Initial Thoughts on Compass S-1

Yesterday, Compass Inc., a real estate brokerage that has been disrupting some of the largest companies in the industry, filed its S-1 to go public sometime this year. I’ve been waiting for this day for quite some time because Compass gives us another traditional split-based brokerage that has to report its numbers.

At first glance, the S-1 seems compelling. On the surface the numbers are way better than I had expected. But are they if we dig in? Let’s explore together.

The Numbers

While there are tables after tables and charts after charts as is common in S-1, let’s focus on two or three I found most interesting and compelling.

What we’re seeing here from the annual financial and Key Metrics numbers is really mind-blowing growth… combined with mind-blowing lack of profitability. That might as well be the story of every traditional brokerage who computes their revenues on GCI.

Revenues went from $884 million in 2018 to over $3.7 billion in 2020. That’s unbelievable growth. And on raw numbers, Company Dollar/Gross Profit line went from $189 million in 2018 to $664 million in 2020. So what’s the problem?

The problem, as my computed numbers show, is that gross margin plummeted from 21.4% in 2018 to 17.8% in 2020. But even that number isn’t telling the whole story.

Sales & Marketing expenses went from $174 million in 2018 to $408 million in 2020. Include those numbers into the “Commissions and other transaction-related expense” line and we’re suddenly at Company Dollar of $15 million in 2018, $68 million in 2019, and $256 million in 2020, on gross margins of 1.7%, 2.8% and 6.9% respectively. Those are terrible numbers, far below the industry average of 15% average company dollar margins.

Now, why would we include Sales & Marketing expense with the Commissions expense? Because of these paragraphs from the S-1:

Sales and marketing expense consists primarily of marketing and advertising expenses, compensation and other personnel-related costs for employees supporting sales, marketing, expansion and related functions, occupancy-related costs for our regional offices, agent acquisition incentives and costs related to administering the Compass Concierge Program, including associated bad debt expenses. Advertising expense primarily includes the cost of marketing activities such as print advertising, online advertising and promotional items, which are expensed as incurred. Compensation and other personnel-related costs include salaries, benefits, bonuses and stock-based compensation expense.

We plan to continue to invest in sales and marketing to attract and retain agents and increase brand awareness with home sellers and buyers. We expect that sales and marketing expense will increase on an absolute dollar basis to the extent that we continue to experience increased adoption of our platform. We expect sales and marketing expense to vary from period-to-period as a percentage of revenue for the foreseeable future and decrease as a percentage of revenue over the long term. We anticipate additional sales and marketing expense during the year in which we complete this offering as a result of the stock-based compensation expense associated with our RSUs as described in the section titled “Critical Accounting Policies and Estimates—Stock-Based Compensation.” [Emphasis Added]

In the industry, we all know that Compass is famous for offering signing bonuses, generous initial splits, and a number of other benefits (paying for office space, paying for admin support, paying for marketing expenses, etc.) that have shaken the competitors to the roots.

I’m fairly confident that not all of the Sales and Marketing expenses can be attributed to incentives to recruit and retain agents, but I am similarly confident that a lot of it can. Because we know this about Compass, and I’ve seen the Compass Independent Contractor Agreements that spell out the incentives, which include incredible splits — 90/10 splits, and more.

Even if we assume that only half of the Sales and Marketing expenses are actually agent incentives, we’re still looking at gross margins of 11.5%, 10.9% and 12.4% from 2018 to 2020. Those aren’t excellent numbers for a traditional split-based brokerage and does in fact raise a real question as to profitability and a path to profitability.

Key Metrics

At the same time, the Key Metrics that Compass posted are quite impressive:

  • Total transactions went from 27,188 in 2018 to 144,784 in 2020, an increase of 433%
  • Sales Volume went from $33.7 billion in 2018 to $151.7 billion in 2020, an increase of 350%
  • Annual Transactions per Principal Agent (team leaders and independent individual agents) went from 10.1 in 2018 to 16.7 in 2020, an increase of 65%
  • Sales Volume per Principal Agent went from $12.5 million in 2018 to $17.5 million in 2020, an increase of 40%

Those are all very, very impressive numbers. In fact, even though we are kind of comparing apples (brokerage) to oranges (franchise), I think these numbers make Compass the new reining king in agent productivity over RE/MAX. They certainly blow the doors off of Realogy (who reports publicly) and its 6.2 transactions per agent, but do fall behind the W-2 agents of Redfin and its 26.9 transactions per lead agent.

(Speaking of Redfin, however, it is interesting to me that in 2020 Compass generated $428K in revenue per agent, while Redfin only generated $307K in revenue per agent. However, even on the very pro-Compass Company Dollar calculations, Compass got $76K in gross profit/company dollar per agent, while Redfin got $118K in gross profit per agent.)

These numbers fit what we in the industry have known about Compass for quite some time. Compass recruits the best of the best agents, often paying them extremely high incentives, which should result in very high productivity numbers, but lower profitability numbers. And that’s exactly what these numbers are showing.

The Path to Profitability

The dilemma for Compass, and the question I expect that Reffkin and team will face most often from potential investors, is how to translate its rapid growth into profit. What is the path to profitability for Compass?

Right off the bat, let me suggest that the idea that Compass will retain its agents past their incentive period and raise the splits is a flat out No. It’s not going to happen. Top-tier top producing agents that Compass has, does and will continue to recruit are not settling for 80/20 splits with no cap, not with the kind of income they are generating.

This is a major reason why I would include some significant chunk of the Sales and Marketing expenses into a Cost of Revenue calculation, to arrive at a more accurate picture of Compass’s gross profit and gross margins. (I think Compass is probably below the 15% gross margin number that the traditional brokerages in North America have had for a while.)

That means Compass, like Realogy, like HomeServices, like everybody else, is going to have to look elsewhere for its profits. Thankfully, Compass spells out that path to profitability in the S-1.

And from the accompanying text:

We have a substantial opportunity in the evolving global residential real estate market. We view our opportunity in terms of a serviceable addressable market, or SAM, which we believe we can address in the short-term, and a total addressable market, or TAM, which we believe we can address over the long term.

We estimate our SAM to be over $180 billion and our TAM to be over $570 billion.

It is the SAM that most people are going to be focusing on, so what is that “serviceable addressable market” for Compass? From the S-1, it looks like:

  • $95 billion in real estate commissions
  • $35 billion in Title and Escrow
  • $30 billion in Real Estate Marketing, which is “comprised of total real estate-related advertising spending according to Borrell Associates”
  • $20 billion in Home Renovations

So much to think about here. In every single SAM segment, Compass is competing against massive existing leaders who have enormous competitive advantages and are orders of magnitude larger.

Despite all of its woes in brokerage, Realogy still did a billion more than Compass in commission income, and almost $35 billion more in sales volume. HomeServices of America with all of their powerful local brands hasn’t gone anywhere. Keller Williams and RE/MAX and eXp are not ignorant of Compass and its tactics. Large local independents such as @Properties are very much aware of Compass. Compass took them by surprise in 2017 and 2018, but by 2021, every single one of those guys — plus every local boutique — knows about Compass and has strategies (effective or otherwise) to combat Compass and its incentive-laden recruiting pitches.

It’s not that Compass won’t continue to grow; I think they will. It’s just that competition has caught up a bunch, and they’re not exactly eager to surrender their piece of the $95 billion in commissions.

Title and escrow might be a $35 billion business, but it isn’t as if Fidelity National and First American are going anywhere. Compass could get a big chunk of the title and escrow income that Big Title shares with brokerages through various ancillary arrangements, but they’re not getting anywhere near the $35 billion in total revenue.

Real Estate Marketing is dominated by Zillow, and the also-rans includes the likes of and CoStar. I’m sure Compass has its strategies, but… really?

Home Renovations… I mean, I guess… but Home Depot, Lowe’s and even Angie’s List are not exactly startups.

The core concept appears to be that by recruiting the best top performers, giving them a fantastic technology platform that has ancillary services baked into it, Compass will reap profits from the “integrated economics.” Huh. Where have we heard that before? Oh, that’s right, from Realogy, HomeServices, RE/MAX, Redfin, Keller Williams, Zillow… literally every single company in real estate.

So it comes down to recruiting as many “great agents” as possible. From the S-1:

We believe that real estate agents are an underserved group of business owners, and by providing them with a seamless, end-to-end platform, we can unlock enormous untapped economic potential.

As we continue to build everything agents need in a single, integrated platform, we believe more great agents will continue to come to Compass. As more great agents join us, our platform helps them provide great experiences to more buyer and seller clients. The ability to create great client experiences drives continued business for agents with repeat and referral clients. This ultimately generates more revenue for the agent, and in turn, for Compass, which enables us to invest more into enhancing the platform. These investments further empower agents to grow their businesses efficiently and effectively. Our platform and business innovations are focused on accelerating this flywheel.

Essential to this strategy and this vision is that single integrated platform, which means technology. And indeed, Compass has a lot to say about technology.

Technology and Compass

Right at the outset, when the S-1 gets into describing Compass’s business, we find:


We envision a world where the experience of selling or buying a home is simple and pleasant for everyone. Our strategy is to replace today’s complex, paper-driven, antiquated workflow with a seamless, all-digital, end-to-end platform that empowers real estate agents to deliver an exceptional experience to every seller and buyer. Our agent-centric platform is at the heart of our mission to help everyone find their place in the world.


Compass provides an end-to-end platform that empowers our residential real estate agents to deliver exceptional service to seller and buyer clients. Our platform includes an integrated suite of cloud-based software for customer relationship management, marketing, client service and other critical functionality, all custom-built for the real estate industry. Fundamentally, we believe that agents are, and will continue to be, central to residential real estate transactions and enabling our core brokerage services. We help agents grow their businesses, serve more clients, save time, and stand out as valued, trusted and professional advisors in real estate transactions.

Further on, we find Compass waxing rhapsodic about its technology:


Built on Amazon Web Services, the Compass platform uses a cloud native microservices architecture that is engineered for high scalability, reliability, performance and security. Our engineering development uses modern agile practices such as continuous integration and continuous delivery (CI/CD), and in 2020 we averaged 160 deployments per day, contributing to a very high pace of software innovation for our agents. In 2020, we launched more than 20 major products and over 100 major feature updates to existing products.

Our systems are engineered for high utilization and efficiency using state of the art technologies such as Kubernetes. Our core software applications are primarily developed with Java, Go, Python, Node, Swift and Kotlin, and our Data and AI applications are developed with Apache Kafka, Apache Spark, PyTorch and Kubeflow.

We are engineered for high reliability and designed to be available 24 hours a day, 365 days a year, so that our agents can conduct business anytime anywhere. We have adopted state of the art practices for fault tolerance, backup and restore, and rollbacks.

We prioritize security and have detective and preventive controls for network traffic, infrastructure auditing, software analysis, phishing prevention, email security gateway, software static code analysis, secure configuration management, and user controls including two-factor authentication and endpoint management.

From 2016 to 2020, we grew our research and development organization from 74 to 661 employees. We have also acquired technology startups in the AI, CRM, and title and escrow space, which has allowed us to further accelerate the development of our platform. Our technology leadership has decades of experience and a successful track record at top technology firms, including Google, Amazon, Facebook, Microsoft, Zillow, Expedia and LinkedIn. In 2019 and 2020, we invested approximately $131 million and $146 million, respectively, in research and development.

That’s all very good, but… as we saw in the numbers, Compass spent $56.7 million, $131.3 million and $146.3 million on Research and Development in 2018, 2019 and 2020 respectively. Expressed in terms of percentage of revenue, we’re talking about 6.4%, 5.5%, and 3.9% of revenue. In other words, the investment into technology is decreasing in terms of percentage of revenue, not increasing… and the investment allocated to Research and Development is… well… not all that impressive for a “technology” company.

To be fair to Compass, every expense is decreasing as a % of revenue… so it may simply be that Compass is so successful that it has operational scale and leverage. It’s making more money with less expense. That is great news. (Do note, however, that Commission expense continues to rise as a % of revenue.)

As a point of comparison, Redfin —  the poster child of the tech-enabled brokerage — spent 11%, 9% and 9.5% of its revenues on Technology and development in 2018, 2019 and 2020 respectively. Over the same period, as a percentage of revenue (which includes the Homes Division, that spits out incredibly high revenue numbers because of how revenue is recognized) Zillow spent 31%, 17.4% and 15.5% of revenues. And Zillow especially in absolute dollar terms spent $411 million, $477 million and $518 million on Technology and development in 2018, 2019 and 2020 respectively.

Is Compass truly competitive in technology? Really? Shouldn’t a technology company — or at least a brokerage that claims to be a technology-enabled disruptor — be spending more on technology? After all, its entire vision and strategy are based on building the seamless end-to-end platform for agents to drive that flywheel which would presumably lead to profitability.

Final Thoughts

I know these are just initial thoughts, and it will be interesting to see how Wall Street receives Compass now that it is formally going public. It has certainly shown meteoric growth in revenues, agent count, and its Key Metrics look really, really good. Compass agents might be among the most productive in the industry, other than W2 agents.

There are real questions to be asked as to what Compass’s actual company dollar/gross profit numbers are, and what its actual gross margins are once you take agent incentives into account. Those are currently classified as Sales and Marketing, and I just don’t think there is any reason to believe that agents will stick around after incentive commission splits are over.

There is a question to be asked whether Compass is really spending on technology like a technology company, one that is looking to build a “seamless end-to-end” platform, should be spending. And there are real questions to be asked about how Compass plans to generate not just revenue growth, but profit and profit growth. The path to profitability is hazy, beset with rocks, and is one that literally every other company in real estate is pursuing, large and small. So how much actual room to maneuver is there for Compass?

I can’t help but note that as of 12/31/2020, Compass had $440 million in cash and cash equivalents on its balance sheet. It lost $270 million in 2020, which was one of the hottest real estate markets (if not the single hottest market ever) in history, with everybody stacking cash left, right and center. On those numbers, Compass has roughly 18 months of runway left.

Is it any wonder that Compass is going public?

Whatever you think of Compass, I’d like to welcome Compass to the ranks of public brokerages that report their numbers. We need more of you, not fewer. I also can’t wait for the first earnings call by Robert Reffkin, so I’m really pulling for Compass to have a successful IPO out of purely selfish reasons of wanting more data, more metrics, and more information about the shape of the residential real estate industry.