Morning, everyone, and thank you for joining the H.C. Wainwright 26th Annual Global Investment Conference. My name is Alejandro Borjas, and I'm an analyst on the investment banking team. We're very excited you all could join us for a productive day of one-on-one meetings, corporate presentations, and panels. For this session, we are thrilled to welcome the iLearning Engines team.
Thank you, Alejandro, and all of you. A real pleasure to be here to talk about iLearning. You know, before I get started with the presentation, you know, I wanted to just say that, you know, over the last few days, we've had a short seller who put out some information about us, and in response to that, we have, I, on behalf of the company, have made a statement and a press release that people can go and read, that speaks to some of the most, you know, some of the biggest misrepresentations that have been laid out in that report, so separately, we have, you know, a special committee that is looking into these things, everything else, so we can get past this.
I'll just pivot now and very excited to talk about iLearning, a company that we have built, very proud to have built over the last decade plus. We are an out-of-the-box AI platform that's really an enterprise-class AI platform that is out of the box and low-code, and I'll get more into it later. That helps companies build learning and work automation use cases inside the enterprise. Every enterprise today is looking for how they can bring AI into the enterprise in a safe and secure fashion, and also to understand what AI really means for them. You know, our journey started a long time ago. You know, we started out iLearningEngines with an intention to help companies make better use of their institutional knowledge and data, and then drive better outcomes.
So if you're a health system, you have a lot of information, but a lot of times patients don't get the data they need in real time. Or if you're an oil and gas company, you have a lot of information on safety and the things that need to be done, but they don't get into the front lines. And so really, that was really the mission of what we did. And then you know, we were very, very much one of the early players in the AI landscape back in the day. You know, we started our product development in 2016. And literally from that point on, we you know, we had three theses on the AI market that we feel have really validated really well.
So one, you know, we felt that it was very important for us to pay for training data for our AI models. You know, what we have today is 50+ AI models for each of the verticals that support all the verticals we are in. So back in the day, in 2018, we recognized that data, and we started licensing and paying for data, and today that's kind of become an industry trend. You know, you see data licensing agreements that have come out from a lot of other AI companies based on the fallout they were receiving from various content makers. Two, we felt that for enterprise AI to really sustain, it has to be on-prem or on a private cloud.
You know, companies in an AI world are, they value their data deeply, and they want to be very protective about it, and so that was the key. So we set out to build our AI models really to support on-prem, AI, and it can make that unit economics work so people aren't spending hundreds of millions of dollars just running these models. And three or three, what we call verticalization at scale, which is, for AI to be really meaningful, you know, you had to be able to have high-impact use cases within a vertical. So fundamentally, what we really do is, for every organization, our...
We are able to take our AI models that come prebuilt with various use cases, sit on top of an enterprise's data and really help them build learning and work automation use cases, overall. Our platform has key, very key features, you know, very specialized datasets, configurable integrations, a low-code approach to building AI HyperApps, and many other capabilities. You know, we are a company at scale. You know, in 2023, we closed our revenue at $420 million, strong net dollar retention. Historically, our contracts are one- to three-year contracts. You know, a wide range of customers, end customers. You know, we work a lot extensively with value-added resellers, and I'll talk more about our go-to-market strategy and others.
You know, we have over 4 million end users, like I said, in 12 verticals. And you know, like I said, we have been, you know, pretty much an AI company, at scale. Like I said, from the beginning, you know, we were selling AI in a world before it became, you know, as famous as it did, today, it's caught everybody's mind. And so really, the only way for us to sell an AI solution was a use case-first approach.
To go to a company and say: "We're not selling you an AI platform, but we're going to sell you a solution that you understand, a learning solution that's powered by AI, or a specific work function solution that is powered by AI, or a workflow solution that's powered by AI." That's how it started, and I think that's been one of our keys to successes. In a world where everybody today, where everyone a model first, use case second type approach. We've been a use case first, product second approach. You know, we play at the intersection of three large markets: the artificial intelligence market that's been growing at a rapid rate, and also the intersection of the learning and the corporate hyper automation markets. We think there's been a tremendous opportunity that we have been able to tap into.
You know, where the AI landscape today is much like where the software industry SaaS market was a few years ago. And I think as this market matures, we hope to be one of those groundbreaking companies in this space. I want to just talk about where we stacked up in the world of enterprise AI. You know, in the world of enterprise AI, we are really a low-code AI platform for learning and work automation. So what this really means is companies are able to deploy a platform and build use cases on top of it in a matter of weeks to months at a disruptive price point, maybe in the hundreds of thousands, because you don't need expensive AI programmers to do this. Our low-code canvas makes that happen, you know.
And in the landscape, what we almost always have seen is custom solutions. You know, custom solutions built on top of a hyperscaler stack, a large language model, a data infrastructure, but having very expensive AI engineers. And so, you know, you're talking about the price points, millions plus, and often, you know, these use cases taking at least a year plus. And so we feel like, we have a very differentiated solution in this market. You know, one of the keys that we have with our low-code and no-code canvas is we are able to make enterprise-wide integrations. So we are able to use this platform to not only automate, you know, simple tasks and processes, but also complex workflows and other areas.
Our models, you know, we have over 50+ enterprise models that have really been designed to make the unit economics for enterprise AI, on-prem AI work, and so these models, when they get deployed with these HyperApps and those use cases, they train on the enterprise's data itself to get better and better over time. Our AI platform has several capabilities, you know, in terms of being able to take content from within the enterprise, being able to convert and generate various learning artifacts out of that. It can automate other functions, and we have a whole slew of other capabilities that we do.
The key thing for us is really, like I said, this no-code, low-code workflow and AI canvas that really allows us to take insights and data from within the enterprise and have them drive through the various workflows of the organization. I want to talk a little bit about our verticalization scale strategy, and I think this has been one of our keys to success. Everything about us is out of the box, which means that our platform gets deployed into an enterprise in a matter of 8-12 weeks on average, and so we're not stuck with long implementation cycles. The other thing is, when we go into any vertical, we take our vertical specific models, and out of that, build pre-built use cases.
So when you think about a vertical, like, say, insurance, everything is out of the box and comes with some pre-built use cases. For example, claims intake, claims processing, smart risk management, early notification of loss, et cetera. Or if you're in healthcare, it could be workforce training, disease management, out-of-the-box use cases. We call these use cases, AI apps or AI engines, but that's really the key. So for each vertical out of the box, we have these use cases. So when you go into a customer who's trying to figure out what AI means for them, they already see that, hey, you know, so I think I can use AI for my, you know, processing, manage, reducing my, risk, you know, processing my, claims, you know, reducing the likelihood of fraud.
Or if it's in healthcare, like I want to manage, expand these capabilities, like disease management, or deliver better workforce training. Or if you're in education, which is another big area for us, is really being able to deliver AI-powered tutoring to students, and we have various target markets. And so this has been a critical case for us. And so this has been one of our keys to success, and I think everything we've built, you know, we have a best-in-class AI platform with over, like I said, 50-plus models. And so these models, again, are designed to not consume a lot of GPU power. It can actually make the unit economics of these things work really, really well.
Like I said, we are in, you know, we continue our growth by, you know, expanding deeper into the verticals we are in and also add new verticals. I want to touch upon our go-to-market approach, which I think is critical to understand. We are an AI platform, and we work heavily with value-added resellers. You can think of the value-added resellers as the appliance makers of AI. You know, I think there's been a lot of talk about how AI sort of is like, electricity, you know, where it was back in the day.
And so you can just imagine a world where, you know, we go, you know, people going into organizations saying, "Hey, would you like to buy some electricity from us?" And people are like: "What do I do with it?" "Oh, well, you can build your own refrigerator, and you can build your own heating systems. Or you could go work with the refrigeration systems makers and the heating system makers, the appliance makers." And the analogy is true here, too. And so one of the things that we saw early, before everybody else, is let's talk and work closely with these value-added resellers who are like the appliance makers of AI. Can we go with them and partner with them, help them build these solutions, and then they can take it to their end customers. So we work very closely with value-added resellers.
We have today over 30 value-added resellers. So these value-added resellers are building use cases on our platform. They are our AI apps, and, you know, we're basically in the process of building out our marketplace. So through these channel partners, we have been able to reach and deploy to over 1,000 end enterprises. So from our standpoint, these value-added resellers either could be the traditional system integrators, they could be domain-specific folks, but they bring a lot. From the outside, they bring a lot of expertise for a particular vertical that allows us to build and tailor our models to those verticals. And then, you know, we work closely with them to sell the solutions to their end customer. So as a result, you know, we do have channel concentration on business.
Today, our top four value-added resellers account for about 52% of our revenue. You know, our top verticals are, you know, education, enterprise automation, workforce, you know, upskilling, and then healthcare insurance that we have together. One of our key targets, why education and healthcare in particular are important, is we target for-profit schools or private schools internationally. You know, there are, you know, over 1 million private schools supporting 250 million students, according to various sources.
But we have, you know, an incredible opportunity pipeline in our pipeline to really work with these schools to create a whole generation of what we call AI schools. So, and, you know, some of the offerings that we are able to deliver to them are things like tutoring offerings built on our platform. That really is a revenue generator for many of these schools. The same thing true for healthcare, but like I said, you know, enterprise automation, healthcare, education, insurance, these are all our top verticals and expanding. Like I said, we have a very robust enterprise customer base, you know, of organizations. I think, like I said, we are embedded because we are a platform on which these solutions are built.
So, you know, for a lot of end customers, the product is either white label, and sometimes they don't know what's the underlying AI platform that is driving it. But like I said, you know, the platform and its unique capabilities are something that we think is a big competitive advantage for us. Today, our revenue is split between the U.S., India, Middle East, and other... North America, India, Middle East, and other areas. And for us, we think that is going to continue to be the case. We are a global company. You know, we have a great team here. Couldn't have done anything without them. And, you know, almost all of them have been with us for a very long time.
I think we are working on many exciting initiatives. The team's very exciting, and a key to retention for us has always been, you know, people who are working there feel very fulfilled. You know, just to summarize some key financial highlights, you know, 97% of our revenue is subscription-based. We're coming on the back of very strong organic growth. We did make an acquisition that was a very successful one for us. We aim to make many more acquisitions. You know, part of our reason to go public was, I think, the ability for us to be able to use the public currency plus a certain amount of cash to make acquisitions.
You know, of course, we didn't anticipate some of the other challenges that come with being a public company, but, you know, we're learning, like with everything else, we'll get past that. And lastly, you know, we've been very capital efficient. I think we have been able to get... You know, when we signed our business combination with for our de-SPAC, which we did in 2023, we had gotten there with less than $3 million in equity. And so one of the key reasons why we were able to really do that is, from the beginning, we were able to work with a technology partner who was able to help us with our product development.
You know, for most companies, during the peak of a product development, you're gonna have, you know, a hundred, two hundred engineers, but once the product is developed, in reality, you probably need less than 10% of that workforce. So it's very hard for a company to hire all of them in-house and then let them all go if they want to survive. And so we kind of saw that from the beginning. You know, I learned from my own past lessons where I've been part of startups where we raised a lot of money and built a lot of product, and when, you know, we had to wait, we're just carrying a heavy fixed cost, and as a startup, one could never, you know, survive that, that process.
So we work at scale very, you know, so finding the right technology partner, working with them, scaling it, and then over time, as you build your revenue and scaling your product adoption, start to get that migrate the process from an outsource model to bringing more things in-house. And that's our goal, you know, from our standpoint, as we continue our path forward. But at this goal, like I said, we are very much, you know, all around product innovation. You know, my background is startup in Silicon Valley, you know, really high-performance microprocessors, building systems that we considered cutting-edge, you know, five, 10 years into the future. You know, I, when I worked on some of the, you know, most successful products in the history of Sun Microsystems.
And so really, you know, we're very excited about the future, you know. And so for us, it's, you know, have strong product, strong customers, strong go-to-market strategy, and, you know, build strong governance on top of it. I think that's kind of what we have, aspired and tried to do always. You know, our top key metrics, like I said, our gross margins today centered around close to 70%. Our sales and marketing costs around, you know, close to 30%, and we spend about close to 30% on R&D. So we'll continue to spend R&D at a high rate, you know, possibly that number over time. In the long term, we think can come down to 25%-27%. And so, really, for us, I think...
You know, one other point I just want to touch on is we have been, you know, EBITDA positive and profitable back from 2020 onwards. So we've been able to raise a lot of, you know, debt and venture debt and funded by that, you know. And so really, on the back of our quality of our customers and performance, we've been able to, to do that. Does have good operating leverage, room for improvement over time. You know, gross margin is going up a little bit, from you know, going from 70% to settle down around 75% in the long term. You know, maybe a few point, you know... Our sales and marketing, we think, are pretty efficient with our strategy.
You know, most of the sales and marketing expenses are really enabling our value-added reseller network, which we think, as we get better, can reduce. Then lastly, on R&D, right? And so, on R&D, we'll continue to have that spend. And like I said, you know, we've from the beginning, well before time, back in 2018, we saw the need for why we had to pay for data. So, you know, we got three pretty, very important things right, I think, in AI back in the day. Like I said, paying for your training data was critical. Understanding that on-prem was going to be the way for AI. You know, there's been a lot of reports about AI being, you know, enterprise, 83% of enterprises really sticking to the private cloud.
So that was critical for building those things. And then really tapping on the verticalization of scale and building, working with value-added resellers. We think those are the key things to success. We've gotten those things right, and, you know, we hope that investors and the public and everybody, you know, will give us a chance to continue to, you know, to bring good value to them in the future. So with that, I'm just gonna pause and open it up for questions.
Quick question, is that all organic growth, the 48% compound, or did you acquire some companies?
So we acquired one company that was, like, not very material. It was like, you know, close to $1 million in revenue, but brought a great deal of expertise to it, but it's all organic growth.
Do you expect to do any acquisitions going forward?
We do want to make a lot of acquisitions. This is because from our standpoint, it's a one of the things. The key drivers for us for acquisition is, you know, our net dollar retention is 130%. So this means that, and, you know, many of the company targets we are acquiring are either where the net dollar retention is either 100% or 90% sometimes. So this ability to upsell where we can bolt on our platform on top of them, kind of really turbocharge them, and then really upsell through the various use cases there. So it's a great customer acquisition vehicle.
Just to keep that, it's been pretty flat for four years. Why is that? I know you talked about increasing your R&D budgets, but where do you see that going once you've normalized margins?
So, like I said, in the long, the longer term, you know, today our growth points are around 70%. We think there's room to get to 75%. And then on the R&D, we are at about 32%, so I think there's probably room for five points to go there. Sales and marketing, you know, for us, I think we, while our sales spending can be optimized, I think we've been traditionally very low on G&A, so we're trying to build that corporate infrastructure. So I think that's an area that will take us longer. So that's the way we looked at it. So for us, and, you know, it's really been a function for us now that we went through a pretty de-SPAC process that you know has that is inherently expensive.
But, you know, the reason we chose that approach was we were backed by a PE company, so we could really bring that private equity, discipline, into the company.
So you think 20% is kind of a normalized point?
Long term, I would say, yeah, closer to somewhere in that range, I would say. Yeah.
Can you talk about some specific end-use cases at the corporate level, maybe like the customer and what they're using it for, maybe one or two examples?
Sure. So for us, I can take many use cases around that. So like I said, learning automation is a very important, you know, was an original and a very important use case for us. So you know, from our perspective, we you know we are really this use case is really around solving mission-critical outcomes. And so you know this is a company that you know had some pretty significant safety issues. And so really the challenge for them was how do they make sure that you know they don't have big safety incidents etc.? So they deployed our platform and really built a learning automation use case which is we helped them build a Knowledge Cloud with all their institutional content.
And our platform was able to take the content from various manuals, like safety manuals, anchor manuals, etc., extract it into various learning artifacts. On the other side, we are extracting from all the different systems the process gaps and competency gaps. And then we were, over time, able to deliver this content to areas where the gaps were. You know, our AI is kind of, you can think of it, the AI is not standalone, it is really supervised AI, meaning that humans are in the process. So the AI is really helping them make the whole process efficient. So in this case, you know, while the companies had accidents, there were a lot of early indicators. They used to have hundreds of thousands of, close to a thousand near misses a year.
So systematically, they were able to bring that down to the low teens, zero time loss to incidents. So that's, like, one example. You know, like I said, in insurance, you know, examples around claims intake, claims processing, so you know, basically, the ability for the platform to, you know. You're basically creating an AI engine that is almost like a human, that basically goes to all the different inputs. It could scan websites, emails, wherever claims are being reported, scan them, and right there, identify if there are any missteps in the claims itself. You know, if there are any indications of potentially, you know, fraudulent claims, or if there are some OSHA-related, just process it.
It can send it to, you know, meet the OSHA guideline reporting, or on the other side, it can quickly send it to processing, whatever it is. That would be another example. Telematics is another thing, you know. So we've been working closely with the hyperscaler, and we built a telematics HyperApp for insurers. So one of the big challenges with telematics is, you know, that there's an old adage: be careful what you wish for, you'll just get it, right? So a lot of insurance companies have been collecting telematics data from all sorts of vehicles every minute, and all of a sudden, they don't know what to do with it, and they also don't have the ability to send it to other outside the premises.
So we were able to take that, generate the insights that they wanted, and really kind of increase the ROI on their data. There are many other examples we can talk. I think I'm running out of time here, but I would love to-
Yeah. Thank you so much, folks. Thank you for the presentation. We are going to have to wrap up here. Thank you for attending, and for those listening in online, thanks to you as well. Have a great rest of your day, everyone.