot learning how to walk by mimicking a dog’s movements using a technique called imitation learning. Separate work showed a robot successfully learning to walk by itself through trial and error using deep reinforcement learning algorithms.

Imitation learning in particular has been used in robotics for various use cases, such as OpenAI’s work in helping a robot grasp objects by imitation, but its use in robot locomotion is new and encouraging. It can enable a robot to take input data generated by an expert performing the actions to be learned, and combine it with deep learning techniques to enable more effective learning of movements. 

Much of the recent work using imitation and broader deep learning techniques has involved small-scale robots, and there will be many challenges to overcome to apply the same capabilities to life-size robots, but these advances open new pathways for innovation in improving robot locomotion. 

The inspiration from animal behaviors has also extended to robot design, with companies such as Agility Robotics and Boston Dynamics incorporating force modeling techniques and integration of full-body sensors to help their robots more closely mimic how animals execute complex movements. 

It’s time to build against pandemics

We’re a few years out from the call to action Bill Gates made in his TED Talk on preparing for pandemics back in 2015, yet the state of scalable software for important workflows like data collection and contact-tracing has greatly lagged expectations during the current pandemic.

The Trump administration’s letter to health agencies regarding data-sharing guidelines asked for daily Excel uploads, and manual contact-tracing efforts without software have proven difficult given the scale of the current pandemic. 

Everything is being built right now. 

Research universities are helping build models used by the CDC for case prediction, and that’s brought to light the dire issues around incomplete data sharing between health institutions and governments. 

Dozens of contact-tracing apps are springing up, surfacing design decisions around privacy, the need for newer technologies beyond Bluetooth for near-field communication and leading companies like Google and Apple to strike partnerships to power cross-platform mobile capabilities.

The good news is that the current efforts are taking seriously the need for better software and driving necessary innovation to help society better prepare for pandemics.

How can detailed case data be shared by hospitals with governments to better predict case and mortality numbers, and used to better allocate medical and labor resources? 

How can software help local and state governments make better policies, and help digitize contact tracing while appeasing privacy concerns?

Software has the ability to power many of these capabilities, and it is creating new opportunities for startups to vet the newly formed appetite for better data and digitized workflows on the part of health agencies, local and state governments and other organizations involved in fighting pandemics.

Investing in the hidden generation

While it’s no secret Hispanics represent unparalleled growth opportunities for the U.S. economy, most startups don’t realize Hispanic youth means an abundance of prime spending years (translation: dollars for businesses). The average age of a Hispanic living in the U.S. is 28. Meanwhile, the average age of their white counterpart is 42. Nearly one in every five people in the U.S. identifies as Hispanic. 

Those few companies that do notice Hispanics and their massive purchasing power (~$1.5 trillion) tend to be legacy companies doing a subpar job at capturing the Hispanic consumer. Furthermore, they don’t target the most valuable member of the Hispanic community — what I call, the “Hypercultural Latinx.” They are where tons of unspent dollars lie. 

As an investor and member of the Hispanic community, I’m confident the startups solving problems for this Hypercultural Latinx member will have the potential to create companies with venture-like returns. 

Who is the Hypercultural Latinx?

The Hypercultural Latinx is a second-generation Hispanic who is 100% Hispanic and 100% American. And while that might sometimes lead to misunderstandings and conflicts with her white counterparts, it also means she excels by creating a pseudo culture where she can thrive best. She brings her unique characteristics to this self-created culture — a culture where her customs, language and values shine through. Furthermore, this person, who often identifies as a Gen Zer or young millennial, is a fanatic of mobile. After all, across socioeconomic classes, their disposable income is disproportionately going to screens (of all types) and tech toys.

I mean, just go into your Hispanic friend’s home: They are likely to have more TV screens than people residing in that household. In fact, a bewildering 29% of U.S. Hispanics planned to purchase a new TV set just ahead of the Super Bowl (guilty as charged). For reference, of the 30% of overall Americans that planned to buy a TV in 2017, only 2.8% purchased in the days before the Super Bowl. Heck, when my family moved, we bought TV screens for every room even before the living room was furnished. Technology — especially newer tech, is significantly more tempting to Hispanics. 

The Hypercultural Latinx should be top of mind for venture investors and founders. She desires to test the untested, and thus, is likely to cross the chasm before the early majority. This makes her an ideal customer segment for consumer startups.

Image Credits: Ilsa Calderon

Startup founders and VCs alike are missing out. As an investor, I often find myself reduced to frustration with the lack of founders and investors committed to exploring audience segments outside cookie-cutter ones. We might not need another consumer vertical product solving a half-felt pain point for the highly educated, white female with a $100,000+ salary living in NYC, SF or LA. However, we do need more products catered toward the Hypercultural Latinx who, by the way, outspend their white counterparts across most categories. In the same way Fenty Beauty exists to solve the make-up needs of primarily Black women, we need that for the Hypercultural Latinx population.

Numbers aside, investors should care about Hypercultural Latinx because they are tech-forward trendsetters who adopt social media at higher rates than their white peers. For example, a Hispanic youth is 87% more likely to use WhatsApp. Additionally, they produce an exorbitant amount of videos on Tik Tok. Several Tik Tok Hispanic-centric hashtags, such as #hispanicmom, are wildly popular and boost over 44 million views. For reference, the most followed Tik Tok stars, like Addison Rae, have just over 47 million followers. In fact, one Hispanic Tik Tok queen, Rosa, has already reached pop culture peak

Facebook ad experiment

Examples of ads I ran. (Image Credits: Ilse Calderon )

If you are more driven by quantitative data, know that paid spend targeting this Hypercultural Latinx could result in lower click cost rates and higher engagement. I ran a two-week experiment on Facebook to prove out this hypothesis. I created a landing page for a fake sunscreen brand, Bounce Skin, with a fake first product, an SPF mist. I created a couple of ads. Then, I ran ads on Facebook targeting two audiences: young Hispanic girls (the Hypercultural Latinx audience) and white girls. The average click cost for the young Hispanic girl audience was $0.06 per click; for white girls, it was $0.33 per click. Of course, my experiment was limited, but it did demonstrate that the Hypercultural Latinx is out there and craving content that tells the narrative of her life. (For more details, please check out this Medium post).

Why is the tech community decades behind when it comes to this Hispanic segment? 

Three key reasons: fear, the subpar state of Hispanic marketing and white men cannot relate to the Hypercultural Latinx. 

Fear. There’s always risk associated with offending the same audience you are trying to captivate. Just take a look at the beauty industry and its frequently associated race problem. The world is not white, and beauty brands that think it is have lived through PR nightmares. Even beyond beauty, tech startups fear negative press cutting short the life of their business. However, it is this gap that creates opportunity.

I encourage the right set of up and coming startups to authentically pursue the Hypercultural Latinx. Even though legacy companies might have heavier balance sheets, they don’t have the clout to lure this young, bicultural consumer. Let’s just say, no 18-year-old is going to be rushing to the Walmarts of the world looking for aspirational goods. They are even less likely to browse Walmart.com for content. 

The state of U.S. Hispanic marketing is ridiculous. In fact, there’s a graveyard of failed marketing attempts to the Hispanic community. Most recently, there was a Mother’s Day Kmart ad that blended two Spanish words (Mama + Namaste) to accidentally create a word translating into a very vulgar and offensive word. Furthermore, given most businesses’ “one size fits all” approach to Hispanic marketing, it’s no surprise they keep getting it wrong. However, if anyone is best positioned to take Hispanic marketing out of the 20th century, it’s small, nimble startups with no history of bad marketing or image problems. 

Perhaps the biggest reason the tech community isn’t approaching the Hypercultural Latinx is because most venture-backed founders and investors are white men. These white men cannot possibly relate to the life experiences of young, biracial teenagers and young adults living in white America. Last year, a measly less than 2% of venture funding went to Hispanic founders — those are the founders best suited to be able to genuinely capture the eyeballs and wallets of this Hispanic youth. On the investor side, it’s even worse with only 1% of venture investors identifying as Hispanic. 

The solution is complex, and frankly, I can’t provide a solution with clarity. However, we can start by building goodwill and non-transactional relationships with those role models Hypercultural Latinx admire. I’ve found that these role models are usually under-the-radar influencers, like Glenda. We as investors can also diversify our top of funnel deal flow to include more underrepresented founders. Lastly, founders with a reach and network of Hispanic youth should consider diving deep into the pain points of Hypercultural Latinx lives.

The new darling of the VC world will be solving problems for the Hypercultural Latinx

In order to become this new VC darling, founders approaching the Hypercultural Latinx should consider two suggestions: a platform play and an army of social guides.

The platform approach entails creating an organization of brands that later spew out new brands horizontally or vertically. An example of this is the company behind my favorite over-priced lemon drink, Iris Nova, or Glossier-team spin-off, Arfa.

The second approach, an army of social guides, means combining elements of affiliate marketing with a kick-ass referral program to create loyal fans that are financially incentivized to sell your products. Sequoia-backed Stella & Dot built out their version of social guides that ultimately became its most defensible strategy. Additionally, in a post-coronavirus world, this strategy is a way for an ever-increasing labor force to get back on their feet. 

At the end of the day, the Hypercultural Latinx demographic is only increasing, and so are its needs. For founders who truly care about the U.S. Hispanic market, pay attention to this hidden generation. For investors, look beyond solutions for your own problems. Winning over the multi-faceted Hypercultural Latinx is not easy, but startups that successfully do so attract my attention and my investment dollars. 

Assessing the potential for a gig economy in education

Over the past few years, personalized learning has established itself as a focal point of innovation in education. Despite the focus, the rate of progress in establishing personalized learning practices in both K-12 school systems and online learning has been slower than expected.  

The Bill & Melinda Gates Foundation and the Chan Zuckerberg Initiative have together invested millions of dollars in support of it, and educators such as Sal Khan, founder of Khan Academy, have spoken extensively about its importance in education.

Personalized learning comprises many aspects of learning: letting students master topics before they move on to higher level ones, giving them agency over their learning based on their interests and goals and using teacher-aided instruction and interactivity, to name a few.

Much of the focus on implementing personalized learning practices has revolved around K-12 school systems, where new initiatives have been met with mixed results, and these efforts will continue. 

Beyond the K-12 school systems however, online education platforms present a large opportunity for delivering personalized learning experiences to students worldwide, and the level of innovation here has lagged expectations.

Massive Open Online Courses (MOOCs) such as Udacity, Coursera and edX emerged in the early 2010s and helped bring quality content online and make it accessible around the globe. However, they haven’t innovated much when it comes to personalized learning, and studies have shown that they have in fact seen declines in completion rate of courses.

In recent years, startups have built platforms that are powering a gig economy for teachers, enabling them to give live lectures in small-group, highly interactive settings. Apps focused on providing personalized learning experiences for users learning domain-specific skills such as math or languages have shown promise, but there’s room for a lot more innovation on this front.

These newer approaches have the potential to democratize personalized learning by innovating on the software teaching platform, enabling better teacher-aided instruction online, and helping students better understand their mastery of topics. 

Latin America Roundup: Grupo ZAP, Grow Mobility, Wavy get acquired; Credijusto adds $100M; Cornershop, iFood brace for delivery boom

As the world locks down borders and capital flows to brace for the impact of coronavirus, Brazilian startups continue to attract international attention. Three large acquisition deals dominated the Latin American tech headlines this month, all coming from the region’s largest country. As investments have waned, these deals offer hope for some increased liquidity in Latin America’s startup ecosystem. 

At the beginning of the month, Brazilian real estate leader Grupo ZAP was acquired by OLX Brasil for $640 million, solidifying the classifieds platform’s position in the local property market. The deal will enable OLX to offer its customers more than 12 million listings from 40,000 agencies and individuals. 

Grupo Zap merged with the property rental platform VivaReal in December 2017, becoming the de facto largest real estate portal in the country. While the brands have operated separately, they jointly receive more than 40 million visits per month to help Brazilians find properties for rental and purchase. The acquisition is still under review from Brazil’s antitrust agency, CADE, and will be finalized later this year.

Meanwhile, Peixe Urbano reported its intention to acquired Grow Mobility, the alternative mobility company created from the merger between Mexico’s Grin and Brazil’s Yellow. Peixe will own the majority share in the e-scooter and bike-share startup, which has recently struggled to turn a profit. After leaving 14 cities in February, Grin Mobility is only active in Brazil’s three largest cities today, as well as in a few countries around Latin America, despite a promising partnership with Rappi in 2019. Grow Mobility raised $150 million in January 2019 when Grin and Yellow merged and seemed to be one of the fastest-growing startups at the time; however, this deal is rumored to be a total write-off for the startup’s investors. 

Finally, a Swedish cloud communications platform called Sinch AB announced it would acquire Movile’s strategic communications company, Wavy, for $68.3 million (BRL$554 million) and more than 1.5 million shares in the publicly traded company. Movile is one of Brazil’s largest tech businesses, a telecommunications company striving to become the region’s Tencent. Wavy is Brazil’s second-largest messaging provider and also operates in Mexico, Colombia, Peru, Chile, Argentina and Paraguay, relaying more than 13 billion messages per year. Sinch will use the acquisition to grow into the Latin American market, where Wavy currently employs over 260 people across nine offices in the region. At the time of purchase, Wavy was growing at 200% year-on-year, hinting at strong growth for the new business over the coming years.

Movile also announced the arrival of a new CEO, Patrick Hruby, in the last week of March. His predecessor, Fabricio Bloisi, co-founder and CEO since 1998, will take a seat as board president and will continue to act as CEO of iFood. Hruby previously spent five months as an Executive in Residence at Movile, where he worked closely on operations with all Movile companies: iFood, MovilePay, PlayKids, Sympla, Wavy and Zoop. Movile is one of Brazil’s least-known unicorns, quietly building a mobile empire for the region with a goal of impacting over one billion people.

Credijusto raises $100M to support SMEs in need

The Mexican credit provider, Credijusto, announced in mid-March that it had received $100 million in debt from Credit Suisse to help the startup extend more loans to SMEs affected by the economic impact of the coronavirus. Small businesses in Mexico already struggle to access financing from banks, and the current economic projections will likely cause financial institutions to hold off on risky investments for the foreseeable future. 

Meanwhile, this credit crunch has caused a surge of interest in Credijusto’s products: online small-business loans. The startup uses an algorithm to rapidly calculate risk and interest rates, providing much-needed liquidity for SMEs struggling in the face of financial turmoil. Credijusto also recently raised a $100 million debt vehicle from Goldman Sachs, alongside a $42 million Series B equity round from Goldman and Point72 Ventures in September 2019. 

Cornershop, iFood: Keeping up with coronavirus delivery demands

While in the U.S., Instacart and Amazon are scrambling to keep up with the boom in delivery orders, Latin American delivery giants Cornershop and iFood face similar challenges. Mexican-Chilean delivery app Cornershop, which was acquired by Uber last year for $450 million, revealed they had just nine months of operating capital left as they face unprecedented order volume. 

Despite the large acquisition deal, Cornershop’s case remains under review by the Mexican antitrust organization, COFECE, which blocked their previous $225 million acquisition offer from Walmart. Cornershop’s co-founder and CEO Oskar Hjertonsson took to Twitter to share the challenges his company is facing as demand for grocery delivery surges due to coronavirus concerns. He notes that grocery delivery has become an essential service in many areas with severe quarantines, yet with the acquisition still in question, Cornershop does not have the resources to serve the current demand. 

Two Mexican regulators are currently fighting over the jurisdiction to review this case, which has been going on for more than six months without a resolution. Cornershop has been at the mercy of Mexican officials since June 2018, when they first announced their Walmart acquisition. On Twitter, Hjertonsson urges Mexican officials to move forward on the decision as soon as possible to capitalize on an opportunity to help millions of Latin Americans who are currently in lockdown, as well as bringing in the resources needed to protect their delivery staff.

At the same time, Brazil’s largest food delivery company, iFood, announced the launch of a new fund to help small restaurants survive the economic tumult brought on by the coronavirus. The food industry has been one of the hardest-hit by the pandemic, as many restaurants live on small margins. To combat this trend, iFood launched a $9.8 million fund that will support small restaurants within the iFood network. 

The company also announced that it would speed up receipt processing during April and May, helping small businesses receive their payments within seven days without extra cost. This measure will inject an additional $117 million into the Brazilian restaurant market. Finally, iFood seeks to support its restaurant partners by returning all fees they receive for delivery during the coronavirus epidemic. Realizing that restaurants must rely on delivery orders to survive this period, iFood has extended these measures to over 120,000 restaurant partners in 1,000 cities across Brazil. 

News and Notes: Vai.Car, ClassPass, Superlogica and NotCo

Despite public health and economic concerns about COVID-19, the Latin American startup ecosystem remained active this month, with startups raising large rounds from local and international firms alike. Brazil’s car rental startup Vai.car raised $85 million from the Brazilian investment platform XP Investimentos, which IPO’d at the end of 2019. The startup targets a young market by enabling medium-term car rentals that are delivered to the user’s door and unlocked with face recognition technology. Vai.car also partners with Uber and 99 to help drivers access vehicles from their fleet of more than 25,000 cars.

U.S. gym-sharing platform Classpass expanded aggressively into Latin America this month through the acquisition of Chile’s Muvpass and Argentina’s Clickypass. These platforms work similarly to Classpass, allowing users to access a network of gyms and fitness classes across the country. Classpass launched in Brazil in December 2019 and became the first unicorn of the decade, with a $285 million Series E in early 2020. 

The Brazilian payments management platform Superlogica raised a $63.5 million round from U.S. private equity firm Warburg Pincus in mid-March. Superlogica helps companies manage recurring payments using a subscription model powered by artificial intelligence. The startup currently serves customers in more than 45,000 rental properties around the country.

Chilean plant-based food tech startup, The Not Company, announced a partnership with Burger King to create a vegan Whopper across the United States. The RebelWhopper is made of plant-based meat and features NotCo’s signature NotMayo, a mayonnaise made without animal products, which rapidly became a household name in Chile. The Not Company raised $30 million from Bezos Ventures and other investors in 2019 and has continued to expand rapidly into Argentina and Brazil over the past year. 

The past six weeks have been characterized by global uncertainty about the future of the economy and international relations as COVID-19 has made its way into every country in the world. However, deal flow in Latin America was still strong in March, bringing large deals and several acquisitions, especially in Brazil, even as the country refuses to lock down to prevent the spread of the pandemic. Notably, despite travel restrictions, many of the deals this month were led by foreign VCs, hinting at a potential for quicker feedback loops in the region as investors disburse capital without traveling first. 

It is hard to see today what the new normal will be globally, and specifically in venture and tech in Latin America. Almost every country has closed its borders, some more forcefully than others, and many are waiting out the pandemic in some level of quarantine. Just Mexico and Brazil, the region’s largest economies, remain adamant about keeping their cities running normally, even encouraging their citizens to visit bars, restaurants and museums as their neighbors shutter businesses. Time will tell how this decision will affect startups and investments, as well as their citizens and political stability, across the region.

The battle to become the Mexican Nubank just started

Banks in Latin America have long dominated the market as oligopolies, becoming highly profitable but not serving well the population.

In the region, Brazil was the first market to have the banks’ oligopoly challenged by neobanks, with Nubank proving that it was possible to break them up. Providing a superior product and exceptional customer support, it was able to attract more than 15 million customers, valuing the company at an astonishing US$10 billion in its latest round.

Although Brazil has been the center of the neobanks’ emergence in Latin America, attention has been shifting toward another country in the region.

The Mexican opportunity

Mexico, the second-largest economy in the region, has become a relevant market for the fintechs in the region.

The reason is not exactly the most flattering; Mexico is a large country with almost 130 million population, but a large share are still unbanked. Indeed, 63% of the adult population still doesn’t have access to financial services, according to the Global Findex database, and banks haven’t been able (or are not interested) to serve them.

Furthermore, Mexicans are very suspicious of banks because of their lack of transparency, as well as recent financial crises.

Because of this skepticism toward banks, together with a cash-based economy, 90% of all consumer transactions are still made in cash, which prompts a rather peculiar situation — twice a month (quincena) there are long lines at ATMs all across the country with people withdrawing all their wages.

On the other hand, Mexico has a digitally engaged population (Mexico is the fifth largest market for Facebook, ninth for YouTube and the third for Uber), with high smartphone penetration (85.8% according to The Competitive Intelligence Unit).

All these elements put together create a rather attractive opportunity for the emergence of neobanks.

Mexico becomes the next battleground

Mexico had a few neobank pioneers in the past couple of years; Bankaool launched its services in 2015, but was too early in the market; later on came Broxel and Albo in 2016 followed by Flink in 2018.

However, the market started to garner more attention in April 2018 when the Iranian Matin Tamizi raised US$2 million from Andreessen Horowitz (a16z) and Kaszek Ventures to create a neobank in Mexico (Cuenca). It is interesting to note that Matin, at that time, had never been to Mexico and only had a slide deck to demonstrate the opportunity.

Neobanks aren’t the only ones trying to get a share of the Mexican wallets.

This event, together with the success of neobanks in Brazil, sparked attention for the potential of the market.

A couple of months after, Albo announced a Series A, raising US$7.4 million. It is currently the leading player in the market, with 150,000 customers and the third debit card issuer in Mexico.

Shortly thereafter, Fondeadora raised a US$1.5 million seed round to enter the neobank market, pivoting from a crowdfunding platform.

Late in September, a new entrant closed a relevant round; Klar raised US$7.5 million in equity and US$50 million in debt financing with the goal to become the “Chime of Mexico.”

Vexi is another player in the market, though it is focused on providing credit cards to people at the base of the pyramid. It has issued, so far, more than 20,000 credit cards and, recently, raised US$2 million in equity and US$1 million in debt financing.

Regional and international players are also becoming interested in the opportunity. The Brazilian giant Nubank announced this year officially that it would be expanding there. From overseas, the leading Spanish neobank, Bnext, announced it would be entering the Mexican market, fueled by a fresh new round of €22.5 million. Different from other neobanks, Bnext partners with fintechs and financial institutions to provide services to its customers via a marketplace.

Nonetheless, there are rumors that other heavyweights, such as the Europeans Revolut and N26, are planning to enter the market, as well as the Argentinian Ualá.

Neobanks aren’t the only ones trying to get a share of the Mexican wallets. Many tech companies such as Cabify, Weex and Rappi are launching digital wallets and issuing debit cards, leveraging their large user base.

To add a final spice to the market, traditional banks are making a significant effort to improve their digital offers — some even going as far as launching digital branchless initiatives. The Spanish Banco Sabadell entered the Mexican market with a full digital strategy, while Banregio (a local medium-sized bank) launched Hey Banco, a new digital account.

On the sidelines, there also are a few neobanks focusing on a different segment. Oyster and Evva are targeting the unattended market of freelancers, startups and SMEs, long neglected by the incumbents.

The stage is set for an upcoming battle

Although the market is still in its early stage with just a handful of neobanks with running services, the stage is set for an amusing upcoming battle. Most players will be launching in the next couple of months, which will trigger a race for acquiring customers and raising more money.

This competition will definitely change the landscape of the financial industry in Mexico, bringing better and more affordable services to its population.

It will be indubitably interesting to watch how the market will unfold in the following years, and the prize for the winners can be quite attractive, as Nubank proved in Brazil.

With Alibaba, Pivotal and Lightbend on board, Reactive flexes its ROI muscle in the microservices world

The Linux Foundation recently announced the launch of the Reactive Foundation. Its founding members are Alibaba, Lightbend, Pivotal and Netifi. So what exactly is this Reactive Kool-Aid, and why are all these companies guzzling it down so fast?

If you buy the premise that developers live in a cloud-native microservices world, then you also understand that most applications are distributed and elastic. The compute is spread across clusters, as is all the data. It could be a few users, or a spike of thousands. Systems need to be architected to handle these spikes. Yet the dark secret of microservices is complexity — the management of resources, costs, performance and latency remain a challenge.

If we break down any application into hundreds of moving parts (such as containers and microservices), then we better have an elegant way to manage those moving parts. These services need to talk to each other, exchange data and ensure that overall performance is reliable, at all times. Easier said than done.

The “big unsolved problem of the cloud”

According to Daniel Berg, Distinguished Engineer at IBM Cloud, “The network is the unsolved problem of the cloud…. We need the network to be a first-class citizen of a cloud system.” Why does the network remain a problem? Is it because we fall back on our old ways, when we need to rethink the new? We have designed the car with the big clunky wheels of a horse buggy. Conceptually, it sounds fine — but it can be a pretty rough ride.

In the layered cake of network protocols, we have the middle layer of transportation (Transmission Control Protocol / Internet Protocol or TCP/IP), and right at the top, we have the application layer. We use a protocol called hypertext transfer protocol (or HTTP) to make sure the web applications can talk to each other. TCP was born in 1974 and is called a “chatty protocol” — it has to go back and forth many times just to do some basic stuff. A TCP joke circulating around proves this point.

HTTP Joke

HTTP came in 15 years later, in 1989, and was used to serve documents in a client-server era. This was when we all had desktops being cooled with whirring fans. We would use a Netscape browser to launch a web page (hypertext) and the web-server would say, “Wait a sec – let me fetch that for you.”

Three decades later, we are trying to make do with HTTP, when the compute layer has exploded. Does HTTP work in the world of millions of interactions with machine-to-machine communications? Our mobile, IoT and edge devices are not quite requesting pages and walls of text. And there is no client-server as much as peer-to-peer exchange. But the network layer is stuck with us and we are trying to make sure these microservices can stay put using some archaic methodologies. “As much as 89% of all microservices architecture is based on HTTP, says Stéphane Maldini, principal software engineer at Pivotal. Pivotal is one of the founding members of the Reactive Foundation. In the process, we are creating a big trade-off in efficiency. We are still using two cans and a piece of string to communicate, when we should use the next iPhone.

HTTP is unsuitable for microservices

If we use HTTP in the micro-services world, we have fundamental challenges. For one, there is no flow control — “which means that data flows from a fire hose,” says Robert Roeser, co-founder of Netifi. Because the data can be dumped at a rapid pace, and multiple threads are opened up, we end up building control features to ensure the application does not crash.

Reactive programming is a paradigm shift at the architectural level. It’s about speed and performance.

Stuff like circuit breakers, retry logic, thundering herd (where a large number of processes wake up, but only one wins, often leading to freezing up) needs to be managed effectively. In HTTP, everything is a request / response, but if we look at a simple notification for an app, we don’t need to keep polling all the time. The request is like a grumpy kid sitting in the backseat whimpering, “Are we there yet?,” when the journey has just begun.

Such inefficient mechanisms cause a huge waste of compute resources when we use the wrong protocol. IBM documented the inefficiency of microservices and concluded that the performance of the microservices is ~ 79% (s)lower than the monolithic model. “We identified that Node.js and Java EE runtime libraries for handling HTTP communication consumed significantly more CPU cycles in the microservice model than in the monolithic one,” conclude the researchers.

Goodbye HTTP, hullo Reactive

The Reactive Foundation sits under the Linux Foundation and aims to accelerate the next generation of networked technologies. It espouses the merits of Reactive Programming Frameworks and builds the community. Ryland Degnan, chair of the Reactive Foundation and co-founder of Netifi, lived the HTTPain while he was a member of the Netflix edge platform.

Ryland understands scale, latency and user experience better than most people. At Netflix, the platform would have billions of requests from over several hundred million members. He says, “We live in a multi-dimensional universe where user experience matters. Developers have to deal with three axes of (a) deployments (b) frameworks and (c) protocols. Spotty connections are unacceptable. Why can’t we pick the stream up from where you left off? If we do that alone, we reduce 90% of our infrastructure.”

Indeed, Facebook has adopted RSocket to reduce the dropped connections over mobile network hops and reduced its edge infrastructure significantly. Steve Gury, a software engineer at Facebook speaking at SpringOne Platform said, “The future is R-Socket.”

Reactive programming is a paradigm shift at the architectural level. It’s about speed and performance. One of the major strengths of Reactive is asynchronous I/O, which allows reduction of edge infrastructure by orders of magnitude.

Andy Shi, developer advocate at AliCloud (a unit of Alibaba), is one of the founding members of the Reactive Foundation. He says, “Alibaba has thousands of developers as we are one of the world’s biggest e-commerce platforms. As we adopt microservices and see that compute is utilized only around 10%, throwing more infrastructure at the service mesh is not the answer. Pods are talking to each other using REST API which is not the way to go.”

REST APIs require multiple endpoints and round trips to get the data. Another founding member of the Reactive Foundation, Viktor Klang, deputy CTO at Lightbend, has been evangelizing Reactive for well over a decade, and feels like the time has finally come. “Our systems need to produce results in the required time frame. Imagine if you could compute an answer to a grand question — like the meaning of life — but if the answer is delivered after you die, the system has failed,” he says.

Comparing service meshes and use cases

While Istio is the 18-wheeler truck best suited for lift and shift, RSocket is the Ferrari — speed and elegance. Experts foresee a world where the two may coexist. Yet there are applications, such as IoT use cases, where RSocket has a clear edge (pun intended). Istio offers load balancing, service discovery, logging and traffic management but with heavy overhead.

In studies, Netifi was able to process 16X more requests and delivered four times higher throughput in comparison while maintaining three times better latency — 372% faster throughput with 300% less latency. “Netifi has the potential to be like a Cisco — the router of the microservices,” says Creighton Hicks, investor at Dell Technology Capital.

Istio was launched by Google, IBM and Lyft, so it is a strong incumbent and with serious brand cachet. But when the likes of Alibaba and Facebook start to showcase the RSocket ROI, the fun has just begun. During a recent presentation in London, the Reactive mafia was in full swing. Ondrej Lehecka, a software engineer at Facebook, and Andy Shi talked about how RSocket is addressing real-world architectural challenges. Shi said, “RSocket is designed to shine in the era of microservice and IoT devices. Projects built on top of RSocket protocol and Reactive streams in general will disrupt the landscape of microservices architecture. The Reactive Foundation is the hub of these exciting projects.”

The budding industry of cannabis tech

From food and drink to health and wellness and beyond, there’s one plant we can’t seem to get enough of: cannabis. It seems like every consumer product nowadays is taking part in reefer madness.

Home cooks are taking edibles to new heights. In places like Denver and California, you can take cooking classes specifically centered around food made with Mary Jane. The editors of Vice’s “Munchies” even put out a cookbook last year called Bong Appétit: Mastering the Art of Cooking with Weed. And it’s only one of many.

But marijuana culture today isn’t all based around the stuff you (er, people you know) smoked in college. Cannabis, long known for its medicinal and therapeutic purposes, is a hot commodity in food tech and other consumer products nowadays. Far more than just a way to get high, cannabis in its various forms has been used medically throughout history and in modern times as a treatment for pain and nausea, and has been found anecdotally or in limited studies to treat glaucoma, epilepsy and anxiety, among other conditions and symptoms. Businesses have caught on, and not a moment too soon.

The food products that utilize marijuana are a far cry from the old classic pot brownies (not that there’s anything wrong with those!). Thanks to modern science, producers are able to separate the two main chemical compounds found in marijuana: THC and CBD. THC has therapeutic benefits, but it’s best known as the part of weed that gets you high. This is because it’s a psychoactive compound. CBD, on the other hand, is not psychoactive — it can (supposedly) provide many of the anti-anxiety, analgesic benefits of the plant without producing a high. For obvious reasons, this gives marijuana a new appeal. It’s now possible to reap the benefits of the plant without experiencing intoxication, so you can lessen anxiety or pain while still functioning normally.

It’s worth noting at this point that many of the health benefits of CBD and cannabis in general are not scientifically proven in statistically significant, peer-reviewed studies. This is for a number of reasons, most significantly that marijuana is still a Schedule 1 controlled substance under federal law in the U.S., making legality an issue in its study.

Clearly, the lack of scientific evidence isn’t diminishing anyone’s desire for herbal refreshment.

But what CBD and other cannabis products lack in evidence, they make up for with enthusiasm. Companies and consumers alike are eager to try CBD in various products, from food to oils to skincare, in hopes of treating anxiety, sleeplessness and other woes. If you live in a place where CBD products are legal, you’ve probably seen them everywhere. Newsweek reported that CBD sales are estimated to grow 40-fold in the next four years, reaching a value of $23 billion. The big business of marijuana and CBD — California-based Arena Pharmaceuticals is the biggest publicly traded cannabis company in the world — is only growing.

You can already find CBD candies and oils at major national retail chains like CVS and Walgreens, and in states and municipalities where it’s legal, green connoisseurs can order CBD-infused lattes and cocktails. Even retailers like Sephora, Neiman Marcus and Barneys are selling curated displays of CBD-infused beauty and skincare products. The aforementioned Newsweek article reports that big names like Coca-Cola and Molson Coors Brewing are among the hordes of companies already working on their own CBD products. Clearly, the lack of scientific evidence isn’t diminishing anyone’s desire for herbal refreshment.

Except for the FDA, that is. The legality of marijuana and CBD is a confusing and often contradictory topic, and a hard one to keep track of because it’s changing all the time at the federal, state and municipal levels. But what can be ascertained is that because so much of the CBD industry is operating outside of any kind of government oversight, legally or otherwise, the quality of products can vary widely. This is something about which the FDA and independent doctors and pharmaceutical experts have raised concerns. Apart from companies making unfounded claims about the effects of their products, the actual ingredient makeup may be inconsistent, with some products containing less CBD than their labels claim. Little regulation and nascent standards of quality mean consumers might not always know what they’re getting.

But given the broad interest in CBD, that’s unlikely to remain the case forever. The FDA may have started cracking down on extralegal CBD product sales, but in the grand scheme of things, that only means that the agency recognizes the significance of the compound. CBD probably isn’t going away anytime soon, and among the food, drug, health and cosmetic industries, the race to do it best and biggest has already begun.

Sequoia shares wisdom with Disrupt SF Battlefield competitors and Startup Alley Top Picks

Editor’s note: James Buckhouse is design partner at Sequoia. 

Last Tuesday, the teams competing in Startup Battlefield at Disrupt SF, as well as founders chosen as Top Picks in Startup Alley, visited Sequoia Capital’s office in San Francisco for a discussion with partners Jess Lee, Roelof Botha, Mike Vernal, Alfred Lin and James Buckhouse. The following is a partial transcript of the session, which was moderated by Buckhouse.

James Buckhouse: We partner from idea to IPO and beyond, but it’s partnering at the idea stage that we love the most — that moment when anything is possible. And it’s happened throughout Sequoia’s history. YouTube incubated in our office. Dropbox was an unreleased demo. Stripe didn’t have a single line of code. Apple was just two dudes named Steve. And so our favorite place to be is in the earliest moments.

We’re not here tonight to share with you lessons of our great wisdom on how company building ought to go. We’re here tonight to say that we understand how hard it is. And the three partners that you’ve got here to talk with tonight — Roelof BothaJess Lee and Mike Vernal — are people who have actually been in the trenches building companies themselves.

Customers

James Buckhouse: Great companies like Apple, Amazon and Zoom all have this one thing in common: customer obsession. That’s an easy thing to think about when you already have a billion customers, and you already have a bunch of money. But what do you do when you’re at the pre-seed stage and you want to be customer-obsessed but you don’t even have a product yet, let alone any customers? How do you even begin?

Jess Lee: I think at the very earliest of stages, all that really matters is product market fit. A common mistake we see is that a founder is only obsessed with the product, and then goes on to think, “I have my product. Let me go find a market that works for this,” when it should actually be the other way around. You should look at the market first, and then get to know the customers in that market by doing customer research.

There’s a great book by Erika Hall where she discusses how to ask the right questions to customers in order to really understand their pain points, their motivations and their needs. That’s a hallmark of some of the best companies that we’ve seen, even at the earliest stages. They spend a lot of time talking to customers and understanding what they want. Something we at Sequoia like to recommend when we work with seed and pre-seed-stage companies is to actually take the time to write down a set of customer personas. Who are your prototypical or your archetypes of different types of customers? In the very early days, you might think, “I know the customer. I can remember this. I don’t need to write it down.” But as soon as you add one new team member, who maybe isn’t as familiar with your customer, a lot of things get lost in translation.

For my company Polyvore, which was in the women’s fashion space, I had a lot of engineers on my team who were men and didn’t understand women’s fashion very well. I would always beat my head against the wall wondering why a feature they designed didn’t quite make sense, and it’s because we did the personas exercise a little bit too late. It made me wish we’d done it earlier. Once we had three very clear personas, I started to notice everything ran more smoothly. I found, whether it was the sales team or the engineering team, people started to clearly communicate the idea of what our customer really wanted. People made better decisions at all levels. That’s why at Sequoia we always encourage even our earliest-stage companies to write their customer research down immediately, way before they think they need it.

Product

James Buckhouse: How does an early-stage startup make sure that they’re on the right track and building the right product?

Mike Vernal: The key thing to me is actually not being data-driven; it’s much more about being hypothesis-driven. The problem is people think about product as art. But I actually think of product as being equal parts art and science. And I think the science part of it, which is really important, especially at an early stage, is being clear about what your hypotheses are, what you think is going to work, why you think it’s going to work and really sort of pressure-testing that on a logical level. And, if you are able to, actually pressure-testing it with real data.

One of Jess’s techniques, which I think is great, is the notion of fake doors. If you want to know whether something’s actually going to hum in the market, whether people are going to care about it, build a landing page for it. Build a sign-up button for it. Run a bunch of ads for it. Test a bunch of different marketing copy and see if people actually want the product. I’ve seen a bunch of companies use this to great effect.

I think that in general the mistakes people make with product is, one, being too artistic and not scientific enough about things. And then two, to Jess’s point, the most important thing before you have a product is finding product market fit. Usually, finding product market fit in a category is a function of two or three important things. Identifying those important things and testing them to get clarity around that first, then designing the full product, is way better than just starting with a masterpiece, and then slowly painting over and over the masterpiece until you get to something that is great.

James Buckhouse: For enterprise companies, Roelof, can you talk a little bit about the Sales Ready Product and Templeton compression approach?

Roelof Botha: If you go to our website and search for Sequoia Sales Ready Product or Templeton, you’ll find very useful content that we put together. The insight came from one of the best leaders that we’ve worked with, in a variety of companies, who argued to not just go for an MVP, a Minimal Viable Product, if you’re building an enterprise company, but what he termed a Sales Ready Product, an SRP.

The difference is that a Minimal Viable Product just gets over the hurdle but doesn’t convince your customer to jump out of their seats to buy your product. When we invested in Cisco in the late 1980s, the first product they shipped had so many bugs it didn’t work. But the product solved such an important need for the customer that they came back to Cisco and asked if they could fix it since they needed the product to work so badly because there was a fundamental problem in trying two networks at the time. And that to me was a Sales Ready Product. You’ve got something that, even if it’s not perfect, really solves your customer’s pain point.

And so to condense the whole theory behind this: Spend a little bit more time, probably another three months, maybe another four, five months, from when you would otherwise ship an MVP to ship an SRP. The reason it matters for an enterprise company is that your sales organization will be so much more effective. Your sales team will ramp up a curve far more steeply and you’ll get sales momentum much, much faster if you sell an SRP.

Culture

James Buckhouse: I’m going to do something a little bit unexpected here and call on Alfred in the back. Could you talk a little bit about what it was like at Airbnb, where they started with culture very early on?

Alfred Lin: Brian, Joe and Nate came and visited Zappos, where we offered tours, to see what the culture was all about (Alfred was COO of Zappos). At Zappos, we started writing down our core values a little late, when we were at about 300 people. And I told Brian, Joe and Nate that that was too late.

After that trip, they went back and wrote down their core values, before hiring their first employee. They knew that they had to create a new category. Home-sharing was not something that people really thought about. And so they needed people who were willing to champion the mission. And that was one of the first core values that they wrote down.

James Buckhouse: Oftentimes, people think that culture is the thing you do later on, once your business has grown large and suddenly you have a lot of people. But that’s not true. Culture matters a lot more than people think. And it matters earlier than people think. Jess, can you talk about your framework on core values?

Jess Lee: This is something we spend a lot of time on with seed and pre-seed companies, who think, “Oh, I already know my culture. I’ll wait to write it down later.” But it’s important to get it right up front. We encourage people to not pick too many core values. Generally, you want a framework that’s a core value and the behaviors you want that exemplify that core value. And most importantly, you need a story. You need some legendary anecdote or example from inside the company that really brings the core value to life.

To use Airbnb as an example, one of its core values is to be a cereal entrepreneur. The reason it’s cereal with a “C” is because at the time, Airbnb was running out of money. They weren’t sure they had product market fit, but they went to the Democratic National Convention to try the Airbnb idea when they were down to the wire in terms of money. In order to just get the word out about the business they made boxes of cereal that said “Obama-Os” and “Captain McCain.” It’s a good example of rolling up your sleeves and doing whatever it takes to get your business launched. Somehow, they actually managed to generate revenue that they put back into the business. The really memorable part of that is the cereal anecdote. Whatever it might be at your company, make sure that the lore lives on. That’s really what brings culture to life. It’s not just the value itself.

James Buckhouse: Roelof, can you talk a little bit about the culture at PayPal in the early days?

Roelof Botha: There are a couple of elements in that. One is this idea of intercept versus slope. For those of you that are fans of math or science, it comes naturally, but sometimes you get to hire people who have a high intercept. They have a lot of experience. In our case, we needed to hire people who knew a lot about financial services, because we as the early, young team didn’t. You hire people with intercept, but then you want people with slope. People who are going to learn very quickly. And at the end of the day, part of what made PayPal successful was that we had a good slope and we learned very, very quickly.

Our culture was very hard-working. We faced a bit of a crunch in June of 2000. We’d raised a bunch of money during the dot-com era, and then we were sitting with seven months of runway and no revenue, burning $10 million a month. It was a “you’re all-in” culture. Management meetings were on Saturdays, because that’s the kind of sacrifice we were going to make as a team to get to the other side. Culture was really important to the success of the company. We had a strong bond between us as team members because we were in the trenches. We had to figure out how to make this business work when the odds were against us and the press had given up on us.

Most people on the outside are going to think that you’re going to fail. Expect that. Don’t be surprised by that. Draw strength from that, and rally your team around your cause. You should ignore that kind of feedback.

Leadership

James Buckhouse: How do you discern a strong founding team?

Roelof Botha: My favorite, especially with companies at the seed stage, is to have no slides and to have a conversation with you about your business. What I find compelling is, the more I dig, the more excited I get, because your depth of knowledge, of understanding the problem that you’re trying to solve, shows itself. There are a lot of people who start companies for the wrong reasons, and they have very superficial knowledge. So as soon as you start to pressure test it, it’s clear that there’s no depth.

The founders who are the best are the ones that are so motivated to solve the problem they’re working on, they’ve researched everything. You would have found a simpler solution to the problem if you could, and you didn’t. That inspired you to start this company. As I ask you questions, you just have this depth of knowledge. You’ve thought about it so many levels deep. Those founders are the ones that keep coming up with new ideas, and that’s why their imitators don’t do so well. We see this in our industry. You come up with a great idea, TechCrunch writes about it, everybody around the world reads about it and now you’ve got 15 competitors in other countries going after what you’re doing. But guess what? They didn’t have the idea, you did. Since you had the original idea, you’ve thought about it more deeply and you can iterate faster than they can.

James Buckhouse: Jess, how about you? What do you look for to discern a strong founding team?

Jess Lee: I do agree, and I think different investors look for very different things. There is probably a notion of founder/investor fit to some extent. For me, I especially appreciate a unique insight and depth of understanding of that customer and that market. But on top of that, the other thing I think about is grit. I think that being a founder is so hard. I felt like I was on the struggle bus the entire time. Either we weren’t doing well, which was a struggle, or we were doing really well and then we were in a state of hyper-growth, and that’s also really hard. Your job changes underneath you every six months. Because even if you’re successful, everything that used to work for you as the CEO or founder is now broken because your team is now 50 people instead of 10.

What is it driving you, to either solve this problem or just driving you in general? Because it’s just not easy, and folks who give up too easily or came into this because they thought being a founder was going to be really cool, it’s not that cool all the time, so I look for that. Sometimes it shows up in the form being really mission-driven, and you have some burning desire to solve the problem. Sometimes it’s just that you’ve been underestimated your whole life and you’re really mad about it, and you want to prove yourself. There are a lot of different ways to suss out grit, but that’s one big thing that I look for.

One thing I also like to see, that is not a must-have but I find very compelling, is if you’re a good storyteller. I think that at the end of the day you have to convince your family that you’re not crazy for quitting your job to pursue this thing. You’ve got to convince early employees to join you when you can’t pay them any money. You’ve got to convince early-stage seed investors to take a chance on you and give you money when there is nothing there yet. And you’ve got to convince customers. Being able to tell a good story, both taking something complicated and making it sound simple, as well as being able to influence and talk about why your approach is interesting and different, not just better than the competitors. I look for that as well. I think that’s important.

One area where I do disagree with Roelof is that I do prefer to see slides. I think it showcases your storytelling ability. I look at a lot of consumer companies and your attention to design and detail is also an interesting thing that you can suss out with slides.

James Buckhouse: How about you, Mike?

Mike Vernal: If you can’t describe the business in a minute or two, then you need to keep iterating. Some bad meetings end up as the following: Someone will come in with 40 slides and want to convey all of the knowledge in the 40 slides in excruciating detail.

I think a couple of things. One is, many investors look at a lot of companies all day long so they might actually know more about your space than you might think. Then two, if you need 40 minutes to explain the business, marketing and all of these other things, then for an investor meeting that might work because you have that time scheduled, but for the random engineer you meet at a party who you want to get excited about joining your company, that’s going to be really hard.

The best pitch is when I’m two minutes in and I’m like, “I get the business. This is super interesting. Let’s ask all these questions.” The tough ones are 40 minutes of being talked at, where there is no real interaction.

Capital strategies

James Buckhouse: Different types of companies need different types of capital strategies. How do you all think about how founders ought to think about their strategy for capital?

Jess Lee: It’s really important to think about three things: First, what is the actual cash you need for your business? If you’re a pure software business you don’t usually need as much as if you’re building hardware or you’re making physical goods.

Second, what is the valuation that actually makes sense? True valuation, when you become a public company, when you do M&A, is actually a function of your free cash flow, or a multiple of your revenue, so just being able to understand in the long, long-term what is a likely five, 10-year-out valuation, and then making sure you don’t overshoot that just because you can. That’s another first principle.

The third thing is ownership. Doing the math, if you don’t need to raise a lot of money, if you don’t need to raise as many rounds, at the end of the day when ideally your company is acquired for hundreds of millions of dollars, or billions of dollars, or you IPO, what is your ownership at that moment? We have founders like Dropbox, that when they went public, Drew and Arash owned nearly 40% of the company. So you have to think — would you rather have 40% of a $10 billion company, or would you rather have 2% of a $20 billion company? That ownership at the end of the day is really important. So you have to think about those three things, which is a pretty complicated equation.

It really hit home for me when my company, Polyvore, went through the M&A process and it suddenly hit me that all the acquirers were not using funny VC math. They were looking at our cash flow and the multiple of revenue. Luckily, we hadn’t raised that much money, as I’d wanted to keep as much ownership as possible. I was optimizing for ownership for the team. Because of that, we actually had a really nice outcome, where everybody made money because we hadn’t over-raised since we didn’t need to. We were a pure software-based, capital-efficient kind of company, but I think not enough founders think about that from first principles, starting from the early days. They just look at who’s raising what, and how much they could possibly get. They want to maximize that, when in reality, it’s not actually the right way to think about it.

Roelof Botha: When you raise money, you’re recruiting a partner. I see too many companies, especially seed-stage companies, make the mistake of accepting funding from whoever shows up, when that’s probably the most expensive equity you’ll ever sell in your business. You could potentially be selling it to people that are not going to be there six months or six years from now, helping you close a candidate, helping you wrestle with an important strategic decision or helping you refine your business model. Those people aren’t going to be there, so it’s a recruiting decision. Take it seriously. It’s also important to check their references. Your investor is going to do references on you. Why aren’t you doing references on them?