Here again for day two of the Ideas Conference. The next company we have for you is Red Violet. They're a leading solutions provider in the ID verification sector. With us today, we've got the Vice President of Investor Relations, Camilo Ramirez. Camilo?
Thank you, Steven. Appreciate it. I'll give you guys a bit of a background about what we do, management background, and then we'll go over our use cases, exactly what we do. Let's make it conversational, so feel free to ask any questions. Stop me if you need clarification on anything, but we'll go from there. I'll quickly flip through the deck. I'll highlight a couple of pages, but preferences for conversation. What do we do? We do all things identity. High level, we aggregate disparate databases, call it liens, judgments, credit header data, anything and everything on every adult individual in the U.S. We aggregate that data and then sell it back out to different industries. We serve five different verticals: collections, financial and corporate risk, investigative, real estate, and what we call emerging markets.
I'll dive deeper into that in a little bit, but let me take a step back and give you a little history about management. In the late 1990s, the management team built a business called Sizen in the same space, data fusion, identity verification. As you can imagine, technology then was very different. We built it in data rooms. There were ad hoc connections. You see Camilo and Steven always present them in an investigation as opposed to a more dynamic presentation where the Camilo result can be applicable only to the situation and don't present Steven. Ultimately, that was sold off to Reed Elsevier's LexisNexis for about $750 million. Non-competes expired. Team got back together. They started a company called TLO. It was being personally funded by one of the founders. He ended up passing away abruptly through the development phase.
Revenue run rate at that time was about $25 million. They were losing anywhere from $1 million- $2 million a month. Ultimately, that was sold off to TransUnion prematurely, right? They were still going through the R&D phase. Non-competes expired. Team got back together, call it in 2014. I'll show you guys a little roadmap of the history. They got back together in 2014. As you can imagine, technology has changed drastically from the late 1990s- 2014. We had AWS, so this iteration is cloud native. It was built in the cloud from the ground up. That's a big differentiator from the competition, the Sizen product and the TLO product that management built. You have LLM models, AI, so it's able to learn upon itself as you feed that data. It's making those connections, building new data points to go out and clear transactions. What do we do?
I'll give you a couple of use cases. We'll start with collections. It's the easiest one. Let's say you buy debt from Capital One. You have a million records. You want to go out and you buy a million records. You want to go out and validate that information. I'm getting kind of an echo in the back. You buy those million records. You go out. You want to validate the information, right party contact information. Have they filed for a bankruptcy? We'll pend that information, send it right back. We go to real estate. We serve real estate in two different manners. On our IDI side, it's going to be more of a marketing play. I want a list of individuals that are 16 and above that live in a two-story home, propensity to sell.
Under our brand Forewarn, which is basically just a skin and application that was layered over our core platform, that's sold directly towards the real estate community. We go after associations. That is going to be, let's say, you're a real estate agent. You get a phone call. I want to see this $3 million home. I'm only in town till tomorrow. My name's John Smith, purposely using a very generic name. I'm going to show up in a Mercedes S-Class. What it does when that phone call comes in, it does a reverse phone number search and it comes up, brings up their background. You don't see John Smith. You see a different name. You don't see that Mercedes S-Class. What you do see is this individual just got out of prison for sexual assault. You're not going to show that home.
What we noticed was that there was a lot of crime committed against realtors. This is more of a safety product for the realtors. We go after the real estate associations. Today we serve about 500 plus real estate associations. There's, call it about 1,100 associations in the U.S. The associations, what they do, they buy this product for their members. It's a member benefit for the users. If you're part of, let's call it down by us, Palm Beach Real Estate Group, the state of Florida purchased Forewarn for all the associations underneath it. They'll just sign up. It's free use, unlimited use, and that's a SaaS product. If you're an individual coming on board, not under an association, that's going to run you $20- $25 a month. Because we sell directly to the associations, economies of scale, those prices are reduced drastically.
On the Florida, Florida is one of the largest associations in the U.S. That's going to be a couple of dollars per member. They have about 400,000 members. It just depends on the scale of the association. Next, we have financial and corporate risk. Under financial corporate risk, we have background screening. I'll give you a couple of use cases there. Let's say you're a Walmart. You have an applicant. They're applying for a job. They enter four addresses. I'll disclose this name because we disclosed it publicly early on. They'll go to a company called Innovative. Innovative is the background screener. Innovative will call out to us, is this information accurate and complete? We'll say it's accurate, but it's incomplete. They left off the fifth address, and that's usually where the criminal record lies.
Innovative will go pull that criminal record at that fifth address and ultimately send that information back to Walmart so they can make their hiring decision. A use case on Innovative. We won that business early on from TransUnion. They're doing low five figures a month. Ultimately, they grew up into well into the six figures. They were sold to Aperis, another identity player. Ultimately, Aperis was purchased by Equifax. As you can imagine, Equifax has all the data in the world, right? They have all that information, but they don't have an aggregation platform. They have the individual data points. At that time, during that acquisition, the Innovative contract was up for renewal. Equifax reached out, hey, can we get an extension on this contract, three-month extension? We said, of course, we understand. We understand you're trying to execute on synergies.
We don't think you're going to have the same throughput or lift on data, but we're here to help if you need anything. They came back, asked for a couple more extensions. Ultimately, they signed a multi-year agreement. They couldn't recreate the data quality or the throughput on their side, even though they have all the individual data points. We have a really good relationship there, multi-year contract. We can go into investigative vertical. That's going to be, for example, law enforcement. You get pulled over by law enforcement. Very easy. Our system or one of our legacy products that we sold off. They'll run a plate, do an investigation. How we differentiate in that space is we have a very neat, mobile application where, let me just flip to another side. A neat application. Let's say you had a witness. I saw this red F-150 in this intersection.
It was a hit and run, something of that nature, crime committed. The investigator can come in, basically do a geofence, drop a pin within that circumference of that area and say, give me all e state, public sector. Public sector, we're really excited about the public sector space. We made a hire, call it just over a year now, Jonathan McDonald. He led the public sector at TransUnion, built it up from zero to where it is today at TransUnion. We built up a team around him as well, call it around 15 individuals- 20 individuals. Public sector, as you can imagine, your typical three-letter agencies doing investigations on individuals. You have ICE doing investigations as well. Those are all opportunities we can win.
Even outside of that, where there's niche use cases, for example, we just won one of the largest toll authorities in the U.S., without naming specifically which one. Down by us in South Florida, we have what they call the SunPass program for the Florida Turnpike. They can do bill by plate. Let's say you're running through the toll, you don't have this SunPass to automatically pay for it. It's going to run, it's going to take a picture of the plate. Then they have to do a search based off that plate to understand who's the owner of this vehicle, registration, search, and then they have to figure out the address, send out a bill to that vehicle owner. If that owner doesn't pay that bill, it goes through a collection process and they have to go through right party contact information.
We won this contract, just call it maybe a quarter ago. We disclosed it on our last earnings call. That's going to be well into the seven figures once that's up and running. That's one of 50, call it. Some other niche use cases are like homestead exemptions. Are these individuals truly living in that home that they're claiming a homestead exemption? Doing an address verification, homeowner verification, and so forth, or even something as niche as, is this child truly living in this public school boundary, right? As you can imagine, there's a lot of situations where parents say, hey, my child's living with the grandparents so they can go to this A-rated school in a public school system, but in reality, they're not.
We have school districts reaching out for that type of data, address verifications essentially to validate that those children truly live in that area and they are eligible to attend that school. Those are some of those niche use cases in the public sector that we're really excited about. Those five verticals, we say revenue is pretty evenly distributed, call it 20% across the board. I'll pause there to see if you guys have any questions. If not, then I'll jump into the financial model. We like to say our data is fixed cost in nature. We go out, do long-term agreements with the data providers. As you can imagine, some of those are going to be credit bureaus, niche data aggregators that are getting data from different municipalities, aggregating that information, we'll consume it. We do long-term contracts on that, and it's fixed cost for unlimited use.
Every additional dollar we're bringing in today is nearly 100% contribution margin. Our last quarter on the gross profit line, we reported just over 80% gross profit. We're starting to approach 40% adjusted EBITDA margins. If you take a look at our financials, our disclosures in the 10K, you'll see that one data provider accounts for 40% of our data costs. That is not just one data asset, its multiple data assets. We have a really good relationship with that credit bureau. Ultimately, we just renewed that contract for another five years here a couple of months ago with minimal incremental increase. The renewal prior to that was flat. We always get the question, hey, if you see these, if the credit bureaus are seeing the type of margins you guys are having, why wouldn't they just escalate the price on you? One, we're multi-sourced on every data point.
If that does occur, we'll just say, hey, we're not going to accept that, and then we'll just bring our tier two data provider up to tier one and then continue to move on. We can give those data points back. If those data points go back, we don't lose the learnings, those connections between brother, son, family connections, those learnings. Those are all proprietary to us. Those learnings stay with us and data points go back, and we just bring up our tier two. The last couple of renewals, we have a really good relationship, and we renewed pretty much flat from that in our last renewal. No concern there.
Are you dependent upon your apps, and do you have some special technology that's not able to do so? Sorry.
No, no, no. No patents, right? We don't have any patents. It's the technology that we have that aggregates that information and links those data points together and continues to learn from that information. A good example that I like to give is, in certain industries, they have what they call the waterfall effect. You'll go into an industry and try to gain a customer and they're going to, we will go in and say, hey, let us come in at tier three on your waterfall. What that means is they're sending a million records to, let's say, TransUnion. TransUnion is going to have a certain hit rate, call it 60%- 70%. Some of that's going to fall out. They don't have information. They'll send it to a second provider, Reed Elsevier LexisNexis Accurint product. They'll be able to hit some amount on that fallout as well.
We'll say, hey, let us come in third. We know we have high confidence in our data assets and we'll get a high hit rate on that. We'll come in even though it's been scrubbed twice already and we'll get a high confidence hit rate. That's how we'll start moving up the waterfall as well. If you have a high hit rate, the customers want you to be first in waterfall because you get economies of scale. As you go down the waterfall, it gets more expensive for them. As you can imagine, they're pushing less, less data through you. It's what we do with that data and how we aggregate that data, as opposed to just hustling or just reselling data points, right? Because a bankruptcy is just a bankruptcy. It doesn't tell you anything. As you're transacting in the commerce today, you're leaving a footprint every single day.
Banks for like know your customer or loan decisions can't rely on data from yesterday or the day before. They have to have the most recent data. They want to understand, did you just file for a bankruptcy? Did something occur? Was there a lien or judgment placed on your property or anything of that nature? You need the most current information.
What about the international market?
Today, we're not in the international space. We have a lot of go-get in the U.S. Today we have about 9,500 customers. The Accurint product has about 400,000 customers, and then the TLO product has anywhere from 60,000 to 70,000 customers. We're just scratching the surface. We're powering seven of the top 10 identity players, to the likes of, call it, Prove, Jumio, Econo, now owned by Mastercard. They all have their own niche way of doing identity verification, how to clear a transaction. Some of them are going to be mobile authentication. Let's say you're logging into Bank of America, Wells Fargo, using facial recognition. What's actually happening behind the scenes is saying, hey, this mobile ID belongs to Camilo Ramirez as a match to the bank side. Here's the PII information. Yes, grant access. If not, ask security questions of that nature.
Or they're going to be doing document verification, take a selfie, load a picture of your ID, and they're going to do the identity verification on that physical document. It's still pulling PII information behind the scenes. Those providers have the technology, the front-facing solution, but they don't have the data in-house. Every time they're clearing a transaction, they have to call out to someone like us or one of our competitors. I think there was a question over here.
Can you talk about the switching costs for your system?
It's a good question.
Is it difficult if someone's there to know?
No, so it's use case dependent, right? If someone's coming on board using our online platform, there's no use cost, no switching costs. They're just going through our credentialing process. We're validating that they have a valid use case, and then they can start using our system. That's how we win business too, right? Because they're not coming in, dropping a team, building out a whole platform. On the API side, our API is very flexible. We like to say it's very customizable, and behind the scenes, it's just turning on and turning off levers depending on what they need. Usually they're going to have teams on their side and they're up and running between 24 and 48 hours.
A good example there on switching costs and ease of use on the API side: a number of years ago, we had a customer, they were doing well into the six figures a month. They were an identity verification, mobile, they do mobile authentication. What they're doing is just the example I gave. They were raising capital. TransUnion ended up leading that capital raise. The TransUnion CEO gave them a call and said, hey, for us to close this transaction, you're going to have to stop using Red Violet's IDI data and move all your consumption to us. The customer gave us a call, told us what was going on. We said, we understand, we would do the same if we needed the capital. We're here if you need anything. We saw that revenue drop off pretty much immediately. Within a couple of months, we started to pick back up.
What came out of it was that once they switched to the TransUnion product, they basically couldn't handle the throughput and the latency on the platform, right? Because they're building a data room as opposed to the cloud. During peak productivity hours, their customers were calling them and complaining, hey, I sent this call through. I'm not, I'm getting false positives or I'm not even getting a response. Ultimately, that business came back to us pretty shortly after. The CEO gave us a call. They said, hey, TransUnion said they can build us a custom API. Just give us a month or two. They said that you guys built us a custom API. We reminded them, hey, you guys were up and running within 48 hours. There's nothing custom about it. To this day, years later, they're continuing to grow with us, even though they have that relationship with TransUnion.
Would it be fair to say that you've got to stay ahead of the technology?
That's a fair statement. Yeah. We're continuing to build that moat around our platform. You have TransUnion, you have Reed Elsevier LexisNexis. They both committed, I believe TransUnion committed about $200 million- $250 million a number of years ago to get their platform into the cloud. We know they're not going to get that platform. We like to give the example, if you take a 757, you want to make it much more efficient, fuel efficient, you're not going to just take parts off, put parts on. You're going to start from the ground up. Once you start removing a couple of pieces from here and here, there's going to be downstream impacts that they're not going to be able to correct within the cloud. They have to start from the bottom up.
We're continuing to build that moat, one through the technology we're using, the throughput, the latency, and then also just from product features as well. Something as simple as, when you're doing a search on the individual, understanding is this relative, is this relative of that individual, is it its mother, father, sibling, cousin, or anything of that nature? We get that granular based off of certain algorithms we have. We're able to tell you what type of relative that is as opposed to the competition. They're just going to tell you it's an associate. They don't have that level of granularity. I think I was going through the financial model. Fixed cost model, nearly every additional dollar is 100% contribution margin. Once you get below the gross profit margin, we like to contract everything long-term contracts to any of our vendors.
Most of our variability is going to be headcount, and that's going to be related to sales and commissions. We verticalize all our teams into subject matter experts. We have a real estate team, law enforcement team, and so forth. There's a couple of verticals we're really excited about for the coming years. One, I alluded to public sector. As you can imagine, that's a very large TAM. With everything that occurred earlier in the year with DOGE, that's a tailwind for us. It worked out perfectly. We weren't deeply penetrated into the public sector market, so we didn't lose contracts. Historically, you have to wait until those RFPs come up. A lot of those contracts were cut, and they're coming back to market for RFP. They cut too much.
Now we have the opportunity to bid on those RFPs that potentially we wouldn't have had that opportunity for a number of years. It's no longer just go with incumbent. Everyone wants to show that they're executing on synergies, reducing costs, and so forth, and having better technology. We're having really good success there on the public sector team submitting those RFPs. These are a much longer sales cycle, as you can imagine. Some of those are going to be year-plus sales cycles, and then there's ramp-up period as well because it follows the budget. On the federal side, budget renewals are around September. On the SLED, state, local, and education, that's usually around the July timeframe for the state and local budget renewals. The RFPs usually follow that as well. Those contracts can be material, as you can imagine.
We've seen contracts, they're going to be the whale contracts, $10 million, $15 million contracts with Reed Elsevier LexisNexis. Even if we get a portion of that, that's going to be material to our revenue. Secondly, we're excited about the background screening space. We won one of the large payroll companies out there for their background screening process. They're in the process of onboarding. When Innovatives was purchased by Aperis and ultimately Equifax, we had a buy versus build decision. Do we continue to just service this industry behind the scenes, or do we go front-facing? We made the effort to build out the products. We already had the data, but how do we productize this and go to market with it? We started productizing it, and this year, we're now ready to go to market in a formal fashion.
We had a lot of beta testing per se, where we got a lot of feedback from the different background screening players. It'd be really nice to do this if you can do X, Y, and Z. Now we're at the place where the product is fully functioning, should service, I call it, 99% of the use cases out there. Proof in the pudding, we won one of the largest payroll processors. We don't, we typically don't disclose names on the IDI side. We'll disclose names on the Forewarn side. If you look at our press releases, they're mostly all Forewarn related. That's by design because there's going to be real estate agents, hey, our sister association offers this product to their members as a member benefit. Why don't you? It's more of a marketing perspective on the Forewarn side.
On the IDI side, as you can imagine, most of our customers don't want individuals knowing how they're doing their identity verification or where they're getting their data from. They're very hesitant on allowing us to press release their name. We also don't want the competition knowing where we're winning. From just a couple KPIs, actually on this slide, from a gross retention number, typically we say these businesses run from 90%- 95%. Last quarter, we came in at 97% on the gross retention side. We continue to guide towards that 90% to 95%. We don't publicly disclose net revenue retention, but as you can imagine, we mimic those information solutions companies. Mid-teens, high mid-teens to upper, to 120%, from a net revenue retention perspective.
What I understand, the revenue, the retention is just a decline. It's not that revenue.
Correct. It's just gross revenue retention. Yeah, we're not accounting for any upsells or anything of that nature in this gross revenue retention.
It is very interesting.
Correct. Any other questions?
Above the flags, could you go, would that be a bus to work or a car? [guess]
Yeah. There are two clear. There is Clear at the airport, and then Thomson Reuters has a legacy product called Clear. That legacy product is mainly in the public sector space. It is a very clunky, dated product, but they have a huge presence in the public sector. We are starting to bump up against them. Yes, Clear like at the airport. They are running identity verification. They are using facial recognition. They are also scanning your ID and doing PII information behind that, validating that it is a valid ID, and then they are pulling up your picture as well with your facial recognition. That Clear would be a potential customer of us or one of our competitors because they need that PII information. It is a good question.
How does artificial intelligence, whether a threat?
Yep. A very valid question. I'll answer it a couple of different ways. We get that all the time, right? Because AI is out there. You can aggregate this, the data. ChatGPT is layered over the internet. Bad data in is bad data out, right? As you can imagine, banks can't clear transactions with ChatGPT. They can say, hey, tell me about John Smith. Who is John Smith? Is this the right individual? How are you leveraging AI today in a couple of different fashions? We have a high confidence data cohort. How do we leverage AI on top of it so we can go out and interact, customers can interact with our platform differently? Today, when you go into our online platform, you're entering a name, date of birth, social, something of that nature, and getting information back. What we're working on today is being more interactive.
Tell me everything about Camilo Ramirez or something as simple, which sounds very simple, but it would require a number of searches. Is there a family member of Camilo Ramirez that has a violent criminal history, whatever you want to call it? Today what you'd have to do, you'd have to search Camilo Ramirez, go into each known associate, look at their criminal background, and that's going to take you time because depending on how many relatives, you're going to have to go through each individual relative. You can keep going down the family tree, as opposed to when you're interacting with the platform. You ask it that question and it'll either say yes, this family member, and it'll give you the family member name, has a criminal history, and it'll give you all that information. It's doing all those searches in subseconds. That's how we want to interact.
Even on the, let's say, let's say you're a CVS. You want to understand the upward and downward mobility of the population within this intersection. We have all those data points of that population. Basically, their analyst is going to say, hey, typical crime rate in this corner is X, Y, and Z. Here's the upward mobility of individuals and here's the house prices that continue to go up. They're going to make a decision, do we build a CVS on this corner or not? It's aggregating all that data and giving you insight into a population as opposed to just individuals as well. From an operational standpoint, how we're leveraging AI, today, let's say for every, call it, thousand customers we onboard, we're going to have to add one or two additional credentialing individuals, right?
How do we automate our internal processes so we can continue to expand on that margin? As we continue to grow, we don't have to add those one or two individuals for every thousand customers. It can be one individual for every 5,000 customers or something of that nature. How do we automate our processes internally so we can continue to extract margin in the future? From a margin perspective, just to note, at maturities, these businesses on the gross profit line, they're generating 90% plus gross profit margins. On the adjusted EBITDA, call it around 60% adjusted EBITDA, just because of that fixed cost nature of the data asset.
That's how artificial intelligence helps you, though.
Yeah, that's a good question, right? There are barriers to entry. As you can imagine, let's say you want to go out and start a business in this nature. You have the credit bureaus. They have the largest balance sheet, and as you can imagine, they have all the data in-house. When they wanted to get into the market, it was a buy versus build decision. Each iteration, they went out and bought the businesses just because of the amount of information that you have to gather and the type of information that's actually beneficial. A good example, early on, we didn't have business data. If me saying that, you're like, why wouldn't they have business data, right? It's very integral to understand who owns businesses, but it's a nice to have. It's a very expensive data set, and there's not much lift. It's low ROI on that business data.
Early on, we didn't have business data. We'd go out and when we would go out to prospect customers, we'd say, hey, they would love our product. They would say, hey, but you don't have the business data. When you really dug in, they were doing a couple of searches a month on the business data. Ultimately, once our P&L supported purchasing that business data, we layered in that business data. Today we're starting to work on KYB, so Know Your Business. If we can understand everything about a consumer and individual as their sole data point, we can know everything about a business as well. LLCs, who are the true owners of these LLCs going down multiple businesses. Being able to come out in the future with a KYB product as well. There's the moat and then the know-how of what data to acquire.
The credit bureaus have to have high confidence in you keeping that data safe, right? As you can imagine, they're not just going to give the whole U.S. population to anyone out on the street. There's a lot of risk associated with that. We're audited by the credit bureaus on a regular basis. We're PCI Level 1, SOC 1, SOC 2 certified, ISO 2700 as well. Safety is at the forefront of our data safety. I believe you had a question.
How many sort of separate real estate deals do you have? How many not renew or in the end contract period? The second point question is, would Uber be a particular customer possible?
Yeah, that's a potential use case, right? I'll answer that one and go back to your other one. Uber, they're doing background screening on drivers as well. They have that high transactions. They're doing identity verification, make sure that's a valid individual, that valid license and so forth. That's a potential use case. On the real estate side, let me flip to the side of it. Today we service over 575 real estate associations. Again, there's about 1,100 associations in the U.S. From a real estate, from a realtor standpoint, we're servicing just about 350,000 real estate individuals. There's about 1.2 million real estate individuals in the U.S. Of those that are truly in the real estate business, call it about 900,000.
Real estate individuals are actually transacting more than one home a year, they just those that have a real estate license just for family and friends and so forth. There's no true competing product in that space. For lack of a better term, most of the competing products are going to be like find the body, so panic buttons once you're attacked and so forth. This is more of a proactive safety solution. Any last minute questions?
Sure.
No, I appreciate it, guys, for making it interactive. Thank you.