Right now, a certain type of investor discourse is taking place in wealth management offices, Reddit threads, and brokerage app comment sections. It goes something like this: Palantir is truly fantastic, its numbers are truly amazing, and the stock is truly terrifying to try to analyze at current prices. The combination of these three factors—excellent business, extreme valuation, and genuine uncertainty—makes Palantir Technologies one of the more difficult decisions in public markets at the moment. It is also one of the more instructive case studies of what happens when growth investing collides with historical gravity.
Let’s start with the company’s true identity and accomplishments. Since going public in September 2020, Palantir’s stock has increased by more than 1,200 percent. In 2020, revenue was $1.1 billion; today, it is close to $5 billion. In addition to an adjusted operating margin of roughly 51% and non-GAAP net income growth of 110%, the third quarter of 2025 saw 63% year-over-year revenue growth, marking the ninth consecutive quarter of accelerating growth rather than slowing.
| IPO & Total Return Since | Went public September 2020; returned 1,200%+ since IPO; ~460% over five years (as of March 2026) |
| 2025 Stock Performance | Up ~150% in 2025; peak price ~$207.52; stock spent periods above $184 (52-week range: $63.40–$207.52) |
| Current Valuation (late 2025) | ~115x price-to-sales; ~431x non-GAAP P/E; most expensive stock in S&P 500 by price-to-sales |
| Revenue Growth (Q3 2025) | 63% YoY revenue growth; 9th consecutive acceleration; adjusted operating margin ~51% |
| US Commercial Revenue Growth | 121% YoY in Q3 2025; Total Contract Value surged 342% YoY to ~$1.3 billion |
| Balance Sheet | ~$6.4–$7.2 billion cash; zero debt; ~$2 billion trailing FCF; 51% adjusted FCF margin |
| Historical Risk Precedent | 7 software stocks that traded above 100x P/S in 20 years — all fell at least 65%; avg drawdown 79% |
| Wall Street Analyst Targets | Average target ~$87–$188 depending on source; range spans $50 (bear) to $255 (bull) per share |
| Governance Risk | Multi-class share structure; Class F founder shares give Alex Karp & co-founders effective corporate control |
| Reference / Investor Relations | investors.palantir.com — Palantir Investor Relations |
US commercial revenue increased by 121 percent year over year, which is the business line that investors are most interested in as a sign of non-government durability. In US commercial contracts, the total contract value increased by 342%. The company has between $6.4 and $7.2 billion in cash, no debt, and trailing free cash flow of about $2 billion. It has a Rule-of-40 score of about 114, which is a standard metric that combines profitability margin and revenue growth. Anything over 40 is regarded as strong. For comparison, the majority of publicly traded software firms that score higher than 40 for two quarters in a row are described as exceptional.
There is a legitimate business case. Genuine enterprise traction has been gained by Palantir’s AIP platform, which enables businesses and government organizations to incorporate large language models into operational workflows rather than merely using them for chatbot-style interfaces. The TCV acceleration and a deal volume that has nearly quadrupled since the end of 2022 for contracts over a million dollars are direct results of the company’s Bootcamp model, in which a team enters a client environment, builds functional AI implementations in days, and accelerates deal velocity. Palantir is ranked first in the AI/ML platform category by independent researchers at Forrester and IDC. The growth is not artificial; the product is not vapor. The company’s eccentric and purposefully provocative CEO, Alex Karp, has created something that organizations, defense agencies, and businesses in several nations are funding and growing.
This is the part that unnerves even the bulls. At the moment, Palantir is trading at about 115 times sales. 115 times revenue, not 115 times earnings. Using the same metric, AppLovin is the second most expensive stock in the S&P 500, at about 44 times. This implies that even if Palantir’s share price dropped by 60%, it would still be the company with the highest valuation. Only seven of the more than 70 software stocks that a researcher examined over the previous 20 years ever had a price-to-sales ratio higher than 100.
By September 2024, Snowflake’s sales had dropped 73% from 222 times in December 2020. Zoom dropped 90% after reaching 123 times. Following their peak multiples, SentinelOne, Cloudflare, SoundHound AI, and Bill Holdings all ultimately fell between 65 and 89 percent. For all seven, the average drawdown from peak was 79%. The only company that has completely recovered from a peak in sales of more than 100 times is Cloudflare. Three are still down more than 80%, and the other six are still down at least 44% from their peak.
A 79 percent decline from Palantir’s August 2025 peak, which was close to $187, suggests a future price of about $39 per share. That isn’t a forecast. It is a pattern from the past. However, the frequency with which software companies have collapsed from these valuation levels is not coincidental, and market patterns are not random noise. Every time, it reflects the same dynamics: markets price in years of flawless execution, then react disproportionately when something upsets the narrative, such as a slight slowdown in growth, a shift in macro rates, or a shift in AI sentiment. Palantir doesn’t have to fail at 115 times sales. It only needs to be marginally less exceptional than what is currently priced in.
For analysts attempting to write a price target, this creates an odd position that is difficult to ignore. Depending on the source, the average Wall Street target varies greatly; some have it at $87, which reflects serious doubts about the valuation, while others push it toward $200 or higher, which reflects the company’s true strength. The fact that even the bullish targets do not anticipate another 150 percent year is telling. In essence, consensus states that the company is excellent, the stock is priced fairly for perfection, and the risk is significantly skewed to the downside if anything falters. That’s an odd position for a stock to be in; it’s not clearly a buy or a short, but rather a hold for anyone who has already made gains and a possible trap for fresh capital chasing momentum.
A layer of risk that is frequently overlooked is added by the governance structure. Regardless of the actual economic ownership held by Karp and his co-founders, Palantir’s multi-class share system gives founder shares essentially permanent voting control over important corporate decisions. This implies that outside investors cannot significantly affect the company’s strategic direction, regardless of how many shares they acquire. Concentrated governance increases certain risks at a company where the government business still accounts for about 55% of revenue and where the CEO’s public persona is purposefully unconventional. This is not unique in the tech industry, as Google and Meta have similar structures. The company and the stock could be impacted concurrently by a change in Washington policy, a reorganization of procurement, or an event involving a key person, with little institutional recourse.
All of this does not imply that Palantir is a failing business. Over several years, as the share price fluctuates, it may grow into its valuation, compressing the multiple through revenue growth rather than price decline. That has previously occurred with outstanding companies. However, there is a sense that the current setup asks investors to believe in a kind of soft landing that software companies at 115 times sales have never really achieved before, based on the stock’s history and the accumulation of comparable data. Perhaps things are different this time. Usually, it isn’t.