CACI International Inc (CACI) Earnings Call Transcript & Summary

All right. Great. Thank you, everybody, for joining our CACI Presentation with CEO, John Mengucci. John, thanks so much.

Thanks for having us, Gavin. Appreciate it.

I'm just going to dive right in and maybe start with a high level. Over time, the battlefield has evolved quite a lot and CACI has evolved quite a lot. Maybe can you just talk about what that evolution has been over the last decade and how you guys have positioned differently than where you were 10, 15, 20 years ago?

Look, so today, we're still in the government services sector. That's what the geeks or the [indiscernible] says. But if you put yourself back to the 2012, '13 time frame, the war in the Middle East was ending. Drawdown had already started. We had a large expertise part of our business. That's how we describe where we provide talented folks to the U.S. government. Commonly known as selling labor hours. About 80% of our business was in the expertise world. About 20% was delivering technology. So if you couple the Middle East drawdown, you're through a little bit of the Budget Control Act and you're through some better buying power 1.0, 2.0 and you sprinkle the rest of that under a low price technically acceptable kind of acquisition strategies. It really was a time where the government was going to change how they were buying support. We looked at that time as a period of a positivity for us. It may not have been great for the 7 quarters, we had negative growth. But it really was about in a 63-year-old company, where do you head next and where do you venture forward. And what we quickly came into is we had two decisions. We could zero investments, we could cut all levels of investment streams, get our rates as low as we absolutely could and then go out there and still compete in the selling labor hour marketplace. The problem is everybody is business school, you sort of learn when price is the only thing you can differentiate on. It's called the commodity and the desire to run the company in the next 20 years selling a commodity, they can only differentiate on price. So we looked at where the threat was going, looked at what the government was sort of telling us, government was telling us that I want to buy people for less. And I don't think you can really differentiate in other than price. That's great. That's the government's view. So we set off on building out a technology arm to this company based on the decades of experience relationships we had built on the expertise side of our business. So we're, again, 80% expertise, 20% tech. In the last 10 years, moved the business to 45% expertise and 55% tech. We are a technology provider of national security solutions. We are looked at very differently by our core customers, that being DoD in the intelligence community, and DHS, which we look at DHS they're part of federal civilian, but they are the sort of paramilitary protecting the homeland. So they buy and they act and they talk about mission, very similar to what DoD and the Intel world did. So as I mentioned, 55% technology, 45% of that expertise. The majority of that expertise is where the government is procuring folks under government direction to deliver software. So we're building a satellite ground station. The customer wants to buy that by asking us to find 100 people who can write agile software to build a satellite ground station. We call that expertise. They're delivering an outcome, but the government is managing them. So a very different looking business and an investment model that has greatly changed. And over the last 5 years, compound annual growth rate revenue north of 7%, generated a little over $2.3 billion in free cash flow of 1.6x book-to-bill. Margins have moved to 9.5% to the high 10s, if you went back a couple of more years be 8.5% to the high 10s in a sustainable level and really focused on free cash flow per share growth of what we think is the ultimate value assessment of a publicly traded company. So that's where we are today.

Great. Two things that you guys have touched on just now at the Investor Day. One is the outcome-based solutions and the other is more software defined. I mean, you've been talking to me about outcome-based solutions for a long time. I'll come back to government efficiency. But I mean, is government efficiency something that actually could accelerate that for you?

Yes, it's -- I guess, I'll answer the second question first. Look, we believed in 2013, '14 time frame that, what the government needed was going to, going to change. The threats we face, the national security threats we face were quite predictable. It was 100% counterterrorism and very little on near-peer. Oddly enough, 3 or 4 years back, we talked about the great threat of counterterrorism is now gone. Applause. Now let's move to the near-peer. Problem is that terrorists don't really pay attention where the government is spending their money or where they say things are done. We always believed that as we built CACI going forward, it was going to be an end case. You got to worry about the counterterrorism threat, you have to worry about the near-peer threat. We really believe it was going to be more about technology and also software. Why software? Because the government today, 5 years, 7 years after we believe that's where we're going to end up, is that the threats and the pace of the threats and the pace of change is too fast for large platform hardware-driven solutions. It's just that simple. If every time United wanted to change their flight app, if you had to go into United and give me your iPhone and get a hardware update versus just hitting refresh or waiting 45 seconds, it's the latter model that the Defense Department -- Defense Department, DHS, Intelligence Community needs. These threats are changing, whether it's Israeli or Hamas or Ukraine-Russia, 24 hours, 48 hours is a long time to have a tactic in place before it's changed. So anyway you're going to get updated protection out to fleets and ground soldiers and space force and air assets are to drive those changes with software. So we doubled down. We like to say software is our superpower, because it is how you provide agility, speed and efficiency. If you roll into the next part of that, which is the inevitable question about Doge, I'm not sure any of us could even spell or pronounce Doge 30, 45 days back. But it is the hot topic and is out there and rightly so. I think it's a theory and it's the case that's frankly created two things within the market. And so it's two things that the market hates. When is it going to end/What is it/how do I measure it? And then it's presenting a lot of uncertainty and the market doesn't like uncertainty, it doesn't like understanding absolutes. And that's sort of where we put all of our investors in. If you look at the last word, it's called efficiency. As a taxpayer, we should want this. In fact, every administration back to Nixon focused on some form of a Doge. We didn't have a Doge name. We didn't have the Internet then. We didn't have high extraverted leaders to it that love sharing things out on the press. So it does create a lot of uncertainty. The fact is we need to always be looking at the efficiency of what the government does, which is provide the American people, both natural security and citizen-facing services. And we should always want to understand can we do that in a more cost-effective manner. Our investors, you all look at our P&L statement. You look at our balance sheet and you have questions every quarter around why is overhead up? And why is SG&A up? So we should be expecting that. I think, the uncertainty as to what's going to happen. So a couple of things that I believe, one it's my 42nd year doing this, supporting the federal government and national security space. So I sort of have some nice folks to go call on and a great team that goes to research this. I think there will be a push to find efficiencies. I think there will be a push to force speed. I think, there will be a push to get greater transparency. And that, nothing is really going to change except possibly for the better. The kind of business that we have every time the word efficiency is used is somebody looking to do something that's more resilient that lasts longer. That is more in the 2020 era versus the 1995 era. All of that begs positive for this company. I can't talk to you about the entire sector. I can't talk to you about aerospace and defense companies. I can tell you for this company, we're well versed and well positioned for that kind of change. It's true that 45% of our business is delivering expertise. But that selected expertise for this company, because we agreed in 2013 and '14 that having an expertise business made sense, because it gets you pretty close to customers. You get to really understand their mission. The trick is what kind of jobs you're going to go after within that part of our business. About 80% of that 45% is customers buying talent from us to deliver an outcome that is software or technology-based. I don't have consultants. I don't deliver consultants. I'm not filling offices with CACI folks to provide support. If we're providing support to DoD, it's embedded folks with groups. The 21 time zones that are not spelled with U.S. That's a kind of long-term enduring support that will continue regardless of any efficiency study, because it's an act on national security. The last thing that I'll share is, if you look at our revenues, we have 5% that is based in the U.K. and other, and 5% of that is pure federal civilian and 90% of that is DoD, Intel and DHS. So it's all around national security. If we find efficiency savings in there, it's going to be spelled with a couple of different words. One is going to be speed and one is going to be efficiency. So we can get more things done in software because the threat continues to change. So a very small portion of our company is citizen facing. We're not managing student loans. We're not processing tax returns. We're not processing healthcare claims. We're not collecting dues to our national parks. We're actually doing things that make or break this nation in term of national security. And that's what makes us different. So with that...

Where I wanted to go with it. I appreciate it. A year ago. I think those is obviously a wildcard. And one thing that's been discussed a lot is what can be done with either executive action or from the head of agencies versus going through Congress. Any thoughts on that setup of where that efficiency push would come either from Congress or from the agencies themselves?

Yes. It's funny that the Republic, and as the U.S. has got a lot of fail safes and a lot of checks and balances, right? At times, we get caught up in the rhetoric of what could potentially happen. Do I see one specific program being driven to cancel something that's major to national security? No. I can't tell you about the rest of the government. I don't watch the rest of the government. It's not of interest to me. Frankly, in my personal life nor as the CEO of this publicly traded company. I'm really watching what this company delivers, and we were very careful in our next strategy back 8 or 9 years ago to really sit down and talk about the business of delivering people to the federal government. It's tough work. It's mostly unpredictable. Things like administration changes can impact that. We went through the Budget Control Act 2011 or '12-ish right? We all learned how to spell the word sequestration, but we didn't know what the hell it meant, right? And now we sort of know that, that's a huge word. It really means some things to some people, but mostly not a lot to many. So the right fail safes and the right checks are in place to prevent what you just mentioned. I also think that we've got to get back to the simple word of efficiency. Is there any taxpayer right now that doesn't believe that the federal government needs to be more efficient? You'd all say no. Okay? So efficiency should be a positive word and the marketplace that we're in and what we do, we make it our business to go in and make things more efficient and allow us to make more exhaustive changes to things in a much more cost-effective manner. So I really don't know how it's going to play out. I don't have my own personal crystal ball other than being in this business 25 hours a day. I talked to customers at all level, talked to senior intelligence and DoD officials. Are they listening to Doge? Yes, they are listening to Doge. Do they see occasional print that goes into this ubiquitous media? They don't quite understand? Yes. Do they know they have a nation to protect and that the Israelis and Hamas and the Ukraine and the Russian issue, what the Houthis do with technology. That's their 101% focus today. And I think things will average out at the end of the day, if we can get network modernization to spend more money -- it's the federal government to spend more network modernization, which we're really damn good at because our networks do need to be upgraded. They need to be able to handle unclassified, classified secret and top secret information. They shouldn't be for separate networks. We have thousands of people to watch. We have an Air Force program called EITaaS. We're trying to return about a 1,500 airmen back to be an airman, an airwoman instead of monitoring an IT network to get e-mail and other back office work done across the United States Air Force. That screens of efficiency. We probably should take Cobalt out of a software program language in the federal government. Unfortunately, a lot of key systems still depend on that. So I'm not going to overreact to Doge. It's going to be done on July 4, 2025. And I look forward to see what they come out with. But again, we're a very different business, okay? We're not in there delivering 8 people to this organization, 48 people for that. We were talking earlier. I don't report on bench strength. I don't report on number of applied heads. I don't apply -- I don't report on direct labor. Those are all things that consulting and people who provide people to the federal government consistently report on. We don't report on that, because it's insignificant part of our overall business. Purposely, not by accident, but a purposeful strategy to make certain we wouldn't find ourselves in position we were in 2011 and '12.

So it sounds like not to put words in your mouth, but your view is that national security will remain a key priority for Congress?

The key priority for Congress as it has, right? We're a 63-year-old company, we probably have seen 15 or 16 administration changes. National Security is, in my experience, one of the only bipartisan areas of the federal government budget. People make a lot of claims on both sides. At the end of the day, if we're at war, we're about to be at war or the world is a dangerous place. We somehow find a happy medium to make certain that we're well situated and well funded. We can debate whether the nation needs 5,000 ships, whether they need 4,218. It's a great debate to have. We should always have that. But this business that I run today, we have 24,000 people that drive growth. We're in that area of signals collection and network modernization and things the nation absolutely needs. We won't put a warfighter out there without signal, without SIGINT. We're not going to put war fighters out there away with an ability not to find drones flying overhead. Those are all narrow deep funding streams across this federal government that if we ever got a call from a group called Doge to explain what SIGINT is, I would be literally shocked. I would be surprised. We're not -- what we deliver isn't $400 billion a copy or $400 million a copy. It's sort of embedded throughout the whole national security plan.

So what's an example of an outcome-based solution where you've gone from offering heads to selling on an outcome?

Yes, a couple of them. One is we're probably close to world class at finding signals, any zero and one that's out there floating around the space, we can find just about everything. You'll hear about 20% of that in the news, or 80%, 80% you'll never hear about, and that's purposeful. So one is being able to collect signals across the federal government, share that across the entire expense of the federal government and make certain that we're building software that can exploit those signals, get that information out and how do we defeat it. Another example is we began a push about 6 years ago, where we were looking for customers that were buying labor hours, buying expertise where they should be purchasing technology. One area that is, there's a lot of apps. There's a lot of applications that the U.S. government uses. One example is customs and border. One could say that their work has become a little more pressing over the last 4.5 or 5 years. But they would traditionally -- that organization would procure 400 to 500 people to keep the software that goes behind all the customs and border apps via viable and making changes to it. The problem is they would, on average, get about one upgrade a year across 140 apps. We were contracted with 5 years ago, they stopped buying 400 or 500 people. They actually bought to an RFP that said I want 140 apps updated quarterly for the next 5 years. That was the statement of work. And we were able to transform them to spend about half of what they used to spend per year. They received all the updates and upgrades that they wanted. So instead of buying an input, buying labor hours, buy an output, which is updated apps on a regular scheduled basis, very similar to way your iPhone works. And that was a transfer from buying labor hours to buying an outcome that cost half as much. And the way that works is you're going to pay half, I'm going to make more. I want you to pay less, I'm going to make more because the risk is more on me. I want to make sure I have the right staff to be able to deliver those. So an agile software development methodology is what we sell. It's why we say software is our superpower. It's very different to deliver software to a platform that is hardware. Why is that? Because every time you need a split second change, we can make that change and send it out to that asset, whether it's a plane or ship, or a tank, whatever that happens to be. So at the end of the day, efficiency has already been in our calculus. We've already been driving how do we make the government more efficient. We're also working on very important things like how do we drive free cash flow per share, how do we drive top line and bottom line growth with a customer that's very rigor around how they purchase. So we were able to deliver a much better outcome to them, upgraded apps on a yearly basis or 4 times a year over a 5-year period. The customer extended that work. We're in year 7 of a 5-year contract, hopefully moving towards a 10-year. At the same time, most customers are going around the federal government saying you should buy what these folks have. So if you looked at a recent award was NASA and CAPS, it was all of the digital apps across NASA. They were -- the government was doing that with a staff of 1,200 people. They got to talk to a lot of our government customers, and they get to understand, I want to buy Agile, I want you to control that. But here's the control [ Agile ] is the customer want. I want to change my requirements as we go along. I don't want to hear that, that's a change. And I want to have multiple modes being done. It sounds really simple. It's really hard. But when you find customers who got tremendous service, they start to sell it on behalf of CACI, and that's a kind of market that we're in, which is tying back to Doge. It's why Doge to me, is a sleeper. I hope they do find some things. I hope they find that the government used to spend more money in updating IT networks that -- networks that are not up 100% of the time to protect troops in the intelligence community, get additional funding. At the end of the day, if you can make a buying organization more efficient, they can procure the same goods and services for less. We should like that. And there'll be some oversteps and we'll have some challenges on the way. No cost takeout process is perfect. But I don't think it's as bad as people have treated it in the market. I do think, again, we're sort of back to nobody likes to be surprised and uncertainty is bad for markets. I will tell you that for this company, what we do, our strategy will not change. It is very founded in the fact that we're going to look at speed and agility and efficiency, and that will carry it and stay over the long term.

I will have revenue and cash flow questions in just a second. I can't wait. But I mean, what you just articulated, it seems pretty obvious just in terms of the outcome based and the take heads and win be more efficient. What's the inertia and what inning are we in, in terms of the government more broadly adopting that?

Yes, great. So the first customer at a large scale we did this, as I mentioned, was Customs and Border, and that was 5 or 6 years back. If you look at how our business development and our awards record is, we've done 11 $1 billion-plus awards. That's part of our business development strategy of bid less and win more and bid larger and longer duration contracts. Customers are spending greater than $1 billion looking at transforming, moving things into software and buying it as -- almost as a buy the bid or by the drink. Then write your requirements, contractor go off and build this, come back so we can tell you that our requirements changed, you can charge us more, and we never actually deliver to the field. We're actually probably in the third inning of customers understanding that. My hope, frankly, is that if Doge is even lightly successful, we skip the fourth inning and we move into the fifth inning, which is just the way you should be buying software if you want speed to the fleet and speed to the need, you ought to be doing that in software. I mean, think about your own life. If every time your car broke down, you have to drive it into the station. What if it took a software diagnostic and a software fix and your car was up and running. It's an absurd maybe example, but sort of what we're talking about. When the ship is being overflown by nation state drones, the only way to get those new signals into that ship to make sure it can find it before Spectral comes out, which is a massive surface ship job that we won. You'll bring that ship into port, you'll put more hardware cards in and you'll go back out there. Problem is the time you're going into the shore and the time you're heading back out to the fight is all lost time on station. We believe in space and the air and the ground and in all of our maritime assets, you should have changes being done in a softer manner across the national security space at large. And that's the marketplace that we find ourselves in.

It sounds like shifting spending privatization. But in terms of your multiyear revenue forecast, is there an underlying budget assumption that you need to see to support that outlook?

No. I mean, for me to watch the '25 budget is probably enough for us to watch, right? So the trick of run a national security company is working on relationships you have, making sure you understand where your customers need to head and making sure that our strategy supports that. And in areas where our strategy doesn't support that, that's not a viable growth area for us, okay? We're not the bluebird company where we just heard a customer X is going to do this, we should go bid on that job. We're much more about that long-term relationship, invest ahead of customer need, let that customer see the art of the possible and then let them buy and procure in the best manner possible. So if I look at where we go out, our multiyear plan was more a function of that our business development machine and our program management skills and the amount of transparency and information we share about our backlog, we're at a point where we should be able to come out there and come out with more than just a 1-year forecast to sort of let people look at the backlog, which today is a little over 4 years. The duration of new business we put into our backyards, our backlog is approximately 6 years. So that means we built duration -- average duration of programs from 3 years in 2017 to 6 years now. So pretty much doubled it. What that means for us is we're going to recompete less. So we have to recompete less, take the book of business we have and have to tell the government to please reaward that to us. We do that minimally today. We take all of that bid and proposal money, the IRAD money we would have spent on that. We flow that all into new business growth. I mean, it sounds like a simple model, but if I recompete less and I want my rates to be the same, I pick up an awful lot of tens of millions of dollars of investments I can put towards other people's areas. One of those big areas was going to where the prime contractors are today, move hardware solutions to software so customers should go be better prepared for the fight and not have vendor lock. Our software is open architecture. There's another company that does signals collection on bottles of water. That's great. Bring us your bottle of water signal code, and we'll go put that into our system and deliver that. So it's not based -- it's far more based on our backlog of business and how our backlog unpacks, which is different from most government services companies. And we believe that we have the transparency and the information to go out there and talk about more of a 3-year financial target all around free cash flow.

You mentioned bid and proposal. I think at the Investor Day, you showed a slide that had been more than doubling over the last 5 or 10 years. There's one big step function here and there. And is this just a -- was that a decision to -- we're going to spend more on B&P and R&D? Or was there one big specific technology or target that drove that step up?

Yes. I mean, there's been a couple of step-up functions, right? One is, if I look at investment overall, it's so much more about -- we believed strategically, strategy is a place where we come from. We put ourselves in our customers' shoes and said 5 years ago or 7 years ago, 7 years from now, what's the government going to look like and they're going to need in the national security space. Okay? If you're collecting parking meter collections, you're collecting that -- there's no strategy for that. That's just bid at the lowest rate as you absolutely can and go sell people to the government, not of interest to us. It really was about, can we make a fist in markets where traditional aerospace and defense primes have been, but the mission packages, not the platforms. We have great aerospace and defense companies. They buy -- they build some of the most eye-watering platforms on the planet. We should be proud we have those companies. But as you look at missions that those assets do, should they be in hardware, should they be in software, we believe it should be in software. Can we do that faster? Yes, we can. To do that, we're going to have to invest, invest more. So we did two things. We left markets that were not profitable to us. We reconstituted our entire rate structure. We went out and retrained PMs of how to grab technology programs versus selling people and we retooled business development. And that really told us we need to make step functions, investments beyond what we decided to do early on. And we're going to pick those areas well. And the areas where other smaller to midsized companies have created those technologies or those capabilities that we think if we had those, we could grow even faster, that's where we'll acquire. If we think we can invent or invest internally, so we own the intellectual property, and we want to build that capability out, then we'll invest. But our M&A strategy is very tied to our overall investment strategy, which we're going to invest, we're going to partner or we're going to acquire. And we've made some really great, not by accident calls of the areas of the market we believe we can make a larger fist than others if we combine the internal investments and investments in M&A.

How does all that translate into margins? You guys have stepped up quite a lot over the last few years. I guess, even though technology hasn't mixed up that much, some of that's been margin accretive M&A. But how do you think about the margin drivers in terms of mix versus execution versus maybe these higher-margin areas that you are pursuing?

Yes. It really starts with what were we starting with. So 2013, better buying power, sequestration comes out, you're sitting there, you're in the boardroom, you're saying, okay, now this is the absolute assault on delivering people to the federal government. Again, it really started with us realizing early on that we had two fundamental choices. We could remove all investment in the company and get our billing rate as low as we absolutely could. So somebody makes $100,000, we're going to build them out $102,000, and that's dollars for us. It does do well for margin. And if the only way I can hold that customer is to lower my price each year, I'm in a market that is worse than a commodity market. So every time a customer ask for a price decrease, they actually get it from someone, either myself or someone else. So a lot of the margin growth has been about repositioning the portfolio this company has. And a lot of that is towards how can I differentiate and be in a market where only five people can bid on this work versus $500 or $5,000. And that was more about consciously moving away from expertise. So the first major part was mix, okay? The second major move was driving a software-based company where all the technology we deliver has to be software-based, must. It's not a question, it doesn't have an exception. It must be software-based. It drives a software and a technology connection to this company. It also does that in the market. It does it with our customers as well. And then once you're in, technology and expertise and you believe you have the right mix, then it's bringing a software-based technology solutions business in place to deliver what people love to call products. I struggle with that term. But it's something that has memory and processing power in it, and it needs very low battery, how do I produce something that goes to achieve that mission. So a lot of it is mix, expertise versus tech. And then within tech, how do we get to high gross margin, high EBIT margin deliveries that we have. So it's sort of a two-step function. We will continue to double down on software period full stop. Every dollar extra that we generate within this company, we're going to put it towards internal investment. It will be in the area of software, whether it's signals collection, network modernization, IT modernization, cyber, AI and ML, those areas are where we're going to continue to invest in. They have the best gross margins. They have applicability across the entire national security space and allows us to continue to grow top and bottom line.

One more for me and then I'll open it up to Q&A. In terms of M&A, how do you think about what your targets are. Some of the acquisitions you made recently, Azure Summit, even if you go back to like [indiscernible] LGS, those were pretty unique niche, high-margin differentiated targets. What's the M&A strategy? What does the pipeline look like for unique assets like that?

Yes. Well, look, we're I can't say it enough. We're a strategy-based company. Strategy is a place where we come from and many CEOs can say that we have so many proof points more than the 5 minutes and 52 seconds are going to allow me to share. Look, we believe there were things that we knew how to create ourselves and there are things that we need to fill gaps in our strategy. We serve seven markets across the federal government. We look at those market strategies twice each year and look for gaps that can fill with investments, partnerships or acquisitions. Our acquisition strategy has been that way long before I came to this company. We've been buying companies for about 25 years. We're somewhere north of 90 companies that we have acquired. We do that in a number of ways. First off, small- to medium-sized companies work best for us. We have a powerhouse shared service center out in Oklahoma City that does the majority of the back office movement. So whether it's 1,000 people or 7,000 people really doesn't matter to us. Whether you're in cost point or financial system X, Y or Z, it doesn't matter to us. We really focus on what technology that company can bring to us. Flexible and opportunistic capital deployment strategy really means that I don't have a budget for M&A. I don't have a budget for share buybacks. I don't have a budget for buying down debt. It is dynamic. It's not predictable. I'm not the CEO who is going to tell you in the next 3 years, I'm going to spend 50% of our free cash flow buying back shares, because I have no idea what the market looks like 35 days from now. So flexible and opportunistic really means flexible and opportunistic. We used to favor M&A over everything that the company could spend its money on. We have been successful in moving that along with the Board's support. That's really a nice -- it really is flexible and opportunistic. Some facts and figures during 2021 -- we bought back 12% of the shares in this company. About 10 years prior, we bought back about 25% of the shares in the company. Why was that? Because we believed our shares were undervalued and underappreciated. There are times we'll do M&As. We have a habit of doing two together. Don't ask me why. It's not a plan. But again, I've learned over time that when I want somebody to be up for sale is never when they want to be up for sale. Every property wipens at a different pace. And we're really good about being a discriminating buyer. The last couple of years, we didn't do many M&As at all. We still have some gaps we haven't filled. But what we didn't do is get involved in a bidding war at unreasonable valuations and reasonable multiples over the last 24 to 36 months that private companies are actually looking for. So we didn't do a lot of acquisitions there. We might have missed some level of growth, but those targets come back around. We also have a large pipeline of those as well. So the strategy in M&A is look for capabilities and customer relationships that are on a time line that I don't have time to build it on my own. And we've said many, many times, we won't buy revenue. So what doesn't -- don't buy revenue really mean? It means that if I'm in a market today and I'm doing technology X, Y and Z, I'm not going to buy somebody who does X, Y and Z to go build market share. I'll beat you. First of all, my business development team enjoys that. Secondly, it's awful lot cheaper. Third, we have a large customer base that we can go out and make those things happen. So we're going to buy companies that fill gaps for us on a very quick measure to allow us to then get to the long-term organic growth that we absolutely love to talk about.

Any questions in the room? Perfect. Well, in the last minute or 2 minutes then, the bid less, win more, how do you make sure you don't miss something small that could ultimately be big?

Yes. So this -- when I talked about reframing business development about 8 years back, I came to the company in 2012, and a lot of that focus was around -- at the end of the day, when you bid more, you lose more. The theory is if I bid more, actually win more. But the fact is people can only put their best effort on so many things. And so the strategy of bid less and win more. During our Investor Day, we actually shared some quantitative financials in the last 3 or 4 years, what that plays out. We're basically bidding about half the jobs we used to bid, and we're up, I think, 55% on awards. So truth that if you bid less and you focus on the things that matter and you're really good at, you win a lot more. If you invest ahead of customer and you give a customer a preference that is called CACI, the customers more apt are able to pick you more often than they'll pick everybody else. So it is this bid less and win more. The next step is bid larger and longer. When the opportunity presents itself, you should always be pressing customers to get ourselves in a longer engagement. But last, there are a lot of small contracts out there. I think our average award over the last 5 or 7 years went from $20 million to $160 million. I talked about we had 11 $1 billion-plus awards over the last 36 months or so. But the small ones are where a lot of our technology starts with. Our software-defined tech, whether it's a handheld, whether it's a rack-mounted started with small dollar value awards. But the promise is, if we do this and we walk this customer through this model that we'll eventually get to $100 billion awards, $300 million awards. So that was the path for Spectral. We started off a very small study. We went to the next stage. We built some hardware with a lot of software on it. We proved we could do that. So there's a lot of small seed programs are out there bidding, but it has to be in the markets that we serve. We're not going to go off the reservation and go bid the work that we're not really, really good at. The last thing I'll say is we're very focused on not subcontracting what is core to us. So we say we're near world-class in the technology, why would we ever subcontract that work to someone else? The answer is we won't. We're going to subcontract to how we're going to fill gaps the same way we do acquisitions. So all of that together allows us to drive top line growth, 7% CAGR, drives free cash flow per share. Our 3-year plan that we just put out is looking at a little north of $1.6 billion of free cash flow. So that in some police departments, it is called a clue, right? We're really focused on free cash flow. A lot of our executive metrics have been changed to be purely based on free cash flow. So what we generate, not only how we collect invoices, but the kind of work that we go after because we really believe that's the ultimate measure of a publicly traded company. It's what we're going to stay focused on.

Great. Covered a lot of ground. Appreciate it. Thank you.

Thanks so much, Gavin. Appreciate it. Thanks, everybody.