Good morning. Welcome to Palantir's Fourth Quarter 2021 Earnings Video Conference. We'll be discussing the results announced in our press release and related materials issued prior to the market open and posted on our Investor Relations website. This morning, we will make statements regarding our business that may be considered forward-looking within applicable securities laws, including statements regarding our first quarter and fiscal 2022 results, management's expectations for our future financial and operational performance and other statements regarding our plans, prospects and expectations. These statements are not promises or guarantees and are subject to risks and uncertainties, which could cause them to differ materially from actual results. Information concerning those risks is available in our earnings press release distributed prior to market open today and in our SEC filings. We undertake no obligation to update forward-looking statements, except as required by law. Further, during the course of today's earnings video conference, we will refer to certain adjusted financial measures. These non-GAAP financial measures should be considered in addition to, not as a substitute for or in isolation from, GAAP measures. Additional information about these non-GAAP measures, including reconciliation of non-GAAP to comparable GAAP measures is included in our press release and investor presentation provided today. Our press release, investor presentation and SEC filings are available on our Investor Relations website at investors.palantir.com.

Welcome to our earnings call from Denver. It's my maiden voyage. And obviously, just a couple of opening remarks, and then we'll jump into questions from the audience as it were. The -- just a five-minute, two-minute -- when Palantir began, people believed that data was worthless, that software was a luxury item, and that we would fail. And one of the very interesting things that's happened to Palantir is we've been able to see how the world has changed dramatically in its perception of software and, of course, of us from a world where software was something that you might want, might be in your car, but de facto would not determine your business to a world where really the laws of finance are going to be rewritten to deal with a world where the only real moat is software. How do you measure it? What does it look like? How does it -- how do you understand when it's creating value? How do you understand when it's declining in value? How do you understand when it's compounding? To what extent is it compounding? What devices do we use to measure that? Are the devices we use to measure it the ones that we used in the past? Clearly, this industry is in its infancy. What's also very special about this industry is it really is, by and large, geographically located in a small section of America, which is odd, and there's lots of interesting reasons for that. but enterprise software is something that America is by far the best at. What we see currently at Palantir is not just the best at building it, it seems to also be the best at understanding what software developments are relevant for the world today and adapting even when it's being offered by a company that in every way looks nonstandard, run by people that are very different, look different, feel different, talk differently and with a CEO that many of you is [Expletive] crazy. And so, just as an interesting preview, I don't want to take a ton of time with remarks because I think when I'm watching these things, if someone talks too long or there's like a lot of canned remarks, I wonder why, honestly gets a little boring. And of course, our legal department and IR department, which are wonderful departments are -- have a cane ready to pull me out if I'm not like a caged animal in the 1950 zoo. As I've mentioned to them, if you want caged animals in 1950 zoo, you can watch any other earnings update. So here, you see an interesting chart. We sent this out. Some of you have probably looked at it. Some have maybe even studied it. Some of you haven't seen it. What's interesting here, what I thought would be very interesting for people who are investors, potential investors, also Palantirians, both current and ex, is our journey and what I actually believe this is a metaphor for the journey of all software companies. We had -- we were very early into whatever you want to call data exploration, building things that would now be understand as useful software for building analytic tools. Some of them were -- and are very, very important for national security and other areas. But then we're not -- obviously, because they were built and conceived in 2004, delivered in 2008, we're not actually able to migrate across the chain into what people need tomorrow, where people are already seeing that they need today, which is essentially not having software as a raw material exploration. You take the data, like it's oil, you pump it out, you churn it, and then you save -- churn the data and you move on, but actually operationally determinative for your business. What you see here is cohort analysis. This is the -- and what you see is from inception of Foundry, the decline of older software products and just the massive exponential growth, obviously, rough math. It's 100% growth year-on-year. And then starting last year, not off of a small integral, which is obviously very important because small numbers can grow quickly even if the software is not strong. Big numbers don't grow quickly if the software doesn't exist, especially given that our sales force is super nascent. We're building it quickly, but we only have 25 fully accredited software. Fully accredited meaning they've been here for nine months or more. And what you see here, obviously, the CAGR here is just unbelievable and in like the 150-plus range, which is super interesting and I wanted to drop the F-bomb here, but I was told that was probably inappropriate, but you could definitely -- we've agreed that this is something like the Phoenix rises. You don't get this. And the other thing that's kind of anti-gravitational about this, that it's easy to forget. This is a company, this is like we've been at this for over 15 years. There are certain laws of nature in business that we are defined, which is that a software company, software usually decays for lots of reasons, decays radically. And so when you see a decline or a software product, it's also basically not part of the law of nature for a software company to build new software. Really, as far as I know, we are the only software company in the world building transformational software this far in. And that's particularly important because most software companies have distribution or they have a product, but they don't have distribution product and the ability to build new products. There's just not a critique. These are great companies. They acquire companies. There are not that many companies to acquire, which is why even relatively weak companies get acquired at a very high price. Let's just look at the next chart, which is USG. USG, you have a very similar phenomenon where you see the inception of Foundry into the USG. This is -- there's a lot here. There's a lot of qualitative stuff here that we can explain, but one of the qualitative things that you kind of can get a sense of is Palantir, the newer Foundry version not only grows dramatically with like -- this is like 65% growth just without looking at it more precisely, but it's over -- it's like over 200% CAGR here, which is also like Phoenix rises kind of thing. But what's qualitatively, particularly important and very protective is these graphs pretty neatly on to what are the programs that are going to grow tomorrow. Where is the future of USG? What do we need in a world where people are recognizing it's very dangerous? And what would the products be that you would need to power that. And so any case with that, I think we should head into questions. A - Rodney Nelson: Great. Thanks, Alex. Our first question is from Brent Thill with Jefferies. [Operator Instructions]

Thanks. Good morning, Alex. When you go back to the sales expansion --

By the way, we can't see you --

…through your overall view on what you need to do on the sales force to get that building in the right direction. And what your plans for capacity are at in 2022?

By the way, it would have been wonderful to see you. But yes, I think the question was what can we do to build on our -- building the sales force next year or this year and the next year. The -- so there are really two parts to the question. One part, I think, is just to look at what's happening. And so like what is growing Palantir now and how can we accelerate it. So if you just -- what I want to do in this call and in future encounters is to both, is to kind of -- what I think is happening and just to frame it, what I think is happening at Palantir is you have what you saw in Silicon Valley 1.0 when Silicon Valley actually produced goods and services that people wanted, which was they built dual product, dual-use products and government, and there was a handoff to commercial. And so that's going very well. So you see -- even if you adjust for SPAC organic growth in commercial last year, U.S. was 80% or just under 80%, it's like 76%, 77%. But to your point, so -- and you see dollar -- net dollar retention numbers that you financial people like, how much is it growing? What I would call actually organically, I guess you guys call net dollar retention. Very, very strong like in the 150 range. We have Glazer here, he'll pipe in with the -- but it's roughly in that range. But what we do not have -- you could look at it as a negative, like we basically are doing this with a very, very nascent -- honestly, if I were not on TV, I would say, bunker small sales force. And so like that means two things. What does it mean on the positive side? It means that net dollar retention number is awesome because my understanding of how this happens at other companies, there's like eight people running around begging the person to expand their deployment. That's not happening at Palantir. The salesperson there is the Foundry product, by and large. So obviously, that means we think we can get lift from hiring salespeople. We're doing that aggressively. I think we have about 150 salespeople now. We're in the market to hire aggressively across Palantir, honestly, mostly not salespeople because we believe we live and die based on our ability to build products of tomorrow delivered today. But we're hiring -- going to hire in the range of 200 people. Hiring and getting value out of them are two different things. And so I think one of the things that we will see going forward is how do we play these salespeople, so that we can get as much lift from them as we're getting from the product itself. And we're at the beginning of that. And so it is the part to people we've fully indoctrinated or as normal people might say, trained, are effective, but we've never done this before. There will be -- that's a process, we're also looking at ways to learn from other companies, but that's going to happen over the next couple of years. I think the primary driver of revenue in U.S. commercial this year will still just be the way in which America, the IT organizations are actually beginning to request without knowing Foundry sometimes what Foundry offers. The CEOs assume that this is available. There is no other product on the market that can actually move from data warehouse, which is like there's like five companies, they all offer the same thing. Foundry interacts with all of them. When you move up to the chain to actually solving a real problem because it's defined your business, Foundry is highly differentiated. So we will layer that. I think it's way too early to know exactly the impact. Obviously, to the extent we get this to work, you could expect even higher growth than what we have, although I'm very happy with it. It was like we've doubled USG U.S. commercial now from 50 to 100, 200, I believe we will double it again this year. So any case, these are -- some of this is unknown, but we are working very aggressively. The other thing I would say outside of the U.S., again, a slightly longer answer, but one of the ways people look at Palantir is if you're going to -- I think the financial analyst likes to -- what we would have called science normalize the data, what you and finance would probably call strip out the inorganic inputs. But if I think if you're stripping out the inorganic inputs, you also have to look at the -- what I would essentially view as an inorganic input of COVID in Europe. If you strip -- and so Europe has been slow for us last year, which is one of the reasons why we were only at 41% last year and growth would be even higher. Europe grew, if you include SPACs at 9%, so slightly above what's probably inflation. But there, we're taking opportunity believing there'll be a handoff from USG to U.S. com to Europe to rebuild with some very, very strong salespeople who've come from the best companies, believing that that's an environment where we can really do that. So like there's two ways to look at this, just on that angle. You could say, well, what could you do to gas charge America? We're working on that and very, very aggressively and most of our sales hires today are focused on that. But there's another more long-term thing. If we're right that the world wants Silicon Valley the way it was, meaning a handoff from government to commercial to the rest -- commercial U.S. to the rest of the world. And with each one with a lag, America to Europe 18 months, the rest of this company only has to grow, everything besides Europe, commercial 38% to get 30% growth in aggregate. So getting Europe to track better, both from a handoff of what happens in America to Europe and with more conventional sales approach is kind of what we're looking at getting done this year and see the fruits of it next year.

And just a quick follow up on the government. I know you had a tough comp. It did decelerate pretty materially in Q1 in terms of the growth. Can you give us your perspective on what's happening on the government side?

Well, there's a couple of things that are happening there. So if you were looking at this more like from a scientific perspective, you had a time series that's 15 years, first thing you would do is say, okay, what's happening in that time series over the 15 years. What you see in U.S. gov is a compounded growth of 30%, but like this, which is the positive of U.S. gov is it's reliable. The sums are big. The quality of the revenue is very high. The -- one of their -- essentially, there are a number of problems, but the biggest problem is barrier to entry, which we've clearly solved and then rebarriers to entry, which we've solved or solving, but that's going very well. And then the second one is lumpiness. Now you -- that lumpiness still exists. And actually, in some ways, it's worse because to get the integral to grow, you need these massive deals. We also have small deals, but the fact that we are on the biggest, most important parts of the U.S. government on our software is. So there's really a twofold answer to your question. One, what will happen this year? Are there -- is there deceleration in actual one over a long time series? The answer is clearly no. But then the question is, if the baseline is 30, how does it get to where we want it, which was like the beginning of last year. And now at the end of last year, and the way that happens is the deals we're already positioned to win actually closed. And so then you get into the granularity of what will happen in the U.S. government, who gets the deals if there's no new budget. There's a lot of granularity there, which we should probably do a better job of sharing. But the short answer is, it's like huge chair gets pulled first. The people are trying to enter the market first, last, so the new start-ups totally screwed. Because the people who are not sitting on crucial programs, partially screwed. The people that have software that is -- or products that are useful in the past, but have the right connections probably -- another version of this is if you just look at that chart, I showed you with the CAGR on Foundry, these are the most important programs for a dangerous world. Now I can't go into all the details, but we used to debate with people, especially my academic friends if the world was dangerous. The danger of the world being clear and present to the U.S. government is very protective. It doesn't guarantee that when this integral actually -- how it behaves, but it makes it much more likely that it will happen here and positively affect our revenue which is another reason why I suspect that we will do well.

Great. Thanks, Brent. Our next question comes from Palantir shareholder, Chase P.

Yes. First and foremost, congratulations on all the hard work. It seems like you guys have a great team and are executing really well. From a retail investor perspective, the most negative sentiment I hear regarding Palantir is in regards to the dilution of shares outstanding over the past 12 to 18 months and primarily in relation to stock-based compensation that's occurred. Other than the remaining shares to be invested that have already been announced, can we expect further dilution in share offerings going forward? Or is it kind of reasonable to assume that the majority of this was from the IPO process and sort of a onetime event for the company? Once again, thanks, and congratulations on all the hard work and business developments.

Thank you. And I really appreciate you, investors. Thanks for investing and the faith you have in us. Okay. So there's like the simple version, which I think it's like -- so there's really -- there's stock-based comp and there's dilution. Dilution thing, that's a red herring. We're not issuing a lot of new shares, I think it's like in the $9 million range. And so it would be a little coy of me to say that's like no issue, move on. The thing to understand about Palantir and then I want to just take this like, it's actually not the result of the DPO, it's the result of the fact that we were completely focused on building product. We had no earthly idea we were going to DPO like right before we did it. And so most companies are quite frankly built so that the -- when analysts look at it, the primary customer of most software companies is not the client, it's the software analyst. So it's like we, obviously, our primary clients are our clients. which doesn't mean -- and then now we're thinking about how do we expose the data in a way that people on the outside like you and professional analysts and others can look at the data and get a better sense of what's tracking, what's not tracking. But the primary source of a lot of these like questions really comes down to look, we built the company to support the U.S. warfighter primarily and then do – take dual, use it for the glory of humanity, particular humanity in the West. That was our idea. And because our primary client was not what someone had a hedge fund would think, we didn't actually think of these things from inception. And so now there's a process of normalization. You're just going to see that in going forward on these calls just like how do you normalize, how do you provide data that people are going to look at, how do you provide data that people can understand that they're used to seeing, while simultaneously staying true to what our mission is. It's like our primary clients are the people we're serving. We're in full align with them. And that's why we survive even with the nascent sales force. You can get things to double, which is in sync. So then you get to stock-based comp, which is like, okay, so -- and there's 2 parts of it. Of course, IRI people kind of don't want me to do any kind of forward-looking math, but if you're smart enough to invest in talent, you're smart enough to figure out. There's essentially -- there's the -- how are we comping people, and there will be a normalization that will get us into a range where you would see in a software company within the next 18 months, latest 2 years. But there's essentially -- and that's going to take a little time. It is going to happen, because it's also very much linked to another question, which is how do you actually run the company so it's profitable someday on a GAAP basis, not stripping out comp. And that was also within eyesight. And those are our goals for Palantir because same reason we have no debt. The same reason we have $2.3 billion on our balance sheet. This is a company built for bad times. Bad times means strong finances internally. And that means at some point, you have to be GAAP profitable. You can't be GAAP profitable if you're diluting people or -- correctly your stock based comp is totally -- is not in conformity with other companies. So you're seeing a normalization. This will change. It will change in the relatively near future. It will be linked to other things that we believe are important for Palantir like having a company that thrives in bad times. And we are -- bad times are very good for Palantir because we build products that are robust, that are built for danger. And then the finances internally are actually built for bad times. And bad times means you have free cash flow, the free cash flow turns into GAAP profit. That means the stock-based comp has to be one that's aligned with our investors also because that's basically -- it's part of a little bit longer philosophical narrative, but like if software is the only moat, then value and gross shares have to be re-evaluated in terms of their value, value only exists if you can actually get a tech node, call it, maybe something besides. And growth only exist if you build a company that is where the technology is strong enough, the business fundamentals are strong enough that the free cash flow actually turns into GAAP profitability, and that's linked to stock return. So this is a priority, both because you care, but also quite frankly, because it is the health of our company, which we care a lot about.

Great. Thank you, Chase. Our next question is from Keith Weiss with Morgan Stanley. [Operator Instructions]

We’re neighbours in New Hampshire.

Also, in New York. So 2 questions. One on the product side of the equation. And the other on sort of the investments into 2022. On the product side of the equation, Alex, maybe hoping you could help us sort of better understand the product road map on a go-forward basis, how you guys are thinking about it from a high level. From our perspective, you guys did a really nice job of better modularizing the platform and made it more adoptable by commercial enterprises. And I think it looks like we've seen that traction in terms of customer adoption. What's kind of the -- on a go-forward basis, is there more activity of that ilk, if you will? Do you create more like prebuilt solutions, if you will, more directly target some of these sort of opportunities that you've been used for, but productize it, if you will? Is that a potential product direction? And then on the other side of the equation, in terms of investment, investing for growth into 2022, I was hoping you could give us a little bit of visibility into the nature of those investments? Is it just sales headcount? Is it the forward deployed engineers? How should we think about where those dollars are being deployed?

Thank you for your question. And so actually, to my perspective, they're very much linked. Our primary investment in growth is product -- is investment in product, and we're doing a number of things in product. The things we've talked about at a general level is kind of making our product more modular. There's a slightly more macro riff here, which is that we were adversarial with IT structurally until recently. We are adversarial because from our perspective, there's a learning process where they had to build these things. Now there's a myriad of companies. They're all honestly, technically hard -- indistinguishable doing data lakes and all sorts of things that help IT people build something that is working for them to do certain things. And we were adversarial because they were like, okay, this Foundry thing, yes, great. But we've already built these things. You would replace this, it may honestly could also make us look bad. No one wants it also. The average sale price for Palantir Foundry across our business, I think last year, it was like $6.5 million. Most IT people prefer a small bite in consumption. You can argue whether that's the right model, but instead of fighting them, it's probably better to figure out a way to get our product more in their hands. So that's kind of the known part. What we've been working on recently, which is less known is what we're really working on is we believe that people are paying a lot now for consumption and compute. No critique. But in reality, that's very much like paying for gas and oil exploration. But what people are really going to want is the ability to use the fully digested product. It's like when you drive your car, that's minerals and oil products and all sorts of chemicals built into your car, finished product. So -- and we're going to build both modules that are reflective of Foundry but also new ones in areas that we understand. And quite frankly, we know will be built in the near future, so the things we know are working, things that we suspect to work in the future so that the nodes not only work separately, but can work together. So de facto, we believe the compute of tomorrow won't be just compute, it will be productized compute. And that's what we're actually working on in rebuilding Foundry. So there's still -- that doesn't mean like the foundry as an aggregate, we can do the whole thing tomorrow thing, massively valuable, and we're working on very, very large deals where companies are like, "Look, we want to transform what we're doing or take what we're doing and export it to every company in our industry tomorrow." De facto, that's a Foundry use case, and I don't think there's anything else that does that, because you can take -- for example, there's a very large company in the health care space, and it's like they have a very interesting way of doing health care. They can't sell to other people without that essentially being a software offering. Building that would be 3 years or it could be 2 weeks. So that, we're very much committed to continuing doing because there's like we're an N-of-1 there. But where we want to be as an N-of-1 on not just making it small, we like that because then we see the IT departments now saying, okay, well, we have all these things. and now we want to migrate here and then selling them something that they can actually bite into. So it's – because the primary resistance to Palantir has never been lack of sales force, like it's been resistance on inside of the IT structure. And so now we're going to give them something they want and then also build an ecosystem around it. So it's what I think people will want going forward. And so people can stop just buying pure compute, they can buy valuable compute. So that's actually a big project. I would say on that end and on these other ends, which is particularly interesting, I mean, it's obvious, but it's not just the dollars, it's like who's spending them, just like the charisma of what we're doing. We're getting -- we're a company that's like 15 years old and say -- like from revenue, 18 years old from concept for the -- is a company like ours should be getting declining talent. The talent we're getting now is the best in the world. It's the best we've ever gotten. And we're getting people who used to be at Palantir. Everyone knows how good our people who are coming back. Just like. no company of our pedigree gets people coming back. The reason they're coming back is because this is just [Expletive] cool. It's like you do this, you change the world. Now there's a lot of other things we're working on the side to actually make sure. One of the things we failed at, honestly, is capturing the value of what we've done. Most of the products you would see on a map in the industry, any company, they're delivering things we built 7 years ago. We failed in capturing the value of that. We're not going to fail again. We failed in capturing the value of that because we were selling to IT and selling to people in an adversarial way. And we – sure, we were 7 years, 8 years ahead, but 8 years ahead, but they can actually interact with you. That's not the right way to do it. And we're not going to do it that way. In Americas, we're building specialized salesforce, -- we’ll take longer to get it to work, because we're working. And in Europe, we have differentiated on the idea that -- so if you just look at the raw numbers, if you look at like where we are investing in, it's like sales hires, very high-end salespeople in Europe and then the rest is just like the best tech engineers in the world because we know we get them, we retain them, and it's just very differentiated.

Great. Thanks, Keith. Our next question comes from Palantir shareholder, Brian L.

Karp, thank you for furthering the ideals of Western democracy around the globe. Of the 1,000 plus roles that you intend to hire this year, how many of those will be focused on sales?

So we're looking to hire 200 salespeople basically and everyone else is just like in the past. So the way I think about it is like x salespeople, it's like still 75% technical. We're going to try and hire 200 salespeople. And then as I mentioned, we just hired some very high-end sales acumen in Europe. Yes. Thank you for your question, and thank you for being an investor.

Great. Thanks, Brian. Our next question comes from Mark Cash with Morningstar. [Operator Instructions]

Kind of going up what you were just talking about. There's been commentary in the past around becoming the operating system for commercial industries. You talked about airline industry. Maybe just talk about health care for a little bit there. But are there other examples, industry examples you could talk about how that's pulling customers to standardize operations around Palantir?

Thank you. Well, there are a lot of -- like a lot of the new deals we're working. So it's like if you look at the time line of Palantir, 2 years ago, it was all kind of -- 3 years ago, it was like analytics and operations. And what you see now is kind of people building off of what -- wanting a standardized productized version of what we did at Airbus, what we've done internally at BP. What we did with William. And so what I can tell you is like, of the very big deals we're working on now, they're almost all this. It's like we used to have to educate people. They didn't believe us. And what's interesting, I do think it's like, obviously, there was also just the COVID distribution thing. I mean COVID distribution in England. England is not one health care system. It's 600 hospitals that our countries. It's like 600 countries -- and this is true. And so just seeing this happen or the networks of people hearing this happen is the reason why you have 80% organic growth in the U.S. ex SPAC and with almost no salespeople. It's because people are now like -- okay, now the caveat here is this is not for everybody. So there's like not everybody wants -- where this is particularly valuable as you have a business that is not protected by a moat, that has -- but has real insights on how to do something and wants to take over their industry. So it's like you're sitting there, you have a product that is maybe the best in the world, maybe the second best in the world, but it's not protected. But you do have insights for your -- in your software, you could take over your market. And that's where we're seeing it. The thing that's really changed is for once, it's not me fighting my way into the person's office and then throwing me out. It's them calling and saying, no, we know this can work. The pilot phase is now or days, and then we're on. And so like -- and we're working on 2 or 3 of these now. It's like -- it's just -- it's very, very exciting. The 1 thing I would say as a caveat though is it's a little bit -- we are both working on this and working on modularization with equal force because -- and they're just not the same thing. Somebody who wants this is not buying modular Palantir, they want the whole Foundry thing. And what they want is, what they want to help with is like, well, how would we identify people to hire the right to your platform? That's actually a big new question, like where can we find people -- they don't have to be Palantir quality software engineers, those are just too rare. But -- and what we're doing with our platform is making it so that just smart people can actually write to it. And so that increases our TAM a lot because smart and smart enough or specialized smart to write code at the level you write it here, those are just completely different TAMs.

Our next question is from Palantir shareholder Juan [Vi].

Alex, my question to you is, in a recent interview, Shyam said that what AWS was for developers of last decade, Foundry really will be for developers of this decade. Can you expand on what paths Foundry has to make available to a broader developer community. Again, thank you for allowing a retail investor like myself to have a seat at the table. I think it's much appreciated. Doesn't go unnoticed.

I see myself as a retail investor, and I have all my assets in Palantir. So I'm very happy to meet another retail investor. I hope -- in any case. So we've done these like founder for builders programs in America and in France, and like it's not charity. It's because we want people in the tech community broadly to learn how to write to Palantir. We also want to learn from them. So that's 1 very important program, not revenue based, but it's essentially very valuable for our tech development and very valuable to get technically literal people on Palantir so they can see what they can do and tell us what they can't do. Then there's the broader commercial. A lot of the companies we've supplied and government agencies side of Palantir, its like they require an ability to write to Foundry or to one of our products as like a core competence. And like so one of the things we know -- so we're obviously figuring out ways how we can train people, how we can make that experience easier, how can we widen the aperture. And by the way, obviously, not just because -- it's like it's not just for the altruistic reasons that are obvious. It's like if you want people who are valuable now to be valuable tomorrow, they must be able to interact with the software platform. So that's like one of the reasons we've had a lot of adoption because if you take a company like Chrysler Fiat, which has very talented people and it's -- they need to be able to write to the platform. And so that's -- and then work with the platform. So we're working on that partly for political reasons and partly because, obviously, it's very good for you as an investor in Palantir and we're proud of that.

Great. Thanks, Juan. Our next question is from Phil Winslow with Credit Suisse. [Operator Instructions]

I do appreciate the cohort data that you disclosed today. That was very helpful. Just to dig in on that a little bit. Alex, you talked about seeing -- starting to see an inflection where Foundry is getting pulled into some of these commercial deals sort of asking for Palantir without even knowing necessarily that's what they needed. You see that in the cohort number on the commercial side in terms of the new for 2021. I'm curious if we can just dig into that a little bit. Are you seeing specific industries really start to have that aha moment get it? Are there certain use cases that they're leaning into? And then just have 1 follow up to that.

What makes Palantir Foundry valuable is that it's not really industry dependent, and this is crucial for us because we -- our sales force is nascent. So we're very dependent on where either there's a crisis or where, again, that’s crisis or where somebody actually wants to enter a market they're not in and wants to expand. So I would say in the past, we were very, very dependent on manufacturing. So like our high-level engineering companies like BP and others because we de facto needed the engineering talent because no one else believed us as the product was differentiated. Now it is much more standard businesses of all kinds. And so just like really the who's who of American business from like people building tractors to building cars to oil and gas to distribution, and there's no one kind of company where we do not do well, it’s like, yes, we're not going to sell to a marketing company. It's like that we're not -- we don't buy their talking to them. They don't call us. We're not good at marketing, like they apparently are. They're so good at marketing. They don't have to quantify their results. We don't -- yes. So by the way, one of the places, Europe -- one of the interesting ways to look at our business, by the way, since you're in Europe and since I have an affinity to your region, Liechtenstein, it's that Europe in general has like grew slower and that's -- if you assume that comes back online, it's just going to be bombastic. One of the places that we're actually very strong in Europe is Switzerland. And there, you have like a lot of the pharmaceuticals, insurance companies, banks, that can't be mentioned using our products. And I think they use it, honestly, because it's like there's a Swiss quality to Palantir. It's like a very high-quality product. We deliver it. It will work. You're not going to get the charming slap on the back or the steak dinner and your software is going to [expletive] work.

And then just to follow up on that specifically because obviously, the net retention numbers you have in U.S. government, government in general is huge, U.S. commercial is huge. That's the 1 thing I noticed is with the disclosure of the net retention non-U.S. commercial significantly lower. What drives that? And how do you sort of inflect that higher?

Well, first of all, I do think there's a hand off function, and I think these things are actually repeating what happened in the earlier days in America when it's like where you have a hand off. Built in government because you have more time to actually get it right, hand off to commercial, go to Europe. I do think, in general, certain Swiss institutions, certain German institutions, certain French institutions are not included, where it is slower. Like if you are building something very, very new, it will be adopted a little later. But then there's also just the COVID. The reaction to COVID in Continental -- in Europe was different than in America. While it slowed things down in America, it didn't really stop them. And so I kind of -- what I suspect is going to happen is that Europe because of COVID reopening and because of basically people copying what's happening in highly adaptive America and sometimes in Switzerland, you will see the European cohort grow. But again, to just make it a little more quantitative, we have, I think, 150% net dollar retention in U.S. commercial. Again, a number which we're getting with basically no one holding it up. So I don't know how it works with other companies, but I think if I were a scientist, I want to normalize that number. But even not normalized, this is a very strong number. What we're going to show is net dollar retention in the U.S., and over time, we're going to show how this expands outside the U.S. And that I think what you'll begin to see is that the strength in the U.S. will go outside the U.S. and we'll make our business very, very robust.

Thanks, Phil. So Alex, we've taken -- we've received over 1,000 questions from shareholders. Obviously, you can't take them all. Any parting wisdom or parting thoughts that you want to offer about the business to our shareholders?

It's really tough times out there, really tough for a lot of businesses. It's -- a lot of things are going wrong in the world, in our world. It -- the obvious danger, the lack of legitimacy of a lot of our institutions. And I can tell you while at Palantir, we are very, very focused on our business and bad times are very super motivational for us. And when we get to good times, we'll be even stronger. And we're a little bit of a like wacky group of guerrilla war fighters, but we're very much in fighting mode. And not just for us and the West, but also for our shareholders. And yes, I hope to talk to you soon. Thank you.