Palantir CEO Alex Karp attends a meeting of the White House American Technology Council on June 19, 2017 in Washington, D.C. Photo: Chip Somodevilla/Getty Images

Welcome to The Ticker, a series that examines everything you need to know about companies going public.

Seventeen years after it was founded, Palantir will make its public market debut tomorrow, via direct listing. Created in the aftermath of 9/11 with the aim of protecting the United States from future terror attacks, Palantir has become one of the nation’s most controversial companies. At the center of the firestorm is the software company’s work with the Immigration and Customs Enforcement agency (ICE), the governmental unit tasked with tackling illegal immigration, its political stances, and its polarizing co-founder Peter Thiel. Despite Palantir’s prevalence in the public conversation — particularly under the Trump presidency — its product has remained relatively opaque until its recent filing to go public.

In advance of the listing, we unpack what you need to know about the company poised to be valued at $22 billion.

At a basic level, Palantir provides data analytics software. The company has two product lines: Gotham and Foundry. Gotham was the first product to market and is focused on public sector customers like U.S. defense and intelligence agencies. Foundry, created later, serves similar products for private enterprises including Airbus, Credit Suisse, and BP. Palantir is coy about the other private companies it serves with case studies referring only to a “major U.S. broadcast network,” and “major payment processor.” In 2019, 53% of Palantir’s revenue came from commercial clients, with the remaining 47% from government agencies.

Palantir’s product lines compete with a coterie of data companies including Tableau, Alteryx, Google Sheets, Snowflake, and more. In this sense, Palantir is fundamentally a software company, serving its 125 customers with differentiated technical tooling.

But Palantir is far from a self-serve software solution. In order to onboard customers, and grow the size of annual contracts, the company invests a significant amount of effort toward sales and customer support. CEO Alex Karp spent 250 days on the road per year pre-Covid, primarily to close contracts and increase their value. This process is pivotal, as most Palantir customers only spend $100,000 in their first year using the product, making their contribution margin negative. In short, Palantir effectively loses money on customers in the early years of an engagement.

With effort and over time, customers steadily increase their spend, growing the average annual contract value (ACV) to $5.6 million a year. Palantir’s most mature customers generated $565.7 million out of $743 million of revenue in 2019, with contribution margins of 55%. Those are a distance from pure-play SaaS stocks (which can reach 75% and above), but nevertheless very solid.

This illustrates how Palantir’s model works: Start with white-glove service, grow ACV, and improve margins. Whether Palantir can scale this model or will continue to rely on the intervention of Karp remains to be seen. In the six month ending June 30, 2020, Palantir incurred net losses of $164.7 million and has yet to turn a profit.

Investors have gotten used to dual-class share structures. Companies like Facebook, Pinterest, and Google have managed their equity such that founders retain control after going public. This is achieved by giving founders one class of shares that possess outsize voting power while new investors receive shares from a different class with less voting power.

Palantir takes this to a new level. Rather than a dual-class structure, the company divides shares into Class A, B, and C. The various ways this might play out are complicated, but the takeaway is this: Founders Peter Thiel, Alex Karp, and Stephen Cohen are firmly in control of the company’s future.

This will remain true even if the founders see their shares diminish in time: With as little as 0.5% of shares, the trio could retain over 68% of voting power. This represents a truly unique structure. Investors in the business should recognize that for better or worse, they’re backing a company beholden to the vision of Thiel, Karp, and Cohen.

Palantir’s first outside investor was In-Q-Tel, the CIA’s venture investment arm. Since that relationship was formed, U.S. intelligence and armed services have remained crucial customers of Palantir’s data analytics platforms. Indeed, in 2019, Palantir secured an $800 million contract from the Army alone.

That goes some way to explaining the brazen nationalism in Palantir’s S-1 filing. In a theatrical opening letter, CEO Alex Karp laid the company’s geopolitical preferences bare:

“We have chosen sides, and we know that our partners value our commitment. We stand by them when it is convenient, and when it is not.”

Explicitly, Palantir is interested in aiding “the West,” noting that its product will help make “America, the strongest in the world, the strongest it’s ever been.” Palantir also committed not to work with the Chinese government, a decision the company recognizes may restrict growth:

“Working with the Chinese communist party is inconsistent with our culture and mission,” the company wrote in its IPO filing. “Our decision to avoid this large potential market may limit our growth prospects.”

In the boldness of its expression, Palantir’s S-1 is a singular piece of corporate writing. But while the company seeks to draw stark black-and-white lines identifying who they will serve and who they won’t, the reality is rather murkier. As it stands, 60% of Palantir’s revenue comes from outside of the United States, with both private and public clients in the U.K. and France contributing to top-line growth. However, a full 34% of revenue comes from the “rest of the world,” a category Palantir does not break out. As political alliances become increasingly changeable, it will be interesting to see how the company decides who is worth serving and who isn’t.

Palantir’s political activity doesn’t just apply to its global positioning but the company’s domestic work, too. A recent Buzzfeed News investigation revealed how the Los Angeles Police Department was trained to use the Gotham tool to track individuals by identifying characteristics like race, gender, tattoos and scars, and by relationships to family and friends. Since at least 2011, Palantir has worked with the Immigration and Customs Enforcement agency (ICE). To date, the company has received over $160 million from its work with ICE. While that work has yielded positive outcomes in the past — including the debilitation of drug trafficking ring “Los Zetas” — it’s also sparked outrage.

Under President Trump, ICE has operated more aggressively, arresting those that have broken immigration laws. Palantir’s Investigative Case Management system (ICM) — a software tool that stores information on individuals — has been used by ICE in this effort. In some instances, migrant parents that brought their family across their border were deported, resulting in the separation of parents and children.

In response to this work, Palantir employees petitioned management to cancel contracts with the agency but Karp refused to back down. This decision was in contrast to much of Big Tech: Google stopped working with the Pentagon after employees protested, and Microsoft changed their policies to allow employees to refuse to work on military projects. Should Trump win a second term, we may see Palantir face further backlash for its work with the agency.

Whether you find Palantir an attractive business or not may depend much on your political and ethical leanings. There’s no doubting the company has built an impressive array of products and is earning significant revenue with strong contribution margins. But Palantir is a company with an agenda: to raise up the West in a manner specifically defined by the company’s founders. Many may take issue both with the partisanship of that vision, and the narrow locus of control.