User blogs

The COVID-19 pandemic has led to people everywhere shopping more online and Latin America is no exception.

São Paulo-based Nuvemshop has developed an e-commerce platform that aims to allow SMBs and merchants to connect more directly with their consumers. With more people in Latin America getting used to making purchases digitally, the company has experienced a major surge in business over the past year.

Demand for Nuvemshop’s offering was already heating up prior to the pandemic. But over the past 12 months, that demand has skyrocketed as more merchants have been seeking greater control over their brands.

Rather than selling their goods on existing marketplaces (such as Mercado Libre, the Brazilian equivalent of Amazon), many merchants and entrepreneurs are opting to start and grow their own online businesses, according to Nuvemshop co-founder and CEO Santiago Sosa.

“Most merchants have entered the internet by selling on marketplaces but we are hearing from newer generations of merchants and SMBs that they don’t want to be intermediated anymore,” he said. “They want to connect more directly with consumers and convey their own brand, image and voice.”

The proof is in the numbers.

Nuvemshop has seen the number of merchants on its platform surge to nearly 80,000 across Brazil, Argentina and Mexico compared to 20,000 at the start of 2020. These businesses range from direct-to-consumer (DTC) upstarts to larger brands such as PlayMobil, Billabong and Luigi Bosca. Virtually every KPI tripled in the company in 2020 as the world saw a massive transition to online, and Nuvemshop’s platform was home to 14 million transactions last year, according to Sosa.

“With us, businesses can find a more comprehensive ecosystem around payments, logistics, shipping and catalogue/inventory management,” he said.

Nuvemshop’s rapid growth caught the attention of Silicon Valley-based Accel. Having just raised $30 million in a Series C round in October and achieving profitability in 2020, the Nuvemshop team was not looking for more capital.

But Ethan Choi, a partner at Accel, said his firm saw in Nuvemshop the potential to be the market leader, or the “de facto” e-commerce platform, in Latin America.

“Accel has been investing in e-commerce for a very long time. It’s a very important area for us,” Choi said. “We saw what they were building and all their potential. So we pre-emptively asked them to let us invest.”

Today, Nuvemshop is announcing that it has closed on a $90 million Series D funding led by Accel. ThornTree Capital and returning backers Kaszek, Qualcomm Ventures and others also put money in the round, which brings Nuvemshop’s total funding raised since its 2011 inception to nearly $130 million. The company declined to reveal at what valuation this latest round was raised but it is notable that its Series D is triple the size of its Series C, raised just over six months prior. Sosa said only that there was a “substantial increase” in valuation since its Series C.

Nuvemshop is banking on the fact that the density of SMBs in Latin America is higher in most Latin American countries compared to the U.S. On top of that, the $85 billion e-commerce market in Latin America is growing rapidly with projections of it reaching $116.2 billion in 2023.

“In Brazil, it grew 40% last year but is still underpenetrated, representing less than 10% of retail sales. In Latin America as a whole, penetration is somewhere between 5 and 10%,” Sosa said.

Nuvemshop co-founder and CEO Santiago Sosa;
Image courtesy of Nuvemshop

Last year, the company transitioned from a closed product to a platform that is open to everyone from third parties, developers, agencies and other SaaS vendors. Through Nuvemshop’s APIs, all those third parties can connect their apps into Nuvemshop’s platform.

“Our platform becomes much more powerful, vendors are generating more revenue and merchants have more options,” Sosa told TechCrunch. “So everyone wins.” Currently, Nuvemshop has about 150 applications publishing on its ecosystem, which he projects will more than triple over the next 12 to 18 months.

As for comparisons to Shopify, Sosa said the company doesn’t necessarily make them but believes they are “fair.”

To Choi, there are many similarities.

“We saw Amazon get to really big scale in the U.S.. Merchants also found tools to build their own presence. This birthed Shopify, which today is worth $160 billion. Both companies saw their market caps quadruple during the pandemic,” he said. “Now we’re seeing the same dynamics in LatAm…Our bet here is that this company and business has all the same dynamics and the same really powerful tailwinds.”

For Accel partner Andrew Braccia, Nuvemshop has a clear first mover advantage.

“Over the past decade, direct-to-consumer has become one of the most important drivers of entrepreneurship globally,” he said. “Latin America is no exception to this trend, and we believe that Nuvemshop has the level of sophistication and ability to understand all that change and fuel the continued transformation of commerce from offline to online.”

Looking ahead, Sosa expects Nuvemshop will use its new capital to significantly invest in: continuing to open its APIs; payments processing and financial services; “everything related to logistics and logistics management” and attracting smaller merchants. It also plans to expand into other markets such as Colombia, Chile and Peru over the next 18-24 months. Nuvemshop currently operates in Mexico, Brazil and Argentina.

“While the countries share the same secular trends and product experience, they have very different market dynamics,” Sosa said. “This requires an on the ground local knowledge to make it all work. Separate markets require distinct knowledge. That makes this a more complicated opportunity, but one that enables a long-term competitive advantage.”

Caesar Sengupta, the long-time head of Google’s Next Billion Users initiative, is leaving the company next month, he said Monday. Sengupta, who additionally also led the company’s payments business in the past three years, is leaving the firm after nearly 15 years.

A regular fixture at Google’s events in India, Brazil, and Indonesia, Sengupta (pictured above) is best known outside the company for leading the company’s Next Billion Users unit, an initiative to make internet and services more accessible to users in developing markets.

As part of Next Billion Users initiative, Google brought internet connectivity to hundreds of railway stations and other public places in India and other markets (then shut down Station), launched Google Pay in India (which unlike Google Pay in the U.S., wasn’t developed atop credit cards) and built several products such as Android Go, Datally, Kormo Jobs and the Files apps.

Prior to Next Billion Users unit, Sengupta served as VP and Product Lead at ChromeOS, the company’s desktop operating system that powers Chromebooks.

“After 15 years with Google, Caesar Sengupta has made a personal decision to leave the company and start something entrepreneurial outside of Google. Through his time at Google, Caesar has played a key role in starting, building and leading initiatives such as ChromeOS, Next Billion Users and Google Pay. We are excited to see what he builds next and wish him the best in his new journey,” said a Google spokesperson in a statement. Sengupta’s current position at the firm is VP and GM of Next Billion Users and Payments.

“To the many, many Googlers working in Africa, APAC, LATAM and MENA, it has been inspiring to hear your voices take more weight in the products Google builds. I know there is so much more work to do,” Sengupta wrote in an email to his colleagues, which he also shared publicly.

“But we are light years ahead of where we were just a short time ago. You’ve helped digitize your economies, made Google feel local and driven Google’s investment into your countries to unprecedented levels,” wrote the Asia-based executive. Sengupta didn’t share what he plans to do next.

Under Sengupta’s watch, Google also made several investments in startups in Asia. Some of these investments include Bangalore-based delivery startup Dunzo, Android lockscreen developer Glance, and popular news and entertainment app Dailyhunt.

M Capital Management founding partners Joachim Ackermann (left) and Mayank Parekh (right)

M Capital Management, a Singapore-based venture capital firm, announced today it has closed its debut fund, M Venture Partners (MVP), totaling $30.85 million USD. It plans to invest in 40 early-stage startups, primarily seed and pre-Series A, with an average initial check size of about $500,000.

M Capital Management was founded by Mayank Parekh, whose investment experience includes launching Grange Partners and leadership positions at Southern Capital Group and McKinsey & Company, and Joachim Ackermann, former managing director of Google Asia Pacific. Other senior team members include Dr. Tanuja Rajah, previously Entrepreneur First’s launch manager, and Chethana Ellepola, former research director at Acquity Stockbrokers.

MVP, a sector-agnostic fund, has already invested in 11 companies, including one, 3D Metal Forge, that recently went public on the Australian Securities Exchange.

Other portfolio companies include behavioral health coaching startup Naluri; AI-enabled lending and credit-as-a-service company Impact Credit Solutions; alternative investment fund aggregator XEN Capital; and Cipher Cancer Clinics, which is focused on making oncological care more affordable and accessible in India.

Parekh told TechCrunch that M Capital Management was launched because “we believe that the early-stage investing space in our region has substantial room for growth. A decade ago there were very few unicorns. This has changed substantially more recently, not only because of obvious advancements bringing online previously underserved or untapped populations, but also because they venture system has developed nicely in Singapore and, for that matter, across the region with support from institutional VCs at various stages of funding need, government agency support, the advent of local accelerators and rapidly growing network of angel investing bodies.”

Parekh added that he expects to see more unicorns and “soonicorns” (or companies expected to hit unicorn valuation in the near future) emerge.

As early-stage, sector-agnostic investors, Parekh said MVP’s focus is on founders, specifically those who have “pedigree professional experience and strong academic backgrounds.” For example, Naluri chief executive officer Azran Osman-Rani was previously founder of AirAsiaX, guiding it from launch to its 2013 initial public offering in six years.

MVP will focus mostly on Singapore-based startups because it invests primarily in B2B or B2B2C companies. “We need a fertile ground for our chosen startups to launch their business models with leading corporate or business partners,” said Parekh. “Singapore provides just that. It’s the hub for market leading institutions and it’s not uncommon to see them creating opportunities for new technology or disruptive ideas.”

Most of MVP’s portfolio companies have “regional or global aspirations, leveraging Singapore as the core launch platform,” he added. MVP has also already made investments in Malaysia and India, and is actively looking at companies in Thailand, the Philippines and Indonesia.

Kate Hiscox is having a moment. Her company, Sivo, founded eight months ago, has already raised $5 million from investors at a post-money valuation of $100 million, and she is in active talks with others who would like her to consider accepting Series A funding from them.

Partly, the attention owes to the fact that Hiscox is part of the newest graduating class of the popular accelerator Y Combinator, along with roughly 350 other companies, and if there’s anything venture capitalists like, it’s freshly minted YC grads.

They also like what Sivo aims to do, which is to strike deals with debt providers for gigantic credit lines that it will then, through its API, work with many companies, big and small, to disburse via their own lending products. Yes, Sivo is making interest off money that it is simply divvying up into smaller amounts. But the real magic, says Hiscox, is in the risk management that Sivo provides. It doesn’t just parcel out debt; it helps its customers that don’t have their own risk management practices figure out who is worthy of a loan and how much.

Hiscox — who has founded a number over the years, one of which she took public on the Toronto Stock Exchange in 2018 — calls it a Stripe for debt. But one question is how Stripe itself might feel about Sivo. Stripe was also once a YC company, it also lends debt to its customers, and it seemingly doesn’t like when its investors fund potential rivals. Another question is how a company like Sivo fares when interest rates rise and the debt it borrows is no longer cheap.

Hiscox suggests she’s not worried about either scenario right now. We talked with her on Friday about the company in a conversation that follows, edited lightly for length and clarity.

TC: You’re building what you describe as Stripe for debt. But isn’t Stripe’s loan business competitive with yours?

KH: No. Sivo is the first YC company that’s building debt as a service.

The reason why [we are] is that it’s very difficult for fintechs and neobanks and gig platforms to be able to raise that capital to be able to lend money to their users at scale; that generally takes a couple of years. What we’re building out is essentially Stripe for debt, which gets these companies access to debt capital on day one. Our team has decades of experience with risk and raising debt and building enterprise tech at companies like Goldman Sachs and NASA and Revolut and Citigroup.

TC: Give me a use case.

KH: So we have more than 100 companies now in our customer pipeline, including Uber. In the case of Uber, they want to be able to offer financial products to their drivers. Maybe it’s to fund a vehicle or provide a payday advance. But Uber really can’t do that because it doesn’t want to look like an employer, and it also doesn’t want to necessarily deal with risk modeling, meaning who in their big driver base has the right risk profile [to rationalize a loan]. You plug in Sivo, and we will cycle through the Uber driver base to figure out to whom it makes sense to make a loan offer, and we do it all this through API.

TC: But Uber is not yet a paying customer?

KH: No, we go live next month; that’s an example of how Uber would use us. There are also a lot of neobanks that are three to five years old and want to start lending and really don’t know have that risk experience they need to get access to debt capital in order to have the money to be able to lend to their customers. So with something like Sivo, they’re able to integrate our service through our API, and we’re able to pretty much tell them who they should be lending to, how much they should lend, and then we offer the debt funding.

TC: Do have any debt deals in place?

KH: We signed a debt deal last week for $100 million and we’re working on another debt deal for close to $1 billion that will be announced next month.

TC: Who is your debt partner and how have you convinced them to lend so much to such a young outfit?

KH: I’m not sure I can say publicly yet who we’re working with, but we source our capital through all the usual suspects — mutual funds, pension funds, banks — and we’re able to do because as soon as we announced that we were going to start doing this as a product, we had tons of customers come and say, ‘I want this. [Trying to do this ourselves] is long and complex and painful, and we want just want to be able to do it in a simple way, like we would use Stripe for payments.’

I also have a lot of experience because I’d taken a company public and have lots of connections in the capital markets, and so does our CFO.

And there are actually a lot of banks that would love more exposure to fintechs and to a basket of YC-backed fintechs in particular because they can get yield, but the check sizes are too small for a bank. There’s also concern that the fintechs don’t really have a lot of risk experience. Meanwhile, our team has a lot of gray hair as far as risk is concerned.

TC: What kind of economic agreement do you have with that debt lender and what percentage of each loan will you charge your customers?

KH: I really can’t tell you, including because it’s going to vary from fintech to fintech; some have more complicated user models, some have bigger user bases, some operate in different regions around the world. What I can say is that it’s an incredible time for us to access debt capital from institutions because interest rates are so low and even negative in some parts of Europe. You just have to have the right team to know where to go and get it.

TC You’re also raised $5 million in seed equity funding already at a post-money valuation of $100 million, including from Andre Charoo of Maple VC, who says he’s written you his biggest check yet. Are you done raising equity funding for now? That’s already a very high valuation.

KH: We’re trying to decide now if we’re going direct to a Series A. This is our first raise, but everybody ‘gets’ our business model, so we’ve had an avalanche of investors, and some very big VCs now have reached out.

TC: Obviously, interest rates will go up. What then?

KH: When interest rates go up, all lending gets more expensive. I mean, there’s a pandemic right now and a lot of cash in the system, and there’s some talk about inflation, but we don’t really see interest rates going up for a few years.

Of course they will eventually rise, but when that happens, everybody’s rates will go up, whether you borrow on a credit card or from a traditional bank or a fintech.

Energy consumption has become the latest flashpoint for cryptocurrency. Critics decry it as an energy hog while proponents hail it for being less intensive than the current global economy. 

One such critic, DigiEconomist founder Alex de Vries, said he’s “never seen anything that is as inefficient as bitcoin.” 

On the other side of the debate, research by ARK Investment Management found the Bitcoin ecosystem consumes less than 10% of the energy required for the traditional banking system. While it’s true the banking system serves far more people, cryptocurrency is still maturing and, like any industry, the early infrastructure stage is particularly intensive.

The cryptocurrency mining industry, which garnered almost $1.4 billion in February 2021 alone, is not yet unusually terrible for the environment compared to other aspects of modern life in an industrialized society. Even de Vries told TechCrunch that if eco-conscious regulators “took all possible actions against Bitcoin, it’s unlikely you’d get all governments to go along with that” mining regulation.

“Ideally, change comes from within,” de Vries said, adding he hopes Bitcoin Core developers will alter the software to require less computational energy. “I think Bitcoin consumes half as much energy as all the world’s data centers at the moment.”    

According to the University of Cambridge’s bitcoin electricity consumption index, bitcoin miners are expected to consume roughly 130 Terawatt-hours of energy (TWh), which is roughly 0.6% of global electricity consumption. This puts the bitcoin economy on par with the carbon dioxide emissions of a small, developing nation like Sri Lanka or Jordan. Jordan, in particular, is home to 10 million people. It’s impossible to say how many people use bitcoin every month, and they certainly use it less often than residents in Amman use Jordanian dinars. But CoinMetrics data indicates more than 1 million bitcoin addresses are active, daily, out of up to 106 million accounts active in the past decade, as tallied by the exchange Crypto.com. 

“We get the total population of unique bitcoin (BTC) and ether (ETH) users by counting the total number of addresses from listed exchanges, subtracting addresses owned by the same users on multiple exchanges,” said a Crypto.com spokesperson. “We then further reduce this number by accounting for users who own both ETH and BTC.”

That’s a lot of people using these financial networks. Plus, many bitcoin mining businesses rely on environmentally friendly energy sources like hydropower and capturing natural gas leaks from oil fields. A mining industry veteran, Compass Mining COO Thomas Heller, said Chinese hydropower mines in Sichuan and Yunnan get cheaper electricity during the wet season. They continue to use hydropower all year, he added, although it’s less profitable during the annual dry season. 

“The electricity price outside of May to October [wet season] is much more expensive,” Heller said. “However, some farms do have water supply in other parts of the year.”

Basically, cryptocurrency mining doesn’t inherently produce extra carbon emissions because computers can use power from any source. In 2019, the digital asset investing firm CoinShares released a study estimating up to 73% of bitcoin miners use at least some renewable energy as part of their power supply, including hydropower from China’s massive dams. All of the top five bitcoin mining pools, consortiums for miners to cooperate for better profit margins, rely heavily on hydropower. This statistic doesn’t impress de Vries, who pointed out that Cambridge researchers found renewable energy makes up 39% of miners’ total energy consumption. 

“I put one solar panel on my power plant, I also have a mixture of renewable energy,” de Vries said. 

In terms of geographic distribution, Cambridge data indicates Chinese bitcoin mining operations represent around 65% of the network’s power, called hashrate. In some regions, like China’s Xinjiang province, bitcoin miners also burn coal for electricity. Beyond cryptocurrency mining, this province is known for human rights abuses against the Uighur population, which China is violently suppressing as part of a broader struggle to capitalize on the region’s natural resources. When critics sound the alarm about cryptocurrency mining and energy consumption, this is often the dynamic they’re concerned about. 

On the other hand, North American miners make up roughly 8% of the global hashrate, followed closely by miners in Russia, Kazakhstan, Malaysia and Iran. Iranian President Hassan Rouhani called for the creation of a national bitcoin mining strategy in 2020, aiming to grow the Islamic nation’s influence over this financial system despite banking sanctions imposed by the United States. 

Wherever nations and organizations offer the most profitable mining regulations, those are the places where bitcoin mining will proliferate. Chinese dominance, to date, can be at least partially attributed to government subsidies for the mining industry. As such, nations like China and Norway offer subsidies that incentivize bitcoin miners to use local hydropower sources. 

As the Seetee research report by Aker ASA, a $6 billion public company based in Norway, said: “The financiers of min­ing op­er­a­tions will in­sist on us­ing the cheap­est en­er­gy and so by de­f­i­n­i­tion it will be elec­tric­i­ty that has no bet­ter eco­nom­ic use.”

The best way to make cryptocurrency mining more eco-friendly is to support lawmakers that want to encourage mining in regions that already have underutilized energy sources. 

When it comes to North America, Blockstream CEO Adam Back says his company’s mining facilities, with 300 megawatts in mining capacity, rely on a mix of industrial power sources like hydropower. He added Blockstream is exploring solar-powered bitcoin mining options as a sort of “retirement home” for outdated machines. 

“With solar energy, if you’re only online 50% of the time, that’s something to consider in terms of the cost analysis,” Back said. “That’s a better option for older machines, after you’ve already recouped the costs of the equipment.”

Due to surging cryptocurrency prices, there’s now a global shortage of bitcoin mining equipment, Back added, with demand outpacing supply and production taking up to six months per machine. Emma Todd, founder of the consultancy MMH Blockchain Group, said the shortage is driving up the price of mining machines. 

“For example, a Bitmain Antminer S9 mining machine that used to cost $35 – $55 in July 2020 on the secondary market, now costs about $275 – $300,” Todd said. “This means that most, if not all mining companies looking to purchase new or secondary equipment, are all experiencing the same challenges. As a result of the global chip shortage, most new mining equipment that is scheduled to come out in the next few months, will almost certainly be delayed.”

Critics like de Vries point out that, due to market forces, industrial miners are unlikely to reduce their power consumption with new machines, which are more efficient. 

“If you have more efficient machines but earn the same money, then people just run two machines instead of one,” de Vries said. 

And yet, because cryptocurrency prices are rising faster than new miners can be constructed, Back said “retiring” old machines with renewable energy sources becomes more profitable than simply abandoning them for new equipment. In addition, Back said, robust bitcoin mining infrastructure can support communities rather than draining resources. This is because bitcoin miners can help store and arbitrage energy flows. 

“You can turn miners on and off if you get to a surge prices situation, you can use the power for people to heat their homes if that’s more urgent or more profitable,” Back said. “Bitcoin could actually support power grids.” 

Meanwhile, just north of the Canadian border, Upstream Data president Steve Barbour said a growing number of traditional oil and gas companies are quietly ramping up their own bitcoin mining operations. 

“Right now it’s hydro and coal. That’s the majority of the big industrial mining. But on the global scale, that’s going to shift more toward any cheap power, including natural gas,” Barbour said. “Oil fields already have cheap energy with the venting flares, the waste gas, there’s potential for approximately 160 gigawatts [of mining power] this year.”

Upstream Data helps oil companies set up and operate bitcoin miners in a way that captures waste and low quality gas, which they couldn’t sell before, totaling 100 deployments across North America. These companies rarely go public with their bitcoin mining operations, Barbour said, because they’re concerned about attracting negative press from Bitcoin critics. 

“They are definitely concerned about reputational risk, but I think that’s going to change soon because you have big, credible companies like Tesla involved with Bitcoin,” Barbour said. 

Even within the cryptocurrency industry, there are many people who dislike how power-intensive bitcoin mining is and are experimenting with different mining methods. For example, the Ethereum community is trying to switch to a “proof-of-stake” (PoS) mining model, powering the network with locked up coins instead of Bitcoin’s intensive “proof-of-work” (PoW) model. 

As the name might suggest, PoW requires a lot of computational “work.” That’s what miners do, lots and lots of math problems that are so difficult the computers require a lot of electricity. With regards to Ethereum, which currently runs on PoW but will theoretically run on PoS in a few years, there are hundreds of thousands of daily active addresses, sometimes half as many as Bitcoin. Like Bitcoin, a few industrial mining projects with facilities in China generate more than half of the Ethereum network’s power. Each Ethereum transaction requires nearly as much energy as two American households use per day. 

“What I like about the Ethereum community is at least they are thinking about how to solve the problem,” de Vries said. “What I don’t like is they’ve been talking about it for a few years and haven’t been able to actually do it.”

The Ethereum ecosystem uses enough energy every year to power the nation of Panama. Like Bitcoin, each Ethereum transaction costs enough for electricity costs that the money could also buy a nice lunch. Both of these networks require enough power to fuel small countries, although Ethereum usually has less than half of the million daily users that Bitcoin has. It’s clear cryptocurrency transactions require more power than Visa transactions. However, a cryptocurrency isn’t just a payments company. It is a whole currency system. 

If the bitcoin market cap were ranked as a country, by the value of the money supply, Bitcoin would come in fifth place behind Japan. And that’s not even considering adjacent ecosystems like Ethereum. In short, power consumption in the global Bitcoin economy is comparable to that of some other industrialized financial systems. It is inefficient, as de Vries points out, as are many of the systems used in emerging economies. Out of millions of users, thousands of people around the world rely on cryptocurrency for income. They are generally optimistic about the cryptocurrency ecosystem, believing it will become more efficient as the technology matures. 

“I see Bitcoin mining increasingly playing a role in the transition to a clean, modern and more decentralized energy system,” said one such Canadian business consultant, Magdalena Gronowska. “Miners can provide grid balancing and flexible demand-response services and improve renewables integration.”