May know us by our former name, ShotSpotter. We changed that last year, approximately in March, for several reasons. But thank you all for being here. Appreciate it. It's always nice to have a large crowd, especially during breakfast. So really do appreciate that. Thank you all. Just a couple of things. My thing isn't working now. Oh. Technical challenges here, folks. We'll get right past this. There we go. So as you know, I will be giving information, some forward-looking statements, talking about not just the history, but some of the future of things we're doing. So just realize that it's best for you to always look at our SEC filings to find out more information as well.
So let me talk to you a little bit about what SoundThinking is, and then one of the reasons that we actually changed the name is when we went public in 2017, we were a one-product, one-software company, and that was the actual ShotSpotter. The ShotSpotter was a gunshot detection. Still is the largest amount of revenue for us, very important for us, continues to grow nicely year after year. Since then, we've done four acquisitions, actually, almost about once every year and a half to two years, and we've added additional software products. We are still about 95%, a SaaS provider or a recurring revenue provider. We do have a little bit of professional services each year that adds to that, that gets to our total revenue.
But in terms of the actual ones that we have, we now have what we're calling the SafetySmart Platform, and the SafetySmart Platform itself, it brings together all of these software solutions to help law enforcement, communities, and private entities protect their communities by ensuring they have the right resources at the time that they need them. I'm gonna cover each one of these a little bit more in details as we go through the presentation, but primarily, what we are trying to do in the main product, the ShotSpotter, and also several of the other ones, is tackle the challenges that law enforcement is having today, and they are significant. At this point, you may know that after the pandemic COVID, a lot of the police agencies have had people retire.
They've been having a hard time getting new people involved to hire new police officers, and that's been becoming incredibly more challenging for them. In fact, if you think about it, there's a public safety gap right now. There's an increased need for law enforcement, mainly because of the rising crime rates that's happening across most of the cities. That's in the first area. You can see that in from 2015 to 2021, the number of deaths went up by over 50%. It's a significant issue. At the same time, the staff shortages that I'm talking about are making it even worse. They got about 18% more in terms of resignation, retirement, 45%. Think about that.
If you had 45% of your people that actually retired, that knew more about what the job required, and now they decide because of the environment that they, they're required to operate in, they're just gonna retire. That's what police agencies are seeing. So there's a lot of shortages falling in most of the cities. Some of them are as high as, like, between 800 officers down to almost 2,000 officers down in some of the larger cities. So what we do is trying to help them do things by bringing in technology that allows them to be more effective with less numbers of people. The one thing that is actually increasing, though, I think, is there is still increasing funding in terms of some of the stuff. At one point, there was a defund the police.
There were cities that literally said, "We're gonna cut off the police budget." Ultimately, that is now passed. In fact, in some cases, the police budgets have actually gone up a little bit, but they don't have the people to hire, so what do they do with that extra money? They're using that to spend it on technology that can help them perform. So in terms of our actual SafetySmart Platform, there are five separate products, and I, I talked a little bit about them on the first slide, but just to go across what those are, for those of you who aren't very familiar with what SoundThinking does, the ShotSpotter is the gunshot detection solution. That's what we were founded on about over 20 years ago, continued to grow. That's still about 65% of our revenues.
The other 35% of our revenues right now, though, are in the other products. When we went public, it was 100%. So you can see, as we expanded, it made more sense for us to talk about what we do from a s- an entire platform of software products than just the gunshot detection. And I'm gonna go down the ARR a little bit in a second. The second product, though, is the Resource Router. This is one of the first acquisitions we did. This is really for community first patrol management. It basically helps them, them being a police agency, determine now that they have less officers, where should they send them?
If they use a lot of information that we provide them, which is, it's AI algorithms that not only uses information from historical crimes that have occurred in their areas, but it's also information that we provide, like, what's the weather doing outside? Where are they in terms of proximity to schools? Where are they in proximity to bars? Did bars just close? All of these things are algorithms we build to send them where they should go in a certain area, and we say, "You might wanna show up in this 5-yard block area and just drive your car around, or you might want to get out and walk around." These are things that we do, in terms of the Resource Router, and we found that very very important.
Still a smaller amount of our revenue, but important from a actual thing that the police need. CaseBuilder is one of our newer ones. We did. This started out as an acquisition that we did in 2020. The primary customer that they had at the time was New York NYPD, basically here where we're at. This is an on-premises-based version used in New York City that we took and built into a cloud-based version. So we took all the modules that were in the on-prem, built them into a cloud-based version, and it's now called CaseBuilder. It's for helping them manage cases. As you know, any police agency, and not just police agencies, but other agencies and other verticals, need to solve things that they're dealing with. That's what we do. We help them manage a case to help solve that.
CrimeTracer is an acquisition from 2022. This is basically, it's what we would call, you know, I've got to be a little careful here, but called Google for cops. Okay? So it's not Google, but it is for cops because you have to have what's called a CJIS certification. It's Criminal Justice Information Services certification to be able to enter into the system and access the data. For example, I don't even have CJIS certification, so I can't get in and look at the data. But if I were an FBI agent, a police agent, or an officer, I can go in there, and I can find out all kinds of information. It's over a billion different sources of data, which helps them connect the dots. I'll spend a little bit more time on that coming up as well.
And lastly, the SafePointe is an acquisition that we just did in August of 2023. This is basically a system to help weapons detection. So if you think about ShotSpotter, an alert goes off, a gun has already been fired. What SafePointe does is has sensors that people walk through as they're going into an office building, a casino, a hotel, or it could be a hospital. And this is where they find out, "Oh, I'm carrying," as I'm walking through the different sensors, "a 9-millimeter gun." That alert gets sent to the security area inside the team within about 10 seconds, maybe 15 seconds at latest, with a picture of me. If I'm wearing a black hoodie, they know that, "Okay, fine. Look for someone just walked in wearing a black hoodie," and then they can get there very, very quickly.
This is hopefully before they shoot the gun, okay? Not just gun, though. It also does, you know, knives. It does other things as well in terms of some kind of explosives as well. Very important. This is the biggest area for us because if you think about it, and I will show another slide on this as well, this is a very, very large TAM. In terms of the first four total is about $2.5 billion in TAM. This alone is about $20 billion-$25 billion itself. So adding that as an acquisition for us was something was significant, not just for adding another software product, but something that we can expand the company. So in terms of expanding the company, you can see down here underneath the red is the actual ARR from last year.
We have not yet announced the total results for 2023 yet, but the ARR, as these continue to grow, you can see that this started with zero 'cause we just acquired. It was really nothing last year, but by the time 2025 comes around, we expect that to be as high as 15% of our ARR because it's a significantly large growing area. We expect CrimeTracer to only be a little smaller because this is getting bigger. The CaseBuilder is one that we just won a very large contract, which we did announce, so publicly, everybody understands, for a Department of Corrections, basically near here in New York City. It's about an $18 million contract, covered over six years. So something that's probably around $3 million a year.
That is based on the cloud-based solution that we took the on-premises space, made it into the cloud, and sold that. That's with CaseBuilder. So that's why the ARR is going up significantly there. ResourceRouter continues to grow, becomes more important and effective. And just ShotSpotter is only going down because of the increase in some of the other ones. It's gonna continue to go year after year. So if you think about this, I talked a little bit about the two reds at the bottom and then the pink, which is the prior solutions that we had in gunshot detection, patrol management, data analytics, and helping case management. That was the $2.5 billion.
When you add the other ones on top of that, in terms of the, you know, both the law enforcement software and the SafePointe that we have, that's a significant expansion and ultimately total law enforcement, which we're not gonna ever get all of it, but as we go into the five products, we're getting more and more of that. It's a very large opportunity in terms of the TAM. Just by the numbers, just thinking of these, and again, we're gonna be announcing our actual 2023 results in just a little over a month. So you're gonna find out some updates on these, but these are some really important and interesting numbers that we have. The TAM was $2.5 billion, now over $20-$2.5 billion. Revenue from last year was $81 million.
We had already given guidance, so I'm not gonna change anything we haven't said, but guidance somewhere between $92 million and $94 million for 2023. That will be, you know, significantly higher again into 2024. In terms of the actual growth rate from 2021 through 2025, we believe that we can have an average of about 23% in terms of the continued revenue growth. That's because of all the products that we now have that continue to grow. Some of them are growing a higher percentage of that, but it's from a smaller baseline. But some of them continue to grow very nicely. Even every one of them is growing in the double digits. In terms of the Adjusted EBITDA margin, 2022 is about 20%.
Ultimately, we believe, and we say ultimately, it's probably about four years away, that 20% Adjusted EBITDA is gonna be about 40%. Okay, so as we continue to grow the revenue from each of the individual software products that we have, gross margins improve. As gross margins improve, where the operating expenses basically don't have to go up near as much as the revenue, all of that flows down to the bottom line. For those of you who do know about us, followed us for a while, you realize we do generate cash quite a bit. In fact, just one statistic, we spent about $60 million in the last four and a half years. That $60 million was about half doing share repurchases that are positive for our stockholders, and the other half was for acquisitions.
Do I have $60 million in debt? No. I only have $7 million in debt. So we generate cash, and by the way, I could pay that $7 million off just about any time if I desired. Something also very important in terms of the actual customers, once we sell a customer, the net revenue retention rate, which counts on the ones that we had the year before, stayed with us, and in most cases, expanded. That's why it's north of 100%. Last year, it was 124%. Our retention rate, which means that if you think about the opposite of that, it's the attrition. The attrition is around 1% or less.
For those of you who look at other SaaS companies or even other really performing companies, if you've got anything with an attrition rate of less than 5%, they're doing amazing. So for us, we've now had three, and hopefully, it'll be four years, where we've had about 1% attrition. That's because when we put a solution in there, the customers realize how important it is, and they keep it. Basically, our idea would be, hopefully, they keep it forever. The other thing that we do, it's very, very different from most SaaS companies, is in terms of the sales and marketing spend to generate that $1 of ACV; we're less than $0.50. Most SaaS companies are gonna spend $0.80, $1, $1.40 to get that new $1 of revenue.
We only spend about $0.40 'cause we're incredibly effective about what we do and how we hire the teams that we use. We are very, very large promoters of Net Promoter Score, so our NPS, which is over 64, that is world-class, especially when you think about most of our customers. For those who understand the Net Promoter Score, we send information out to people and ask them a bunch of questions. The most important question is: Would you recommend this product for someone else? If you get a nine or a 10, it's a positive. If you get something down in the six or seven, it's neutral. If you get something below that, it's negative. In order to get a score over 60, you have to have almost everybody up in the nines and 10s, and that's what we have. We've got large number of paying customers.
That coverage area is going up significantly. The data records is in the CrimeTracer, and just in terms of the patents, we have a lot of intellectual property as well. I'm gonna talk a little bit more about the coverage area, but, you know, here's the thing. Let me just say it right now. We just announced publicly this week, the beginning of 2023, we said we're probably gonna go live, and this is in our ShotSpotter product, which is miles covered. We normally, l et me put it this way: We added 50 new miles in 2020. We added 100 new miles in 2021. We added 101 new miles in 2022. We just announced that in 2023, we added over 150 miles.
So we started the year saying we were gonna do 120, and then mid-year, we said, "Well, it could be as high as 140." We just announced publicly that we went actually over 150 miles, which is a significant number of miles, go live miles, which helps not only build the ARR, but build the other customers that are out there as our potential new net promoters for us. We added 21 new cities through Q3 alone, and out of our current customers, 18 of them expanded. So that's just through Q3. We haven't even told you what we've done in Q4. All very positive, though. So in terms of the competitive advantage, there's a lot of things. It all starts with the IP in the middle.
Okay, but even if we gave all of our IP out of those 34 patents to someone else and said, "Go ahead and compete with us," what they wouldn't know is the experience curve. They wouldn't know the things about, Okay, if I'm in San Francisco, how do you put those sensors and use them on hills? If you're in New York City, you've got large buildings, how do you use them effectively between the buildings to avoid the shadow zones? Those aren't in the patents. That's in the 20 years of experience curve that is the main reason we have no real competition at all on the ShotSpotter product itself. Trusted relationships is incredibly important for us. We have over a 160 cities where the police agencies are working with us.
It's not only because they're working with the ShotSpotter, they're now saying, "Okay, now I understand why I need the ResourceRouter. Oh, I need the CrimeTracer, too." So we started doing bundling last year, and so now many of our customers not just have one of our products, they have multiple products.
But the relationship that we have with them is even more important because one of the reasons I'm able to have that $0.40 in terms of the sales and marketing cost is because a chief of police here that's been with us for two years, talks to a chief of police over here, doesn't have us and says, "Hey, look, if you wanna deal with gun violence, if you wanna help yourself with lower officers, you need to go with SoundThinking." Okay, so they're better salespeople than our salespeople, actually, and they cost me no commission, and we don't do donations like that either. So it's very clear. Our Net Promoter Score is also very positive 'cause they talk to other people about that. And then, of course, the total addressable market was originally at $2.5 billion.
It's a strong competitive advantage 'cause you're not gonna have Google, or you're not gonna have, you know, Yahoo, or anyone else go after a $2.5 billion market. Okay, now that we've added the SafePointe, so it's larger than 20, we're still, it, it's strong from a competitive advantage, but we there will be some competition in that particular product that we have. So when you think about how we grow in terms of what we've done, starting with new customers, cross-selling other products, which is the bundling I just talked about, expansion, where I said 18 customers expanded last year. The year before, that was about, you know, 15. Innovation, product innovation. As we add new software products, we do bring on some more R&D people to our team.
But even more than that, if you think about ShotSpotter, we're not competing against anyone else. So what happens? We compete with ourself. So we make sure that our team realizes they have to do something better and better and better. And over the last, let's just say the last five years, improvements that we've made in many cases are related to the AI algorithms that we use to test our database, which helps us, it become even more important. One example of that might be, you know, five years ago, we would get a certain number of incidents or actually potential incidents and realize that those are helicopters that sound like automatic weapons. Well, now we know that they're helicopters, not automatic weapons, so the amount of incidents that we have to send out as a potential have gone down.
We also introduced a brand-new sensor, which is the fifth generation sensor, which is more. I wouldn't say active, but it's more accurate in terms of some of the things that we're doing. We've already started deploying the new sensor. We didn't have to do that. The other sensors were working just fine, but we compete against ourself to make sure that it's always gonna be hard for someone else to compete against us. We had international expansion. It went down to almost zero during COVID, but it's now ramped back up. We're back in Cape Town, South Africa. We did an expansion in the Bahamas. We literally just went live in Uruguay. Montevideo, Uruguay, is a brand-new country for us, so that's actually pretty exciting.
We have done strategic M&A acquisition, but I will tell you this, although we have done four acquisitions, we probably looked at 40. So we are very, very, you know, interested in making sure it's the right acquisition that adds to things that we need, can add to the SafetySmart Platform or other things that will help the company in terms of the growth. And then best-in-class customer retention as well, with an attrition rate of 1%, it, there's no—pretty much doesn't get any better than that. This is sort of that within a, and I realize I'm looking at my timer here. I'm gonna have, probably have to speed up a little bit, but we've been around for a while. But you can also see how the revenue has increased over the last years.
When we went public, we went public at a $15 million run rate. That would be incredibly hard to do even now, but it was positive then. We've continued to grow significantly since then through all the things that we've talked about, especially every year, adding a couple new cities, and those cities help us add more cities. So ultimately, when you think about it, we're gonna continue to land and expand, do the international new product, tuck in acquisition to get us to the future. That's what we're already starting to do. I'd like to talk a little bit about the SafetySmart because of the large TAM. The $20 billion-$26 billion is areas that are not the municipalities that we're selling to. It's not necessarily the federal government, and it's not the states.
It's things like schools, it's things like casinos, it's things like hospitals, other financial institutions that we haven't sold before, because what it does is sets the sensors out so that when someone goes between them, they know they need to do something because of all the active shooter events that have happened over the last couple of years. Many, many agencies, not just agencies, but any one of these realize they just have to do something to lower that. And it may be that someone may still get in, but at this point, someone inside, the security officer that is in that particular building, is already going towards that person, knows what they look at, 'cause our solution gives them a picture and tells them what they might be carrying, that they may still get a shot off, but they might not get 20 shots off.
So it's incredibly valuable, and we're finding that this is something that we know needed to be done. We know that it's basically a market that's still growing. There's a company out there called Evolv that's actually doing a quite good job at this. We're, we're cheering them on. We actually like them in terms of what they're doing. They're growing the market and helping other customers, potential customers, know that they need to do something for their particular building or facility. So from the gunshot detection, and if anyone here. Is there anyone here that does not know sort of what we do from the gunshot detection? I'll just go ahead and just give a real quick sense or a summary of this.
So basically, we have sensors that are generally on the top of buildings or light posts, and there's about somewhere between 15 and 25 sensors per square mile , depending on the proximity, the location of buildings, and things like that. When someone shoots a fire, shoots off a gun, between three and eight of those sensor hear that at a different time of arrival, and you use math to calculate exactly where that shot has been fired. So if I did the shot, the police would get that alert within 45 seconds or less. Actually, our SLA is less than a minute, but generally, it's about 45 seconds. They can get here in a couple of minutes, save the person that I shot because they're bleeding out, and if they didn't know about it, that person would die.
Ultimately, if I'm gone, at least interview other people who saw who I was. So it's something that is a game changer for the communities that we serve. It's really important because the communities that we serve here, in most cases, and when I say most, it's about 80% plus, don't call 911 at all. Now, you think about that. If you don't know about 80% of the gunshots that are going off in that community, who's gonna help them? The police aren't gonna show up because they didn't know. No one called them to tell them where it was, and when they do call, they don't know where to say, because the sound of a gunshot echoes, so they don't really know.
They say, "It's in my neighborhood." When you get a ShotSpotter alert, it tells you within about 25 meters, like 80 feet, almost like in the middle of this room, that's where I was shooting. So they know where to go, they get there very quickly, and we save lives. It's something that we're seeing. If you were to follow SoundThinking or, or ShotSpotter on, on, on X or some of the other ones, you'll hear, you know, every week, people are talking about lives that were saved, guns that were collected. When they do that, maybe there were drugs also in the building or something like that. So it's something that helps them out a lot. This is just an example of what it looks like. This is what the police would see in their particular patrol car. This might be on their smartphone.
So you can see how many, this one happened to be, you know, 10 shots were fired in this particular location. That 10 was one of six that were going on in this particular time period, and they get additional information up here on the left that might say, "Vehicle moving 3 miles an hour on the, to the east." It might say, "Multiple shooters." If you get something as multiple shooters, that's incredibly information to keep the officer safe, 'cause they're not gonna send one patrol officer, they're gonna send four. Okay? Or it could be a high-capacity weapon, like a machine gun, things like that. All of that additional information is provided. The ResourceRouter is the patrol management. This is where I talked about using the AI algorithms and other data to figure out where for the, to send the officers.
It's not about gunshot detection, although it can be used for that. It's also about robberies. It could be about other things as well. And if you look at some of the, some of the, you know, quotes, I'll just pick one. I'll pick the second one. "ResourceRouter has helped our community by putting officers in areas that are most needed. We have staffing challenges, like a lot of agencies, and right now, it helps make us more efficient. We know where to put the officers at what time." It's just unbelievable how important that can actually be. Moving on, in terms of the CaseBuilder, this is the one that we just sold to the Department of Corrections, which is worth the $18 million contract. That's a prison, okay? It's the first prison.
So in terms of a TAM expander, it's something that's incredibly important for us because that's our first prison. There's hundreds of prisons, just domestically. There's thousands internationally, but it's something that they have to do to manage the cases that they have. We went to this one because of all the work that we had done with the on-premises version for NYPD. That's why the Department of Corrections here in New York City gave the contract to us. So it's something that's incredibly important. There's a lot of data inside of it. Helps them with different modules. You might have a module that deals with internal affairs. That would be the same thing the police would use, that the prison would use. It might have one that has to do with prisoners.
Well, that's not used by NYPD and theirs, but it's a new module that will be used in the actual prison, things like that, or people that are out on parole. That would be a different module. So there's the software itself has very many different modules that are used based on what the use case of the customer needs to be. CrimeTracer is a search engine. This is the one where it has all kinds of information that I can't get to as a civilian, but someone that has a CJIS certification can. In this particular case, you can. H ard to see over here, but what you might see is, okay, this was a report over here.
This report had this person, this building, that type of gun, and then you connect that person to that person because they say, "Okay, they know that Alan Stewart has a wife named Jennifer. She lives over here. They have two kids that live over here. I'm affiliated with a gang. I have a tattoo on my left arm." These are kind of information that, you know, when you have, it helps you solve the crime. This is something that, we acquired now two years, almost to the day, that, ago, and that's, it grew about 20% last year, and it continued to grow. The SafePointe is the intelligent this is the, in terms of notifying with the weapons coming in.
The best part about this is if you go through these, you don't even know that you're getting tested for, if you have a weapon or not, okay? It's, it's not going to hurt someone if I had, like, a pacemaker. It has nothing to do with that. It does with the magnetic imaging of the actual image that's going through there, as well as a camera that they, that the person, that is there, the security officer, can see. So they're coming up here, they're still 50 feet away from this. The alarm goes off, picture takes a picture of that. The security person who's inside gets to go to the front, look for this person, and then figure out, "Hey, you know, you apparently have a concealed weapon." "Oh, well, I'm an off-duty police officer.
You know, I'm allowed to do it. Here's my stuff." They get information. "Yes, sir. Okay, go ahead and go on." Or they find someone that's carrying an M, an M1 5 or something else like that, and then they can get that. That would be like a red alert. So you can see there's boxes down here. So you've got, like, yellow alerts, information, green alerts are something that, you know, no one's worried about. An orange or a red, though, is something that the security person inside might want to know or find out more about. Okay, this is basically what the solution does.
And after they go through that, within about 10 seconds is when the security person inside, if they're not right at the door, they can come up and very quickly, wherever they are in the building, get there very quickly and still find the person because they have an image as well. So very effective and something that we're incredibly excited about where it's going. This needs to be just about everywhere, based on what we know in terms of the active shooter and the instances that have occurred in the past. So our growth strategy, we normally start with something very small, and then once they get going, they expand. This is just a couple examples of four different cities in terms of the actual initial ARR and where the ARR is now.
Okay, you can see hundreds of % increases, and in fact, Albuquerque started down here, went through here, and now they're talking about doing another expansion even greater than where they are, you know, even at this point. So this is what we do in terms of getting in there, prove that it's effective, and then they expand from there. There's just a couple other case studies. This is, this is actually kind of more talking about the Net Promoter Score. So when we started over here, we were in Louisville, just outside of the state, but the Cincinnati police chief heard about it, got called from this officer, and said, "You need this for gunshot detection." Then he talked to someone in Columbus. They talked to Toledo, Mansfield.
So you expand because one's saying, "It's working for me, you need it." These are just two states that you can see. This happens in many other states as well, but examples of where it started and where it went to from that and how it's continuing to expand. Talked a little bit about the cross-selling opportunities. Four years ago, we didn't have this. Now we do. So if you think about this, each one of the products that we have, you know, for example, the SafePointe won't necessarily be used for domestic police, although it could be. It could be where a police department still wants sensors outside of their departments, where they actually have police, but we're not necessarily focused on that. All the other ones, you can see that in some cases there's X's across the board.
We now have, after 2023, which is the first year we started bundling the sales, several of our customers have two of our products, a couple of them have three, and I think one of them, at least, has all four. That was before we acquired the SafePointe. So this is something we're gonna continue to do because it adds cross-selling opportunities for us. In terms of financials, I already covered a little bit, most of these, so I'm not gonna cover them again. We can. I'm running out of time, so I gotta leave some time for some Q and A. But I will say that the net revenue retention rate continues to be north of 100. Last year, it was 124%. That's because the customers expand. And then lastly, I would say the sales efficiency ratio, again, way lower than most SaaS companies.
But where we are, 40%, it might go up a little bit for the, for 2023, but again, it's way lower than what most SaaS companies have. So, with that, if I were to summarize it, look, we, we are, you know, the leading SaaS platform for precision policing and also data solutions. We have a lot of stuff in terms of the actual gunshot detection technology. The customers that we have stay with us forever. That's why the attrition rate is really low. Massive opportunities, especially for some of our new products, consistent customer growth, and, the, the people that we hire are seasoned operators who help us improve what we do in every one of our products. So with that, I'll go ahead and I think I have a couple more minutes.
I'm turning 60 next week, so my eyesight's gone down a little bit. I can't see how many more minutes I have, but I think it's a couple more. So if there are any kind of questions that you have. Yes, sir.
Yeah, can you just talk about SafePointe and the revenue model there? There's a hardware sale or lease, just a service model, or are you still on the equipment?
Yeah, great. The question was, can I talk about the revenue model for SafePointe? The best thing that we found out when we found out about SafePointe, they actually came to us, said, "We have a very large customer we're going after, and we'd like you to partner with us to do that." Well, ultimately, that led to the acquisition. Then what we found out is they have been following us for five years and built their model directly around ours. So it's a SaaS-managed services. They own the sensors. All they do is, is sell it on a SaaS monthly or annual recurring fee, so it's very similar to exactly what ShotSpotter does. So that'd be the reason we really liked it.
What sort of payback do you get on that CapEx investment when you're building, owning the equipment, so you get back the cash flow?
Yeah, so what is the payback there? So, we have said in the past that in terms of a ShotSpotter, guys, basically costs us less than a year's revenue in order to break even. So even if they canceled after one year, which we've never had anyone do, we would still make money. Same thing with SafePointe. Less than one year of revenue to actually implement and deploy the solutions.
How long are typical contract terms in that business?
It's still a little early for us to determine. Most of them are still about a year. Some of them they've had for five years, so they just renew year after year. But it is something also that we are investing in. I mean, we know how to do this. We've done it for a long time. So we bought a company that was incredibly capable, very strong people in terms of technology, but they also were somewhat bootstrapped, didn't necessarily have the time to spend or the money to spend. We have a little bit more of that and are able to help them do everything that they're doing from the technology, from the customer relations, from the sales and marketing, better than they were doing in the past. Great questions. We got about-