Part I

In part I, we established that Casper is loosing money, but has relatively strong margins when compared to their strongest competitors. We left off with the open ended question - will Casper be able to scale retail and online channels and achieve profitability while maintaining margins?

Visit this link for part I of the Casper S1 analysis.


Casper, founded just seven years ago, was originally a D2C only brand. Today, Casper considers brick & mortar a critical sales channel and a major linchpin of growth. In fact, according to Casper's S1, in 2018 Casper sourced 13% of revenue from retail channels and increased that share to 17% in stub year 2019. As of the S1 publication, Casper has 60 direct stores, 18 retail reseller partners including Costco and Target, and line of sight into 140 additional stores in North America.

I'm interested in Casper's retail strategy for several reasons. While I understand that a retail presence is needed to test high-touch, personal products, I'm curious as to why Casper has expanded beyond a handful of stores in major metros. Typically, DC2 brands open just a few flagship stores in major metros. These stores are generally just as much advertising as they are sales channels. In Los Angeles, where I live, popular touristic thoroughfares on Melrose Ave and Abbott Kinney are dotted with D2C brands like Glossier and Outdoor Voices. Do they lack faith in their online channels or is there real opportunity?

Further, throughout the S1, Casper emphasizes how online and retail sales complement growth. As a consumer, it is helpful to have both channels, but I am suspicious of Casper's claims that online and retail sales have a significant and synergistic effect on each other. Is this a classic case mistaking correlation with causation? Lastly, I want to understand the forward-facing risks along with their store financials. Relative to competitors, are sales per square foot strong? Knowing these figures is critical to help us understand if Casper is on the road to curbing losses and achieving profitability.


Let's start with the inherent risk associated with a go-big retail strategy. [Actually, this post was originally supposed be an overview of Casper's most salient risks (marketing efficiency, manufacturing risk, etc), but I found the store strategy warrants it's own post.] If only based on the sheer number of bullet points listed in the S1, the inherent riskiness of retail store expansion seems to be one of Casper’s most important risk factors.

To pursue our retail store strategy, we will be required to expend significant cash and human capital resources prior to generating any sales in these stores. Delays in new store openings or an inability to generate sufficient sales from these stores to justify such expenses could harm our business and profitability.

  • Identify suitable locations, including our ability to gather and assess demographic and other related data to accurately determine customer demand for our products in the locations we select

  • Assess the potential profitability and payback period of potential new retail store locations

  • Hire and train skilled store operating personnel, especially management personnel, and our ability to immerse such personnel in our culture

  • Understand and assess the demographic profile of, and provide a satisfactory mix of merchandise that is responsive to the needs of, our customers living in the areas where new retail stores are established;

  • Establish a supplier and distribution network able to supply new retail stores with inventory in a timely manner

  • Scale our differentiated in-store experience that is unique to our brand, attracts customers, and builds deeper relationships

  • Create a technology infrastructure that serves our retail, e-commerce, and customer service channels connecting customer data and operational data to deliver a seamless user experience

Executing a major national store growth strategy is an extremly ambitious enterprise. Picking the right locations, the best products, training staff, building out systems, and optimizing supply networks must all occur, often simultaneously, in order to execute a strong retail expansion play. These challenges are almost is completely distinct from the business of selling online directly to consumers. In the next section we'll touch on the format, function, and footprint of the Casper store.


The Casper store on the fashionable Melrose Ave is painted with bright, candy-colored hues. The inside is even more colorful. The 3,330 square foot store looks like more like whimsical dorm room and less like a mattress store.

Thanks to the pictures featured in this blog post, we can see a test pod where a customer can evaluate the comfort as a mattress in a semi-private crisp, clean setting.

While we can clearly see why Casper sets itself apart from incumbent competitors, I wonder if Casper is able to harness the retail opportunities. Here are the store unit economics and goals for future store build out that Casper highlights in their S1:

Current Unit Economics:

  • Across our retail channel, our AOV increased from $437 in 2017 to $720 in 2018 and to $820 for the nine months ended September 30, 2019.

So AOV has almost doubled between 2017 and 2019, which seems promising.

  • Stores that have been operating for one year or longer have averaged approximately $1,600 in annual net sales per sellable square foot, which we believe is reflective of our high volumes of consumer traffic, our ability to successfully engage with consumers to drive sales, and an effective pricing strategy.

Future Targets:

  • We expect that our typical new stores will have between 1,750 and 2,250 square feet of selling space.

Some Thoughts

If Casper opens up one store at 2,000 square feet and $1,600 in annual net sales per sellable square foot, they would generate $3.2 million. In the first 9 months of 2019, Casper generated about $53.1m from retail sales. I dug up the 10K of competitor Sleep Number and see that their most recent sales per square foot is $998.00. Casper's higher sales per square foot is most likely driven by their simple product offerings. Casper currently offers only three editions of the mattress - The Wave, The Casper, The Essential - and thus needs less space. I purchased a mattress at a standard mattress store in 2013. There must have been at least 20 different types of mattresses. I see how the Casper store has the potential to generate even more revenue as they bump up AOVs with complementary products, optimize prices, and devise clever store experiences to induce purchases. However, I still wonder if 140 more stores is a sound investment.

Remaining Opens

To my earlier point, it's hard to model out the COGs/OPEX associated with the retail store as Casper doesn't break that out, but they do mention how stores are profitable after 1 year excluding build out costs.

  • Existing stores that have been operating for one year or longer are all four-wall profitable, calculated as gross profit, less operating expenses (excluding one-time build-out costs and non-allocable marketing and overhead expenses), for each store.

  • Consistent with our experience to date, we target for our future retail stores a cash-on-cash payback period ranging from 18 to 24 months.

Given that Casper is starting to look more and more like a traditional retailer, does Casper expect to be valued like a fast-growing technology company or more like an old-school mattress company?


Casper leads their S1 with this thought-provoking data point:

  • D2C sales have grown on average over 2X faster (or 100%) in markets with ecommerce and retail stores than in markets without out a retail store

In other words, Casper believes that their D2C and retail stores are synergistic - every dollar of retail sales drives $2 dollars in online sales. Initially, this could tell us that Casper stores produce a halo effect - e.g my favorite Chloé perfume drives my affinity for higher price point items such as bags and apparel. In this case the store is the perfume. Keep this assertion in mind as we dive into their store location strategy.

In keeping with the original goal of the post, here is how Casper views the risk surrounding opening new stores:

  • Identify suitable locations, including our ability to gather and assess demographic and other related data to accurately determine customer demand for our products in the locations we select

Can Casper find profitable, suitable locations for an additional 140 stores? Let's start by understanding the locations of their current roster of stores. Note, I could not find one master list of stores on their website, so I went to their store finder on the Casper website and typed in the names of states one by one. I may have missed a few.

Glance through the list, perhaps paying attention to regions that may be familiar to you, and scan for patterns.

Source: Casper Website

What Did you Notice

Casper stores appear to be located in affluent areas within major population centers. For example, in California, my home state, there are a total of ten stores concentrated in the most economically productive areas of the state. The Bay Area is home to three stores in the Bay's distinct subregions: SF, Pleasanton (East Bay), San Jose (South Bay). There are also six locations in Southern California including Melrose Ave, the South Bay of LA - close to affluent neighborhoods including Manhattan Beach, Palos Verdes, and the San Fernando Valley - an urbanized valley in LA with nearly with 1.7 million people. World famous La Jolla, San Diego is also home to a store. Interestingly, they also have an outlet concept in the Camarillo Outlets in Ventura county, a coastal county north of Los Angeles. The sole Northern California location is located in a high-end Westfield mall in the thriving suburbs just outside of Sacramento. This pattern repeats itself on a slightly smaller scale in other states like Texas, Georgia, and Florida. For example, I colleagues in Austin who shop and eat at The Domain, which is close to many tech offices. You do not see store saturation - e.g only one store in the South Bay Area of San Jose. Multiple stores in one municipality do occur only in major metros.

When there are smaller populations to support, Casper will open a single store as is the case in Denver, Colorado, Portland, Oregon, and Bellevue, Washington. Although each market only supports a single store, those cities are filled with professionals earning high incomes. In other words, you don’t see a Casper store in Rutland, Vermont. [Sorry to anyone reading this from Rutland, it were the best example I could think of off the top of my head. I still love the Green Mountain State!]

Now that we’ve conducted a qualitative analysis of store locations, let’s revisit 2X synergy statistic, which posits that Casper’s online and direct retail channels somehow interact and influence each other.

  • D2C sales have grown on average over 2X faster (or 100%) in markets with ecommerce and retail stores than in markets without a retail store

Casper views this phenomenon as a major competitive advantage and a key pillar of their expansion strategy as evidenced by the number of mentions in the S1. As an economics major with formal training in statistics, understanding correlation vs. causation is a way of life. To truly test this statistic, one would need to build an econometric linear regression model (another project for another day). Before embarking upon this model, one would ask, what are the a prior expectations - a fancy way of asking - what do we think is going on? My theory is that the selection effect is at play. Casper stores are located in several of the most populated cities in the United States. The stores are located in bigger markets so there are simply more people to sell into.

  • Ranked by population size, Casper has stores in these “proper” cites: NYC (1), LA (2), Chicago (2), Houston (4), San Antonio (7) , San Diego (8), San Jose (10).

  • Casper has stores in the suburbs just outside of two major metros: King of Prussia - Philadelphia (6) - and Southlake - Dallas (9)

In other words, the only top 10 city without a Casper store is Phoenix.

Source: U.S. Census Bureau.

Updated: Jan 19, 2020

I was recruited to work for Intel after business school in Los Angeles. I grew up in Sacramento, only about a 3 hour drive northeast from the heart of Silicon Valley, but the region was a world away from the people and places that were familiar to me. Shortly after I relocated to a sprawling apartment complex in North San Jose, I read through the seminal works on the key players and technological movements that made Silicon Valley what it is today. These books aren’t just for the earnest business school graduate. I passed these books back and forth with my father, a learned man, who loved the rip roaring biography of Noyce - basically a cowboy disguised as an electronics engineer - and the tragic mismanagement at Xerox’s PARC.

The PC is now considered a commodity and the internet is thought of as a basic utility. Today’s innovations stem from an abstracted layer in AI, machine learning, computer vision, and neural nets. But Terman and Moore’s law are the reason why many of us plug away at Big Tech Companies or work breathlessly at a seemingly paradigm shifting start-up. Further, by understanding the history of the Valley we can better predict and decipher the technology of tomorrow.

There have been hundreds of books written on each of these topics, but these are considered to be some of the most important works on these subjects, with several written twenty to thirty years ago, reminding us that some of the best authors are not making the rounds on today’s tech podcasts. Start with this group and read, re-read, and keep as a reference. As I progress in my career, I often find myself thinking back to lessons from these writings.

Listed in rough chronological order:

The HP Way: How Bill Hewlett and I Built Our Company (Collins Business Essentials)

HP Way is essentially a manifesto. One of the first strong culture companies. Andy Grove looked to HP to fortify Intel’s culture. Similar to the Netflix culture deck, H/P were the first to articulate these unique concepts. This is the closet thing I have to a "primary source" on this list.

Core Values:

  • Profit for Shareholders

  • Employees as human beings

  • Well-being of customers

  • Support community at large

The Chip: How Two Americans Invented the Microchip and Launched a Revolution

This book, the story of Fairchild Semiconductors, Noyce, and Kilby, explains how the microchip was developed and why it was so revolutionary. The book is both science and story.

The Man Behind the Microchip: Robert Noyce and the Invention of Silicon Valley

Robert Noyce is one of the greatest Americas of the 20th century. There is a reason why Intel named the main HQ building RNB - Robert Noyce Building. I believe this book isnt’ even in print, but can be downloaded. If you don’t read the book, at least peruse the wiki page for Robert Noyce. Fun fact - Noyce married Ann Bowers, the first VP of HR at Apple.

Dealers of Lightning: Xerox PARC and the Dawn of the Computer Age

The infamous story of Jobs and the computer mouse. The origins of the ethernet. How to go (or not go) from idea to commercialization. While in the Bay Area, I met someone in the Bay Area who worked at PARC. He was impressed I even knew what it was.

Fire in the Valley: The Birth and Death of the Personal Computer

Heard of the Homebrew Computer Club? This book could be renamed, “How Hobbyists Changed the World.”

Accidental Empires: How the Boys of Silicon Valley Make Their Millions, Battle Foreign Competition, and Still Can't Get a Date

Reads like a novel

Where Wizards Stay Up Late: The Origins Of The Internet

From the author’s website, “At a time when computers were mainly used as processing machines, one man, J.C.R. Licklider, a psychoacoustician from MIT, saw their potential as communication devices. With the help of a small group of engineers and researchers, Licklider’s vision eventually yielded the Arpanet, which in turn created the foundation for the Internet we know today.”


The only book you need on Jobs. I picked up this copy at this really Readers Bookstore at Ft. Mason in SF. I recall they had a great selection of similar titles. I still remember that overcast San Francisco day!

Only the Paranoid Survive

Read this to prepare for my Intel interview. Groves dissects a critical inflection point in Intel’s history. Groves blessed the world with his writings on his pioneering management philosophies.

Updated: Jan 16, 2020

Casper has the honor of being one of the first S1’s of 2020. According to Crunchbase, Casper has raised nearly $340M of financing from investors such as NEA - a $13.1m series A in 2014 - and a $55m 2015 series B round with investors including Slow Ventures, Norwest Venture Partners, Pritzker Group.

A Priori Questions

  • Competitors have proliferated - how does Casper handle this and what is the impact on the financials?

  • What does their marketing spend look like as a % of revenue?

  • Repeat customers? Customer replacement cycle?

  • Gross Margins? Are they higher relative to competitors b/c foam seems inexpensive?

  • Omni-channel strategy?

Intro Casper’s Highlighted KPIs & Introductory Positioning

In the first few colorful pages of the S1, Casper positions itself as part of the broader wellness category, mentioning how the US sleep economy tops $79B. Like many others, I did not realize a “sleep economy” existed.

Casper then lists several financial and operational KPIs. These KPIs warrant an additional level of scrutiny, especially in light of their prominence in the intro section of the S1. Here are a few that struck me as important.

  • 45.5% net revenue CAGR - full year ‘16 through ‘18 - revenue more than doubled in this period. Nine months ended September 30, 2019 and 2018 net revenue grew 20.3% year-over-year

After looking at the sequential revenue growth for ‘16-18, CAGRs may be emphasized b/c revenue grew by 48% between ‘16/’17 then slowed to 42% growth between ‘17/’18. In order to achieve sequential YoY growth of 45% in ‘18/’19, they would need to land at about $519m. This would amount to a 75% CAGR between ‘16-’19. You can see how elastic CAGR is as a revenue measure. I have never personally been a fan of using that figure for simple P&L analysis. Note, revenue in 16-18: $161Mm, $251m, $358m, with my forecast of $519m.

More on the revenue forecast figure: $519m seems to be at least within the realm of possible as Casper has banked $312m in revenue in the FY19 - 9 month stub period. So 40% of their revenue would need to materialize in Q4. Given that Q4 is a slower month in their business, this is probably a stretch. Later analysis may reveal how attainable this figure is.

Here’s a table to exemplify:

  • 50.7% gross margins for 3 months ending 9/30/19

It’s mentioned later on that margins are 42.8% for the year ended December 31, 2016. However, it’s hard to know if these are strong margins without looking at competitors. Casper is able to compete with both Temper Sealy and Purple, while all three lag far behind Sleep number with margins of at least 60%.

  • Each dollar of marketing returns $3 of revenue

This figure seems easy to game. What are the compositions of both sides? We will dive into this more in part 2. But as you read the financials, ask yourself if this ratio makes sense? What multiple would they need to be breakeven or profitable - e.g. each dollar of marketing would need to return at least $5 to break even?

  • 20% repeat customer rate for the first 9 months of 2019

This also seems easy to game. Core products only or core plus a low AOV (average order value) product?

  • D2C sales have grown on average over 2X faster in markets with ecommerce and retail stores than in markets w/out a retail store.

Initially, this could tell us that there could be a halo effect of the Casper stores. This also gets me thinking about their store strategy. In 2018, Casper had plans to open up 200 stores. However, it seems like Casper got ahead of their skis. The 2020 S1 states, “Currently, we distribute our products directly to customers in seven countries through our e-commerce platform, 60 Casper retail stores, and 18 retail partners.” It’s mentioned later on that 48 stores were opened in by September 2018, so Casper opened about 12 stores in 2019.

Highlights from Market & Industry Data & Prospectus Summary

The most surprising facts new facts that we learn in this sections are trends around customer behavior and the the synergistic effects between the two sales channels - direct retail (Casper stores) and ecommerce sales.

  • Over 16% of customers who have purchased at least once since Casper's inception have made a repeat purchase, despite the fact that the traditional replacement cycle of many of our products is longer than Casper's existence.

  • Direct-to-consumer sales in cities where we have opened retail stores have grown over 100% faster on average than cities without a Casper retail store. Store strategy analysis in part 2.

Selected Consolidated Financial and Other Data Analysis

We’ve already touched on some of the key financial metrics that underlie the performance. Now we can holistically view the P&L.

9 months view AKA Stub Gross margins land at 49.7% and 44% in years 19/18 respectively - GMs ticked up by a respectable 5.7% points. However, S&M expenses increased by 23% YoY. Despite this, Casper maintained an almost constant ratio between marketing dollars and revenue. S&M as % of revenue was 36.5% and 35.7% in years 19/18 respectively. Similar effects are at play with G&A spend - relative to revenue, Casper is spending about the same % on G&A. If Casper has solid Q4’19 performance, they would be on track to beat their ‘18 results (below) by about 5-6% - assuming their Q4 is as strong as Q1’Q3’19.

Here is how Casper explains their gross margins, S&M, and G&A for this period.

The increase in gross margin was largely driven by higher margin products among our new offerings as well as implementing supply chain initiatives designed to reduce product unit costs, which as expected has had a favorable impact to our cost of goods sold. Sales and marketing expenses increased as we continued to invest in driving traffic to our e-commerce website, marketing our products to consumers and building our brand. General and administrative expenses increased as we invested for growth in hiring to support our growing business, particularly in retail stores and product development.

Full year view - ‘18/17

Gross margins at 44% and 46% in years 18/17 respectively - slightly declining margins. S&M as % of revenue is 35% and 42.6% in years 18/17 respectively. Does the decline in marketing spend relative to revenue indicate success in their marketing programs? S&M as a % of revenue declines by about 7-8% while GMs decline by only 2%. G&A as a % of revenue was nearly flat in each year - 35% and 32% in years 18/17 respectively. Part of the increase in G&A can be explained by, “Increased investment in product development 89.2% to $12.3 million in 2018 from $6.5 million in 2017”

Here is how Casper explains their gross margins, S&M:

The decrease in gross margin was largely driven by increased sales to retail partners, which are at a lower gross margin than direct-to-consumer sales. We also recorded a charge for slow moving inventory of $2.3 million in 2018 which is included in cost of goods sold. Casper explains the improvement in S&M as a % of revenue. This improvement reflects more efficient marketing returns as Casper expands its product offerings in existing and new categories, increased consumer awareness, a higher returning customer mix, growth in our retail partnership and retail stores and more effective marketing models as we continue to improve our understanding of our consumers.


Casper has relatively strong gross margins, but the below the line costs (S&M/G&A) are driving substantial operating losses. In part two we will start to contextualize their strategy including retail store expansion, international growth, margin improvements initiatives to understand if Casper is on track toward profitability. We will revisit some of Casper's favorite KPIs including returns on marketing spend. We will also dive deeper into risk factors including manufacturing and vertical vs. horizontal integration risk.