Is Palantir's Government Business Struggling?
A New WSJ Article makes us ask the question
Welcome back to DailyPalantir! In this episode — we’ll primarily be diving deep into the question a WSJ article posed yesterday: Is Palantir’s government business in trouble? Let’s get into it!
The WSJ Article
So, this WSJ article definitely made some bold claims about where Palantir’s government business was going. I talked to some people, did some research, and ultimately concluded:
This article was not a hit piece.
It’s not the best article in terms of understanding Palantir’s government business, but it also isn’t going after Palantir in an unfair way. However, the solution the article implies may work at Palantir’s dismay is absolutely not the right way for the government to be procuring software (in my opinion) and I’ll explain why below.
First, what did the article say?
Core points:
Palantir faces slowed growth in U.S. government contracts, impacting stock performance despite diversification efforts.
U.S. agencies are exploring cheaper, more flexible tech alternatives, reducing Palantir's federal market share.
Customer dissatisfaction over data access and analysis charges has led to increased competition from firms like Deloitte and Booz Allen Hamilton.
International expansion is hindered by data-privacy concerns, although commercial business growth and recent profitability show potential for recovery.
The article makes the claim that because of slowdown in government revenue and number of clients (we are growing these parts of the business, just not at high rates), investors need to be concerned about the multiple that is being given on the stock when the gov aspect is 43% of revenues and not growing strongly.
Quotes from the article:
“Government agencies are changing the way in which they work with Palantir,” said Rishi Jaluria, a software analyst at RBC Capital Markets. “They don’t want to put all their eggs in one basket because Palantir is expensive.”
Palantir Chief Executive Officer Alex Karp told The Wall Street Journal that the U.S. government is a “cyclical and difficult” business that doesn’t generate revenue in the same steady manner of a commercial business.
Some U.S. government customers have grown frustrated with Palantir over limits to how they can access their own data. The company’s practice of sometimes charging customers when they want to run a new analysis using existing data on Palantir software has driven the cost up for some agencies, said former Palantir employees.
A couple of responses:
1.) I agree with Karp — the government business is cyclical and CAGRs over years, not quarters. 2024 may not be the easiest to land more deals because it’s an election year and there will be instability within the White House until we resolve the election, but over the years, there is enough growth as the government procures more software. Some years grow 40% while others grow at 10%, but it does balance out.
2.) Even if the gov business slows down — US commercial should cover for that. Look, if Palantir guides for sub 20% growth this year, it won’t even matter that government is down. We need to be at 22-25%. The gov biz was never going to get us there — US commercial will. International commercial is still struggling, so really a bet on Palantir right now is a bet that US commercial will grow fast enough to give us breathing room on the gov/international commercial side until we figure those out. The valuation is stretched for the growth, but if the US business can grow enough for us to make sure the rest of the business properly accelerates, it’s not in some ridiculously overvalued territory that the journal suggests, especially given the need for funds to have AI exposure, S&P inclusion, and TITAN contract being awarded soon.
3.) The article goes on to suggest that Open Source solutions that give the government more control over their data and allow them to contract consulting firms like Deloitte, BAH, etc. would be better than Palantir’s software.
I think this is a bad, bad argument. Open source software in the defense space, historically, has sucked. It’s why Palantir recently got a $115M Army Vantage extension or a $250M Military AI research contract — you need stuff that works. If the government is going to be using tax payer dollars on open source software that ultimately does not work, it’s a waste of money for the tax payer. Now, these open source softwares are cheaper, and the US is facing a lot of debt, but if they don’t work — it doesn’t matter how cheap they are, because they aren’t getting the job done.
It’s not easy to do this work, and while consulting firms may argue for cheaper prices with open sourced tech, that doesn’t mean the tech will work.
Remember why we got a $458M Army Vantage contract…
Palantir's software is designed to handle large and complex datasets, making it well-suited for the Army's needs in consolidating and analyzing disparate data sources for informed decision-making.
Palantir has a history of successful partnerships with government entities, demonstrating their ability to meet the stringent requirements and security standards necessary for military operations.
The Army Vantage platform was promised to improve the Army's data-driven decision-making capabilities, thereby enhancing readiness and operational efficiency across the force. The Army recently stated that they saved $4B over the past 4 years because of Palantir’s software.
Shyam’s Tweet
It seemed like Shyam Sankar, CTO of Palantir, knew this article was coming out, and put out his on thoughts on the matter a day before the article released:
Shyam basically is making an argument for why open sourced software is ONLY valuable to the consulting firms that want to create more billable hours.
Yes, it is technically cheaper than buying high end software that works.
But the value of those dollars only go so far — especially when the incentives of these companies are not in providing value, but in figuring out how much they can bill you for their time.
This is EXACTLY what Karp has critiqued over the years — software is not about how much money you spend, it’s about who got the money and did the work get done correctly. If not, you can pay a lot of money to companies that won’t actually solve the problem, and end up wasting more money in the long term.
So, how should investors feel about this?
I think we need to heavily monitor the government business and make sure Palantir is in the running to win the contracts that they are going after. The 1.5B TITAN contract is still yet to be awarded, which should be awarded in the next 2 months, and that is a must-win for Palantir.
Yes, the government wants to spend less — but they also just issued $886B for defense spending, have Palantir’s head of AI, Anthony Bak, on one of their advisory firms, and ultimately want to make sure their products work vs losing our competitive edge in AI/software to China or Russia.
While defense tech may become more competitive, I see Palantir’s FedStart product as a hedge against more competition, and ultimately just believe their products will win out long term, even if quarter over quarter we don’t see massive growth.
That’s it for today — I’ll see you tomorrow in your inbox!