Good morning, good afternoon, everyone. Thank you again for joining us. My name is Luv Sodha. I'm part of the software equity research team here at Jefferies, and we're very lucky to have Alan Stewart here, who is the CFO of SoundThinking. Alan serves as the CFO right now. Prior to that, he was the Managing Director at RA Capital Advisors, a private investment bank. Prior to that, he was the CFO and then Chief Development Officer of Epsilon Systems Solutions. Great to have you, Alan.
Thank you.
We'll let you take it away.
Thank you very much. I appreciate it, thank you for being here, everyone that's participating and listening online as well. I'll try to make it through the slides as fast as I can, leave some time open for some Q&A. just as you know, I am gonna be going through some forward-looking statements, so please take a look at all our SEC documents, including our 10-Q and our 10-K as well, so you can figure out more information for things that we also provide here. for those of you who know SoundThinking, you may realize that about a month ago, we changed our name from ShotSpotter. Many of you may know ShotSpotter, but not so much SoundThinking.
The reason we did that was because we shifted from only one software product to four software products over the last five years since we went public. We are a leading integrated SaaS, you know, platform for precision policing and data solutions. We have proprietary widely adopted gunshot detection technology. That is still the majority of our revenues. It's the. We've been doing that for over two decades at this point, but we have a very loyal customer base and a minimum churn down sell, which makes us even more important for those customers. In fact, we'll talk a little bit about attrition later, but for those kind of loyal customers that we have, we've had less than 1% attrition for the last three years in a row.
99% + are staying with us, and that's 'cause they realize that the solutions we provide are working and important for them. There's a large opportunity that we're still facing. We're only in about 10% of cities in the United States, so that's domestically. We're also expanding internationally, and we have multiple growth levers, not just within the gunshot detection solution, but the other software products that we have as well. We've had consistent growth with strong and improving margins. When we went public in 2017, we said we'd grow at a CAGR of about 30% for five years. If you take a look at what we did, we actually grew almost exactly 30%.
We try to tell everybody what we're doing, how we're doing, and why we're doing it, and why we think we can continue to achieve those goals. We've seasoned operators, a track record of exceptional performance. We tend to hire the right people and keep them for a long time. Just a couple numbers real quick. In terms of the TAM, we've got over $2.5 billion of our TAM. That's between the four different solutions that we have. Last year's revenue was $81 million. We've already given guidance this year for between $92 million and $94 million, which mid-range would present 15% growth as well. In terms of what we expect the CAGR to go for our revenue, it's about 23% between 2021 and 2025.
Our adjusted EBITDA margin last year was 20%. We've already given guidance as well for 2023 of our adjusted EBITDA to continue to go up and increase to between 24% and 26% this year. You'll see a slide that I have later that that's gonna continue to grow as the company does as well. Our net revenue retention rate, 124%, significantly higher than 100%. Gross revenue retention, also 99%, and we spend very, very little money in terms of compared to other SaaS or software companies, only $0.40 per $1 for each new annual contract value, $1 of annual contract value. Most SaaS companies you might see are spending $0.80, $1 or $1.50 even, to get that same kind of solution that we get.
We're a big Net Promoter Score provider. We have a 56 in our Net Promoter Score, which is world-class. In the gunshot detection, which is the biggest one of our solutions, we have over 1,000 mi already covered with that. We have some other products that we'll talk a little bit about as we go. One of them is a data solution, sort of like a Google, a database for cops, and that has over 1.3 billion data records in that, which helps them figure out a lot of information. I'll cover that product a little bit as we go on. From an IP perspective, we have 35 patents and continue to add those as we go.
If you think about the timeline, and I'll go through this fairly quickly here, we started, and the company was actually founded in 96. We went public. I'll go through a couple of these fairly quickly. 2017, we went public. You can see that there was three acquisitions that we did after that, about every two years. We did one in 2018, 2020, and 2022. It is something that as we continue to grow our core business, which is the gunshot detection, still named ShotSpotter, we're adding the other solutions as well. It doesn't mean that we're necessarily going to add more or wait two years for other M&A activities, but it is something that we do, but we do it very carefully.
What we don't do is acquire a company just to staple on revenues, so the revenue growth is there. But you can see in the future, land and expand, we'll continue to do that. International is something that slowed down during COVID, but is now starting to expand again. We're adding new products, and we may do tuck-in acquisitions as well. In terms of the TAM, when we went public, it's the 1.5 pre-acquisitions, number one, gunshot detection you see at the bottom. We've added the patrol management, data analytics, and investigative tools. Those are the other three software products. If you think about where we're heading or where we're going, it's not just staying where we are.
There's things in number three that might be additional law enforcement software solutions, but might be things outside of law enforcement, like court management. Could be something with digital evidence management, other things that continue to expand our TAM. That is something that we hope to do and will continue to do as well. One of the most important things, and the reasons we changed the name, was because instead of just doing gunshot detection with a ShotSpotter, we've actually introduced, called the SafetySmart Platform. This has all four of the software solutions, and it basically brings together all of those and allows us to still help law enforcement and civic leadership to better protect their communities, ensure the right resources are provided when and where they're needed most.
If you think about those, you can still see the ShotSpotter still on the left. We're gonna go over the CrimeTracer, we'll go over ResourceRouter, and CaseBuilder as well. Those names, all part of the SafetySmart Platform that we have. If you think about it right now, and you can see down on the very bottom, you can see the left in the red box, the ShotSpotter and the ResourceRouter is really just for cities and local law enforcement. The CaseBuilder and CrimeTracer is also for those, but can also be used for federal, state, and local law enforcement as well. You can also see the actual ARR down at the bottom as well, in terms of where we are now versus where things are going.
You can see that in 2022, the ShotSpotter was still 76% of the ARR, which gonna continue to grow, but as we get into 2025, we're estimating that to go down, be closer to 70%, as the other ones are continuing to grow. Just to go over what those are, ShotSpotter is the gunshot detection solution. The Resource Router has had a couple names over the last couple years. The last one was Connect. It's a community-first patrol management solution, and I have another slide that shows you exactly sort of what that looks like. The CaseBuilder is one that I'm most excited about. This is the one that we have developed after the acquisition that we did of LEEDS in 2020. They built a crime or a case management solution, an on-premises-based version of this for NYPD.
NYPD is the largest, most capable law enforcement agency in the world. Basically, we believe they spent almost a decade building that solution, those on- premises. One of the reasons we acquired that company is we wanted to take that on-premises case management solution and put it in the cloud, into a cloud- based, which we did. That's now CaseBuilder. Just, I don't have a separate slide on this, but I'll tell you, we just started selling that last year, and although some of the sales and the products were related to that are somewhere between, you know, $50,000-$200,000, we're getting ready to execute and sign a contract with CaseBuilder that is almost $16 million over five years. That's several million dollars each year, as you can think about that.
That is also to a prison. It's a Department of Corrections customer, which is not related directly to a police agency, but also adds things to our TAM and is incredibly important as we think about where we're going next. We might be looking at other prisons to be able to sell that to as well. This is something that is not only a fantastic software solution for us, but it's also a big TAM expander, and it is possible that may even have other verticals for managing and solving cases as we move forward into the future. Lastly, the CrimeTracer is related to what I kind of call the Google for cops. It's basically a law enforcement platform. It's a leading search engine for law enforcement.
This is basically what we acquired last year when we acquired a company called Forensic Logic in early 2022. We're already seeing that expand from 2022 into 2023, it's something that we're also excited about. If we just go through some of the solutions itself, those of you who know what ShotSpotter is, you've seen this slide before, probably for the last couple years, actually, but this is what it looks like. We actually put sensors on top of buildings, on light posts, about 25 sensors per square mile, and each one of the sensors listens for a loud pop, boom, or bang to figure out if it's a gunshot detection. Okay, if it is an actual gunshot that goes off, the time difference of arrival between the different sensors allows us to mathematically figure out where that was shot.
We also send that through an Incident Review Center, which is people that listen to the algorithms and the actual audio that comes through to determine if it's actually a gunshot or if it might be a firecracker or a car backfire or things like that. We have a very low false rate, like in 1% - 2%. We're very efficient in terms of what we do. We send that to the police. They get that within 45 seconds to one minute from the actual time that gun is fired, so it allows the police to know where to go and save someone's life. They might find someone that got shot, they still might find someone that was shooting there, or they might have people that they're able to interview, such as witnesses. That's basically the ShotSpotter.
This is kind of what it looks like in terms of the actual police officers in their cars or in the maybe a 911 dispatch center. You can see some of the things, shows where the shots are, shows how many they are. If there's additional information, like it's a full auto that shot, or it's a multiple weapons or multiple shooters, we put that information as well, so the police know what they're going into, and it's also it's a safer use of solution for them, so they know what they're going to be approaching as they go in. The ResourceRouter, this is just something where it's not just tied to gunshot detection. This is something where, as you may understand, police are having, after they went through the defund the police, a lot of police retired.
They're having a hard time getting police agencies to grow, so they have less resources. We're helping them route those resources in a way that's more effective using AI algorithms, historical data, what happened in the past. This particular one that you see right on the, on the left, is for a robbery. This is where we would tell the police, based on where it might be prior events, it might be, where there's bars that are going up, it might be schools, it might be due to weather. There might be all kinds of things that tell why we would tell police to go in this particular area in one of those boxes, and then do one of the things that we say down there, like visit the businesses, park, walk, and talk to people, or do a high-visibility car patrol.
We help them send the police resources they have to place to help them become more effective. The CaseBuilder, which is the case management solution, is the one that I talked about that was developed originally as an on-prem version in NYPD, and is now going to outside of just NYPD to other cities, other police agencies, also the Department of Corrections, these are the prisons. This is the solution that we have there. That software, also, as I mentioned, is the thing I'm most excited about in the future. The last one is the CrimeTracer. The CrimeTracer is related to the used to be called COPLINK X. This is something that is where we got when we acquired the Forensic Logic acquisition last year in 2022.
Basically, if you think about it, you can see on the left, this would be... It's a little hard to see here, but this would be, like, two people. You could see me and my CEO Ralph Clark. For me, they know who I'm married to, where I live, where my kids are, all the information that's provided. It might be, what kind of tattoos do I wear? Who are my friends? There's all kinds of information that's in a database that you have to be Criminal Justice Information Services certified, CJIS certified, in order to get access to that information. Police detectives, they're allowed to get that information. People like me, I can't even get access to it, but the information is there to help them solve crimes. There might be two people that are involved in terms of a particular crime.
They manage to you know, take a look at the data that's there and figure out exactly how to solve that actual crime solution. I have to move a little faster here. Some of our competitive advantages, I think we've covered most of these already. In terms of our go-to-market strategy, you can remember that I talked about in terms of the land and expand, and we have things that are for gunshot detections, we call them Tier 1 through Tier 5. Tier 1 is the biggest city, like Chicago, New York, even Detroit, Puerto Rico. We've got Tier 2, Tier 3. These are cities like Albuquerque, Birmingham, Fresno. We also sell to very, very small cities, which we literally just started last year in Tier 4 and Tier 5.
We hadn't sold to those until last year because we didn't think people were gonna need 1-2 sq mi of coverage. Well, actually, we hired a salesperson last year, gave him a quota of 10 mi he had to cover. He ended up selling 27. What do you think we did? We went from one salesperson for Tier 4 and Tier 5 to now we're gonna have four people, Tier 4 and Tier 5, the TAM is expanding there as well. A lot of potential there. This is what we typically see when we start with the city. For example, you could pick any one of these, and let's take a look at the far right with Albuquerque. Albuquerque started with a very small footprint of only about $400,000 in terms of ARR.
They're now over $1.7 million. A city will start with one area, realize it works, and expand to more miles of coverage. This is for the gunshot detection. In terms of our growth strategy, you can see that instead of just having the ShotSpotter, we've got the other, three solutions as well. You've got the ResourceRouter, CaseBuilder, and CrimeTracer, and what we are now able to do is bundle actual sales of multiple products to one particular customer or different customers. If you think about domestic police, they could buy all four of those. If you think about Departments of Corrections, that might only be CaseBuilder and CrimeTracer, 'cause they're not gonna need to do gunshot detection.
What we're able to do is continue to expand the cross-selling opportunities related to the four different products, and we're already seeing that happen quite nicely. We'll talk more about this in the future in terms of our earnings calls, but I can say that just without giving more details, we've already had, just within the last couple of months, a couple of customers who have gotten, two, three, and I think one of them actually is looking at acquiring all four of the solutions. This is something that we're continuing to do, and we believe will not only help us expand our revenue, but expand what we're able to do with the existing and future customers. In terms of where we're going, you know, new customer acquisition, always important. We're doing that. Cross-selling, as I just described.
Square mile expansion, that's with the gunshot detection. Product innovation, these are things that aren't necessarily tied to what people even know we're doing. Each of the products that we have, we continue to develop and make the next version of that product based on information we get from the customers or information that our own technology team knows needs to be added. This is something that we continue to do internally and then give that to the customers, and they're like, "Oh, that's exactly what I needed." We have what we call a customer success team, which four years ago, it was like two people.
We now have, like, 13 people. The customer and success team are the ones that make sure the products that we provide for them are used effectively, and if they use best practices, they get better results. We learn from them in terms of things that they say, "Well, this works well here, but maybe if you could do this, that would help us even more." That helps them, it helps us sell additional things as well. We're constantly doing product innovation. We will expand internationally, primarily starting with the gunshot detection, but also some of the other ones. Strategic M&A is possible, of course, keeping our Net Promoter Score as high as it can, keeping the best in class for customer retention.
In terms of our financials, I've already covered a lot of this already, but if you think about it, like, just over 90% of our revenues is SaaS-based or subscription revenue. When you think about it in terms of, a lot of people say, "Well, you're using sensors. You're not really a software company. It's managed services." We own the sensors. The customer oftentimes doesn't even know where they are. We provide them the service, so it's a solution that we provide. There's still about 10% of our revenue that is tied to professional services, either ramping up a new product that they will acquire or additional things that they ask us to do, but still less than 10% of our total revenues. Last year, revenue growth was 39%.
I think I've already covered all the other ones real quick, we're running out of time here. I think this is a very cool slide because it shows you where we were in 2020 to where we expect we're gonna be from an ARR basis by product in 2025. You see in the bottom, the ShotSpotter gunshot detection solution will continue to grow. That's gonna continue to add $7 million, $8 million, maybe $9 million a year in new ARR with the new miles that we are selling and hopefully keeping with the low attrition rate that we've had and for the past, we expect to have that continue. You also see the other solutions that we have starting to add additional ARR as the business grows.
I think this is probably one of the most important slides that I'll, and I'll probably stop and then leave it open for some Q&A. We are not only growing revenue. You know, this year, guidance is mid-quarter is about 15%. We're expecting that to stay 15%-20% for the next five years, so we're gonna continue to grow top-line revenue. As we grow top-line revenue, though, we expect we're gonna become more profitable all the way to the bottom, down to the bottom line and adjust ed EBITDA. If you think about what we did in 2022, we had 20% adjusted EBITDA, which is pretty good for a company our size. We already gave guidance in terms of 24%-26% for 2023.
We know that over the next couple of years, depending on how a couple things would change, we can be at 40% or more in terms of adjusted EBITDA. as the revenue continues to grow, there's a couple things that are happening. The gross margins are improving because we've already, you know, put or hired the people that are helping us build those solutions. As revenue grows, we don't need to continue to add significant costs in terms of the COGS. The other things, in terms of the operating expense, sales and marketing, R&D, and G&A. Sales and marketing, we expanded significantly over the last three years. Went from about 15-20 people to over 50 people.
We've already added the cost in there in terms of it's not gonna need to continue to grow at the same rate that the revenue's growing. R&D, we will continue to add a little bit as revenue grows, but again, at a rate lower than the revenue growth is gonna be. G&A is also gonna grow, but at a rate lower than the revenue's gonna do. As you see revenue growing, gross margins going up, OPEX staying relatively low and lower in terms of the growth rate versus the actual revenue, all of that adds to our adjusted EBITDA growing. We know we can get there, whether it's gonna be 2025 or the end of 2025 or whatever, we know we can get to 40% or 40%+. I have two minutes left. I think that's my last slide.
I did save a couple of minutes for Q&A. If there's any questions, I'll be glad to answer them.
Cool. Thank you. Any questions from anyone out here? Maybe I'll ask one. You know, one of the big topics that a lot of companies are dealing with is artificial intelligence. Could you talk a little bit about how artificial intelligence is integrated within your portfolio of products?
Sure. The question is, how is artificial intelligence integrated into our portfolio of products? I would say it's two of them that are mainly using that right now. The first one would be the gunshot detection. The actual algorithms that are built are using a lot of information from the audio that we've collected over the last 20 years. You have audio that you know is gunshot, you have audio that you know isn't. You can run those through AI algorithms that help you determine a better solution for actually what you're hearing. That's the first one. The second one is the actual ResourceRouter.
That's the one where the solutions that have what happened in the past for particular crimes, so what might happen in the future if you tie those crimes, they happened because it was rainy or wasn't rainy, it was sunny, or it happened because the bars just shut down, or the bars didn't shut down, you know, where you're at. It's something that we use all of these to help add that together to figure out where the solution might send the resources to help them be more effective.
Got it. Go ahead. Maybe I'll ask one more on, you know, the funding environment for government agencies and how that impacts potentially your revenue streams. Any color on that?
Sure. The question is, in terms of the funding environment, how is that affecting our products and our sales? Right now, the funding is the best that we've ever seen it before, at least in the last couple of years. It's not been a challenge for our customers. As you think about what's happened for the solution that we provide, especially the gun violence, it's gone up, right? A lot of the customers or potential customers trying to figure out how to solve that, and they're using the funds that they have available, either from the federal or the state or even some other grants that they have. We can also help them tie some of those in.
Funding environment is strong, probably as strong as it's ever been for us.
Got it. We'll leave it at that. Thank you so much.
Thank you very much.