By David Hartley | Senior Equity Analyst — Technology (AI / Data Center) | 2026-05-05
Rating: Hold ($160 PT)
Sector: Tech & AI > AI/ML Software
Tickers: $PLTR, $SNOW | Related: $MDB, $NOW, $NVDA
If you opened your phone Tuesday morning and saw Palantir trending, you weren't imagining it. The numbers Palantir put out Monday after the close were the kind that get screenshot and texted around — revenue up 85% year-over-year, US business up 104%, contracted backlog up 134%. For an AI software company, those are the loudest quarterly numbers anyone's posted in 2026.
And yet, looking at the same release, our rating on Palantir (PLTR) is still Hold, $160 price target — about 11% above where the stock trades today.
That gap is the whole story of this article. The numbers are real. The price is the question. At roughly 108 times forward earnings, the market is already paying for the next four to five years of execution to go right. Below, we walk through what actually happened on the quarter, why the engine is running this hot, and whether buying after a print like this makes sense for a regular investor — or whether the smart move is sitting in your seat.
What Palantir Reported on May 4
Palantir released first-quarter fiscal 2026 results after the market closed on Monday, May 4. The headline figures came in well ahead of consensus on every line that matters:
| Metric |
Q1 FY26 Actual |
Consensus |
YoY Change |
| Revenue |
$1.63B |
$1.54B |
+85% |
| Adjusted EPS |
$0.33 |
$0.28 |
beat |
| US revenue |
$1.282B |
— |
+104% |
| US commercial revenue |
$595M |
— |
record |
| Remaining Performance Obligations (RPO) |
$4.45B |
$1.9B |
+134% |
| Net Dollar Retention (NDR) |
150% |
— |
+1,100bps QoQ |
| Commercial customers |
1,007 |
— |
+31% |
| Adjusted operating income |
$984M |
— |
60% margin |
Two of those numbers deserve to be read twice.
US revenue grew 104%. That's the first time Palantir's domestic business has cleared the 100% growth line since it went public in 2020. For a company that did $1.3 billion of US revenue in a single quarter, doubling that base is genuinely unusual.
RPO — the contracted future revenue not yet recognized — jumped 134% to $4.45 billion. Consensus had it at roughly $1.9 billion. RPO is the cleanest forward-looking number in software because it's already signed paper. A 2.3x beat on RPO means customers aren't just trying Palantir; they're locking in multi-year commitments.
Management also raised full-year guidance to $7.65–7.66 billion, up from a prior range below that, against Street consensus of $7.27 billion. That's an implied +71% growth rate for the full year — not a slowing-from-here re-acceleration, but an "already running this fast" trajectory.
The stock's reaction in the after-hours and Tuesday opening session was strong. By Tuesday's close, PLTR was trading near $144 — call it 11% below our $160 fair-value anchor.
Why the Engine Is Running This Hot
The mechanism isn't a single contract or one-time win. Three things are firing at once.
The Artificial Intelligence Platform (AIP) finally has a commercial flywheel. AIP is Palantir's chat-and-agent layer sitting on top of its older data-ops engine, Foundry. For 18 months, the bull case has been "AIP will turn pilots into paying enterprise contracts." Q1 is the first quarter where the math actually shows that. US commercial revenue hit $595 million — bigger than the entire commercial business was as recently as 2024. AIP "bootcamps" converted to paid deployments at a noticeably higher rate than last year, and average deal size keeps stepping up.
Government contracts are still compounding, not plateauing. The US government segment, which many investors had written off as "saturated," accelerated. Recent DoD ecosystem expansion — multiple eight- and nine-figure awards across Army, Space Force, and intelligence community customers — is showing up in revenue now, with even more in the pipeline. Government remains roughly half of total revenue.
Net Dollar Retention reset higher. NDR of 150% means the average existing customer is spending 50% more than they did a year ago. That jumped 1,100 basis points quarter-over-quarter, which is rare in mature software businesses. Customers aren't just renewing — they're aggressively expanding seats, modules, and use cases. This is the single best operational signal in the print.
Add it up: 11 consecutive quarters of accelerating revenue growth — a sequence genuinely hard to find in public software.
The Valuation Question — and Why It Anchors Our Hold
Now the part that's quieter, and the reason we're not at Buy.
At roughly $144, Palantir trades at about 108 times forward 2026 earnings. For comparison, the S&P 500 trades around 21x. High-growth software peers like ServiceNow trade in the 50–60x range. Snowflake and MongoDB sit between 80x and 110x on forward earnings, but they aren't growing at 85%, either.
A 108x forward multiple is doing a specific thing: it's pricing in roughly four to five years of continued elite execution before the multiple normalizes to anything resembling a premium-software peer. That's not bearish — companies that deliver four years of accelerating growth do exist — but it leaves no room for the second derivative to slow.
To make this concrete, here's what the market is implicitly saying when it pays 108x:
- AIP commercial growth needs to stay above 50% per year through at least 2028.
- Government revenue can't decelerate from current double-digit growth.
- NDR has to stay north of 130%, which would be a multi-year record at scale.
- Operating margins need to keep grinding higher from the already-impressive 60% adjusted level.
If even two of those four hold but one slips — say AIP commercial growth lands at 35% in 2027 instead of 55% — the multiple compresses fast. A move from 108x to 80x on the same earnings is a 26% drawdown before any earnings miss.
This is the asymmetric problem with buying after a blowout print into a richly valued name. The good news is mostly already in the price. The bad news, when it comes, hits a stretched multiple that has further to fall.
Three-Scenario Price Target
Our framework reflects this asymmetry directly. We're modeling the next 12 months across three paths:
| Scenario |
12-month PT |
Probability |
What it requires |
| Bull |
$200 |
25% |
AIP commercial accelerates further; another DoD mega-contract; FY26 revenue beats raised guide by 5%+ |
| Base |
$160 |
50% |
Current trajectory holds; FY26 lands in-line with $7.65B guide; NDR stays near 140% |
| Bear |
$110 |
25% |
AIP commercial growth decelerates to 40-50%; one quarter of NDR contraction; multiple compresses to 80x |
Probability-weighted fair value: ($200 × 0.25) + ($160 × 0.50) + ($110 × 0.25) = $157.50
That blended number lands within a couple of dollars of our $160 base-case target — and crucially, only about 9% above current trading. For a stock that can move 10% in a day on a single comment from management, a 9% expected return is not the kind of risk-reward setup that warrants a Buy. It warrants holding what you own and waiting for either a better entry (a bear-case correction) or new evidence that the bull case is more likely than 25%.
Implications for Different Holders
If you already own PLTR: the thesis is intact and the fundamentals support the position. We're not saying sell — we're saying don't add aggressively at this multiple. A pullback into $115–125 on any market wobble would be a more attractive accumulation zone.
If you don't own PLTR and are deciding: the print is a poor entry. You're being asked to pay 108x for a company whose numbers, however good, are now widely understood. Consensus has caught up; the asymmetric upside that existed when PLTR was $40 in 2024 is gone. A better playbook is to set a price alert at $115 and revisit.
If you hold AI software peers: Palantir's print is a directional positive for Snowflake (SNOW), MongoDB, and ServiceNow — all of which trade at meaningfully cheaper multiples. Smart-money rotation may flow there if Palantir gets too crowded.
Spillover Across the AI Software Stack
This print tells us something about where AI capex is landing. We've mapped the physical infrastructure layer — the chips, racks, generators, cooling — carrying the $527B AI capex cycle. Palantir sits one layer up: the software that lets enterprises use the compute they're buying.
The read-through: demand is moving up the stack. After the chip names had their 2024–2025 moment (NVDA, MRVL) and the power names had their 2025–2026 moment, application-layer AI software is now showing the same growth profile. Names worth watching alongside PLTR: Snowflake (SNOW, Cortex AI), MongoDB (MDB, Atlas Vector Search), and ServiceNow (NOW, Now Assist). All three trade meaningfully cheaper than PLTR. If the blowout raises expectations for the cluster heading into their prints, multiples could compress on disappointment.
What to Watch Over the Next Two Quarters
Three things that will tell us whether the bull case or our Hold is closer to right:
- Q2 NDR. Does 150% hold, or was Q1 a one-time spike from a small number of mega-expansions? A pullback to 130–135% would still be elite, but it would dent the bull thesis.
- AIP commercial growth rate disclosure. Management has been selective about isolating AIP from Foundry in disclosures. If they break it out cleanly in the next two quarters and growth stays above 60%, the bull case strengthens. If they keep it bundled, the market should be more skeptical.
- Government contract concentration. Roughly half of revenue still comes from US government customers. Any sign of budget delay, contract restructuring, or political turbulence around big DoD programs would expose how much of the bull case rests on commercial AIP momentum alone.
For deeper modeling and our updated forecast curves, see the Palantir forecast page.
Frequently Asked Questions
Should I buy Palantir stock after the Q1 2026 blowout?
Probably not at current prices. Palantir trades at roughly 108 times forward earnings — a multiple that already prices in four to five years of strong execution. Q1 was excellent, but the Street has caught up. Existing holders can sit tight; new money is better off waiting for a pullback into the $115–125 range or evidence that the bull-case scenario (sustained 50%+ AIP commercial growth) is becoming the base case.
Why is Palantir stock so expensive when the numbers are this good?
Investors are extrapolating: revenue grew 85% in Q1, RPO grew 134%, and NDR jumped to 150%. The market is paying today for those rates to persist. The catch — any deceleration, even to "still very good" 40-50% levels, typically triggers a multiple reset. At 108x forward earnings, the price already assumes the next several years go well.
Is Palantir really an AI stock or just a government contractor?
Both, and the mix is shifting. About half of revenue comes from US government customers — DoD, intelligence community, civilian agencies. The other half is commercial, driven by the Artificial Intelligence Platform (AIP). In Q1 FY26, US commercial revenue grew 104% to $595 million — the fastest-growing segment and the reason the bull thesis exists. Palantir is a defense-tech business at scale plus a high-growth enterprise AI software business stapled to it.
What does Net Dollar Retention of 150% actually mean?
NDR measures how much existing customers spend today versus a year ago. A 150% reading means the average customer paying $100 last year now pays $150 — through new modules, more users, expanded contracts, price increases. Above 120% is considered very strong in enterprise software. 150% at Palantir's revenue scale is exceptional and is the single best operating signal in the print.
What's the biggest risk to the Palantir thesis right now?
Multiple compression on a growth deceleration. The stock is priced for sustained acceleration; if any single quarter shows AIP commercial growth dropping below the mid-50% range, or government revenue stalling, the forward multiple could re-rate from 108x toward the 70–80x peer range. That's a 25–35% drawdown on the same earnings — without anything breaking in the business. Valuation risk, not business risk, is the main thing to watch.
Bottom Line
Palantir's Q1 print was the strongest quarter we've seen out of any large-cap AI software company in 2026. US commercial up 104%, RPO up 134%, NDR at 150%, and 11 quarters of accelerating revenue is a genuinely rare run. None of that is in question.
What's in question is whether the price already reflects all of it. At 108x forward earnings against a $160 base-case target, the math says yes — about 11% upside in the base case, with a left-tail to $110 if the growth story takes any hit.
That's why our rating stays at Hold, $160 PT. Own what you own. Don't chase a great quarter into a great multiple.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. All investments carry risk. Readers should conduct their own research and consult a financial advisor before making investment decisions.
