Name: Thriving in the Channel Q2 2026
Uploaded: 2026-04-28T19:04:05.383Z
Duration: 47 min 44 s
Description: Thriving in the Channel Q2 2026

Transcript for "Thriving in the Channel Q2 2026": Hello, Paul. Good afternoon. to Jiffy, how are you today? I'm fantastic. I am fantastic. Thanks for leading us today, my friend. Let's get going here in about thirty be. here. Yep. I would just give it thirty seconds. We will dive into the fun. Sounds like a winner. Good luck we should have background music, you know, like the the the the the the the the totally. should. Actually, should fill in that on everything going on in Charlotte, North Carolina right now. Your Your crazy weather patterns are probably gonna prevent me from getting home. How about that? I'm sorry, dude. We've been having crazy thunderstorms all morning. So you know what? It's just because Charlotte's excited that you're coming here. That's what it is. It's just bursting with joy. I will say the weather is probably better here. I'm in Traverse City, Michigan. If you haven't been, I highly encourage everybody to come visit. It's a great spot. It's a little chilly today, but other than that, it's, it's awesome. Fantastic. Alright. Well, looks like we have thirty seconds in, so I think it's time to begin. For all the folks out there that have joined, thank you so much. Hey, Adam. Thank you for saying hey to us. Welcome to Thriving in the Channel. If you want to know the the method of the madness, we get together the first, the first month of every quarter on the last, the fourth or last Tuesday of the month. And we give you all the good updates for all the things that need you to know going to the future. Let's see here. And there we go. Now okay. Here's my more most important job, the agenda. Get you guys all excited about what's coming. First of all, I'm I have a few, quick high level, piece of information I'm gonna give out to everyone. And then number two, we have our fearless leader, JP. He's gonna go through all the updates, especially about a particular event that we just, came back from. Then the stars of the show are really happening at this point as well. We're gonna go into a Managed AI workshop, a crawl a a crawl, walk, run, or crawl. I did the words together. Crawl, walk, run approach to, just makes me stay on AI. Very excited about that, Included in a demo today. Thank you, Matt and Emily for being a part of that. And then, of course, we have a case study. And I know you guys love case studies. We love to get into an actual challenge we solve, how we solve it, and, of course, the the partner benefit as well. Then of course, we'll come in if you have any questions along the way, we will dive into those. Without further ado, I'll dive right in. Q2's fifth. Definitely, if you've not heard about this, I hope I'm the first to tell you. If not, I hope you've been selling stuff all year and been seeing the benefits of your award all along. But right now we do have for Q2, which also matched Q1 in this case, we do have a 1xMR spiff. Basically, if you go out there and you sell a $10,000 deal, basically the high level is you would get $10,000 in addition to your regular recurring monthly. Of course, as my disclaimer will be based on commissionable services, blah, blah, blah. But high level, it's pretty direct. You sell something to MRR, you're going to get rewarded for that. The next thing, and this is like I'm super excited to tell you about, is the introduction of Amelia Suarez. Amelia Suarez is our new channel marketing manager. I have been very fortunate with to work with her at another organization, including this one. So this is my second time being able to work with Emilia. She is fantastic. She's gonna help all of the folks out there, not only with our messaging, but to be able to basically help be able to work with you to be able to drive more growth together. And I'm gonna introduce Emilia to you now. So Emilia, tell us about yourself. Yeah. Hello. Thank you so much, Paul. I am Amelia Suarez, channel marketing manager, and I am thrilled to be here at Thrive. And I will keep this very short and sweet because we have some amazing content for you today. So most recently, I worked in the MSP space, and previous to that, I have fulfilled any variety of marketing roles within the channel ecosystem. I'm looking forward to working across our entire channel partner community to continue building our shared momentum and making it easy to do business together. So with that, I will hand it back to you, Paul. Yes. I think Amelia is going to be a game changer for all the stuff that we're doing for you guys in our, partner to Ecosystem. And without and just like she said, I'm going to pass it on to JP. JP, it's up to you. Absolutely. Well, thank you, and welcome, Emilia. We are totally excited and stoked to have you here on the team. And, you know, obviously, it was very, important that everybody picked up on one thing that Paul said, and that is spiffs. That gets a lot of attention. Pay pay attention to that pay attention to that. Obviously, we want everybody to sell. We wanna pay on what you sell, so let's go all do great things together. I wanted to just jump on here, though, briefly, and talk through, obviously, the largest by far, it is our largest channel event that we do, year over year. And we are we go to channel partners, which is obviously the large channel show in in Las Vegas. Right? That all 4,000, 5,000 of our closest friends all settle in on the strip. And we obviously collaborate. We obviously meet with clients. As you can see here, we had over 80 plus partner meetings in person, which is such a great feeling that you're not over Teams. You're actually in front of people. You're working through strategic initiatives. You're talking through client experiences, etcetera. We also just had four strategic TSD meetings. And if you know us, like, I think that you do, we obviously leverage a lot of suppliers in our own platform. Right? So we had tons of good intentional meetings with them as well. But I wanted to bring back to everybody a couple core concepts that really just, like, hit home for me as I left there. I I I took a bunch of notes. The concepts, and it goes without saying, everything everybody wanted to talk about was all around the things of AI, and we're gonna come back to that here in a minute. However, we're we spent a lot of time speaking with partners on the collaborative strategies and just synergies around what their cloud posture looks like, what their security posture looks like in the client experience. And with all that comes an element of collaboration. So I just wanna share a couple things with you all. And as you all know from a security standpoint, you know, we at Thryv, we have led with a security mindset day in day out. And one of the reasons and I don't know whether all of you know this, but I'll share this. One of the reasons why I'm passionate and I came here to Thryv was because we could serve as somebody's managed security services provider, but we could also be the managed services provider for a end user. Right? Obviously, being sold by our partners. But at the end of the day, it gives us the ability to triage something both from a network ops and a security operations team perspective. So we spent a lot of time working and talking with our strategic advisors and our partners, talking through what that benefit means to you. And what that definitely means is we're not sending alerts outbound for somebody else to triage, we're triaging those things proactively on your behalf. From a cloud perspective, obviously, coming off the heels of last year, there's a lot of buzz around Broadcom. And now the question is, what do we do with all this cloud infrastructure that I've either procured or maybe I have housed in a cloud, or maybe my public cloud, you know, our costs are too expensive? I want all of us to know that Thryv has a three pronged approach when it comes to our cloud strategy. Number one, we do have the ability to help people through their Broadcom journey. Secondly, from a public cloud initiative, obviously, we have a huge bench of folks here at Thryv that can help you with your journey to mitigate, obviously, cost structures and risk when it comes to public cloud consumption. We have a whole practice around Azure. We'd love to talk to you more about that. Thirdly, if anybody wants to have a hybrid cloud strategy conversation, we wanna help you be resilient in that department. So we have partnered with HPE Morpheus to help us deliver upon that. And then from a collaboration and just total experience, end user experience, one of the things that you will continue to hear and you will probably more than likely hear this on the next webinar, which will be next quarter, is we wanna talk about all things digital employee experience. And what do I mean by that? So we are rolling out decks in an effort to collaborate with the end users to be proactive with all things that they can consume and do from a support perspective and ongoing just interaction perspective. But at the end of the day, what it does is it will allow us to go deeper and be more intentional within each one of the clients' experiences that we're gonna obviously be a part of. And lastly, AI. And from an AI perspective, if I was in 80 meetings, I was in 8,000 meetings where we all talked about artificial intelligence. And as you can imagine, there is so much noise around AI right now. There is a lot of conversations. It is the shiny object that's in the room. There is no doubt about it. However, what we've seen, there is a lot of talk, but there's not a lot of strategy. So what I would like to do is walk you all through what our strategic initiatives are around AI, launching it through our channel community so your clients can embrace it and adopt it. And that is the critical thing here. And by the way, I'm gonna introduce a couple of friends that have just joined me. Number one, Emily Steen. She is our AI solutions developer here at Thryv. And then from a product standpoint, Matt Taber runs our AI product initiative here at Thryv. And so, Matt, Emily, I'm gonna turn it over to you all, and I'd love for you to be able to share what our strategy is. What does that crawl, walk, run initiative really mean, and how can we help our partners collaborate and, obviously, through adoption, help their end users go to market with that? So without further ado, Matt, take it away. Thanks, JP. So happy to be here with everybody today. Gonna go ahead and talk a little bit about how we're doing AI at Thryv, how we're, how we're helping customers get, through this challenge that is not gonna stop. So we need to have a plan and need to move forward. So let's start off with where we are today. So employees inside of organizations are already using AI. Your clients are using AI. Everybody is using AI at this point. The the problem is is from what we've seen through the many conversations that we've had, is a lot of times this AI use is not necessarily safe. It's not strategically delivered, and there's lot usually not a lot of governance that goes along with it. Amongst our client base and and just, prospects that we've talked to through all kinds of different industries, We're finding tons of use of shadow AI. And if you've never heard that term before, that's unsanctioned AI. That's when you're allowed to when somebody's using their own ChatGPT account or using their own Cloud account or whatever it might be. And so the problem with this is is that if you're not familiar with any of these personal accounts, what happens is when you're using a free AI, the free AI, the reason why it's free is because they're using that interaction. The the providers are using that interaction, to to be able to, train on on the train their models on your interaction. It's not just the data that you're putting into it. That's a huge part of it, but it's also your your interactions, how you're how you're working with AI. All of that stuff is going into training models. When you when you buy, professional, subscriptions, even, like, your own personal subscription, there are still loops that you have to jump through in order to make sure that you're not training the model on that information. So that's a huge concern, across all fields. Right? And there's even some paid accounts, that still do some element of user interaction training even though they're not pulling your data in. So you need to be very, very careful as you're starting to roll AI into your environment. And this creates urgency and but there's a better path forward around how what what you can do with this. So what Thryv has launched, and we launched it earlier this year, we have a managed AI workspace solution. What this solution delivers is it's a secure hub that replaces shadow AI. There's 60 plus LLMs. I think we're up to 70 now, Emily. Right? And then, 50 plus connectors that are automatically built into the platform that allows you to connect to business apps and different tools, that basically make AI a lot more useful inside of the business world. We also provide governance over permissions and data, and then we have enterprise security relationships with all of the LLMs so that there's zero data retention, from anything that you put into this, into the solution. So we we in addition to all of those cool things, we've also realized that we can't the biggest problem that we found when we were talking to customers early on when trying to build the solution was that they were focused entirely on the million dollar problem, but we didn't know how to get them there because they couldn't trans they couldn't talk the talk of AI. They weren't able to translate their business needs, their requirements, their processes in terms of AI, from the very beginning. So it made us very hard for us to go in and solve big problems right out the gate. So what we did is we took a step back and we said, well, you know, it's gonna be on Thryv to help figure out how to train users that are onboarded onto this platform as to how, to even use it at an early level and then quickly increase their skills until we're able to help them move forward. So we created this crawl, walk, run approach to AI. In the crawl section of this, we're mainly focused on widespread adoption. That's lockdown, safe use, governance plans, all the stuff that we would need in order to to make sure that the AI is useful but tightly secured. When we get into this walk, portion of it, the next five to ten weeks with us, we're teaching them how to how it interacts with other data. We're basically bringing in interactions. We're bringing, integrations with other systems. We're bringing in, other tools. We're talking about how to build agents that'll help them do processes better. And then finally, once they're confident with those those initial concepts, we can then move on to the run, which is the fun part. This is where we can automate entire business processes for their user, and that's all on a per case basis. So the first two of those items are delivered under a professional services. Right? So we have a project that comes along with any sale where we'll get the customer through the crawl and walk stage. All along, they have a managed service that's attached to the to the system that we talked about, AI workspace, where we can help, provide, solutions or we can provide support for anything that they might have a problem with while it's going on. And then at the end, when they get to this walk stage through that, we're able to create, projects using people like Emily Steen, who's a fantastic AI developer, to get in there and build out, high value solutions. But you know what? I can talk to you about it, but it's a lot easier if I actually tell you about it, right, in terms of a customer. So let's just look at an an a customer that we've already worked with and how we talk them through this crawl, walk, run solution. So this customer that we're gonna talk about today, they were let's just call them a general use user. They're a PR firm. They work with a lot of high profile financial service customers in the, in that industry. And what we did for them was basically what have them walk through this entire process, until they've gotten to the run phase, and now they're starting to develop, a bit more advanced solutions. But in the early stages of crawl, we basically, taught them how we started with chat use. We showed them how to use the system across the entire solution. We we introduced really cool interesting tools into the mix like Perplexity and Firecrawl and Tableau, which, Emily will show you in a couple of minutes on how that all works. We created prompt guardrails for them, and we we basically taught them how to select the LLM for how they needed to use the the tool. When we got to the walk stage, this is where things got really cool. We created, an agent for them called Lyra. They were having trouble with, like, prompting out things properly. So we built this Lyra tool for them that helped them optimize prompts so they could put in what their thoughts were, and it would spit out the right prompt to be able to push forward whatever they were looking for. They thought that was amazing. We also built some cool apps like a transcript formatter, which would pull, format, which would basically pull transcripts from conversations that we'd have and turn them into scripts. And then they also had us an agent that was an agent in our workflow that were built entirely around research for their podcast that they were doing for with different people from their their their clients. So basically, what this all happened is we shared this entire model. We we started them on this the crawl walk crawl walk run journey by helping them understand the initial platform. We built, on top of that with different agent use, team enablement sessions, all the things that you would need. It was basically over a course of ten weeks. I think we were with them two three to three times a week working on different, training sessions, getting them up to speed on what they needed to do until they were finally at the run stage, where we're now talking to them currently about all of the really cool things that we'll be able to do more moving forward to help them, work work smarter. So just an idea of what we do as a customer. Like, when when the customer is going through these processes, we'll work through and deliver them, you know, what do we call the AI report card. And so this is what the customer looked like. I'm just gonna show you a snapshot of this. This is what the customer looked like seventy days on the platform. They had 16 active users on the platform. They had 2,000 total interactions. They had spent about 2,000, credits with us around the platform. They had deployed three agents, and we were in the process of building four workflows for them. They saw increased value, 95% time savings on this transcript formatting, 91% time savings on podcast research, five use cases are now in the pipeline for the post thirty day run because now they understand how to talk to us and we're able to move forward. So a lot of amazing, I don't know what I'm trying to say. A lot of them a lot of amazing work happened here during this time frame in the first seventy days, and we're ready to move forward here. Just another couple highlights real quick. This is part of the report card that the client sees. It shows their chats, how many chats they ran, the agents they were working with, the workflow runs. It was only a few of them, but it was powerful. We're starting to see how they're working through it. You can see user credits, how they're used, who's using them, how they're working through these solutions. It's very cool. We're providing all the usage back through this. And then soon we'll be adding it to our portal so clients can actually see directly what's going on in real time. And then finally, last but not least here, the business real this is the the money part right here. Right? This is where it all comes down to play. Just with this transcript formatting, which they were doing several times a week, we took a forty five minute process and got it down to about two minutes. Then the podcast research element, that was about sixty minutes of time, give or take, and we got that down to five minutes. As you can see, we're not only helping customers on a platform, but we're also showing value on that too, and we're helping them show their business how much time that they're actually saving through this use of AI. I think that's the challenge that a lot of people are missing. What I'm going to do is stop talking right now and hand this over to Emily Steen, our AI solutions developer, and she's going to show you guys what this looks like in real time. Emily, your show. Thanks, Matt. Okay. I'm gonna stop sharing the slides here, and then I'm gonna share my screen. Okay. This looks good, I think. Okay. So thank you, Matt. Kinda set the the stage well for what I'm gonna show. So I'm gonna show a walk through of our AI workspace platform. I'm gonna show an overview of the platform itself, kind of what you see out of the box and what the configuration items are, and then I'll walk through some of the cool things that we've built on the platform. So what you can see on the screen right now is our AI workspace platform. It should look pretty familiar as in similar to other AI platforms that I'm sure you've seen by this point, like a chat gbt or a Gemini or cloud interface, but there is some additional functionality as well. So the first is that you can select any large language model. We have over 70, as Matt was saying, like, over 70 now available on the platform, and everything you could think of that you would ever need is here. So we have all the anthropic models, OpenAI, Google, Meta, etcetera. When a new model comes out, it's usually available on the platform within twenty four hours. So it's really quick to get all the new models after they're released to the public. You can see we have Opus 4.7. We got it like the day after it was released. I like to point out here that there are two main benefits to having all the large language models as options. The first is that you're not committing to any one company or any one LLM. So this landscape, the AI landscape is constantly changing. The new and the best model is constantly changing for specific use cases and functionalities, and so it's super beneficial to not have to commit. An example being, there was a time, back in the fall when OpenAI had the best model for image generation. Then Gemini came out with their nano banana model being the gold standard. And now I think OpenAI has come back with a new image generation model that's supposed to be very good as well. So there's constant change within the industry, and not committing to one model gives you the flexibility to switch, when a new and better better model comes out. The second benefit, which will make more sense as I get further down the demo, is when we build multi agentic workflows. So workflows that use multiple agents and multiple large language models, you can use each model to its best features and best capabilities at that relevant step within the process. So for example, you can use Gemini as a document processing step because it has a very large context window. You can use Cloud in the middle step for analysis and reasoning, and then you could use, like, OpenAI or anything else that you want as the output step. So you can take advantage of each of the capabilities of each model at each step of the workflow, and you'll have an overall highly optimized end to end workflow taking advantage of all the model options. The second configuration item that we have here are tools. There are two subcategories of tools here. The first are tools that add functionality pass what a large language model can do naturally. I'm going to point out a few and then I'll scroll down the list. Arguably, the most powerful on this whole list is the Python code execution tool. Connecting to this will allow you to do anything that you can do in Python. Creating PDFs, creating word documents, generating code, running code, any complex analysis or reasoning, you can do with this Python tool connection. We have some incredible ones for web research. Those would be perplexity, Tivoli, Exa, and Firecrawl. So Tivoli and Exa are AI optimized search engines. So you could think of, like, similar to Google search, but specifically for large language models, as well as perplexity too. Firecall is a web scraper. So Firecall, if you connect it to a website, it will scrape the entire site and pull all the data from the site. So a really powerful web research workflow involves using perplexity to Verily and Exa to find relevant sources and then using Firecall to extract all the data from the sites. It's a very, very powerful method of doing, web research. I'm gonna scroll down a little bit and then we get to the second category of tools, which are data connection tools. Our most popular one in the platform is the Microsoft three six five connector. So you can connect to Outlook, Teams, SharePoint, etcetera, all using this Microsoft three six five connector. And then I'll scroll down the list. These are all of our native connectors. If you if there's a platform that you would like to connect to that's not on this list, there are two options for that. We offer offer custom MCP connections. So if your platform offers an MCP server, we can connect to that. If not, you'll see on the bottom here, we have n a n, make, and Zapier as connection options. We can build out a connection workflow using those intermediate steps. Those are the baseline configuration items just off the bat. Again, this is the prompt window here, so you can start a chat in that prompt window. So now I'm gonna talk about the types of products that we can build in this platform. The there are three. They're called agents, apps, and workflows. So I'm gonna go through each of these options and give an example of what we've built. The first option here is an agent. So an agent is the only product they can build that exists within the chat window. So it allows for a back and forth chat with the LLM. The difference between an agent and the chat window that you see right now is that an agent allows for configuration on the back end to serve a specific purpose. So it allows for this assistant prompt configuration, data connections, tool connections, etcetera. So that on the front end, from the user point of view, you have to do very little configuration and very little prompt engineering, and the agent just knows what it's supposed to do. So I'm gonna show an example of this. You can see in the platform, instead of selecting a large language model like this CloudOpus model, I'm selecting the agent that I've previously built, which is my financial analysis agent. So this agent is configured on the back end to connect to my SharePoint, like my demo SharePoint instance. And in that SharePoint instance, I've created a folder where I extracted or I downloaded data from the Internet on Apple, Zoom, and Cisco, and all that financial data from those three companies. Then I can ask a question. My question is very basic. It's a very basic prompt. I'm not even telling it to go to SharePoint. I'm not telling it what kind of analysis to do because I've already told it on the back end to go to SharePoint. I've already told it what type of analysis I want to do, and I've already told it what type of output I'm looking for. So I can just run this. This does take a few minutes, just because it is going to SharePoint, finding the relevant data, downloading the relevant data, and then running the analysis. So I'll let this run for, like, ten ten more seconds, and then I'll just switch to the output, that ran right before this meeting. Okay. Now I will switch. So this is what I ran right before this call. Again, the exact same prompt using the same agent. You can see it's running, it's doing the tool calls to extract the relevant data from SharePoint. It's running the analysis, and then you can see it's using all that information, and running the analysis and showing the output in the exact format that I requested on the back end. Very little prompt engineering, if not none no prompt engineering, and I didn't have to tell it exactly how to run this, as the user. So those are agents. The second two types of products you can build are apps and workflows. I'm gonna skip apps and we're gonna jump straight to workflows. Apps and workflows look the same from the user point of view, but workflows allow for more complexity on the back end. So I'm just gonna skip to that and show. So a workflow is allows for multiple agents to work together to generate a final output. So there is no chat. There's no back and forth with the LLM. It's just a one time run, and it's configured to have multiple agents. Each agent has its own configuration with its own LLM, its own prompt, etcetera. And so you can generate a really, really highly complex process that generates a fully complete product and a complete output. So I'm gonna show two examples of workflows. The first one here that you can see on the screen is a preliminary review memo generator. This is something that would be built for a private equity firm. And the firm, what they would do is they would upload their documents relevant to, a current deal or a current company, and then it would automatically create that preliminary review memo based on the documents uploaded and based on their current manual process. So what we try to do here is we try to mimic the manual process as exactly as we can so that the output looks exactly the same as what's being done in the company manually, focusing on just saving a ton of time on all the analysis and all the formatting, etc. You can see here I've uploaded a SIM file for the analysis And what it's running is it's running, one agent. Each box here is one agent that's connected to its own large language model, and then it's sending the outputs down the line through the various agents. The ultimate output that this creates is this file here. It's the preliminary review memo based on the company data. And you can see it covers all the sections that you would expect to see in the preliminary view. This is again done totally automatically, including using Python to create and insert the charts. Something that I like to point out here is the way that I designed this is that I have one agent that does each section of the preliminary review memo and then one agent at the end that formats the report based on the output of all the other agents. But we have one agent here, for example, for company overview, one agent for management background. There are a lot of reasons for this architecture design, but one I really want to highlight here is that by having an architecture that's designed like this, you can limit what agents have access to what tools depending on what you need them to have access to. So for example, we built something, similar to this for a client. And what the client requested is that if the data was not did not appear in the documents, for the competitive landscape grid here, they wanted to go to the web and fetch that data and populate that field obviously with appropriate sourcing. So I could implement that in this competitive landscape section, but in the company overview section, we can leave that to strictly pull only from the documents. So you can have that segregation between tools, which can be very valuable, when you're pulling that when you're trying to pull the appropriate information for each section. The second and last workflow I wanted to share was one that was built for an asset management, client. And this is, let me just share a quick slide here. So what the asset management firm wanted to do is they wanted to analyze a loan portfolio of about, 4,500 mortgages. And the data room for this loan portfolio had about 15 Excel files and a bunch of PDFs. And the normal process that they would go through to analyze this is this seven stage process that you can see in yellow here. So what we built for them was a multi agentic workflow that used seven agents to build the committee ready investment memo just at the click of a button. And the way it was architected was that each agent, served the same functionality as one step of this manual process. So we can ensure that the entire workflow aligned exactly with their manual process and the output was exactly what they would expect. So I'm gonna show the output of the, of the workflow here. So you can see this is a 22 page investment memo, which includes the investment recommendation, the target purchase price, and then the results of that entire seven stage analysis. So this was created, I wanna say, like, two months ago. At that time, Gemini was a front runner for this basically for all types of output formatting, specifically for Word documents like you can see here. But since then, Claude Opus 4.6 came out and obviously made quite a splash, specifically in the fact that they were the first large language model that could create PowerPoints. I could really quickly go in that workflow and go in that last agent, the output agent, and select Claude Opus 4.6 as my model instead of the original Gemini one, and it can make the output be, a pretty good looking PowerPoint. Again, this was with very little, if not no prompting on my end. I just switched the LLM, and it created a pretty nice looking PowerPoint. Also, just to point out, this slide here was also created using CloudOpus 4.6, and connected to our Thrive branded PowerPoint template. So, you know, now it really shows the advantage of having access to all the LLMs and being able to pivot to choose when a newer and better model comes out. So that I'm gonna stop sharing. So that is an overview of platform we covered, what exists out of the box, the configuration items, as well as the the types of products you can build within the platform. And and, Yeah. Emily, I I just wanna jump in here real quick. I wanna kinda connect a handful of things for the for the advisors that are on here. Couple things. Number one, everybody is having the AI conversation. We want you to be a part of it, and we would like for you to bring us with you so we can help you navigate the waters. There's no doubt about it. So just as if the rest of our services where we're selling with you, we're here to help you and your clients' journey. So we'll help you navigate what works for them, what is the right strategy. And by the way, from an adoption perspective, I think what you heard from from Emily was we want that high rate of adoption because that will be the continuous use case throughout the business. Secondly, and I wanna land the plane here, as those conversations are being had, as she and Matt were both talking about governance, when I hear governance, I think security. Your customers are asking for a better security posture. We're here to help you with that too. So please get us engaged early and often. We're here to help, and we'll help you with any, questions that you may have. Matt, if you could do me a favor, could you answer and Kathy asked this question. Could you. talk about how credits are managed? Yeah. So, it's there is a two step to this. There's now and then about three weeks. Right? So currently today, the way that it works is credit this platform uses a credit system which equates to tokens. And the the problem with tokens is every provider has their own tokening system and how much things are worth. So a credit system was created that's universal. It's almost like prepaid dollars. Right? So every month, the client's going to pay for a certain amount of credits. The entry level point is the 100,000 credits per month, but we have plans that go up into the millions. The way that that works is we watch them very closely when they hit up to a certain threshold. We'll let the customer know, hey, you're getting close to your tokens, you're getting close to your credits. What do you want to do about it? The way that the platform works is there's no spin overage on this. Once you get up to a certain point, the platform will let you go about 25% over. Then once you hit 50% over your limit, it'll start slowing down the models and stuff like that. We do have a couple of things in the works today to be able to offer overage utilization on that. But currently, right now, we just navigate the customer through that. We could help them figure it out. What we are planning on doing about three weeks is we'll have everything available inside of our client portal. Customers will be able to manage their own credits. They'll be able to see what's going on. There'll be a big ticker that'll show when their credits are getting close. They could push a button, buy more, all that stuff like that. So that's how it's working right now, and we'd be happy to update you as as these new features come out in the next couple of. weeks. Yeah. Thanks for the question, Kathy. And Matt and Emily, we appreciate you you both. Sure. Thank you. and with that said, I let's kick it over to, to Pat and Chris, if we can. Thanks, JP. Speaking of governance but unrelated to AI, we wanted to give everyone a quick use case and win in the channel specific to CMMC. For those of you who are not familiar with CMMC, it is the cybersecurity maturity model certification that's been rolled out by the Department of Defense. So any organization in the defense industrial base, whether you're a prime or subcontractor, has these requirements that are being rolled out by the DOD. And before going to the specific use case that takes over on that front, what we are seeing in this space is the DOD has their defined rollout for the cybersecurity maturity model certification, which is mapped to NIST eight hundred one seventy one. There are about a 110 controls, 320 control objectives, all tied to IT relevant and IT security relevant controls. And the, Department of Defense prime contractors are rolling it out quicker than the DOD had planned. Because they have supply chain requirements, we are now seeing a lot of our our clients that are in the DOD supply chain serving as subcontractors have these requirements rolled out to be certified quicker than expected. And this is a a specific use case and an example of that. I'll hand it off to you. Yeah. So, this was a good one to to kinda review because it started off as a, consulting engagement. So we were actually doing, VCs with them, and that started about eighteen months ago, and this was through Accelerate Partners. They brought us into this account. The account is a physical security company based in the New England area. They do work with commercial, but specifically around this area, they do, DOD contracts as well. Through the VCISO engagement, we had a lot of conversations around building out their cybersecurity platform and building out best practices, policies, and procedures, and realized that there was a need for CMMC compliance and that they did not have the team inside their organization to meet that need or or meet that requirement. That's when our VCs, so myself engaged Chris over at would that specialize in the CMMC side, to start the conversation. And I'm gonna kinda pass this back to Chris so that he can talk through what that process looks like. Yeah. Thanks, Pat. So we're we're focused on all things compliance readiness, whether that's CMMC, SOC two, ISO 27,001, you name it, we can help. But this is specific to CMMC, and this day under one seventy one. And we are a registered practitioner organization under the CMMC accreditation body, and we do have CCPs and CCCA on staff. So CMMC certified professionals as well as a few CMMC certified assessors. And our goal coming into engagements is to not only help organizations identify, where their CUI resides, so they're controlled unclassified information, which is what the DOD wants you to protect, what the data flows are around that, and how you're currently protecting it against all 110 controls and 320 control objectives. So, essentially, we've created a program here where we we came in, and, performed that assessment. And right away, we identified that they had a negative negative 147 score. The lowest you could score you could have is negative 203. In about three months into the engagement, we've already, progressed through that and brought them to a negative 32 score, a perfect score being 110. And so there's a lot of things that we've identified as gaps and a lot of things that we're advising on as a part of the continuous compliance program. It's not just identifying gaps. Right? It's continuous advisory based on the gaps that were identified and developing all the required documentation specific to CMMC, inclusive of inclusive policies, procedures, your system security plans, your plans of actions and milestones, reviewing the evidence and validating that it meets the requirements. And, ultimately, I think, you know, the bright side of this this opportunity is through that assessment, we we identify more opportunity based off of the specific control requirements. And, Pat, I'll hand it back off to you to talk about some of those things that we've. already identified and already have, upsold within the account. Yeah. No. Absolutely. So, taking a step back, we did VC so originally. So the account started around $5,000 a month billing. That then built that then grew to a total of, like, right around 12,500 right now. That includes the vulnerability scanning, CMMC compliance, program, and autonomous pen testing. The vulnerability scanning and pen testing were added after the CMMC piece. This was one due to the fact that they didn't have a program in place. They needed to have something, but it was kinda low hanging fruit. Like, it was they didn't have it. They didn't have the team to deploy it, and they didn't have the tool in place right now either. So, this was one that was pretty easy for them to pull the trigger on, to get started. And it also helped with their, you know, in in improving their their spur score as well. Yeah. In in addition to that, I think the big takeaway here is is all things compliance lead to larger opportunities, whether it's CMMC or any other framework, they're all requiring you to do stuff around security. Right? So specific to vulnerability management and and pen testing, there's a lot more things that are required within CMMC, like logging and monitoring. So think twenty four seven security operations, having a backup and disaster recovery, an incident response plan. So at the end of the day, yes, this started out as a five k opportunity, but this opened up the door for not only vulnerability scanning and pen testing, but opens up the door for a lot more than just that. So, we're we're we're able to help from a continuous compliance perspective all the way up into if if the organization needs help and support with the certification or auditor assessment that they have to go through throughout the year, we could sit there right by their side and support them with that being that we've helped develop the program from the ground up. Yeah. And. this if I can add one more thing, it does, play across multiple different frameworks as well. This example is CMMC, but this does play well across all frameworks. Right? We do have the VCISO practice, and we have the expertise in house to assist you get to that specific regulatory compliance journey that you're going on. So, Awesome. Well, thank you so much, Chris and Patrick. Fantastic information. I stole this from CGA the other day, but if you think about it with with when it comes to compliance and CMMC, think of like the SAT. We'll help you prep for the test. We'll be in a room with you when you take the test. We just don't certify that you passed the test. We'll get you all the way up to the finish line and help you every which way we can. I know we are close to the end of our time. We have tried to answer questions throughout. I do have one for you guys here. It looks like John had asked, do all, Department of Defense or Department of War suppliers need CMMC? Basically, he's asking, does it go all the way down to, like, uniforms or, you know, hair every every single part of it. Yeah. That's a great question. They will all have CMMC requirements. However, there are three different levels to CMMC. So it's really important that you read your contracts and what the requirements state. So CMMC level two is specific to, protecting controlled and classified information or CUI. And level one is a lot less stringent. There's a lot less controls, and it's specific to federal contracting information. However, what we've seen is organizations and private contractors specifically, blanket level two requirements across their supply chain. So you might think you have FCI and level one requirements, but you need to pay close attention to your contracts to determine what your actual requirements are. And that's something that we're helping organizations with. If you don't understand what your requirements are today, we could peek at your contracts, look at the specific DFARS clauses, look at the specific language to determine what are your real requirements. Love it. Well, guys, just to be respectful of everyone's time, we are right at the end of our time today. If, if we didn't answer your question, just know that we do track these and we'll be able to try to respond after the call. Also too, Tory and our marketing team always does a great job of sending you the slides and the recording after the session. Definitely thank you for your time today. We sincerely appreciate it. And if there's any to dos out of the, questions, we will definitely address those as soon as possible. Have a wonderful day and keep thriving.