Remember me?

A few months have passed since my last newsletter. Every now and then, I get into a pattern at work that so consumes my schedule that I don’t have time to sit down and write this newsletter each week.

I’m definitely in one of those patterns now. I can’t recall a time when I was busier. That said, I did announce a new investment this week in All Day Kitchens and wanted to share some of the thought behind it with all of you. I’d welcome your inquiries and feedback.

I hope to get back into more of a regular writing cadence before the end of 2021. Until then, Firehose is going to lean heavily on the “ish” in “weely(ish).”

One Big Thought

Local restaurants form the character of the communities they serve. Because food is the only we retail product we ingest, restaurants are deeply personal establishments. The late, great chef, author, and food culture documentarian Anthony Bourdain spoke eloquently on this topic:

Food is everything we are. It’s an extension of nationalist feeling, ethnic feeling, your personal history, your province, your region, your tribe, your grandma. It’s inseparable from those from the get-go. Even before we get into food professionals or food bloggers or food nerds, you’re already talking about something that people identify very closely with their identities.

A dark cloud and a silver lining.

The COVID-19 pandemic has wreaked havoc on public health and the global economy. Less discussed is pain it has inflicted on the social fabric of communities, of which restaurants are a critical part. When restaurants shut down, communities suffer too. From the beginning of the pandemic through the end of 2020, 17% of all US restaurants (over 100,000 establishments) closed their doors permanently. 16% of those restaurants had been open for more than 30 years. Local heroes like Pok Pok of Portland, Mission Chinese of NYC, Cliff House in SF, and City Tavern in Philadelphia were among the list of casualties.

The silver lining during these dark times for restaurants has been a booming e-commerce business. Online delivery had admirably grown to 8% of US restaurant sales in 2019, but COVID-19 pulled its market penetration forward by 2–3 years, according to Morgan Stanley. We now expect 20–25% of the of the US restaurant industry ($900 billion at its peak) to be fulfilled through delivery by 2025, with 53% of Americans now saying that purchasing take-out or delivery is “essential to the way I live.”

The demand-side of the food ordering equation would not be possible without delivery networks like Doordash (DASH, $74B), Uber Eats/Postmates (UBER, $89B), and more. Self-ordering is also possible thanks to Square (SQ, $121B), Toast (TOST, $28B), and other vendors who allow restaurants to manage those orders themselves. Middleware vendors like Olo (OLO, $5B) allow restaurants to connect those solutions into other systems of record, creating an omni-channel experience for restaurants.

Yet, despite this new source of e-commerce demand, restaurants have struggled to keep up. The dominant reason is that restaurant demand is inherently “peaky.” It’s even more peaky now that delivery and pick up are turning restaurants into omni-channel retailers. Restaurants have fixed labor costs that do not scale up or down with volatile demand throughout the day and week. Kitchens also have operating constraints defined by physical space, in addition to staff levels. While some QSR chains like Chipotle and Wendy’s have built out their own large commissary kitchens to serve the mass market, the roughly half of the US restaurant market that is independently owned is not operationally set up to scale like internet businesses.

A scalable back-end for independent restaurants.

Ken Chong and Matt Sawchuk saw these problems first-hand during their time at Uber, where they helped grow Uber Eats’ GMV from $0 to $6B in GMV run-rate in only a few years. They understood that the critical bottleneck to the growth of delivery was the kitchen itself.

The duo left Uber to start All Day Kitchens to address this problem. Today, we’re announcing that Lightspeed is leading the company’s $65M Series C financing. I delighted to join the board on the firm’s behalf.

Ken and Matt quickly noted a few problems with the conventional “ghost kitchen” model. First, traditional warehouse kitchens require significant scale in real estate, so the cost per square foot needs to be much lower than a typical restaurant. Those spaces don’t typically exist near where people live. So delivery orders fulfilled by warehouse kitchens in cheap, industrial zones often take significant time to reach the end-customer — resulting in a poor customer experience. Second, restaurants do not want to manage additional staff in a warehouse kitchen in order to scale. They have enough trouble hiring line chefs at competitive rates and can only manage so much labor complexity in a small business. Third, each restaurant manages its own menu and assortment in a traditional warehouse model, working with a restricted menu to streamline operations. It therefore lacks the ability to cross-sell into a longer tail of products, and participating in multi-restaurant baskets is impossible.

All Day Kitchens was designed with a novel operational model in mind. Instead of large centralized warehouse kitchens, it operates a dense network of distributed small kitchens that sit in neighborhoods close to the customer.

Restaurants prepare food in their own kitchens during down time and ship product “nearly finished” into the All Day Kitchens network. Each All Day Kitchens location (today in SF, Chicago, LA, and Dallas) serves food from dozens of these “local heroes,” and consumers can order through their favorite delivery app like DASH or UBER, or directly on Alldaykitchens.com. Users can add their favorite CPG products to orders, and even place multi-restaurant orders in the same cart. Ever feel like a bit of curry or dumplings with your burger, maybe with a bit of ice cream to finish? All Day Kitchens has got you covered!

All Day Kitchens has performed remarkably well, growing 4x YoY with industry leading profit margins and service levels. The average All Day Kitchens order arrives in 20 min — less than half of typical delivery times. Its newer locations are performing even better than its older ones, a sign of network effects from density of locations and restaurant partners, as well as growing operational sophistication. The company has built significant tech to operate locations remotely, an essential element for fast and profitable new market entry.

All Day Kitchens marks Lightspeed’s 12th global investment at the intersection of food and commerce across the US, Europe, India, China, Southeast Asia, and LATAM:

We are thrilled to partner with Ken and Matt on their journey to help the best local restaurants operate at internet scale and delight their customers with tasty, satisfying cuisine. We are grateful to work alongside friends and co-investors, Andrew Chen at Andreessen Horowitz, Keith Rabois at Founders Fund, Adeyemi Ajao at Base10, and Jeremy Kranz at GIC, all of whom are investing in this round.

All Day Kitchens is hiring aggressively to hit some ambitious goals. If you’re excited about using technology to scale restaurants and want to get on a rocket ship, check out the job listings page: https://apply.workable.com/all-day-kitchens/

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in a number of public equities, including, but not limited to AAPL, ABNB, ADBE, AMT, AMZN, BLK, CCI, CRWD, FB, GOOG/L, HD, MA, MCD, MELI, MSFT, NEE, NET, NFLX, NOW, NVDA, PYPL, SE, SHOP, SNAP, SNOW, SPOT, SQ, TMO, TWLO, TXG, V, and VEEV. (Last updated: 10/3/2021)

One Big Thought

I’ve been spending time lately thinking about how operating leverage functions differently in product- vs. services-oriented marketplaces.

Operating leverage measures the degree to which a company can increase its profits by growing revenue. Companies segment their costs into both variable and fixed costs. The former grow proportionally with revenue, while the latter do not — at least in the short term. The split between variable and fixed costs in a business determines its ability to achieve operating leverage.

For example, consider a SaaS company that puts fixed R&D dollars into a new version of its product. Whether it ships that product to 10 or 1,000 current users, the marginal cost of doing so is small relative to the up-front investment in R&D. While the company certainly needs to spend some amount of R&D to maintain its code base, a big chunk of that initial R&D should be considered fixed costs. The SaaS company therefore has strong operating leverage on its R&D spend.

To the contrary, consider a consulting firm with hundreds of employees billing their clients on an hourly basis. If the firm wants to onboard 1 or 100 new clients, it must hire a proportional number of new consultants. It can make those consultants marginally more efficient by standardizing Powerpoint templates, or building an internal knowledge base; but, the cost of these new consultants is almost entirely variable, and therefore the firm has low operating leverage in its core service.

Why does operating leverage matter? Because high leverage businesses show faster operating profit growth per dollar of additional revenue. Operating leverage is akin to the mechanical advantage you get from a physical lever. In a high leverage business, you push down with a little more revenue and you get a lot of profit on the other side:

Operating leverage can be applied to various line items on the P&L. Typically most of the costs included in COGS are variable (although not always), and most of the costs “below the line” in OpEx are fixed (although, again, not always). In the simplified case that all COGS are variable and all OpEx are fixed, then gross profit margin is a good proxy for operating leverage. Software businesses, which have high leverage, also have high gross profit margins, and the opposite is true for consulting businesses.

Where this gets interesting is in complex “low margin” businesses like marketplaces. By definition, the vast majority of marketplace GMV goes to participants in the marketplace. What is left behind is there to cover the remaining variable and fixed costs. These marketplaces can generate operating leverage from several different sources, with a significant difference between product- and services-oriented marketplaces.

Product-oriented marketplaces (e.g. EBAY, TDUP*, Faire*) primarily match supply and demand in goods and then enable the payment for and fulfillment of those goods. The way these marketplaces achieve leverage is by driving the cost of the transaction down. For example, TDUP lists a line item in its OpEx called “operations, product, and technology,” as shown below from the company’s S-1 filing:

This cost line has declined from 52% of revenue in 2018 to 50% in 2019. It went back up in 2020 to 54%, but that’s likely related to COVID complicating operations. Importantly, TDUP also breaks out its non-GAAP contribution profit per order, which increased from 23% of revenue in 2018 to 36% in 2019 (and back down to 27% in 2020, again likely due to COVID impact). In addition to payment processing costs, contribution profit per order deducts so-called “distribution center operating expenses” from gross profit:

“[Distribution center operating expenses] includes inbound shipping, inbound labor, distribution center fixed costs and management labor [emphasis added], excluding stock based compensation expense, which are included within our operations, product and technology expenses.”

We can see the overall calculation here:

So, contribution profit per order breaks out fixed costs of running distribution centers. As TDUP scales more volume in its distribution centers, contribution profit should go up disproportionately. Moreover, its inbound labor costs benefit from fixed investments in automation. While those investments are likely treated as CapEx, and are therefore not included in the contribution profit calculation, the benefits of those investments should improve the inbound labor costs included in contribution profit:

We believe that average contribution profit per order, when taken collectively with our GAAP results, including gross profit per order, may be helpful to investors in understanding our order economics. We believe that if we are successful in scaling and automating our platform pursuant to our strategy, our results will show (i) an increasing average contribution profit per order over time and (ii) a growth rate in average contribution profit per order that exceeds the growth rate of average gross profit per order due to our ability to reduce inbound processing costs [emphasis added]. Such results would likely mean that our average order economics are becoming increasingly attractive and that our investments in technology and automation in our distribution centers are having a positive impact on our average order economics.

So, this line item is crucial to understanding operating leverage at TDUP. The company believes that contribution growth should outpace gross profit growth because the former includes the benefits of operating leverage, which is exactly what the “lever” analogy would imply. Indeed, the former grew 83% from 2018 to 2019, vs. 44% over the same period for the latter.

The nature of operating leverage in a business like TDUP is that it comes from a lower cost structure. Investing in automation, for example, should enable inbound processing labor to be more efficient and therefore amortize more volume over the fixed cost of distribution center employees. TDUP can then choose whether to allow those gains in profitability to flow to the bottom line, or to reinvest them, for example, in better payouts for sellers, or lower unit prices for buyers. Regardless, the source of the leverage comes from lowering the unit cost at the contribution level.

Services-oriented marketplaces (e.g. FVRR, UPWK) primarily match supply and demand in services provided by humans. Because humans in a given field and market require a certain pay rate, less flexibility exists with respect to lowering the cost of a transaction (other than typical geo-based labor arbitrage). Instead, these marketplaces put an emphasis on high quality matches and lower cognitive overhead of running a search process for labor.

For example, in its F-1 filing, FVRR discusses its innovative approach to productizing the “Gig”:

At the foundation of our platform lies an expansive catalog with over 200 categories of productized service listings, which we coined as Gigs. Each Gig has a clearly defined scope, duration and price, along with buyer-generated reviews. Using either our search or navigation tools, buyers can easily find and purchase productized services, such as logo design, video creation and editing, website development and blog writing, with prices ranging from $5 to thousands of dollars. We call this the Service-as-a-Product ("SaaP") model. Our approach fundamentally transforms the traditional freelancer staffing model into an e-commerce-like experience. Since inception, we have facilitated over 50 million transactions between over 5.5 million buyers and more than 830,000 sellers on our platform.

Still, FVRR’s take rate in these transactions has remained around 25% for several years. Its gross margins are some of the highest of any marketplace, at over 80% today. The main way in which FVRR achieves operating leverage is not by lowering costs, but by increasing the spend per customer meaningfully over time. At the time of its IPO, FVRR had more than doubled that figure over a 6 year period:

Increasing spend per buyer essentially provides operating leverage against a fixed marketing expense. The shape of the curve means that older cohorts contribute more and more to FVRR’s overall quarterly revenue, without significant additional marketing support. The below demonstrates this methodology on a cohorted basis:

One other method to increase operating leverage in a services marketplace is to invest in labor differentiation and education. I’m not aware of a public company that does this yet, but expect to see more in the coming years.** Services marketplaces will eventually provide education and additional resources to help their workers “level up” to higher pay rates — perhaps in exchange for some sort of temporal exclusivity. The trade school of the future, in fact, may not be a school at all. It could be a labor marketplace with embedded education. Because the marketplace in question earns a fixed % take rate of billings, the incentives of all parties would be aligned.

In summary, operating leverage works differently for product- vs. services-oriented marketplaces:

• A product-oriented marketplace should focus on driving more volume to cover fixed costs at scale, lowering overall unit costs.

• A services-oriented marketplace should focus on both driving spend per customer up in a given customer cohort and investing in supply differentiation and capabilities.

• Some overlap exists in these methodologies, but firms will get the most “bang for their buck” by prioritizing the top source of operating leverage in their respective categories.

Thanks to Lenny Rachitsky for reading a draft of this post.

** If you hear of any good examples, please let me know!

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

The last year or so have spared me from all manner of the normal colds/flus that afflict a household with two preschool age children. However, this past week our entire household got our first cold since COVID started. The kids are fine now, but this weekend I was down for the count. Father’s Day basically consisted of hanging out in bed listening to podcasts — not so bad actually.

As such, I don’t have a newsletter for you today. I also plan to skip next week because I have some unrelated family obligations. So I’ll be back in your inboxes in a couple weeks!

One Big Thought

Marketplace IPOs are commonplace today, but just 5 years ago they were a rare commodity. To quote myself in Firehose #187:

Of the publicly traded, independent US marketplaces that exist today, only 6 were public companies in April 2016 — EBAY ($30B), GRUB ($1.8B), GRPN ($1.8B), ETSY ($715M), CHGG ($321M) and OSTK ($234M). These companies were worth a combined $34B, 88% of which was EBAY — a company that was already 20 years old back then.

Since then, the public markets have embraced 16 new consumer marketplaces ranging from travel (ABNB*), logistics (DASH, UBER, LYFT), apparel (POSH, REAL, TDUP*, FTCH), real estate (OPEN), e-commerce (WISH), automotive (CVNA, VRM), events (EB), and services (ANGI, FVRR, UPWK). EBAY has shrunk from 88% of aggregate marketplace enterprise value to only 10% today. Aggregate enterprise value has exploded to $474B across these 22 listed companies. Of the $440B of gain in enterprise value over the last 5 years, only 15% came from the original 6 stocks. These 16 new marketplaces have been responsible for nearly all the incremental value creation.

With all these additional public companies, we can finally make some generalized observations about the marketplace business model. The most important characteristic of a marketplace is its flywheel. The flywheel dictates not only how fast a marketplace grows, but what constitutes its moat over time. Below are some of my favorite examples from DASH, UBER, Faire*, FTCH, and ABNB*.

Flywheels give marketplaces inherent momentum. Like a powerful locomotive, they are painful to get started, but difficult to slow down. The fact that Craigslist — a product that has made very few changes since the late 1990’s — still dominates certain markets in the US is evidence that liquidity, which governs the flywheel, is the most important component of a marketplace. Everything else is secondary.

Some flywheels spin faster than others. I spent time this weekend looking at the path to IPO for a number of publicly traded marketplaces. I found that the companies can be split neatly into two groups based on rate of growth since inception.

The grouping is tighter when we plot annual gross profit instead of revenue, since it puts businesses like OPEN (6% GM at IPO) on level footing with those like FVRR (79% GM at IPO). I’ll call the upper grouping in the gross profit chart the Mega Marketplaces (EBAY, DASH, UBER, LYFT, OPEN, WISH, ABNB*). They roughly track the ascendence of EBAY in blue. I’ll call the lower grouping Steady Marketplaces (GRUB, ETSY, CHGG, POSH, REAL, TDUP*, FTCH, VRM, EB, FVRR, CARG). They roughly track the path of GRUB in red.

The differences between these two groups are significant:

• Time to IPO. The average Mega gets to IPO two years earlier (8 vs. 10 years after founding).

• Gross margin at IPO. The average Mega has a significantly lower gross margin at IPO (49% vs. 61%). That’s because the Mega’s are mostly in “commodity” industries like food, transportation, and real estate.

• Faster to $1B+ in gross profits. The average Mega achieves $1B in gross profits 8 years after its founding, at which point the average Steady has around $100M of gross profit. That order of magnitude difference is counterintuitive since the former actually has lower margins on a percentage basis than the latter.

The qualitative differences between Mega’s and Steady’s are a bit murkier. At the risk of seeming reductionist, it might all come down to a market size measured in trillions, not billions. UBER declared in its S-1 that its current serviceable addressable market (SAM) was $2.5 trillion:

ABNB had a similar feature in its S-1:

We have a substantial market opportunity in the growing travel market and experience economy. We estimate our serviceable addressable market (“SAM”) today to be $1.5 trillion, including $1.2 trillion for short-term stays and $239 billion for experiences. We estimate our total addressable market (“TAM”) to be $3.4 trillion, including $1.8 trillion for short-term stays, $210 billion for long-term stays, and $1.4 trillion for experiences.

OPEN positioned itself to address the 87% of US homes within the $100-750K buy box, which constitute a $1.3 trillion addressable market:

DASH cites third party reports in its S-1 that peg $1.5 trillion of food spend in the US, of which $600 billion is out-of-home:

We are the category leader in U.S. local food delivery logistics today and have an enormous market opportunity ahead of us in food alone. In 2019, Americans spent $1.5 trillion on food and beverages, of which $600.5 billion was spent on restaurants and other consumer foodservices.

While Steady Marketplaces must also address sizable markets, their respective paths to scale are likely slower because those markets are “only” measured in billions. FTCH, for example, is an admirable company with nearly $2 billion in net revenue, but it addresses 10-25% of a Mega’s addressable market. See below from its F-1 filing:

According to Bain, the global market for personal luxury goods was estimated to reach a record high of $307 billion in 2017, growing at a 6% CAGR since 2010, and in 2017, the personal luxury goods market experienced growth across all regions. Bain also states that online has become a larger percentage of the overall market, growing at a 27% CAGR since 2010.

Obviously, everyone wants to go after a big market. If you had to choose a trillion dollar market or a billion dollar market, why would you ever choose the latter? Well, smaller markets often have less competition and, therefore, higher margins. Those margins can present an attractive entry point for an online marketplace. It’s much more common, and probably more advisable, to attack a fat margin pool if you see one. Yet, the biggest markets of all (food, transit, healthcare, energy, etc.) are very competitive and low margin. The data clearly shows that an effective flywheel in one of these markets produces the biggest outcomes fastest.

To paraphrase a famous Jeff Bezos quote, two kind of companies exist — “those who work hard to try to charge more, and those who work to charge less.”

The Mega’s are the latter.

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TDUP, TMO, TWLO, VEEV, and V.

One Big Thought

Latin America (LATAM) is the fastest growing e-commerce market in the world. It has 640M residents with $5.8T of GDP (6–7% growth). GDP per capita is $9K, or roughly the same as China. The region has near ubiquitous mobile and internet penetration, yet it has a fraction of the tech investment dollars of comparable developing regions like Southeast Asia. One sign that capital is flowing into the region now is the meteoric rise of MercadoLibre (MELI)*, which has increased its stock price by 4x in the last 3 years, buoyed by LATAM’s digital transformation during COVID.

Private capital flows have accelerated too. 2021 is tracking to an all-time high of ~$15B of venture capital in LATAM companies, a 3x increase since 2018. The secret is getting out that LATAM will be the next global region for breakout tech startup growth.

While the macro factors are attractive, one of the more practical aspects that makes me bullish on LATAM is a new generation of experienced talent emerging to start their own companies. MercadoLibre (founded in 1999) is the poster child of the 1st generation. Companies like Nubank (founded in 2013) and Rappi (founded in 2015) exemplify the 2nd. This 3rd generation cut their teeth at these foundational startups and got an invaluable, real-world education in company building. Their market timing is even better than that of their predecessors.

In late 2019, I met a compelling founder from this 3rd LATAM generation. Fabián Gómez was one of Rappi’s earliest employees and top performers. He had recently founded Frubana — a wholesale marketplace for local restaurants. In our initial meeting, I was blown away by Fabián’s incredible passion, market knowledge, hustle, and attention to detail. I had some good context for his business since I led Lightspeed’s initial investment in Faire in 2018. My experience with Faire, as well as our portfolio companies Udaan (India) and Ula (Indonesia), informed my belief that B2B/wholesale would be the next wave of online marketplaces.

While we weren’t prepared to invest in Frubana at that time, my partner Mercedes Bent and I set out on a mission to map the LATAM ecosystem and build relationships with entrepreneurs and investors in the region. We met hundreds of companies in the following 18 months. In early 2021, we finally decided the timing was right to make our first few investments.

We happily announced our first LATAM investment in Stori a few months ago. This week, we announced that our second is Frubana!

Frubana* is a wholesale marketplace serving the restaurant industry in LATAM. 1.5M restaurants in LATAM spend over $100B procuring goods each year. The market is transacted today through a complex supply chain of middlemen in local stalls in town plazas. The below is a typical produce market in Mexico City, for example. Farmers and restaurant owners meet at these markets several times a week to exchange goods and money. To reach these markets, product passes through multiple layers of middlemen in a highly inefficient process with little transparency or consistency.

Frubana has vertically integrated the supply chain for produce, giving farmers more value for their products, reducing waste, and delivering transparent pricing to restaurants. Frubana purchases directly from farmers, aggregates produce in regional distribution centers, and fulfills deliveries using intelligent routing. While building a produce supply chain is no easy task, doing so has enabled Frubana to create a highly recurring purchase behavior with its customers, leading to best-in-class retention for a wholesale marketplace.

Frubana also sells proteins and packaged goods through a 3rd party marketplace where sellers compete for Frubana’s customers. While most customers start buying produce from Frubana, over time they grow their “share of wallet” by adding the other categories to their deliveries. Frubana has some of the best retention we’ve seen in the wholesale category. In cities like Bogotá, Mexico City, and São Paulo, tens of thousands of restaurants essentially run on Frubana.

Compared to in the US, the restaurant industry in LATAM is significantly more fragmented. 86% of restaurants are independently owned, similar to developing nations like China (81%) and India (81%). The US restaurant industry is only 47% independent, and as such large food distributors (Sysco, US Foods, Performance Food Group — in aggregate, $59B market cap) have grown to serve corporate restaurant chains. Because of their local economies of scale in distribution and ability to cross-sell, these distributors are incredibly hard to disintermediate in the US. However, in LATAM, no such oligopoly exists. By building a supply chain for produce first, Frubana has a unique opportunity to aggregate LATAM’s restaurant market.

COVID hit the restaurant industry hard in LATAM. Approximately 70% of establishments shut down or slowed operations. Despite these headwinds, Frubana grew active customers 6x and sales 3x in the last year — establishing its position as the “everything store” for the LATAM restaurant industry. This resilience in the face of adversity demonstrated how capable Fabián and his team truly are. With little digital penetration in local restaurants, we think Frubana will be well positioned to bring these businesses online, help facilitate a transition to electronic payments, provide working capital, and in general drive efficiencies throughout the entire food ecosystem.

Frubana’s mission is to make food cheaper and more accessible in LATAM. It reduces food waste thanks to its innovative supply chain. Frubana only looses 1–2% of tonnage to waste compared with over 50% in the existing supply chain. That means more money in the pockets of farmers, more efficiencies for the restaurant owner, and cheaper end-product for consumers. Any great platform gives away more value than it captures, and that’s certainly the case for Frubana.

Lightspeed is delighted to partner with Fabián and the Frubana team to build a category defining company in LATAM. We’re excited to co-invest with friends Carlo Dapuzzo at Monashees, Hans Tung at GGV, Tiger, Softbank, and more.

If you’re interested in hearing more of my thoughts on Frubana’s market opportunity, and global wholesale marketplaces in general, check out this interview I did last week at a virtual customer event for Frubana. It’s even translated into Spanish!

And, if you’re interested in working for Frubana and reside in Bogotá, Mexico City, Buenos Aires, or São Paulo, check out their careers page.

Note: I originally published this post on the Lightspeed Medium here.

Share

Tweet of the Week

Links I Enjoy

#commerce

Pumping up Peloton. →

Peloton was about as counterintuitive an early stage consumer investment as you can find. CEO and founder John Foley famously struggled to raise the first $400K angel round. A $2,000 exercise bike, with negative margins, in an industry built on breakage? Conventional wisdom said, “No way.”

But, Foley led Peloton through its IPO and built the most addictive exercise service ever to exist, with 90%+ annual retention of members. Its margins are world class on both hardware and subscription, and its market position continues to grow.

This week I read an insightful bull case on the company. It hits on some of the same factors that make Netflix such a successful business, namely that Foley regards the subscription price of $39/month as “sacrosanct” and that over time he wants to “make it irresponsible to not have a Peloton membership.”

The management has stated that the goal is to have 4 or 5 fitness platforms that are instructor-led, music, and socially driven. What initially started as a bike to democratize access to boutique fitness classes will transform into a multiplatform fitness ecosystem that will radically improve our approach towards fitness.

In addition, the post touches on another element of Peloton’s strategy, which the author calls “premiumization”.

Our day is composed of hundreds of experiences and if any given company is able to improve a relevant aspect of our life for a reasonable (but higher than other alternatives) price, we will eventually make the transition towards its products or services.

The author is making the argument that gyms used to be gritty and inconvenient (Gold’s Gym, etc). Then they got fancier, but still inconvenient (Crunch, Planet Fitness). Then they got social and fancy, and slightly less inconvenient (SoulCycle, CrossFit). Peloton is the inevitable end-state: a premium, social gym experience with all the convenience of working out at home.

This evolution of the fitness industry from gritty gyms to $2,000 connected home bikes reminded me of the premiumization coffee industry. My mother drank Folgers’ brand, freeze dried coffee every morning growing up (and so did her mother). Then Starbucks came along and turned coffee into an “experience” — charging multiples per cup of what you could make at home. Keurig made similar quality coffee even more convenient at home. Then came along even higher end home coffee machines like those made by Nespresso. Finally, we saw the rise of the “third wave” coffee companies like Blue Bottle that make a truly artisan product available commercially through all points of purchase. The price of coffee for a consumer can literally span an order of magnitude today. It turned out the market for coffee was much bigger than anyone expected as it became de-commoditized, or “premiumized.”

#media

Anime is going mainstream. →

One of the oddest trends in media during COVID has been the western rise of anime in popular culture. According to The Information, Netflix has seen viewership of anime double over the last year in the US. In Chile and Peru, anime shows are consistently in the top 10. Netflix apparently pays 50% more to anime studios compared to its usual deals, due to the demanding technical nature of developing these shows. The best IPs are also proven out by the comics (or “manga”), which means they are highly competitive and generate bidding wars. Netflix acquired the first season of Demon Slayer, for example, on a non-exclusive basis — highly unusual.

The subsequent Demon Slayer movie, released in late 2020, has now grossed over $500M globally, making it the highest grossing film of 2020! For those of you playing along at home, that worldwide gross is more than Christopher Nolan’s Tenet brought in at the box office. Furthermore, the movie starts chronologically exactly where Season 1 of the anime stopped, with no recap whatsoever. It makes no sense at all if you aren’t familiar with the series. And it’s in Japanese with subtitles! I gotta believe that everyone who saw that movie was already a severe Demon Slayer fan.

I fully expect we’re going to see more anime movies opening in wide release post-COVID and, given the prior anime TV series and manga popularity, we’ll see them beat out major live action releases in US theaters.

#tech

Boom and bust (and boom again?). →

NFTs are one of the most exciting trends in tech this year, but it’s hard not to notice the rapid decline in enthusiasm in nearly every NFT category in the last month. I point to the data in this post not as a nay-sayer, but as a conscientious observer who is interested in the underlying long-term trend.

One nevertheless bullish observation is that, compared to June 2020, active NFT wallets are still up YoY in most categories. In addition, the biggest category of decline is collectibles, which I’ve personally viewed as the most speculative category. The time it takes to establish an item traditionally as collectible is at least a generation (nostalgia needs time to set in), so a whole raft of creators claiming they’re creating new collectibles in 2021 feels premature to me. On the other hand, NFTs that are attached to experiences like games, meta-verses, or sports, have retained much better than collectibles.

My prediction is that we’re about to enter the “trough of sorrow” for NFTs, but that more productive use cases will rise from the ashes.

#science

Twisted black holes. →

Black holes are the mystery machines of physics. They sit at the edge of what we know and what we want to know to understand the cosmos. The scale and energy of these systems simply defies all intuition. This article is about a particular, supermassive black hole called M87 that its at the center of the Milky Way galaxy:

“The black hole in M87 is about the size of our solar system,” Issaoun said, yet it produces a 5,000-light-year-long current of white-hot plasma. That’s like the Statue of Liberty popping out of a marble. Some 3 trillion trillion trillion joules of energy flow up the jet each second — 500 trillion times more energy than the entire human population burns in a decade. “How could something so tiny be so powerful?”

In particular, the article discusses recent explanations of black hole jets and goes back to review old theories. The above video does a good job reviewing the basic concepts.

#culture

Here’s Johnny! →

Like many “old millennials,” I grew up watching MTV. Jackass debuted in 2000 and ran for 3 seasons on the air. It redefined the age old art of the daredevil in an irreverent, vulgar modern context. The first few seasons of the show were addictive, and Knoxville, along with his peers like Bam Margera and Steve-O, became iconic symbols of the Gen X, IDGAF, skater punk generation.

This GQ profile of Johnny Knoxville is nostalgic for me, and I bet it will be for some of you. Crazy to see your childhood heroes get a full head of grey hair.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

How big can a single brand be?

Here’s one data point — for about a week this May, the richest person in the world wasn’t a tech magnate like Bezos or Zuckerberg. It was Bernard Arnault, the majority shareholder in LVMH, the parent company of luxury brands like Dior, Louis Vuitton, and Tiffany. LVMH has a market cap over €320B and produces around €50B in annual revenue across 75 premium brands. Fashion & leather goods are 47% of revenue, while wines & spirits and perfumes & cosmetics make up 11-12% of revenue each. The remainder is comprised of watches & jewelry, as well as other non-branded activities like retail and publishing.

LVMH doesn’t disclose the revenue of individual brands. However, if we remove the retailing, publishing, and hospitality revenue, we get around €40B, or ~€500M per brand on average. At current exchange rates, that’s $600M+ per brand. The leading brands in its portfolio almost certainly produce over $1B in annual sales each. For example, Louis Vuitton, which was founded in 1854 (!), is estimated to produce over $15B of revenues at over 30% profit margins. Even at a conservative 10x EBITDA multiple, Louis Vuitton should be worth over $45B with that kind of performance. Despite its scale, only ~10% of LVMH sales come through e-commerce. Even in 2021, the world’s leading luxury brands are predominately offline businesses.

Newer brands have accumulated similar revenue and value over an even shorter period of time. One secret is owning their own distribution — first retail, then e-commerce. Founded in 1964, Nike (NKE) produced $39B of LTM revenues and is worth $215B today in market cap. 35% of its revenue, or $13B, was sold through its own website. Founded in 1998, Lululemon (LULU) produced $4.4B of LTM revenues and is worth around $51B today in market cap. 52% of its revenue is DTC and 38% comes through its own network of retail stores. Founded in 2006, Yeti (YETI) produced $1.1B of LTM revenues and is worth around $7.6B today in market cap. 53% of its revenue is DTC, and that segment is growing faster than wholesale.

The trendline suggests that $1B+ revenue brands are becoming more common and are scaling revenue faster. The one common characteristic of those who reach scale is a high percentage of e-commerce sales. As articulated by Andy Dunn of Bonobos* years ago, the digitally native vertical brand (DNVB) thesis relies on a direct relationship with an end customer to accelerate the discovery and continual optimization of product/market fit. The flywheel of building brand love simply spins faster with online transactions. As such, these brands can get to $1B+ of revenue faster than their offline predecessors.

DNVBs have finally started to go public. Founded in 2014, Casper (CSPR) had its IPO in early 2020 and is currently worth $389M in market cap on close to $500M in trailing revenues — not exactly stellar performance. In Firehose #154, I wrote about some of the problems with Casper’s strategic choices prior to its IPO. However, we should also note that its direct competitor Purple (PRPL) is worth $1.9B on nearly $650M of revenues. So, the mattress-in-a-box business actually works, just not for CSPR unfortunately.

A similar dynamic played out in the meal kit space with Blue Apron (APRN) and Hello Fresh (HFG). The former was the pioneer of the model, but the latter got the implementation right. APRN is now worth only $90M in market cap on $488M of revenues, while HFG is worth $16B on $5.2B of revenues. It crossed $1B in revenues in its first 5 years, and $5B in revenues in a decade — a record for a DNVB. In addition, The Honest Company (HNST, a Lightspeed portfolio company)* had its IPO last month and is currently worth $1.4B in market cap. It disclosed over $300M in revenues in the last fiscal year. We discussed its path to IPO in my partner Jeremy Liew’s blog post here.

All these companies were built on custom technology platforms. The newest cohort of companies are building on 3rd party e-commerce platforms instead. For example, medical apparel company Figs (FIGS) IPO’d last week with remarkable success. FIGS is itself a business that few people “got” at the beginning; yet, it clearly serves an under-appreciated niche extremely well. This statement in the company’s S-1 stood out to me:

Our mission is to celebrate, empower and serve those who serve others.

Who does that sound like? I’ll give you a hint.

In Jobs’s incredible speech upon his return to Apple in 1997, the Apple founder talked about how he was rearchitecting the company’s marketing strategy:

To me, marketing is about values. This is a very noisy world. And we’re not going to get a chance to get people to remember much about us. No company is. And so we have to be really clear about what we want them to know about us….

One of the greatest jobs of marketing that the universe has ever seen is Nike. Remember — Nike sells a commodity; they sell shoes. And yet when you think of Nike you feel something different than a shoe company. In their ads, they never talk about the products. They don’t tell you about their air soles and why they’re different from Reebok’s air soles.

What does Nike do in their ads? They honor great athletes, and they honor great athletics. That’s who they are; that’s what they’re about.

In honoring those who serve others, FIGS is carving out a unique space for consumer brand values and claiming it as their own. They also are the first company to go public explicitly with the DNVB moniker front and center. Again, from its S-1:

We Are a Digitally Native Direct-to-Consumer Brand.

We are a digitally native DTC brand that utilizes technology to deliver a differentiated customer experience. We disrupted the industry’s historical distribution model, which required healthcare professionals to physically travel to brick-and-mortar stores to purchase their uniforms. We have built the largest DTC platform in healthcare apparel, leading the industry in the shift to digital. By selling directly through our digital platform, we control all aspects of the customer experience. Further, we are able to engage with our community of healthcare professionals before, during and after purchase, through our digital platform and numerous other channels. This direct engagement enables us to establish personal relationships at scale and provides us with valuable customer data and feedback that we leverage across our organization to better serve our community.

FIGS has grown over 100% annually since 2016, with over $260M revenue in 2020. FIGS took a while to find its groove, but really took off in 2017:

The company’s gross margins are over 70%, with 28% adjusted EBITDA margins. It has simultaneously increased LTV and lowered CAC over time — although COVID likely had something to do with falling CAC in 2020. Cohort retention is very strong for an apparel business, likely due to the non-discretionary nature of the category.

FIGS is currently worth around $5B on its $318M of trailing revenues and $57M of trailing net income (~100x P/E). Much of that value was created in the last 5 years. If we started the clock 5 years ago, it’s possible FIGS could be the fastest DNVB to $1B in revenues. If so, what enabled it to move so quickly?

One anomalous factor could be that FIGS is the first public DNVB to build on a 3rd party e-commerce platform. The Shopify* CEO Tobi Lutke himself pointed this out on Twitter this week:

When Tobi said his goal was to “arm the rebels,” most of us assumed that the rebels would all be individually tiny. But, what if that assumption is false? What if building on a 3rd party platform like Shopify actually increases your odds of building a $1B revenue brand? Will we see DNVBs grow even faster to $1B+ in revenues because of the infrastructure Shopify and its competitors/partners provide?

My view is that it’s become so much easier to build iconic brands with these tools, and that many of those brands will start online with platforms like Shopify. They’ll also be able to access retail at scale much more easily thanks to wholesale platforms like Faire.* Omni-channel omnipresence is within the reach of the youngest companies now. With digital platforms, it shouldn’t be surprising to us that the some of the “rebels” are now themselves multi-billion dollar companies in record time.

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

A popular position in this newsletter is that small businesses are underinvested in relative to the critical role they play in the economy. Two weeks ago, I argued that the opportunity in SMB tech was hiding in plain sight, like mobile and cloud were a decade ago. At Lightspeed, our investments in Faire*, OYO*, and others have shown us that aggregating existing, offline small businesses can deliver massive efficiencies to the entire ecosystem. This is one of the reasons I’ve predicted a next-gen wave of B2B marketplaces in the coming decade.

In aggregate, small, offline businesses have inherent defensibility because people live in the real world. That might sound like a trivial statement, but our offline behavior is markedly different from our online behavior. The former is taking a stroll down Main St, popping your head into various shops along the way. The latter is hanging out on only a few, very popular websites or mobile apps. Case in point: we all start our searches on Google*, our purchases on Amazon*, and reflexively browse our favorite feed-based social products to kill time (Facebook*, TikTok, etc). It’s the offline equivalent of hanging out with friends on the highway median.

On the internet, near zero marginal costs of distribution and powerful network effects make this handful of companies very powerful. If you haven’t read Ben Thompson’s canonical post on Aggregation Theory, now would be a good time to check it out.

If you want to sell a product online, publish a game, or distribute a new app you built, you must pay a tax to one of several dominant aggregators. “Real estate” on the internet can therefore be much more competitive than it is in the physical world, where you pay a market-based rent, but foot traffic comes for free. A few digital companies have been able to grow initially by hacking distribution on someone else’s platform (e.g. Zynga on Facebook, Foursquare on Twitter, Airbnb* on Craigslist), but the bigger platforms eventually stop the fun. Very rarely, inherently viral digital products like Snapchat* or WhatsApp figure out an independent growth model.

This dynamic has begun to change in recent years. With its “arm the rebels” maxim, Shopify* has demonstrated a path for small, online businesses to get on the same footing as Amazon in their own niches. Wix and Squarespace have followed in this direction too. Substack is attempting to do the same in media. Vertically-oriented marketplaces like Outschool* have built platforms for individual creatives to find and engage audiences.

Yet, it’s still quite hard to build an audience online when you (a) don’t have name ID from another platform, (b) aren’t a naturally talented social media promoter, or (c) are creating in a small niche. Even if you are able to build an audience, it’s really hard to monetize it on a big platform unless you’re absolutely massive. For these reasons, Li Jin of Atelier Ventures argued that the creator economy needs a middle class:

The current creator landscape more closely resembles an economy in which wealth is concentrated at the top. On Patreon, only 2% of creators made the federal minimum wage of $1,160 per month in 2017. On Spotify, artists need 3.5 million streams per year to achieve the annual earnings for a full-time minimum-wage worker of $15,080, a fact that drives most musicians to supplement their earnings with touring and merchandise. In contrast, in America in 2016, 52% of adults lived in middle income households, with incomes ranging from $48,500 to $145,500.

A recent post by Cooper Turley and Kinjal Shah advanced my own thinking on this topic. They wrote about online “micro-economies” and what it takes for niche communities to develop an audience and a business model. The vision is that “the shift from Web2 to Web3 [will] create revenue streams which prioritize community ownership over individual ownership.”

Web3 is a new vision of the internet. It takes us away from today’s centralized internet, where data is stored in servers managed by trusted counterparties like Facebook and Google, to a decentralized internet, where a trustless network of peer-to-peer servers manages and verifies data. The network is mediated by various, composable blockchains working together towards a common goal. As Patrick Rivera explains in this presentation, the gravity of the data inside of these trusted counterparties in a traditional client-server architecture leads to monopolistic behavior, centralized control, and a single point of failure for these services.

According to Patrick, Web3 solves these problems by decentralizing governance, pushing data ownership to the individual participants in the network, and establishing a global state. The main challenges to Web3 are technological, but at the rate of change we’re seeing now in blockchain tech, many of those problems will be solved in the near future.

What would solving these problems mean for creators? Most importantly, it would allow fans to invest behind individual creative projects. For example, the aforementioned “Rise of Micro-Economies” post was written on Mirror, a decentralized blogging platform that employs one of the authors. You can see the ownership of the post is split 40%/40% between the two authors, with the remaining 20% split between other contributors — both named and anonymous.

When the scale of investment is proportional to the addressable market of the project, even the smallest creators can bootstrap. Furthermore, their success is financially aligned with the goals of their audience. Equity isn’t just defined in a corporate context; it can be defined on a creator-, or even project-level.

Some readers know that I played music pretty extensively in high school and college. My friends and I would get into bands often before they were widely known. For example, my bandmate Andrew saw John Mayer perform at a local record store (when those existed) in 2001. Mayer’s iconic debut album Room for Squares came out that June. I’m fairly certain that Andrew purchased the album at that event. Imagine if he were able to roll his purchase into tokens tied to the future royalty streams associated with “Your Body is a Wonderland” or “Why Georgia?”. As a fellow musician and fan of acoustic guitar music, Andrew was just as much of an authority on Mayer’s future success as any big label A&R executive.

If the micro-economy vision comes to fruition, college kids hearing their favorite bands play in local coffee shops will be able to participate in those bands’ future success. Doing so will make them even more likely to promote those bands to others — completing a virtuous cycle that funds more artistic production. The same principle could apply to writers, artists, and any manner of creator. The power shift from aggregators to the creators and their fans will be a massive value transfer away from middlemen to those who make things people love, and to those who enjoy them.

I’m excited to participate in this shift as an investor and, most importantly, a fan.

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

Hello, all.

I sacrificed a few recent weekends to work, so today I decided to spend the day with my family to celebrate Mother’s Day. That means no newsletter today, but hopefully I’ll be back in your inbox next week.

If you’re looking for interesting stuff to read, check out my last 3 posts on various marketplace-related topics:

• #187: A new marketplace wave

• #188: Payment tech is motor oil

• #189: Hiding in plain sight

Best,

Alex

One Big Thought

Great investments tend to exploit information asymmetry.

While some information does a lot of work hiding from you, requiring years of research to unearth, a surprising amount hides in plain sight. Case in point: you could have invested in Apple* the year after iPhone came out (2008) and beat the NASDAQ by a cumulative 31% CAGR return through today. Or, you could have also invested in Amazon* when it first publicly broke out the financials for AWS (2015) and beat the same index by a cumulative 23% CAGR. On a risk and fees-adjusted basis, these returns are comparable with top quartile VC fund performance. Anyone with a brokerage account could have benefited from these investments.

Those who invested in Apple and Amazon over this period likely held a contrarian viewpoint on the potential of mobile and cloud computing, respectively.

How do you like ‘dem Apples?

At the end of 2008, the App Store was only 6 months old, with only a few thousand apps. The top free apps included a flashlight utility, several casual games (Tap Tap Revenge, PAC-MAN Lite, etc), and an HTML-5 based Facebook* app that the company later scrapped and Zuck called “the biggest mistake we’ve made as a company.” All the top paid apps were casual games.

With this app lineup, it would have been relatively easy to relegate the App Store to the junk bin of tech. Furthermore, not only was Apple a minority in PC sales, but its late entry into the smartphone market earned the infamous derision of Steve Ballmer, then CEO of Microsoft:

“That’s the most expensive phone in the world, and it doesn’t appeal to business customers!”

Ballmer’s critique turned out to be irrelevant. iPhone was a blockbuster for consumers, who later brought it forcibly into the enterprise with the “bring your own device” movement. The toy-like apps that littered the early App Store charts gave way to Uber, Foursquare, Instagram, HotelTonight, and many of those early native iOS apps that extended the platform’s utility. The rise of the iOS ecosystem was hard to predict looking at the App Store charts in 2008, but, if you were paying close attention, you could have identified the growth in apps and bet on the Apple stock.

Head in the clouds?

The same contrarian setup applied to cloud computing in the mid-2010’s. Below is the Gartner Hype Cycle for Cloud Computing, published in mid-2012, which shows cloud computing in the so-called “trough of disillusionment” on a 2-5 year journey towards the “plateau of productivity.” Other similar concepts (elastic multitenancy, PaaS, dbPaaS, public cloud storage, IaaS, etc) were at similar or longer lead times.

IT professionals threw a ton of shade at the cloud back then. I recall hearing a lot of arguments like this one around the security model of cloud computing. We also saw regulatory pushback around the geographic residency of data, or specific industry requirements around the collection and storage of personally identifiable information that the public cloud couldn’t easily support.

You needed to be one of the minority of IT professionals connected to the startup ecosystem to see the potential of the cloud at the time. Almost no one in Silicon Valley purchased servers. As startups grew into large companies, they set the standard for how progressive, flexible infrastructure would be deployed in the enterprise.

I recall a watershed moment, in 2015, when Mark Pincus returned to Zynga as CEO after a brief ouster. In cost cutting mode, he announced that the company was moving its massive computing infrastructure to AWS from its private cloud. He said Zynga would spend $150M with Amazon to do so:

“There’s a lot of places that are not strategic for us to have scale and we think not appropriate like running our own data centers,” said Pincus.

“We’re going to let Amazon do that.”

That last sentence was particularly telling. Here’s how some more progressive industry observers viewed AWS at the time:

It won't be long—maybe not in 2016, but soon—before any computing project that doesn't happen in the cloud will have to spend many cycles justifying what will be seen as an old-fashioned and inefficient approach. 

A few companies still generate their own electricity, but they're rare and special cases. That's the kind of market dominance cloud computing is headed for. And we're likely to get there faster than we think.

Of course, this was the right point of view, but given the concerns about information security and regulatory prohibitions, even in 2015 the public cloud’s dominance in IT infrastructure was not an obvious call. The information was out there, but you needed to speak to the right IT people to appreciate its gravity.

Small is the next big thing.

Small businesses are quite literally everywhere.

According to the SBA, the U.S. has 20 million businesses with fewer than 20 employees. In sectors like food services, construction, wholesale trade, real estate, and agriculture the majority of businesses are SMBs. Yet, while SMBs are ubiquitous, they are notoriously difficult to monitor, track, and analyze.

For example, when we made our initial investment in Faire* in 2018, I spent months looking for data on independent retailers. Because of the industry’s fragmentation and informality, I could only rely on official U.S. Census data from 2012! Beyond that, I had to rely on my intuition and past knowledge of the retail ecosystem. To most outside observers, while independent retail was admittedly everywhere, few thought of it as a large market. I knew enough to believe it was.

Working on a new SMB-focused investment opportunity this week, I realized that the lack of data on this category might, however, be a temporary problem.

If we more see companies like Faire, or Slice, or Toast, succeed, we will necessarily start mapping out the surface area of the SMB ecosystem for the first time. It will be like exploring the ocean’s depths with a new submersible that can dive deeper and reveal the unknown. We’ll make new discoveries, and some of them will influence the way we think about the world. I suspect that these discoveries will validate what has mainly been intuitive to me so far — that SMBs are still vastly underinvested in from a technology perspective and present an economic opportunity similar to mobile and the cloud did in their respective generations. In the context of data yet to be unearthed, what will be seen as contrarian today will therefore be seen as consensus tomorrow.

Share

Tweet of the Week

Links I Enjoy

#climate

Drowning in vegetable oil.→

More land globally is devoted to vegetable oil crops than all fruits, vegetables, legumes, nuts, roots, and tubers combined. Oil crops are second only to cereals in their global land area, but are growing significantly faster — doubling since the 1970’s.

Oil crops are some of the most land inefficient crops in existence, requiring an order of magnitude more land per kg of food product than barley or potatoes, for example. Higher land use leads to more deforestation, which furthers the outsized climate impact of these crops. And for the benefits of their caloric density, oil crops are some of the most nutritionally poor crops available. They account for 30% of global crop lands, but 0.01% of the world’s essential nutrients.

#media

It’s a vibe. →

Like many in my generation, my friends and I grew up making and trading mix tapes. We would often capture songs off the radio, or later rip full CDs of bootlegs from concerts downloaded over the internet. These mixes could be a flex — showing that you had that rarest of rare performance of a Pearl Jam B-side — or the expression of a vibe — unrequited love, teen angst, etc. Sometimes a list of songs can say something that is hard to put into words.

All that’s old is new again, as “oddly specific mix tapes” are starting to emerge on Spotify* and Facebook.* They appeal to a similar need that I found in my middle and high school days, but with the context of a global pandemic:

All of the admins I spoke to say that they have seen the group’s membership grow over the past year—not just a natural incline, but one that they believe is a result of the coronavirus pandemic. As increased loneliness and stress have contributed to declining mental health, people have turned to online communities to seek reassurance and companionship. “I guess loneliness is, more than ever in recent memory, a common experience for many,” says Eli. “This has driven perhaps more extroverted people online to find social interaction, and the chance to find like-minded people worldwide has, and speaking from a personal perspective, really helped me through some dark times.” While he doesn’t believe that everyone in the group is looking for this kind of deep connection, even the posts that are only intended to “get a giggle” out of their audience still inspire and drive the community.

Enjoy “Eating Painkillers at McDonalds” to get a sense of what they’re talking about.

#tech

Who disrupts the disrupters? →

Packy McCormick at Not Boring wrote an important piece about what comes after the web aggregator phase of the internet. We’ve heard a lot recently about how crypto will enable a new set of disrupters, but not so much how this magic will occur. Packy leans on Clay Christensen’s theory of disruptive innovation to illustrate how crypto tech can enable “better than free” platforms:

While the Aggregators definitionally serve nearly everyone, they actually do overlook a segment of their users: the people who aren’t happy to just use the products for free, but who want to earn ownership, privileges, and money for their participation, and who are willing to accept a lower-quality user experience to get those things. This is where they’re susceptible -- while they generate tiny profits on each user, they’re not willing to go negative and actually pay those users. 

So how do you low-end disrupt a high-quality, ultra-low-margin or even free product? How do you get a price lower than the $0 they’re paying today? You go negative; you actually pay people to use the product, in a currency that gets more valuable as more people join. 

#science

Speed running Pi. →

At age 23, Isaac Newton completely changed the way we calculate the irrational constant Pi. For 2,000 years, upper and lower bounds for Pi were approximated by subdividing polygons into distinct shapes. Newton played around with the binomial theorem and used the calculus he invented to create an infinite series that converges to Pi much faster than the polygon method. This video by Veritasium walks through how Newton derived his brilliant insight.

#culture

Peak shounen. →

Anime and manga (the Japanese comic books from which anime are derived) have very specific genres that are designed to be marketed to different groups of fans. The “shounen” genre is marketed towards high school age boys and often follows a teenage protagonist through a progression of battles and obstacles towards a stated goal. Many shounen anime can be fairly predictable — they all have similar plot notes and progression dynamics.

Within the last year, however, anime fans have been blessed with two genre re-defining shows that will become classics. One is Attack on Titan (I’ll write about this one later because it’s incredible), and the other is Jutusu Kaisen. I wouldn’t watch this video unless you’ve spent some time on the series (spoilers!), but if you have it does a great job laying out why Jutusu is the future of shounen in so many ways.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

Inspiration for this newsletter comes from all over the place.

I save articles all week long to read over the weekend, and those form most of the topic base for Firehose. In addition, I sometimes draw inspiration from a tweet, a blog post, or a conversation I’ve had with a founder. Occasionally, one of my partners says something so insightful in a meeting that I just have to write about it.

This week’s inspiration comes from a stranger source — a piece of marketing material from a wealth management firm. Bessemer Trust sends out a Quarterly Investment Perspective, and this past one was on the topic of “The Future of Money.” It’s quite good and can be accessed online here.

A particular section stood out to me, which I’ll quote verbatim:

E-COMMERCE AS THE PAYMENTS INCUBATOR

A virtuous cycle exists when an online marketplace is met by robust digital payments infrastructure. In many cases, the former precedes the latter. In those instances, the marketplace platform has served as an incubator for a payments solution that ultimately extends far beyond the confines of its initial host. This playbook has been adopted globally, much to the benefit of shareholders given its track record of significant value creation, often stemming from expansive growth of these once-incubated payments platforms.

The piece then goes on to discuss the canonical example of the intersection of payments and marketplaces — EBAY/PYPL*:

In 2002, eBay was processing ~$13 billion in merchandise annually. However, 60% was paid for by check or money order, and the sellers usually waited for the checks to clear before shipping their items, resulting in a transaction process that could take as long as two weeks. PayPal’s technology, by contrast, allowed consumers to make and accept payments over the internet without using a credit card, with the bulk of its business coming from eBay auctions. eBay acquired PayPal for $1.5 billion that year, removing friction from the eBay buying process and supporting the virtuous cycle of marketplace network effects. While eBay continued to scale its auction platform, PayPal expanded to the other merchants and has participated directly in the large and growing addressable market of online commerce.

If marketplaces are engines, then payment tech is motor oil. Removing friction from an economic engine lets it work harder and faster. PYPL accelerated EBAY’s network effects by lubricating transactions, and that lower friction translated into more profits and growth.

The article goes on to discuss other global examples:

• BABA* launched Alipay’s escrow service in 2004 to solve trust problems in its B2C marketplace Taobao, which launched a year prior. Alipay eventually enabled users to do more with their account balances, supporting lending, wealth management, insurance products and more today.

• MELI* launched in 1999 to a largely unbanked population in LATAM. Over the next decade, its MercadoPago product rolled out to absorb 1/3rd of transaction volume on the platform. Today it covers nearly all transactions on platform and also supports a significant number of off platform transactions.

• SE* launched in 2015 in Singapore, and its Shopee platform has grown into an e-commerce leader in Southeast Asia. 70% of the region’s population is unbanked or underbanked, so SE launched SeaMoney, a mobile wallet that has grown to 30% of gross orders last October.

Money has gravity. Once users start to accumulate a balance of local currency, the friction is lowest to spend it on the platform itself, or on services directly linked to the platform. That’s why BABA, MELI, and SE have all shown success bringing tightly integrated, non-marketplace use cases to their digital wallet users. While payment tech starts as the lubricator of the marketplace, eventually it forms the foundation of new use cases. That means consumer payments, when incubated inside a marketplace, often become bigger than the marketplace itself. As noted in the Bessemer Trust report, PYPL is now 7x more valuable than EBAY, 6 years post-spinout:

These observations also have implications for the emerging wave of B2B and wholesale marketplaces I discussed last week.

SMB payment tech is quite far behind where consumer payments sit today. We should expect that much of the innovation in lending, insurance, factoring, and other financial services for SMBs will come from marketplaces, not neobanks. In fact, due to the difficulties some B2B marketplaces have extracting sizable take rates (again, read my post from last week!), you might expect digital payments to be the dominant monetization model in the category. Fintech infrastructure providers like Finix* and Synctera* (both Lightspeed portfolio companies) should see strong uptake in the B2B marketplace category as these businesses arm themselves with payments features.

If you see great examples of B2B marketplaces deploying payments tech, please send me a note. I’d love to feature them in the coming weeks.

Share

Tweet of the Week

Links I Enjoy

#commerce

Hip to be Squarespace.→

Squarespace filed its S-1 after nearly two decades as a private company. The prospectus identifies over 800M small businesses globally in its target market, and notes that half a million of these are started each month just in the US. 46% of small businesses are not online today.

Squarespace has 3.6M unique subscribers at a $187 ARPU. When your ARPU is that low, you need to establish a hyper efficient go-to-market strategy. With over $152M of free cash flow and a “Rule of 40” number of 52%, Squarespace has both growth and profitability dialed in.

#media

The media clock. →

Someone on Twitter shared this old Wired article by Aaron Zamost of Square*. It’s just as relevant in today’s media environment. Aaron’s metaphor is a clock. In my conversations on this topic, I’ve often thought of media as a pendulum that swings from friend to foe. Because media needs to create conflict to drive engagement, it often builds up a company in order to knock it down later. Then it loves to tell the redemption story.

Unfortunately, it’s quite hard to opt-out of this model. So instead startups need to embrace the cycles and understand where they are in it. Planning for the inevitable crisis also helps navigate through the fog of war when it eventually happens.

#tech

Apple’s dominance with teens. →

Airpods launched in 2016 to confusion and jeers. I think I might have called them “ear cigarettes” at some point. But when I tried them, I instantly fell in love. Airpods are the best product Apple* has created since the iPhone.

The teens agree. A whopping 70% (!) own a pair (up from 52% a year ago), slightly less than the 88% who own an iPhone. Only 34% of teens own a smart watch, but 84% of those who do own an Apple Watch. Apple’s utter dominance with teens, and the seamless interoperability between its devices, means that it’s quite unlikely they shift to Android in later years.

#science

Muon along. →

We don’t encounter new physics very often, and when we do, the results are often attributed to experimental or human error. Scientists recently observed muons in a particle collider oscillating at a rate not explained by the Standard Model of physics. That would suggest the presence of a particle or force not yet predicted by the Standard Model. The measurements have only a 1 in 40,000 chance of being a fluke, or something like 4 standard deviations from the mean.

#culture

Edible art. →

I’ve been baking a lot during quarantine. It appeals to my inner chemist and somehow breaks up the monotony of sitting in my house all week. My kids also love getting their hands into the ingredients and getting to consume the results of their labor!

As the YouTube algo has adjusted to my new passion, I come across videos like these which are mesmerizing. I would never have the patience to make something this intricate, but admire those who do.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, ADSK, AMT, AMZN, BABA, BRK, BLK, CCI, COUP, CRM, CRWD, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MELI, MSFT, NFLX, NSRGY, NEE, NET, NFLX, NOW, NVDA, PINS, PYPL, SE, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

Consumer marketplaces in the US have exploded in value over the last 5 years.

Of the publicly traded, independent US marketplaces that exist today, only 6 were public companies in April 2016 — EBAY ($30B), GRUB ($1.8B), GRPN ($1.8B), ETSY ($715M), CHGG ($321M) and OSTK ($234M). These companies were worth a combined $34B, 88% of which was EBAY — a company that was already 20 years old back then.

Since then, the public markets have embraced 16 new consumer marketplaces ranging from travel (ABNB*), logistics (DASH, UBER, LYFT), apparel (POSH, REAL, TDUP*, FTCH), real estate (OPEN), e-commerce (WISH), automotive (CVNA, VRM), events (EB), and services (ANGI, FVRR, UPWK). EBAY has shrunk from 88% of aggregate marketplace enterprise value to only 10% today. Aggregate enterprise value has exploded to $474B across these 22 listed companies. Of the $440B of gain in enterprise value over the last 5 years, only 15% came from the original 6 stocks. These 16 new marketplaces have been responsible for nearly all the incremental value creation.

Consumers have been the prime beneficiaries of the marketplace model of innovation. Connecting fragmented populations of individual buyers and sellers has created a surplus of economic value for consumers, a fraction of which has been captured by the platforms themselves in the form of a take rate (ranging from 5-50%, depending on a variety of factors, but typically settling in the 10-20% range for most):

Offline buyers and sellers face all sorts of issues that an internet platform can solve: trust & safety, logistics, financing, etc. Solving these problems creates a wedge to disrupt incumbent middlemen like travel agents, auto dealers, real estate agents, ticket brokers, and department stores. While much of the low hanging fruit has been picked in consumer marketplaces, we still see new examples popping up, including in education (Outschool*), coaching (BetterUp*), and entertainment (Cameo*).

While nearly half a trillion in enterprise value created by US consumer marketplaces in the last 5 years, we may see an even greater wave over the next 5 years in wholesale marketplaces.

Wholesale value chains have similar problems. They lack price transparency, suffer from inefficient logistics, and struggle to collect and distribute payments on time. Furthermore, the dollar values are significantly higher and margins are often lower — which means that errors have significantly greater consequences. As such, many wholesale markets have built-in redundancy to deal with inefficiency and minimize risk.

The most challenging aspect of wholesale marketplaces is how you fight disintermediation. If two businesses continually work together, they will have low willingness to pay to a marketplace that connects them. Two consumers who want to coordinate an Uber ride (one in the other’s car) have few options to practically arrange a pick up. But, a grocery store might purchase bread from the same bakery it has been working with for years. At the end of the day, the owner can always pick up the phone and place an order the old fashioned way.

Creating a repeatable purchase pattern between two recurring buyers is therefore the main challenge and opportunity for a wholesale marketplace. Extracting a 10-20% take rate is only possible when each transaction has more value than the matching itself. Here are a few ideas that could help:

1. Take risk away. Trying new products is risky in a variety of ways for wholesale buyers. New products might not sell through. New vendors can have variable quality. A buyer might not have a liquidation channel to recapture value in the event an item does not sell. To solve this problem in its own business, Faire* guaranteed free returns on new brand orders and, over time, managed the risk itself as a feature of the platform. This innovation allowed retailers to shop without worry for new items and find new ways to grow their businesses. Because retailers want to maintain “freshness” of inventory, there’s a recurring need for new merchandise and, definitionally, these new brands are incremental to the platform.

2. Come for the logistics; stay for the network. Some wholesale industries are limited by delivery times and availability of product. Produce, chemicals, and construction materials are all industries where products must arrive when expected. Fruit can rot, chemicals can expire, and steel beams can miss their construction window. Providing more reliable logistics can be the reason that a buyer will reorder from the same vendor and still pay the required take rate.

3. Business-in-a-box. Since I wrote my original post on this topic in 2019, I’ve seen more examples of companies that are giving away vertical software for free to SMBs in order to insert themselves in the flow of payments and services. Many SMB sectors have some form of order management, point-of-sale, or other horizontal software. But few have software truly built for their own vertical. If you can build the system of record for an industry and make it freely available, SMBs will be inclined to order through your system as well.

If a wholesale marketplace solves the disintermediation problem, it can often generate multiple orders per month. Few consumer marketplaces can come close to this level of purchase frequency, so the success metrics for wholesale marketplaces tend to look much different. The north star metric I follow in wholesale is “share of wallet.” Often a wholesale marketplace will “land” a customer with a single SKU line that has high purchase frequency and low margin. The goal is then to “expand” through cross-sell into the other buying needs of that business — ideally the less frequent, high margin purchases. The marketplace can cross over into a very defensible position once the customer is purchasing a meaningful amount of their volume through the platform. I often see a share of wallet tipping point after which churn become negligible (other than a customer going out of business).

Dollar retention is therefore very strong for successful wholesale marketplaces. For SaaS companies, valuation multiples are strongly correlated with net dollar retention, and I expect wholesale marketplaces will exhibit the same dynamics as future public companies:

Increasing scale benefits these marketplaces on the customer acquisition side as well. As product catalogues grow, conversion rates go up. While cost of leads might go up over time, the increased conversion rate can provide a buffer to CAC and keep the unit economics of the business strong. We see similar effects in consumer marketplaces like DASH, which has been able to increase its LTV/CAC over time:

In short, consumer marketplaces have become relatively mature, but wholesale marketplaces are just getting started. I’m excited to meet more entrepreneurs building these businesses in the coming years and back more of them.

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ABNB, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NFLX, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TDUP, TMO, TWLO, VEEV, and V.

One Big Thought

I’m attempting to the take the week off with family, but I wanted to write a short post to congratulate the team at thredUP on its IPO on Friday!

thredUP holds a special place in my heart both personally and professionally.

On a personal note, co-founders James Reinhart and Chris Homer were friends and classmates from HBS. I have fond memories of brainstorming with them after class. I remember multiple coffee meetings with James over the first few years when he was testing different business models. I recall visiting a small, early office in SF, where boxes were piled everywhere, with clothing spilling out. And I remember walking the floor of the company’s first distribution center in Pleasanton, CA, when it was an empty shell. A year later it was filled with machinery, fixtures, people, and bags of clothing piled high. Getting to see friends build an idea into a finely tuned operation with thousands of employees was both thrilling and gratifying.

Professionally, thredUP is a reminder that valuable platforms take time and considerable effort to build. At my former firm Highland Capital, we first invested in 2012 and celebrated an IPO 9 years later. thredUP was founded 12 years ago. James’ vision has been the same since nearly the beginning, and much of the team has stayed together for the entire journey. Strong leadership and long-term thinking were essential to reach this important milestone.

Here a few other lessons I learned along the way:

• Retail is detail. When you’re moving and selling physical products, every penny of cost matters and must be counted. Obsessing over the details is critical.

• Improve 1% every week. You can’t just “flip the switch” in operations and see significant unit economic improvements overnight. Nor do you want to jump from manual processes to fully automated right away. The best operators chart a path for consistent, periodic improvements and chisel away at inefficiencies.

• Marketplaces must balance. Supply and demand are invariably linked in a marketplace. You can’t grow demand arbitrarily fast because you’ll run out of supply and vica versa. thredUP had to develop sophisticated regulators and incentive systems on both demand and supply to keep the marketplace in balance as it scaled.

• The board is a braintrust. James recruited a board of directors with vastly different personalities, backgrounds, and opinions. He viewed the role of the board as a place to test his ideas, get feedback, and ultimately inform his decisions. He managed the board accordingly and held us to high expectations for intellectual output.

For more thoughts on the thredUP IPO, check out my tweetstorm from Friday:

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NFLX, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

Physicists are trained to look for invariants. These are variables that stay the same under some transformation. The speed of light, for example, is an invariant across all inertial frames of reference. From that observation alone, the theory of special relativity magically falls out of the math. As such, physicists know that if you find an unchanging variable, you have likely uncovered a deep truth about the universe.

Invariants are less rigorously defined in business or investing, but they still possess significance and inform the creation of useful principles. One of my favorite examples is the “Rule of 40” that illustrates the fundamental tradeoff between growth and profitability. The Rule of 40 variable (call it R) is equal to the sum of the annual revenue (or ARR) growth rate % and the company’s EBITDA (or FCF) % of revenue. You can change one of the latter two variables, and in theory, the other should increase or decrease to offset — yielding a constant R, ideally above 40%:

A company with R>40% is a more efficient grower than roughly half of the public comp set. In the above graph, while only 42% of public companies in 2016 had R>40%, that 42% represented 77% of public company value for SaaS companies. R is therefore a highly useful metric to separate great performers from also-ran’s.

Other invariants can be industry specific. The slope of Moore’s law, which predicts the doubling of the number of transistors on microchips every two years, has been an unrelenting invariant over five decades of progress in semiconductors:

Dunbar’s number — the cognitive limit to the number of people with whom humans can maintain substantive, stable relationships — is another invariant in social networking. Social products are routinely built around the Dunbar limit and its related quantities:

I’ve been searching for invariants in the arena of climate tech for the last few months. Climate is uniquely tied to the physical world, and therefore its invariants have some relationship to physics. For example, Carnot’s theorem states that a heat engine (a device which does work by harvesting energy moving from a hot to a cold reservoir) cannot operate more efficiently than the maximum provided by this formula:

T_C and T_H are the absolute temperatures of the heat engine’s cold and hot reservoirs, respectively. The most straightforward way to make a heat engine like a power plant produce a higher output is to increase its efficiency by increasing T_H, the combustion temperature. (You could also lower T_C, but if the cool reservoir is the ambient environment that’s less practical.) A fuel like natural gas burns at a fixed temperature at a certain pressure and air concentration (e.g. 1,950 degrees C in standard atmosphere), but you can increase pressure or oxygen concentration to make it burn hotter. Both paths require bigger pressure vessels and more safety precautions and, as such, have diminishing economic returns to scale. Yet, due to Carnot’s theorem, there is some invariant proportionality between the size of a thermal engine (below in MW output) and its efficiency:

I credit my former Highland Capital partner Paul Maeder (who is a brilliant investor and mechanical engineer) for pointing out to me that this invariant implies that the business model of thermal power generation is centralized. Carnot’s theorem means bigger power plants are more efficient; therefore, you build them as big as possible (until you see diminishing returns), and transmit their electrons over wires long distances to where they are used. This single invariant relationship explains the architecture of our centralized power grid completely.

Solar photovoltaics (aka solar cells), however, adhere to a different set of physical laws. Solar cells are not heat engines. They convert photons into electrons directly. The maximum efficiency of a solar cell depends more critically on its structural properties (especially a quantum mechanical property called its “bandgap”) than the size of the solar array. Specifically, the so-called Shockley-Queisser limit is the maximum efficiency of a single junction cell in typical solar illumination and varies with the bandgap as shown below:

The business model implication is that solar arrays can be as small as an application warrants, without suffering losses in efficiency. Solar is therefore a disruptive source of power because it can be distributed. Efficiency is invariant with size. The two variables are decoupled. This is why nearly 2,000 of the 2,500 non-residential solar facilities in America in 2018 were only 1-5 MW:

Solar constituted 43% of new grid capacity in 2020 and shows no signs of slowing down, thanks to precipitous drops in cost per watt to below $1 on a module level. As shown below, solar costs have followed their own version of Moore’s law, with the exception of a mid-2000’s supply shortage:

Solar is, therefore, disruptive to incumbent thermal power sources due to its independence from Carnot’s theorem and its exponential economies of scale. If you were to build a grid from scratch in the next 5-10 years, you’d be more likely to do so in a distributed manner — with solar and batteries working together. If you had noticed these invariants and invested in enabling tech for distributed solar like micro-inverters (SEDG*, $15B market cap, up 5x in 3 years) or enabling business models like solar financing (RUN*, $11B market cap, up 7x in 3 years), you’d have done quite well.

As such, I believe that the business of energy, the environment, and climate change are inherently tied to physical invariants. We will see more examples in the coming years as the current climate tech boom pushes forward.

If you have more ideas for new invariants for climate tech, please send them along!

Share

Tweet of the Week

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NFLX, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

Coupang is synonymous with e-commerce for a big chunk of the 52 million people living in South Korea. The company is largely unknown in the US, but is nonetheless a fascinating foil to our own e-commerce ecosystem in the US. In today’s newsletter, I will compare Coupang’s model to the prevailing ones in US e-commerce, and note potential changes on our domestic horizon.

One housekeeping note: no links again this week. Ran out of time. But will hopefully get them back in the next few weeks!

End-to-end logistics

Americans often take for granted the incredible logistics infrastructure that has been built since WWII. Without that infrastructure, Jeff Bezos has noted that it wouldn’t have been possible to start Amazon* in the 1990’s:

"I didn't have to build a transportation network to deliver the packages," Bezos told Norah O'Donnell. "It existed: It was called the post office."

The comments were designed to contrast with the work Bezos is doing to commercialize space travel. Bezos said he was investing up front to build infrastructure, which should enable future generations to enter the field more easily — a bit like what the Postal Service did for Amazon.

"It takes hundreds of millions of dollars to even just get started with something interesting," Bezos said. "What I want to do is reduce that price of admission with reusable space vehicles so that the next generation you can actually have two kids in a dorm room build a great space company."

Coupang did not have access to an equivalent system early on, so it built its own with the almost impossible design constraint of next day delivery. From its S-1:

“Our first watershed decision was to build an end-to-end logistics network powered by the latest technologies. The lack of existing fulfillment capacity and limitations of third-party logistics in the market forced us to build our own from scratch. It was an immensely difficult effort that required significant investments of time and capital, but we never wavered in our conviction that it was the right decision for our customers. We can now do what we believe no one else in our market can — guarantee one-day delivery on all Rocket orders — without reducing our selection or charging for delivery.”

South Korea happens to be the 13th most densely populated country in the world, with 1,338 people per square mile — that’s 14x the USA. It’s less dense than a city-nation like Macau or Singapore, but its population is significantly larger. Bangladesh, with 170M people and 3,063 people per square mile, is both larger and more dense, but its population is much poorer — with 1/17th the GDP per capita of South Korea. As such, if you wanted to pick a country where the cross product of population, density, and wealth were all in your favor, you’d pick South Korea. By this metric, South Korea is the perfect market for e-commerce.

70% of South Korea’s 52 million inhabitants lives within just 7 miles of a Coupang fulfillment center. It has over 25 million sq ft of logistics space, or ~2 sq ft for every South Korean. In comparison, Amazon has 125 million sq ft. in the US. If we just count its 100 million prime subscribers, that’s 1.3 sq ft per person. Relative to Amazon, Coupang has over-provisioned logistics resources per person by more than 60%. That additional marginal investment, along with South Korea’s inherent density, allows Coupang to deliver nearly all products next day, and many same day — a feat even Amazon can’t replicate today in the US.

Coupang’s investment in end-to-end logistics allows it to offer unique services that we don’t see in the US. For example, it uses reusable packaging on 75% of orders — eliminating wasteful cardboard boxes. Coupang also offers effortless returns through its own infrastructure:

“If online shopping for more and more categories of items were so convenient, why couldn’t online returns be as well? That’s why we introduced effortless returns. Customers make returns for Rocket deliveries simply by tapping a few buttons on our app and leaving the item outside for pickup — no repackaging, no labels, no post office trips required.”

In the US, startups like Boox are attempting to build a similar circular commerce infrastructure by working directly with retailers. Similarly companies like Happy Returns have attempted to solve the reverse logistics problem again by working with retailers. In an ideal world, Amazon would itself provide services like these to its customers, but has likely decided that such they are too difficult to implement without owning the last mile.

A super app for commerce

Coupang leverages multiple business models on top of its infrastructure. Most of its revenue comes from core e-commerce (à la Amazon), but the company also offers Coupang Eats (à la Doordash/Uber Eats), Coupang Pay (à la Shopify* Pay or perhaps Paypal/Venmo), and Coupang myShop (à la Shopify’s core product).

In the US, Amazon has failed to execute a last mile food delivery strategy. It importantly owns 16% of Deliveroo in UK, but Amazon missed out on the opportunity in its home market. Meanwhile, Doordash has announced grocery delivery and has signaled intentions to enter last mile parcel delivery in major cities. Could it go backwards up the value chain and eat Amazon’s lunch (pun intended)?

Similarly, Amazon chose to centralize 3rd party sellers in a marketplace. Coupang does that too, but it took Shopify to give Amazon sellers their own space on the web. Shopify today is approaching 40% of Amazon’s 3rd party GMV. Coupang gives sellers the best of both worlds: either sell in Coupang Marketplace, or build your own space online through myShop with 3.5% commissions.

Perhaps most disruptive is the Coupang Pay product. Amazon has its own payment service Amazon Pay, but Shopify’s alternative Shop Pay has quickly become more popular in the US — allowing consumers to check out instantly on Shopify-enabled sites. Because Coupang has bank account or credit card info for nearly 15 million customers (29% of South Korea’s population), its Pay product could become the de facto consumer payment product in South Korea for e-commerce. Of course, it has strong competition on the P2P side from both Naver and Kakao, but as we’ve seen with Paypal/Venmo and Square Cash in the US, the consumer payments market can support multiple players coming from different angles.

Operating cash flow focus

Part of the decision to IPO now could be that growth in 2020 pushed the company into positive operating cash flow territory. As I’ve noted before, Amazon was operating cash flow positive for almost its entire history:

Similar to Amazon, Coupang has begun investing its entire operating cash flow (actually a bit more) back into CapEx. That’s a good thing when you’re building the delivery infrastructure for an entire nation! The flip to positive operating cash flow was almost certainly aided by COVID tailwinds, but the result is that Coupang went from needing $1B/yr from the capital markets to less than 1/10th of that in 2020.

More addictive than Amazon?

Coupang’s net revenue per active customer was $256 in 2020 — doubling in the last 2 years. Amazon’s customers spend more today — $600 on average, and over $1,400 per year for Prime members. But, if you adjust what Coupang users spend for differences in GDP per capita ($32K in South Korea vs. $65K in US), they spent only 13% less than the typical Amazon user. With Coupang’s rate of growth in spend, and its smaller base to build on, there’s little doubt that, adjusted for GDP/capita, a Coupang user will spend more on its platform than an American on Amazon in a few years.

Part of the dynamic here is nearly 100% YoY growth in cohorted spend. That spend drifts down to ~50% by Year 4, but given recent CapEx investments, I would expect that number to increase in future years.

All this data points to a highly addictive platform with a growing set of use cases on and off the core e-commerce platform.

The all-in-one e-comm platform

Overall, Coupang is a unique platform in the global e-commerce market — birthed from a market which is uniquely large, dense, and rich. These factors enabled Coupang to build out a lucrative core e-comm business with best-in-class logistics. And those logistics have enabled it to build other services to extend its reach in commerce. It’s a playbook that Amazon would love to replicate, but instead we’ve seen other companies nibble around the edges of its home market. Amazon’s loss, in not building Coupang domestically, is a gain for those startups, who have the opportunity to build complementary businesses and eventually encroach on Amazon’s established turf.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NFLX, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

Hello, all.

After a few months’ break, I’m back to my weekly writing habit. I will try to stay on track this time. I’m admittedly a bit rusty, so please forgive verbosity, typos, etc while I’m getting back into my rhythm.

This week’s Firehose will just feature the “One Big Thought.” I need to collect enough material for the other sections, so give me a few weeks to do that. I also plan on adding a new #climate section to reflect an increasing interest of mine.

For now, enjoy! Excited to return to your inbox.

One Big Thought

Private markets have been more bullish on tech than public markets for most of my career in VC.

In May 2009 (my first month on the job), the news leaked that Facebook* was raising $200M from an obscure Russian investment group called DST at a whopping $10B valuation. A few months later, DST initiated a tender offer for another $100M of Facebook common stock at a $6.5B valuation.

While few in the industry knew of Facebook’s incredible numbers at the time, the chatter was highly skeptical. But, DST’s Yuri Milner was confident in this Techcrunch interview with Mike Arrington at the time:

MA: Why did you invest in Facebook?

YM: Because it’s a great business.

MA: You’re comfortable with the $10B valuation?

YM: Absolutely. And, you know, I can repeat the reasons why. Basically, I think they have a very unique perspective on social network monetization, that other investors don’t necessarily see. You see how social networks have been monetized in our part of the world, and we’re just doing our math and coming up with numbers that we feel very comfortable with going forward. We don’t really value this business on (fee?) basis 2009 but rather on a longer term, based again on our experience, and we’re very confident that, you know, those numbers can be achieved.

Of course, history proved Yuri’s investment to be prescient. He had seen global social networks monetize and realized that Facebook’s ARPU should be bigger given American’s size and GDP per capita. He also saw Facebook’s engagement numbers in the US and was able to make assumptions around potential ad volume in the network. In short, his experience from other parts of the world was portable and allowed him to see a bit farther into the future than most other investors at the time.

In its S-1 filing, Facebook ended up disclosing $777M of revenue in 2009, with incredible 185% YoY growth. EBITDA was $340M. The blended cost of Yuri’s $200M preferred shares (at $10B) and his $100M common shares (at $6.5B) was $8.8B. Yuri’s entry multiples were therefore 11x 2009 revenues and 26x 2009 EBITDA. Facebook continued to grow rapidly in the following years — to $2.0B revenues and $1.2B EBITDA in 2010, and $3.7B revenues and $2.1B EBITDA in 2011.

At the time, Facebook listed the following companies on its list of competitors:

Amazon.com; Apple; Cisco Systems; eBay; Google; LinkedIn; Microsoft; Netflix; Oracle; salesforce.com; VMware; Yahoo!; and Zynga.

Here’s what (most of) that group’s trailing revenue multiples looked like, both historically through 2011, with the May 2009 deal announcement highlighted below:

Google hadn’t traded north of a 10x trailing multiple since the early 2000’s. At the time of the DST deal, the top comps (Google, VMWare) were both trading at half of Yuri’s multiple. The EBITDA multiple chart below is messier, but shows a range of 6.5-40x, with the median closer to 10-15x (Google at 13x), and a few outliers like Amazon and VWMare at the top of the range. Yuri’s deal wasn’t priced at the top of the market on an EBITDA basis, but its EBITDA multiple was notably 2x Google’s.

In summary, Yuri bought Facebook at 2x+ the revenue and EBITDA multiples of its nearest competitor (Google) and at a broad premium to the market. However, Facebook handily beat the market on both growth and profitability. With respect to the former, its top comps above grew at 30-40% CAGR in 2009, as shown below:

Because Facebook was growing 4-5x faster than its market comps, what began as a 100% premium on the stock vs market multiples turned into a 50-70% discount in just a few years:

Fundamentally, this is the math that made Yuri’s growth investment in Facebook such a legendary move. Based on his past experience investing globally, he saw that Facebook’s potential growth was so much greater than any other US internet company at that moment in time. He could have paid up to double his blended cost and still have been “in the money” by 2011.

After the Facebook IPO validated the DST investment, a flood of private growth stage capital entered the tech markets. Mark Suster of Upfront Ventures published a post that showed $100+ million rounds grew from 13% of financings in 2013 to 47% in 2018. These so-called “mega rounds” became the “private IPO” of the day, allowing companies to push out their actual IPOs by many years. 2015 was the first year when more mega-rounds occurred than traditional IPOs:

While these private growth investors gained increasing conviction to pay higher multiples for market leading technology companies, the public markets demurred. According to Meritech’s Alex Clayton, other than a brief spike in the 2013 time frame, median NTM revenue multiples for SaaS businesses hovered around 7-8x fairly consistently from 2011-19:

My own recollection is that high quality, private SaaS companies regularly traded at 8-10x ARR for a good portion of the last 5 years. Since COVID’s acceleration of the software market in 2020-2021, we now regularly see significantly higher private multiples. Some companies are even raising at 100x ARR:

Even so, public markets today are often valuing tech companies more aggressively than private investors. We can see this development in part by looking at the dispersion between public and private valuations within a given sector (see below for SaaS). An analysis by SaaS Capital shows that the dispersion is growing, driven by a handful of superb performers accelerated by COVID tailwinds:

Adding to the public market’s enthusiasm is the SPAC surge of 2020. According to Bloomberg, SPACs raised nearly $26 billion in January 2021 alone. This capital will compete to bid up prices on the best assets, and maybe some of the mediocre ones as well:

The surge in valuations definitely makes an investor’s job harder. You can’t simply “get lucky” by investing below the market price. Market timing is working against you.

That’s why I go back to Yuri’s insight in 2009 as an inspiration. He had a simple insight, informed by his own experience (“You see how social networks have been monetized in our part of the world, and we’re just doing our math and coming up with numbers that we feel very comfortable with going forward.”) in a market that few appreciated at the time. He was willing to pay a significant premium (100%+ vs. market comps) based on that insight because he was confident that future growth would be multiples of the slower growing comp set.

Even in today’s “crazy” market environment, I still believe this analysis works for the right companies. Plenty of deals that look crazy today won’t in retrospect.

Share

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V. I own some BTC as well.

Hello all,

The last newsletter I sent out was a few months ago. A number of you have reached out to ask when the next one is coming.

The short answer is: “hopefully, soon.” As some of you know, I write this newsletter predominately on Sundays — which are essentially a work day in venture capital. I’ve had childcare on Sundays for years now to enable me to have time to get work done. Unfortunately, towards the end of last year, we lost that childcare. As a result, I haven’t had time to write in the manner and depth that I’m used to.

At the same time, venture capital seems to have accelerated its intensity and pace, far beyond anything I’ve seen in my decade+ career. So I have intentionally focused on performing my day job in the more limited hours I have during the week.

Finding childcare during a pandemic is uniquely challenging. One reason I’ve gleaned from talking to various agencies is that, because international travel is restricted, fewer au pairs are coming into the US, and as a result there’s more demand for traditional nannies as a substitute. Marketplace aficionados might call it a “supply constrained” market. Moreover, there are unique, COVID-related considerations now. Does this person live a fairly contained lifestyle? Will they be a responsible member of your home “bubble”? Those concerns run in both directions, so it’s much harder to match with someone.

I’m sure many of you are going through similar circumstances now. I am optimistic that we’ll find reliable childcare again soon and that I can get back to writing for you all every week. Thanks for your patience during such an unusual time.

Grateful to all of you for subscribing, and hope to be back in your inboxes soon,

Alex

One Big Thought

Subscription has become the default way that businesses purchase software. The SaaS ecosystem was worth $640B in 2019, up from next to nothing just 15 years ago, and is likely worth over $1T today.

Subscription has become an increasingly prevalent consumer business model as well. I don’t have a neat graph of market cap over time for consumer subscriptions, but the market cap of these businesses in aggregate is likely larger than for B2B. Legacy consumer utilities like Verizon ($250B market cap) and Comcast ($236B) have been subscription oriented for years, but internet native companies have built even more value with subscriptions in the last decade. Amazon* alone is worth $1.6T today, and its high margin Subscription Services segment is currently on a $24B revenue run-rate. 126 million Americans subscribed to Amazon Prime as of last month. Apple*, at nearly $2T in market cap, has grown its Services segment to $54B revenue in 2020 — larger than any other segment except iPhone, and at more than twice the gross margin of Apple’s legacy hardware business.

The categories of media, food, gaming, and fitness have been more recently transformed by subscriptions, thanks to companies like Netflix* (73 million US subs), Doordash (5 million Dashpass members), Microsoft* (15 million XBox Game Pass subs), and Peloton (3 million members), respectively. Even old school categories like insurance (Lemonade with 739K members), education (Chegg with over 4 million subscribers to its “Services”), and healthcare (One Medical with over 500K members) are susceptible to a subscription disrupter. In aggregate, these players and their competitors likely add up to or exceed the value of Amazon’s Prime member base.

Subscriptions are attractive because they are predictable, but not every product or service is a natural fit for subscription. First principles would suggest a few things must be true for a subscription to stick and grow rapidly.

Spinning the data flywheel

First, the subscription service must get better the more your customers use it. The data exhaust from transactions should lead to a more appealing service in the future. If this condition is true, a virtuous cycle driving more transactions is really beneficial to the customer, not just the company selling the service. This quote from the S-1 filing of Stitch Fix* in 2017 lays out the argument well:

We believe we are the only company that has successfully combined rich client data with detailed merchandise data to provide a personalized shopping experience for consumers. Clients directly provide us with meaningful data about themselves, such as style, size, fit and price preferences, when they complete their initial style profile and provide additional rich data about themselves and the merchandise they receive through the feedback they provide after receiving a Fix. Our clients are motivated to provide us with this detailed information because they recognize that doing so will result in a more personalized and successful experience. This perpetual feedback loop drives important network effects, as our client-provided data informs not only our personalization capabilities for the specific client, but also helps us better serve other clients. For example, a client may tell us in her feedback from a Fix that a size small blouse was the right size, but fit too tight in the shoulders. Based on the detailed information we know about that client’s body proportions and fit preferences and the item specific merchandise data we gather about the blouse, our stylists, informed by our algorithms’ recommendations, will not recommend that blouse to clients with similar body proportions and fit preferences. We believe our client feedback loop and its corresponding network effects ultimately contribute to better personalization for all of our clients.

Stitch Fix was inspired by Netflix*. In the Netflix model, success is measured by low churn. Netflix has industry leading churn, as shown below.

What’s most impressive is that Netflix has increased prices many times in the last decade, on the order of ~10% each time, with little evidence of increased churn. The most recent price increase in October 2020 raised standard plan to $14 (+8%) and premium to $18 (+13%).

If you can keep churn low while raising prices over and over, then you have solid evidence that your service is better than it used to be. For Netflix* and Stitch Fix*, that’s all about aggregate customer data.

Tapping into recurring behavior

Second, the subscription must tie to a recurring behavior. The causality must flow from the recurring behavior to the subscription, not the other way around. A simple change to subscription pricing, without addressing the underlying behavior, will not be successful.

Consumables like food are great opportunities for recurring subscriptions. Nearly 30% of Doordash customers opt-in to its Dashpass service because doing so unlocks better economics for those customers, who have a recurring use case for takeout food. Peloton’s subscription works because the benefits of exercise only accrue if activity is repeated multiple times per week. Social ties can even increase workout frequency. Pandemic aside, Peloton has meaningfully grown monthly workouts per connected fitness subscription. Calm* has done the same in the mental health space.

To the contrary, many of Netflix’s competitors claim to have a similar recurring behavior of media consumption, but in reality consumers sign up for a single show and shut off their subscription the next month. These services are not really subscriptions; instead, they are employing a transactional model in a subscription pricing scheme. Similarly, some of Peloton’s online fitness competitors do the equivalent of “channel stuffing.” In January, when New Year’s resolutions are top of mind, they deeply discount their monthly plans in an “annual” offering and pull forward the year’s revenue up-front. Most of these services churn out users at high rates once users hit a subscription cliff.

Path to upgrade

Third, a subscription offering should self-select your best customers and give others an aspirational path. Amazon Prime* is the canonical example. As of 2018, Prime members spend $1,400 annually vs. $600 for non-Prime users, shopping 26 times vs. 14 times per year. Amazon is happy to subsidize shipping because it makes more contribution dollars on these customers in aggregate. Faire* has a similar program called Insider, in which the company subsidizes shipping for its highest volume customers.

Upgrade paths can even exist without a de facto subscription. Both Chewy and Stitch Fix* offer an auto-ship feature, which gives meaningful discounts to the most active customers in exchange for predictable revenue. In 2019, nearly 70% of Chewy’s revenue came through auto-ship. At the time of its IPO, Chewy attributed its stellar net dollar retention to the high percentage of auto-ship customers, who spend 6% more than the average non-auto-ship customer in each purchase. That adds up over time!

In short, subscriptions are not a panacea. Price changes do not solve product/market fit on their own. But, if your product gets better with more usage, addresses a naturally recurring behavior, and has a potential “upgrade tier” of users, a subscription model could be very successful. I’m excited to see more companies add subscription offerings to their consumer services in creative ways.

Share

Tweet of the Week

Links I Enjoy

#commerce

Airborn.→

So many IPOs; so little time! Between Wish, Doordash, Roblox, and others it’s been quite a couple weeks for consumer tech.

Airbnb filed its S-1 the week after I published my last Firehose on Doordash. I decided rather than do another S-1 post to share some work from others. This balanced post by Mauricio Prieto compares Airbnb’s journey to that of Bookings Group. As shown above, gross bookings is nearly identical on a time lag. The companies respective bottom lines look much different, however. Bookings Group achieved similar growth with very healthy EBITDA margins, while Airbnb burned several hundred million in 2018 and 2019. That’s despite a marked advantage for Airbnb in organic traffic (~20% of revs vs. 30%+ for Bookings Group).

Airbnb is no doubt a special company. It has invented a category of travel and now has the challenge of driving that model into the mainstream. Time will tell if its inventory advantage, which Prieto observes is receding, and its organic brand halo will ultimately win out against a more horizontal, commoditized competitor at the top of the funnel. Some of my favorite people in tech work there or have done so in the past. I am rooting for them all.

#media

Newsletter-ception. →

I joined a newsletter co-op called “The Type House” a while back. It features some friends and colleagues from other popular publications, many of them on Substack.

We collaborated recently on a “how to” post for newsletter writers. If you’re interested in getting into the game, I’d encourage you to check out tips from yours truly and others!

#tech

Bloxheads.→

Roblox filed its S-1 and it’s a doozy. The company has built a dual flywheel business that’s both social network and content development ecosystem. On one side sits 36 million DAUs. On the other sits 7 million active developers with 18 million unique experiences, of which 12 million were active in the last 12 months. I calculated that a DAU spends an average of 2.6 hours in Roblox, which is so large it feels incorrect, but not crazy given how immersive Roblox is. Its monetization is still lower than I would have expected at $39 ARPDAU, but with 23% FCF margins the company is going to invest heavily in increased monetization. While pre-COVID growth was slowish, the pandemic really bolstered this online community.

By the way, of the $1.2B in bookings Roblox saw in the last 9 months, more than half was deferred revenue! Consumers essentially float half a billion to the company in pre-paid in-game currency and then keep a meaningful chunk of that currency in game. It’s a massive float that makes me wonder if we should think of Roblox as a neobank! Along those lines, the company’s biggest cost bucket is payment processing (~25% of bookings).

#science

The long haul.→

A new study (disclaimer: not yet peer reviewed) shows that those infected with COVID-19 have enough immune cells to fight off the virus 8 months later. Some patients could potentially have immunity measured in years, but the response was not super consistent across patients.

#culture

“Home Alone” for the holidays.→

Paint can to the face? Crowbar to the chest? VSauce broke down the traps from iconic Xmas movie Home Alone and demonstrated that those burglars definitely had their work cut out for them.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NFLX, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

One Big Thought

Doordash dropped its S-1 this week. I read through the filing this weekend and found it to be one of the most transparent, well argued prospectuses for a consumer business in recent memory. This week, I wanted to highlight a few elements of the S-1 that I found compelling and potentially useful for others constructing pitches.

A case study in cohorts

Doordash (soon-to-be DASH) reveals incredible detail in its cohort data. This contrasts sharply with Uber (NYSE:UBER), which revealed almost no cohort data in its IPO filing in 2019. Perhaps the most important chart in the entire S-1 is the cohorted contribution graph:

DASH defines contribution profit (CP) as gross profit less sales & marketing costs, with a few non-cash items added back. CP is negative in Y1 due to spend on customer acquisition. In subsequent cohort years, DASH levels out its re-engagement marketing spend to ~2% of gross order value (GOV). That implies gross profit (GP) rises to 10% of GOV as cohorts age.

DASH therefore invests around 7-10% of GOV to acquire a customer in Y1. That customer then generates 5-8% of GOV of CP in each year thereafter. This implies a 12-18 month payback on customer acquisition, and 3-4x LTV/CAC over the first 3 years.

In addition, DASH gives us some insight into the compounding GOV growth of its cohorts with the following table:

At this rate, the cohorts compound at least 160% annually. A 3-year compounding period would put the cohort at 4x cumulative GMV, which roughly corresponds to the aforementioned unit economic ratios.

The consequence of these stable cohorts and attractive marketing payback ratios is that DASH will be able to compound its growth for years to come, with older cohorts baking in a “same store sales” baseline of organic growth. Over time, this dynamic can result in significant operating leverage, as I tweeted about earlier this week:

A granular view of unit economics

DASH puts together an intuitive breakdown of the economics of a single order. Because it serves 3 parties (consumers, merchants, dashers), the flow of cash between them is more complex than it is for most marketplaces:

The filing also breaks down the granular components of contribution profit and tracks changes from 2018 to 2019:

Here you can see the increase in take rate from 10% to 11% from 2018 to 2019. You can also see that DASH traded off improvements in COGS (-8% to -6%) for additional investments in sales & marketing (-5% to -7%), maintaining a constant CP of -2%. Watching these tradeoffs gives one a sense for how management runs the business. In this case, management is investing additional cash flows from operational improvements in growing the business faster. You’ll see why in the next section.

Crystal clear strategy

DASH boils down its business to 3 interlocking, virtuous flywheels:

The explanation touches upon the 3 key strategic elements of DASH’s business:

• Local Network Effects: Our ability to attract more merchants, including local favorites and national brands, creates more selection in our Marketplace, driving more consumer engagement, and in turn, more sales for merchants on our platform. Our strong national merchant footprint enables us to launch new markets and quickly establish a critical mass of merchants and Dashers, driving strong consumer adoption.

• Economies of Scale: As more consumers join our local logistics platform and their engagement increases, our entire platform benefits from higher order volume, which means more revenue for local businesses and more opportunities for Dashers to work and increase their earnings. This, in turn, attracts Dashers to our local logistics platform, which allows for faster and more efficient fulfillment of orders for consumers.

• Increasing Brand Affinity: Both our local network effects and economies of scale lead to more merchants, consumers, and Dashers that utilize our local logistics platform. As we scale, we continue to invest in improving our offerings for merchants, selection, experience, and value for consumers, and earnings opportunities for Dashers. By improving the benefits of our local logistics platform for each of our three constituencies, our network continues to grow and we benefit from increased brand awareness and positive brand affinity. With increased brand affinity, we expect that we will enjoy lower acquisition costs for all three constituencies in the long term.

With this in mind, you can see why DASH would plough its margin gains into increased spending on growth. More growth drives the flywheel faster, which brings more future margin improvements.

The chart also implies that brand is its own flywheel. “Brand” is one of those fuzzy concepts that Silicon Valley people tend to shy away from. It’s often viewed as subjective, or at best difficult to measure. After a decade of working with consumer-facing businesses, I can say that brand is very much a real thing and can build on its own success. However, brand is not an input to the flywheel; it’s an output. A great brand is what happens when you operate the business well. It’s a promise to the next cohort of customers that the service will be excellent. DASH gets this concept and features brand prominently in its strategy.

Distinctive insights that drive performance

The S-1 is littered with little insights that show the distinctive way in which DASH operates. Here’s an excerpt from the cultures/value section:

Operate at the lowest level of detail: Averages in our industry are meaningless, it’s the distribution that matters. No consumer cares if our average delivery time is 35 minutes if they received their food in 53 minutes. At DoorDash, we go to the lowest level of detail to understand every part of our system, looking for “and” solutions that fight false “either/or” dichotomies. This is one reason why everyone at DoorDash, including me, tries to step out of our day-to-day roles once a month to do a delivery or engage in customer support, menu creation, or merchant support—staying very close to the needs of those who use our platform is key. We attribute our category-leading spend retention and capital efficiency, in part, to this obsession.

Getting obsessed with the tails of the curve, not the average, is a great way to encapsulate what’s necessary to succeed in consumer-facing businesses. Fans are earned or lost at the margin. It’s not the average performance that counts.

DASH also revealed the core insight that differentiated it from its competitors in the early days:

We believe that suburban markets and smaller metropolitan areas have experienced significantly higher growth compared to larger metropolitan markets because these smaller markets have been historically underserved by merchants and platforms that enable on-demand delivery. Accordingly, residents in these markets are more acutely impacted by the lack of alternatives and the inconvenience posed by distance and the need to drive to merchants, and therefore consumers in these markets derive greater benefit from on-demand delivery. Additionally, suburban markets are attractive as consumers in these markets are more likely to be families who order more items per order. Lighter traffic and easier parking also mean that Dashers can serve these markets more efficiently. As a result of our early focus on and experience with suburban markets and smaller metropolitan areas, we are particularly well positioned for continued growth in these markets.

It’s hard to recall how unintuitive this insight was at the time Doordash started. Everyone assumed that the superior customer density of cities, selection of restaurants, and availability of couriers would lead to superior unit economics. In addition, Uber was relatively underpenetrated in the suburbs, so when it launched Uber Eats, it had fewer drivers to deliver and compete with DASH in those areas. When you start with mobility, you need to focus on cities. That requirement allowed DASH to win in the suburbs with superior margins.

In summary, I found the combination of quantitative and strategic insights in the S-1 to be illuminating. I hope that more companies in this IPO season reveal this level of transparency and insight into their respective businesses, and that some of this analysis inspires earlier stage companies to do the same.

Share

Tweet of the Week

Links I Enjoy

#commerce

Not all SPACs are created equal. →

The rapid influx of capital into SPACs has resulted in dramatic changes to a once fringe component of the capital markets. Since 2015, SPACs have seen far fewer liquidations (10% vs. 20% previously), the average size has doubled from $200m to $400m, and mainstream investors from other asset classes have entered the fray.

Among these new SPAC investors, performance is not evenly distributed. A new study by McKinsey shows that SPACs with operators at the helm outperform those without operators by over 40% over the 12 month period following the merger. In fact, operator-led SPACs outperform the market index of the underlying company by 10%. For the first time, we have evidence that, with the right team and governance setup, the SPAC structure can produce superior results.

#media

Podcast and chill. →

I almost missed this short article a few weeks ago about Netflix* entering the audio-only space. Historically, Netflix has gone through two product phases. The first was the original DVD-by-mail business. The second is the current, streaming service. Philosophically, Reed Hastings has stated that he would rather expand into new types of content with a single platform to deliver it, rather than enter new content types (think gaming, music, VR, etc).

By testing audio-only stories on Android, he’s potentially showing a desire to start competing more directly with Spotify*, Apple*, and Amazon* in the podcast and audiobook category. If Netflix chose to do so, it could have a significant advantage given its content library and significant, global subscriber base.

#tech

Intel out.→

James Allworth, Head of Innovation at Clouflare and co-host of the Exponent podcast alongside Ben Thompson, wrote a piece discussing the recent breakup of Apple* and Intel. Apple just released its first Macs with its own line silicon, abandoning Intel’s x86 chipset for good.

James shows how the move off Intel is yet another example of Clay Christensen’s disruptive innovation framework. The irony is that Intel’s long-time CEO Andy Grove deeply understood Christensen’s framework, and that Intel itself benefited from it back in the day:

The causal mechanism behind disruption that Grove so quickly understood was that even if a disruptive innovation started off as inferior, by virtue of it dramatically expanding the market, it would improve at a far greater rate than the incumbent. It was what enabled Intel (and Microsoft) to win the computing market in the first place: even though personal computers were cheaper, selling something that sat in every home and on every desk ends up funding a lot more R&D spend than selling a few very expensive servers that only existed in server rooms.

Similarly, Apple’s initial foray into chips didn’t produce anything that special in terms of silicon. But it didn’t need to — people were happy to just have a computer that they could keep in their pocket. Apple has gone on to sell a lot of iPhones, and all those sales have funded a lot of R&D. The silicon inside them has kept improving, and improving, and improving. And their fab partner, TSMC, has gone along with them for the ride.

The roundtrip story of how Intel both benefited and was disrupted according to Christensen’s theory is another demonstration of impermanence of market leadership in technology.

#science

Gut call.→

Scientists have discovered a strong linkage between the quantity of amyloid plaques in the brains of patients with neurodegenerative diseases like Alzheimer’s and certain bacterial products of microbiota in the gut. The association opens up new lines of therapy for these progressive illnesses, potentially by altering the gut microbiome of patients.

#culture

Mind the partisan gap.→

The aftermath of the U.S. elections has been as contentious as the election itself. While some polling data gave the impression that the country’s partisan divide had begun to close over the last 4 years, new data from the election suggests otherwise. This article from The Economist shows a significant correlation between Democratic vote margin and population density. This is essentially the urban/rural divide in America, which has increasingly become self-reinforcing:

One possible explanation for this trend is the tendency for Democrats and Republicans to live among their own kind. Americans are increasingly sorting themselves into politically like-minded communities. For liberals, this means diverse, densely-populated cities; for conservatives it is places that are mostly white, working-class and sparse.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

Header image credit: Techcrunch.

One Big Thought

We had a lot of portfolio news this week, with no fewer than 7 financing announcements across the Lightspeed portfolio. They form an interesting cross section of our investment activity. Now, I normally try to avoid making this newsletter a commercial for my firm; but, given this unusual amount of deal volume, I wanted to pause and reflect on how these investments reflect the trendlines in technology at this moment.

Faire* announced a $170M Series E round at a $2.5B valuation led by our friends at Sequoia Capital with significant support from us and other insiders. While COVID-19 hit small retailers particularly hard back in Q2, the entire industry rebounded in Q3 when cities began to open up and adopt to the pandemic. Faire accelerated growth to over 3x YoY with the vast majority of customers coming through organic or referral channels. The palpable market pull is evidence that the wholesale supply chain has seen a dramatic leap forward in digitization — similar to what we’ve seen broadly in consumer-facing online commerce. I expect we’ll see other wholesale industries transformed as well in the coming years.

Theta Lake* announced a $13M Series A round led by us. Every time a technology shift occurs, a new attack surface emerges in the enterprise. Today that shift is real-time audio, video, and chat. Even before the pandemic, organizations were shifting many of their critical operations to remote work. Post COVID-19, nearly every organization is rolling out some sort of remote strategy. With that, they need solutions for security, risk, and compliance. Theta Lake designed software that uses AI and prebuilt workflows to constantly monitor these streams and flag suspicious events. It’s yet another example of how AI can be used to solve specific applications in the enterprise.

Salto* announced a $27M Series A round led by us and our friends at Bessemer and Salesforce Ventures. The company applies modern best practices in software development to the field of business operations (aka “BizOps”). It is currently focused on how Bizops workers interact with various SaaS tools in the enterprise. CEO Rami Tamir explains it like this:

“Defining a company’s business logic as code can make a fundamental change in the way business applications are delivered. We like to think about it as ‘company-as-code,’ much in the same way as ‘infrastructure-as-code’ transformed the way we manage data centers.”

Salto feeds two existing trends we are tracking. The first is the proliferation of SaaS products in the enterprise. A typical Salto customer may have several hundred of them deployed at various degrees of scale, all with decentralized decision making and constantly changing configurations. The second is the movement towards digitization of workflows in various horizontal job categories. We’ve already seen this in primary work categories like sales, marketing, and HR and are now seeing it more in supporting categories like BizOps, IT Ops, etc.

Curve Health* announced a $6M Series A round led by us. When I was in business school at HBS over a decade ago, I recall Professor Clay Christensen discussing his new book The Innovator’s Prescription. It applies his famous theory of “disruptive innovation” to the healthcare industry. His basic assertion in the book is that the only way to lower the cost of healthcare delivery is to use technology to move it out of hospitals and into cheaper care environments — eventually in the patient’s home.

Curve aligns with that vision. It has developed a platform which merges telemedicine, smart billing, revenue cycle management, health information exchange, and predictive analytics in order to provide care for seniors. COVID-19 is a critical inflection point for telemedicine, and no population needs it more during the pandemic than seniors. Curve has a chance to become the connective fabric to the healthcare system for this growing population of Americans.

Hubilo* announced a $4.5M seed round led by us. The company had built an event management platform for physical events prior to COVID-19, but quickly pivoted to supporting virtual events when the pandemic hit. The pivot was inspired by what our team saw in China’s COVID-19 recovery — that the majority of offline conferences went online, and stayed that way when restrictions lifted. We think that the event space is at a critical inflection point. Those corporate gatherings that do occur when life returns to “normal” will have a significant digital component as well.

Betty Labs*, makers of the Locker Room app, announced a $9.3M seed round led by our friends at Google Ventures with participation from us. Lightspeed first invested over a year ago, prior to Locker Room’s launch. We continue believe that audio is the most unexploited medium for a social platform and that passive consumption is a major use case. Sports talk radio is a strong legacy analogy for what Locker Room is building, but with so much more interactivity and engagement.

Eightfold* announced a $125M Series D at a $1B valuation led by our friends at General Catalyst with strong support from us and other insiders. The company has broken out as the leading AI-based system of record for talent. My partner Peter Nieh, who sponsored our seed investment in the company 4 years ago, writes:

The three most important assets a company has are the product/service it provides, its customers, and its people. Software companies that have built systems of record for and helped manage each of these core assets have become category-defining and highly-valued. For product/service, there is ERP/MRP (SAP, ORCL). For customers, there is CRM (e.g., Salesforce). For people, there is HRM (Workday, SAP/SuccessFactors). However, HRM is just about the administration of employees. There is no system of record for talent (people’s skills, capabilities, experiences relative to company needs). This is fundamentally Eightfold’s opportunity.

Importantly, Eightfold’s AI system pays careful attention to talent diversity. It removes implicit recruiter bias by using NLP to hide evidence of age, gender, ethnicity, or equivalent categories in candidate profiles.

Reading these announcements, I am reminded that resilient founders find opportunity even during the most unconventional times. COVID-19 has clearly accelerated transformations in commerce (Faire), telemedicine (Curve), digital events (Hubilo), and media (Locker Room). In addition, the ongoing application of AI to various parts of the enterprise stack (Theta Lake, Salto, Eightfold) continues to change how we work. Whether in consumer, enterprise, or healthcare, the cutting edge of technology has never been so exciting or ripe with potential.

Tweet of the Week

Links I Enjoy

#commerce

Markets as Veblen goods. →

Packy McCormick of Not Boring wrote a must-read piece that ties together the surge in public equities with the rise of a new class of “digital asset” platforms. His key insight is that financial markets are behaving more like a Veblen good — where demand increases with price. During COVID-19, luxury goods spending plummeted and savings rates expanded. Without spending money on fancy bags or cars, some consumers have extra financial reserves and are looking for ways to express their preferences, and maybe flex a bit in front of their friends.

Packy extends the idea beyond conventional stocks to other types of assets: art, classic cars, real estate, and cryptocurrencies. He claims “the ultimate digital asset might be investments in traditional assets with new digital interfaces.”

Reap what you SOW. →

In “Firehose #178: North Star,” my One Big Thought was the north star metric, and how many companies choose a concept based on share of wallet (SOW). Julia Morrongiello of Point Nine Capital reached out to me to share her own piece on how to calculate SOW. It adds a lot of useful detail that I left out. If you enjoyed #178, I’d encourage you to read her piece as a follow up.

#media

Crave the wave. →

The world has plenty of reasons to be stressed out right now. For a few moments, Nathan Apodaca collectively chilled out the world when he uploaded a TikTok vibing out to Fleetwood Mac on his skateboard and guzzling Ocean Spray cranberry juice. The moment was such a pure expression of Ocean Spray’s brand ethos that the company quickly hopped on the opportunity to get its CEO on a skateboard, responding to the meme. Now Apodaca has a new car, courtesy of Ocean Spray, and doesn’t need to skateboard to work anymore!

#tech

Developer dropout.→

My Lightspeed partner Nnamdi Iregbulem wrote a phenomenal research piece that asks the question, “Where are all the developers we need?” He shows with data that software development skills depreciate faster than other technical skills and, therefore, the industry has something of a leaky bucket problem. Fascinating!

#science

End of a paradox.→

At the heart of physics, lies a deep conflict between Einstein’s theory of general relativity, which governs the very large, and the theories of quantum physics, which govern the very small. In normal circumstances, these theories don’t have the opportunity to come into conflict, but in black holes they do. The attempt to resolve this conflict, in the context of a black hole, is one of the most intractable problems in all of physics.

Recently theoreticians made progress on resolving a paradox regarding how black holes handle the information associated with the objects they consume. This article describing the development of these theories is difficult to parse, but is worth reading for nuggets like this one:

But almost everyone appears to agree on one thing. In some way or other, space-time itself seems to fall apart at a black hole, implying that space-time is not the root level of reality, but an emergent structure from something deeper. Although Einstein conceived of gravity as the geometry of space-time, his theory also entails the dissolution of space-time, which is ultimately why information can escape its gravitational prison.

The idea that “space-time is not the root level of nature, but instead emerges from some underlying mechanism that is not spatial or temporal” makes the imagination run wild with possibilities.

#culture

Ghost in the shroud. →

Where did the modern image of a ghost come from? This article traces its origin to the burial shrouds used to cover dead bodies hundreds of years ago. Wealthy people were buried in coffins, while the poor were buried in a simple shroud.

The author also traces a sharp uptick in the iconography of the white shroud to an incident in London in 1803, where a hoax regarding a local ghost resulted in the accidental shooting of a man dressed in white. He then supposedly started haunting the town. Children began to dress in bedsheets to imitate the form of this particular ghost, but eventually the imitation (the bedsheet) began to replace the original (the shroud). The bedsheet image has stuck in popular culture through today.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, TWLO, VEEV, and V.

Header image credit: Credit to “30 Rock”.

One Big Thought

My interest in education stems from a personal connection. My mother was a teacher for many years. My wife is also a professional educator. After college, I saw myself entering the physics profession as a teacher and researcher. During my first year of graduate school at MIT, I had the pleasure of teaching several sections of “Introduction to Solid State Chemistry” — overseen by the legendary Professor Don Sadoway. He’s the kind of guy who wears a tuxedo and gives a champagne toast to celebrate his last class of the year:

As an investor, education was one of the first sectors I “deep dived” into during my time at Highland Capital Partners in Boston. I worked alongside one of the firm’s founders, Paul Maeder, to underwrite our investment in 2U in 2010.

Our core thesis at the time was that for-profit, online higher education providers like Apollo, Strayer, and Cappella were misaligned with their students. Having just gone through the mortgage crisis in the U.S., we saw similarities in how these schools signed students up, extracted meaningful tuition dollars (often taxpayer subsidized), and then failed to graduate the majority of them. University of Phoenix’s online program was a particularly egregious example, with a graduation rate of 5%! Students who left these programs went back into the workforce worse off, saddled with heavy debt and no greater job prospects than before.

The founders of 2U took a different approach. They envisioned 2U as a service provider to non-profit universities. In a model later dubbed Online Program Management (OPM), 2U offered universities marketing, technology, program management services, and up-front capital in exchange for a cut of tuition. Universities, however, would still do all the teaching and admissions. As such, 2U would only make money when it recruited students who were a strong fit for its university partners and when it retained them through graduation, which requires a high service level. The OPM model attracted some of the best non-profit universities in the country, including USC, Georgetown, and UC Berkeley. At the time of its IPO in 2014, 2U (then NASDAQ:TWOU) had over 14,000 enrolled students across its various programs. It was the first of the recent generation of edtech companies to go public.

While my interest in education continued, it took me almost a decade to make another investment in edtech. I struggled to find another great business in higher education, and, while K-12 saw a ton of innovative products going viral in schools, the path to monetization proved challenging. The market was waiting for a catalyst to release the potential energy of entrepreneurship in this important sector of the economy.

COVID-19 changed everything with respect to education in the U.S. It was the accelerant the industry has been waiting for. I’ve written extensively on its impact on higher ed since the crisis began. Here are 3 different pieces from this newsletter in May 2020:

• Reimagining online learning: I discussed the need for digital transformation in universities and how edtech is essentially risk mitigation for these schools.

• School’s out: I discussed the economic impact on higher education of COVID-19 and the coming budget crisis.

• Fix college first: I shared an article by Flockjay’s* CEO Shaan Hathiramani, in which he calls on university presidents to use COVID-19 as an opportunity to invoke job-ready criteria, reduce inactive tenured faculty, grant regulatory reprieves to endowments to free up restricted dollars, and enact several other key reforms.

My Lightspeed partner Mercedes Bent, who has a deep background in edtech, wrote a two-part series on the impact of COVID-19 on early childhood education, as well as on corporate education and career mobility. Both are worth reading in depth.

Most importantly, I’m thrilled to be investing in edtech again after nearly a decade of sitting on the sidelines. In the last 18 months at Lightspeed, we’ve led three rounds in U.S.-based in companies touching education: Flockjay* (seed), Forage* (Series A), and Outschool* (Series B). These companies join a global edtech portfolio which includes Byju’s* (India), Handshake* (US), Whitehat* (UK), Clever* (US), and several more. The announcement blog posts for these companies explain why they are special and how they address opportunities in career mobility, early career hiring, and interest-based K-12 education, respectively:

• Come fly with me: Announcing Lightspeed’s investment in Flockjay, an online academy to train and support top sales teams.

• The future of job search: Meet Forage, the first major platform combining career discovery, learning, and recruitment all in one, straight from top employers.

• Every kid deserves a great teacher: Announcing Lightspeed’s Series B investment in Outschool — the leading marketplace of live, online classes for K-12 students.

It feels good to be investing in education again at such a pivotal time for the industry. We hope to continue to support entrepreneurs finding new channels to market with exciting products like these. Feel free to reach out if you want to connect.

Tweet of the Week

Links I Enjoy

#commerce

The COVID-19 bottleneck. →

With several vaccine candidates for COVID-19 on the horizon, the next issue to address is how to distribute those vaccines to billions of people around the world. With global travel down significantly, fewer commercial planes are flying. These planes traditionally ship around 50% of pharmaceuticals in their cargo holds. Existing freighter aircraft are attempting to make up for the volume, but there are simply not enough of them. The cold chain makes the distribution task even more complicated. Everything has to be specialized in a cold supply chain, and each vaccine sitting in the cold chain must be used quickly after shipping. Several of the vaccine candidates require storage at even lower temperatures than normal.

Even if a vaccine is available in the next few months, if you’re young and healthy, you’re likely going to be at the end of the line, perhaps getting it in late 2021.

#media

A brief history of audiotech. →

As usual, Matthew Ball dropped a comprehensive, thoroughly researched piece on the history of media — this time focused on audio. He draws inspiration by comparing the effects of technology development on audio vs. video, gaming, and other media sectors.

Audio stands out for its early innovation, but general sluggish pace thereafter. Ball speculates on why:

This dynamic stems from audio’s technological simplicity compared to other media types. We can see this in how much earlier recorded audio emerged than recorded video or video game arcades, and live audio (radio) versus live TV or online gaming. Or how much easier it was to make and record music than shoot and press a film, or design a video game.

One could even argue audio is simpler than text. Although the printing press emerged centuries before radio, printed text was difficult and costly to distribute, and could not be delivered live. Live audio, meanwhile, is even easier than recorded audio and radio broadcast technology meant it could reach every single American household simultaneously and at no marginal cost.

While audio’s simplicity provided it with a head start on other categories, it has also held back its growth. As a general rule, media categories that are strongly affected by technological changes are advantaged over those that are not. We see this through the ways technology changes have increased the diversity of content, delivery and monetization.

Ball sees opportunity for new format innovation in both audio-only storytelling and virtual concerts. Tour revenue is still a massive component of the music market, and with COVID there’s a strong accelerant for bands to try new formats. We’ll certainly see more performances like Travis Scott in Fortnite* in the future.

#tech

Follow the LIDAR. →

Apple’s latest iPhone 12 Pro is the first consumer device to feature a LIDAR scanner. The baseline pitch is to improve lowlight photography, but having a sensor that natively understands depth is going to lead to a ton of new AR use cases. Snap* is a launch partner for this initiative and will allow developers to build with LIDAR in its latest Lens Studio release.

#science

Glitches in the matrix. →

Elon Musk popularized the idea that we’re likely living in a simulation, but the idea itself is almost 20 years old now. I was personally familiar with some of the base arguments, but hadn’t previously considered the experiments one could run to detect a potential simulation. The collapsing of the quantum mechanical wavefunction upon observation, for example, never made sense to me:

For Owhadi, the most promising way to look for potential paradoxes created by such computing shortcuts is through quantum physics experiments. Quantum systems can exist in a superposition of states, and this superposition is described by a mathematical abstraction called the wave function. In standard quantum mechanics, the act of observation causes this wave function to randomly collapse to one of many possible states. Physicists are divided over whether the process of collapse is something real or just reflects a change in our knowledge about the system. “If it is just a pure simulation, there is no collapse,” Owhadi says. “Everything is decided when you look at it. The rest is just simulation, like when you’re playing these video games.”

#culture

Rock the vote. →

Pearl Jam has always made a hobby of advocacy. Back in the 90’s, they fought against the Ticketmaster monopoly (and lost). In 2000, they promoted Green Party candidate Ralph Nader (that didn’t go so well either). Let’s hope they’re lucky this time around with their “get out the vote” campaign.

I loved this quote from the article. Come on, NYT! You can’t print the F word? Not very rock ‘n roll of you.

Asked how their feelings about this moment could be best summed up in a Pearl Jam song, Mr. Vedder immediately offered the opening lyrics to “Porch” from the album “Ten,” an anthem that cannot be quoted here.

Enjoyed this newsletter?

Getting Drinking from the Firehose in your inbox via Substack is easy. Click below to subscribe:

Subscribe now

Have some thoughts? Leave me a comment:

Leave a comment

Or share this post on social media to get the word out:

Share

Disclaimer: * indicates a Lightspeed portfolio company, or other company in which I have economic interest. I also have economic interest in AAPL, ADBE, AMT, AMZN, BABA, BRK, BLK, CCI, CRM, GOOG/GOOGL, FB, HD, LMT, MA, MCD, MSFT, NSRGY, NEE, PYPL, SHOP, SNAP, SPOT, SQ, TMO, VEEV, and V.

Header image credit: https://www.bridgemi.com/talent-education/which-comes-first-tourism-dollars-or-students