Episode 119: Lopsided Biceps

 

Erick and Rich discuss why anyone using or selling LLM tokens badly needs FinOps help, as well as why an MSP’s onboarding process determines how long new clients stay, how much they spend, and how profitable they become. Then they’re joined by Rev.io’s Evan Rice for a conversation about why MSPs are the logical candidate to own the agentic layer at SMBs, and how they can go about doing it. And finally, one last thing: We’ve had it with these &*$%-ing bees on this &*$%-ing plane.

 

Discussed in this episode:

Balancing AI innovation and cost: The new FinOps mandate

Rev.io Doesn’t Fear Claude Code. It Loves It.

North Carolina flight delayed by bee swarm on turbine

 

Some guests on this podcast are clients of Channel Mastered. Compensation plays no part in their appearance or the content of the discussion unless the episode they appear on is a “bonus episode” explicitly labeled as sponsored.

 

Transcript:

Rich: [00:00:00] Think your client’s cloud data is fully protected. Think again. Microsoft 365 and Google Workspace leave critical gaps, ransomware, accidental deletion, insider threats, and no guarantee of full recovery. Join MSP Chat co-host Erick Simpson on Thursday, April 23rd to learn how to turn SaaS backup into a must-have security layer and a powerful revenue driver.

You’ll discover how to position backup is non-negotiable, overcome objections and standardized protection across your client base. Plus, see how top MSPs are boosting recurring revenue, improving retention, and reducing risk with SaaS backup. Register now at bitly slash secure. Backup. That’s BIT ly slash [00:01:00] secure SaaS backup, all one word.

And now on to the show. And 3, 2, 1. Blast off, ladies and gentlemen, welcome to another episode of the Ms. P Chat podcast, your weekly visit with two talking heads, talking with you about the services, strategies, and success tips you need to make it big and manage services. My name is Rich Freeman. I’m chief analyst at Channel Mastered at the organization responsible for the show.

I’m joined with you as I am every week by your other co-host, who he is, also our CEO and chief strategist at Channel Mastered. At his name is Erick Simpson. Erick, how you doing?

Erick: Doing pretty well, rich. No technical issues on my end today. It’s interesting, we are always pulled back into the world of dealing with technology issues and challenges right before important events.

Just like a lot of our audience. Happen to have, to deal with from time to time care to share?

Rich: Yeah. We are starting this recording session a good 20 minutes later than we were [00:02:00] supposed to because of a whole series of technical issues on my end, including a complete inability to get audio outta my my laptop.

Yeah. The funny thing is Erick, you expect that there are going to be technical issues in, in life, but the frequency with which it is the most basic stuff and the frustration level where it’s like I, there’s a tablet I use to do my end of the show. Couldn’t get that to call up documents, couldn’t get audio on, it it’s not the crazy sophisticated, my LLM my MCP interface isn’t working, it’s just, why can’t I get sound on this MacBook?

Erick: Yeah. It’s the bane of workers trying to maintain business continuity to do their jobs, hey like everybody else, we are human and subject to the whims of technology.

Rich: I’ll also say we are recording this at the beginning of a workday, and so I feel confident, Erick, [00:03:00] the worst part of my workday is behind me now, and it’s all downhill from here, which is which is good news.

And let’s begin the positive momentum here with our story of the week. Which concerns finops. So if I use the word finops people are gonna be familiar with that word and that concept and the context of cloud computing it’s a pretty familiar use case. Basically. There are a lot of cloud solutions and platforms, our consumption based, usage based they are on demand very difficult to predict and therefore control and manage how much you’re gonna wind up spending on those tools.

And there are many solutions. The tech industry has done a good job of coming up with cloud finops tools that businesses and partners who support businesses can use to track and control and manage all of that. What though, Erick, about finops for ai. Let’s just think about this use case a little bit.

We’re more familiar with the cloud one. So first of all global spending on AI is gonna exceed [00:04:00] $2.5 trillion this year. It’s gonna be $2.52 trillion this year, up 44% according to Gartner. So there’s a lot of spending happening a lot of fin to control in AI right now. Let’s consider the implications of that from three perspectives all of which matter and are complicated.

And so there is the MSP who is either using an AI powered service internally or delivering an AI powered service to their customers. They’re relying on LLMs to some extent for that. They’re the vendors. Selling AI powered LLM powered solutions to and through MSPs, which is to say every vendor out there to some extent or another.

And then obviously there are end users who are either running an AI powered solution they created themselves or an AI powered solution you created for them. All three of these constituencies, Erick, very much want some predictability in terms of [00:05:00] what it’s gonna cost them to consume tokens out of those LLMs.

And the unfortunate fact is there is no predictability there. It is exceedingly difficult to anticipate and therefore budget for AI on the consumption side of things. So there, there’s variation on per token costs across models. There is change across time. Gartner is anticipating that, lLMs in 2030 will be up to a hundred times more cost efficient than the earliest models developed late in 2022. So the costs are going down, which is good, but that just means there’s constant flux and shift. And again, budgeting becomes very difficult. There is complexity in terms of the workload that you’re looking to power.

And therefore the kind of GPU you wanna connect with. Are you looking for high volume, lower cost tokens? Are you looking for low latency, but higher cost token? This is gonna matter. You gotta factor this in somehow or another. And then, [00:06:00] obviously on the the end user side, there’s shadow AI that’s adding all new dimension to this.

By the way, all of this complexity precedes the real mainstream agentic era. There. There’s plenty of agentic AI happening now, but we’re really just on the cusp of what’s gonna happen. IDC anticipates, there will be. 2.3 billion agents out in workplaces globally by 2030 doing various things. And that in 2029 agents will be performing close to 217 billion actions per day, largely autonomously.

So if you think it’s difficult to track and manage and control AI and LOL consumption with humans, imagine having thousands of agents running around in your business and your client’s business, and they’re all doing things autonomously and you don’t know what they’re doing in terms of consuming tokens via the olms today, right now, because of how [00:07:00] difficult AI finops is.

41% of companies are wasting more than 15% of their total AI spend according to IDC. So there’s a lot of waste happening right now because of the difficulty of AI finops out there. All three of those audiences that I’ve talked about, Erick, are really gonna have to get their hands around this problem in one way or another.

And there are not actually very good tools. If you are an an MSP in particular, I’ve done a little bit of research looking for multi-tenant MSP friendly finops for AI tools. I found two one called Vantage, another one called Maverick, M-A-V-V-R-I-K. If you’re curious. We’re gonna need more of those out there to to help out with this.

But but everybody there and in particular the vendors and the MSPs are gonna be stuck in this situation where very little predictability in terms of what they’re [00:08:00] spending on one side of the equation. And they’re gonna have buyers on the other side of the equation who very much want fixed rates, predictability in what they’re paying.

This is gonna make locking in and anticipating and controlling margins. Difficult for people. And of course, to the degree that this AI spending can get out of control on the end user side. This is a service that MSPs can and should be providing to their end users being their AI finops partner to help them control, manage and budget for AI spend.

So it’s this big issue that I think a lot of people are aware of Erick, but it’s not an easy one to solve right now.

Erick: Yeah, you’re right, rich. And it kind, it, it absolutely reminds me of Cloud finops and, the opportunity that MSPs latched onto in helping their clients lower their consumption costs for things like Azure [00:09:00] and these other cloud platforms.

I think what’s missing here is what you just mentioned is the application or the platform or the process whereby it makes it easy. To understand the burn of all these tokens. Like where is consumption happening through all of these different LLMs and how do you get a handle on it?

So that’s so that also reminds me of the SaaS challenges, like how do you track, all the SaaS applications, the shadow SaaS, and now you’ve got the shadow AI going on. So I think it’s a, I think it’s an opportunity, but it’s a, it’s gonna be a problem until there is some way to identify and measure this consumption in order to provide strategies to reduce that consumption.

Because I’m sure there’s lots and lots of GPTs and things out there that people are subscribed to that they may not even be using. I’m, I am a, an offender of that. I’ve got three, GPT, apps that I use and I use 1 [00:10:00] 99 0.9% and the other two I don’t really use anymore. ’cause yeah, they didn’t really do it for me, but guess what?

Still paying for that. So this is a huge problem. It’s a huge opportunity though. And the other thing that kind of jumped out at me was when you, how many of these agents will be running around, it’s for someone that might be new to the show and is going how is that even possible?

How are we gonna have more agents than people? That’s exactly what’s going to happen. We’ve been talking about this on the show will have agents, like every employee will have multiple agents to do different things to help them do their job. And then the organizations themselves, business units, might have agents running to do business unit processes.

The organization overall will have agents to to do other things. So I don’t remember what the stat was, rich, but I think you mentioned it in a recent program where it was like, eight agents per person or something like that. It was just gonna continue multiplying. And then the last kind of funny thought I had [00:11:00] was like if someone’s not getting their work done in the age of mature ai, everybody’s got their agents, it’s, it’ll be like, Hey, why are your agents slacking?

What’s going on? Get on that.

Rich: Yeah. We’ve talked before about this concept that came outta Microsoft about a year ago at this time, of the agent boss basically where people who are individual contributors today at some point soon down the road, are going to be managers of, of agents and agent bosses.

And, yeah. You’re, the management you report into is gonna be criticizing your management ability and and effectiveness if your bots aren’t performing right. It exactly the same way. If you manage a team of people and they’re not getting the job done, you’re gonna hear about that too.

Erick: And I remember also there was a staff of a staff that was from a survey that you called up and it was about what percentage of people would want to be managed by an agent instead of their existing boss. Remember that story?

Rich: I do vaguely I can’t [00:12:00] quote the exact numbers, but a shocking percentage of people were happier working for the ai than a person, which again not a good sign for the the management talent out there right now.

Erick: No. That just leads to Skynet and everything else just accelerates it. But anyway.

Rich: So folks if you bring this issue to the attention of your clients and better yet bring them a solution and help them understand, control and manage it, you are gonna add a lot of value to to their business and strengthen your relationship with them.

But you, your tip of the week, Erick has to do with something else a little more familiar you can do to really kinda lock in lifetime value with an end user.

Erick: Oh boy, rich you know it, and it’s it’s all about onboarding. I was on a call with a new MSP partner that we’ve started working with at MSP Mastered yesterday, and they’ve invested heavily to fix their sales and marketing challenges.

But it’s like [00:13:00] when you improve, it’s almost like if you’re, exercising and you just work on one muscle, then it becomes, now you look lopsided ’cause you got this huge Arnold Schwartzen and your bicep and the rest of you it looks like, not, weak, let’s say.

So if you work on one business unit, improve it, what you’re doing is shifting the bottleneck or the friction to the next business unit in the process of performing your services for your clients. So I’m taking the next obvious task that happens when a new client is sold. NMSP, they need to be onboarded.

And, when I work with MSPs and the team works with MSPs the one area that we focus on first is the kind of service delivery to make sure that we’re identifying what’s happening in that unit and making sure that we’re improving the efficiency. But one of the low hanging fruit opportunities.

I wouldn’t say it’s the easiest one to solve, but there’s process and there’s training that can go a lot faster to improve new [00:14:00] client onboarding than there may be in some other areas because it’s basically, it’s a project. Rich. So I guess my overarching theme or message here to partners is the, you set the expectation of the entire relationship with you, with your new client based upon their onboarding experience.

I’ve spoken to MSPs who have lost clients during the onboarding process, it’s not, it doesn’t happen often, but it has happened, and that’s a wake up call. It’s they are just expecting something completely differently and they’re not getting it. So there are a lot of things that can go wrong here.

So MSPs, clo, focus a lot on closing the deal, but they’ve got to then figure out what’s the next, the bottleneck in the process, the slowest point. Getting a client to becoming profitable for the MSP and solving it along the way. Three three tips. Map your onboarding process, end to end.

Get that template out. They’re available everywhere. [00:15:00] Your vendors that support you can provide you with onboarding processes, but don’t treat it as a list, a checklist of things. It is a project. There is a, there is phasing that has to happen during this project. And when you do this correctly rich, you’re also setting the expectation with those new clients of how every project is going to go with you.

So not only are you setting the expectation for the relationship during your first onboarding engagement with the client, getting them onboarded to your systems and processes and preparing them for receiving your services against your SLA, but also it sets the expectation of when you’re trying to pitch them at project a migration and upgrade something.

You can call back and say, Hey, it’s a project, just like the onboarding project. And so they will be more apt and willing to say yes if they’ve had a positive experience during that first onboarding project. Project. So it is a project. You’re gonna phase it out, you’re gonna do your assessment, then you’re gonna [00:16:00] begin deployment.

And the third phase will be training. Training will be where you train your internal teams and the clients users and the client on how to receive this service. What qualifies as a, a priority one instead of priority two party three. What are the SLA expectations? So everybody knows how do they open tickets, right?

All of that standardize that first 30 days so that you’re delivering the same experience to every single client. And then you’re gonna schedule phase reviews during that process. Okay we completed our assessment. You’re having a kickoff meeting, just like. A project with everybody that’s involved, explaining how it’s gonna go.

You’re gonna do the assessment, then you’re gonna begin deployment. You’re gonna do a phase review at the end of deployment. We’re ready now for training. You’re gonna train everybody. We’re gonna have a phase review at the end of that. And then you’re delivering service. And then 30 days after the client has been onboarded, you are now having another meeting, which sets the [00:17:00] stage for your strategic business reviews and how you conduct yourself during that first meeting also sets the tone for what they should expect moving forward with you.

So make certain that you are focusing on outcomes and not focusing on technical details and getting in the weeds. You’re focusing on business outcomes. You’re focusing on the positive benefit of completing each phase of this onboarding, and then the value of your strategic business meetings that you will have with your clients moving forward.

If you do that, rich, you’ll keep clients longer and they’ll be much more satisfied. But you have to be consistent, and it has to be a repeatable process, which means it has to be documented. Everyone has to be trained on how to execute it, and we have to execute it and deliver against our KPIs as we measure success moving forward.

Rich: Yeah, great stuff. I particularly like the call out that this is a project and you therefore have to structure it and manage it like a project. It’s not something that you do on a haphazard [00:18:00] basis. There’s some discipline to project management, obviously, and tools that can help you with that.

But that is, I think, very much the right mindset and approach to take. The other thing that occurred to me, it’s really interesting ’cause I often say only half jokingly that 80% of success in business is managing expectations. And normally what that means is, if something you’re gonna do for a customer is gonna take you a week, you want them to understand that.

And if anything, tell them it’s gonna take me two weeks. Because when you come in at one week, you’re gonna be a, a hero. And what you really don’t want is for them to think it’s gonna take two days a thing. So it, it tends to managing expectations tends to work on protecting yourself on the downside.

But you use the term setting expectations a number of times in there. And it really just reminded me managing expectations can be a positive thing. If at the beginning of the relationship with the customer, the expectation you set is these guys are on it. That’s a great thing. And it’s part of that [00:19:00] bigger managing expectation kind of question.

You wanna make sure nobody expects more of you than you can deliver. But you also wanna make sure people understand exactly how much they count on you to deliver great work.

Erick: You’re absolutely right, rich. And a couple of just, follow on thoughts. When I say it’s a project, I’m, I so assuming that everyone understands that during, project management, right?

There’s, there are these check-ins, there’s these phase reviews, there’s these approvals that move forward. The most important thing is managing that expectation. And the most, and the biggest point of failure that I see during onboarding with partners is this is managing that change. Management process, right?

Because everybody knows that has been an MSP, right? I’m speaking to you MSPs. You know that things, you’re gonna discover things during onboarding that you didn’t realize. You’re gonna find hidden devices and applications that weren’t, evident during your initial assessment and [00:20:00] things like that.

You have to manage change. And that’s why when I say you’re sitting down and having that kickoff meeting with the client, you’re also letting ’em know how you manage change and risk. Hey, if we start deploying something and something goes sideways, here’s how we roll back. If we find something that’s out of scope, something we didn’t identify or prepare for it, here’s how we’re gonna manage that change.

We’re gonna identify the need for change. We’re gonna document it. We’re gonna talk to you about it. If it requires an adjustment of schedule, we’ll agree to that. If it require, it requires an adjustment of budget we’ll agree to that. The objective here. Is to get you onboarded. And just like with any good project Rich, when you involve the client and let them know that they are a, an active participant in the project, not just somebody that signs off and pays the bills, they’re actively engaged and their team is engaged in the successful outcome of that project.

This project happens to be onboarding, so everyone is involved and dedicated in making certain that we achieve that onboarding outcome within a, the timeframe that we scoped. And if we’re not gonna meet it, [00:21:00] that’s when change management happens. And I’ll tell you, rich, change management can be an awkward conversation for folks that don’t do a lot of projects.

But when you Mastered it and are good at it and are dedicated, you can have five different change orders during an onboarding project and have that client be absolutely thrilled at the end of it if you’re doing it correctly and involving them in that process.

Rich: Okay. That’s a skill to cultivate for folks in our audience.

So 2.3 billion agents coming to the workplace by 2030, according to IDC. A lot of those are gonna be running at SMBs. The arrival of this agentic layer is a phenomenon that’s headed to the the market that we care about here on the show. And our audience cares about. The folks at Rev io really believe MSPs are logically the people who should help SMBs Mastered the ental layer.

We, after the, are gonna be joined by their president and COO Evan Rice for an interesting conversation about [00:22:00] why MSPs are the logical people to own the agent layer, and more specifically what that means, how to go about it, how to profit from that, how to help your customers benefit from agent ai.

It’s all coming your way. And for a quick break, stick around. We’ll be right back.

And welcome back to part two of this episode of the MSP Chat podcast, our spotlight interview segment where we are extremely pleased to be joined by the president and COO of Rev io. His name is Evan Rice. Evan, welcome to the show.

Even: Hey, rich, Erick, great to be on. How are you guys?

Rich: We are great, and we are looking forward to getting into a conversation with you about the implications of Agen AI for MSPs and how they can own that ag agentic layer that’s emerging in the SMB space right now.

But for the benefit of folks in our audience who have not met you before or for some crazy [00:23:00] reason or new to Rev io te, tell ’em a little bit about yourself, your role at the company and about Rev io.

Even: Yeah, absolutely. I’ve been with the company now for about 12 years. Rev io we started as a.

Billing provider for communications service providers. Many of those are also, MSPs and we’ve been doing this since 2002. So really a, enterprise platform designed to manage the back office for a service provider from quote to cash as well as ongoing account management payments as well.

And just over the course of the last several years, we’ve really seen these industries coming together where more and more of our communication service provider clients were coming to us and saying, for the managed services side of the business, we are using, one of the legacy platforms that you guys are familiar with, right?

The Connect Wises of the world, the Autotask, the Kase, and we need you to integrate. Rev io into this so that we can have a cohesive workflow and that [00:24:00] our client can ultimately get one bill, have one place to go to, pay for those services, singular portal. So we set out to build those integrations and it was, candidly fairly challenging.

Pretty frustrating trying to deal with some of those technologies. And, in 2023 we really said we need to bring this in-house. And we had the opportunity to at that time acquire a company called Tiger Paw, which was one of the first PSAs in the industry. Predates those other companies that I mentioned.

And we didn’t re-adapt the technology. We really, adapted and integrated the team into Rev io and set out to build an all new platform. And that’s what we’ve been working on for the last two years, is to really integrate. All of the, ultra complex billing workflows, usage rating capabilities, everything that you need to bill any type of service offering regardless of how you wanna price it and package it.

And then integrate the other areas which are obviously robust [00:25:00] ticketing, dispatch, field service contract and agreement management. And then put all of that inside of a singular platform. We are now the only, the industry’s only end-to-end PSA communications, billing payments. And then we also did a deal recently with the Kronos where we have now integrated their cybersecurity and RMM stack to where we believe we have the full, end-to-end workflow.

And then the vast majority of the time that we are spending now is really how do you wrap the agentic wrapper around all of that to allow, our MSP clients to get more efficiency inside of their business and then be able to extend those workflows into their end customer. To really deliver the value that AI is going to deliver for all businesses, but so many as, particularly the s and b segment just have not been exposed to, or maybe they have in their personal life, but they certainly have not harnessed any of that power in their business.

And I think that’s where we got to the topic today.

Rich: Yeah. Yeah. [00:26:00] And it’s worth noting you folks are thinking very hard about how MSPs can use Agent Agentic AI internally and with their customers. And that’s what we want to talk about with you here. But the last time you and I met, you were telling me a little bit about the role AG agentic AI played in that transf post Tiger Paw acquisition transformation.

So you have some direct experience at Rev io with. Putting AG agentic AI to work and accelerating process processes that would otherwise take a lot of time. But I always, I always feel obligated in any kind of conversation like this about AG agentic ai to just make sure we’re all talking about the same thing here.

So give me your definition of Ag Agentic ai and if from a your sense if from an MSP standpoint, there’s any kind of nuance here. ’cause I’ll hear the word ag agentic applied sometimes to what say Claude Code does, but when MSPs are thinking, or at least when people in the tech industry are thinking about AG agentic ai right now they’re often more focused on [00:27:00] the kind of thing that open Claw does.

So when we’re talking about agentic AI here, what are you talking about specifically?

Even: I think all the above count, right? I think the. Experience that most of the country and most of the planet have had with AI is limited to chat or two with free GPT, right? So the concept that, AI is chat, I think is what a lot of people think.

I think, this community that’s watching this podcast is a lot more savvy, right? In particular managed IT. Providers are thinking about this. They’re trying to think about how does this affect my business? And ultimately, how does this impact what my customers are doing? And I think a lot of them are looking at it from a security perspective.

To me, agen AI is really AI that can do things, can it do things on its own? Whether that be write code, like you mentioned with Claude Code you mentioned Open Claw, right? That kind of harness and wrapper that goes around the [00:28:00] LLM to really give it tools and extensibility to actually do things either for you personally or in your business.

And that. Ability to harness those connections and really to me is almost just a new type of integration that we need to be thinking about. Integration is no longer system to system, just data flowing. It’s how can I now have an intelligence layer that actually reaches out into my application stack and does things for me?

And ultimately how can I have it, run and automate my workflows, whether that be writing code, whether that be taking a call in my office and updating a CRM and, dispatching a tech automatically, whether that’s stitching together for disparate platforms and making sure that data flows and that it’s, scrubbing and optimizing how the data flows between those platforms.

But to me, agentic AI is AI that does not just AI that talks.

Erick: Yeah. It’s interesting. We talk a lot on the [00:29:00] program about systems of record, systems of action and how. The and how things are evolving. Evan, so we’re talking on this conversation mostly about how AG agentic AI can help MSPs externally in terms of services and revenue.

But I wanna touch on some of the things that you just described, how, what it means to an MSP to integrate AI internally into their business. And you mentioned a few things that you’re doing at Rev io, so can you kinda share some of those things what you think MSPs can leverage, and then maybe some of the experiences that you and your team at Rev io or seeing especially with like agentic coding and things like that.

Can you dig a little bit deeper for us?

Even: Yeah, happy to. And I think I think it’s, I think that’s the right way to be thinking about it is, you absolutely want to put this in your own business. Before you try to implement it into your [00:30:00] customer’s businesses, right? You need that firsthand experience.

We’ve talked a little bit about, the coding and ultimately, acquiring a 40-year-old extremely robust platform with almost a thousand customers on at our own platform being, 20 years tenured enterprise application with several hundred, communication service providers, some of them very large, carriers.

That, to rebuild all of that in a timeframe that we set ourselves, which was, two years impossible to do the original way. We have over 30 developers still impossible to do that, it would’ve taken us. A decade if you’re using the traditional workflows, right? So we have certainly implemented spectrum driven development.

We’ve gone through multiple different models internally. But, and now is a hundred percent of our code gets generated through agents. There are absolutely humans in the loop at every step of the way. Our process has evolved [00:31:00] from product managers meeting with clients, documenting those requirements, creating written documents, which we called PR FAQs.

If anyone’s interested, you can check that. Amazon Working backwards methodology out, that’s pretty publicly available. And then we started realizing like, we’re using a process from before the iPhone came out and we’re as far as just traditional documentation and then we’re feeding that in, downstream where we are having agents produce, code Claude Code is.

The primary, LLM that sits underneath our custom agents that leverage that. So we have, use custom markdown files to give them instructions of how our particular development process works. But what we’ve really seen change is before there was this distinction of the product manager meets with client’s documents those requirements, puts that in a written document.

Developers are taking that, and then they’re breaking that down, and they’re building out user stories and they’re developing those [00:32:00] features. And then you have a separate QA team that’s doing QA and testing. We have a designer in the middle who’s designing the UI flow. And really now what has happened is every one of those roles can do everyone else’s job because really we’re piloting, agents that we have tuned to do this. You absolutely have to have humans overseeing it. The agents still need help, right? They get stuck. Sometimes they need additional context. I don’t know that it’s gonna be that long, that it’s gonna stay that way. But for right now, they very much like to check in, with their human.

I think these models are designed that way and they were trained that way. But now we have product managers building designed mockups and checking in code. And we have developers that are, running agents to do the QA testing and we have designers building code. It’s really now everybody can do these different pieces.

So it has really changed the way the workflow happens. And we’ve seen what has [00:33:00] happened in this company in particular in our tech teams. We had a strong initiative in 2018 to specialize our roles to make sure people were doing. Very specific functions due to the fact that we had seen some people get overly general and it caused, challenges in terms of effectiveness and output.

We’re now going through a transformation process back in the other direction, which is really des specialization now due to the commoditization of these skill sets and, the ability to leverage these tools now, what you had at your disposal is a lot different than the skill sets that you just had personally.

And it’s really around do you have vision? Do you know what the outcome that you want is, can you be articulate? And if you can’t, then pre-pro it, right? Have AI expand on your thoughts. And we’re really, now what we’re working on internally is how do we have vy, our own internal AI really be the orchestration layer across all these other agents and to do things like [00:34:00] pre prompting.

For our users to really say, Hey I’m missing some of this context before I feed this down into a downstream agent. You’re having collaborative conversations with a front end orchestrator agent to really flesh these ideas out before it even gets fed into the downstream process. It’s really allowing more and more of our individuals to contribute.

Including myself. I built a demo grade integration not that long ago. It was like, Hey, I want to go see if we can do this, how quickly can we crank integrations out? And if someone like myself who has zero technical ability, lemme make sure I’m very clear, maybe if there’s negative technical ability, then you know, I’m on that side of the spectrum.

But I do know what outcomes I want. And I do understand our client’s business, and I certainly understand our business. So now, when you have that type of context. To be able to have these tools that allow you to, see your vision through to reality and to be able to cut through, hey, we not, I don’t need to get to every [00:35:00] single member of my team.

I can actually work this workflow all the way through, and then we can go get together and collaborate. And we’re looking at, like house analogies. If you’re remodeling a room, you can put the tile on the wall and see if everybody likes it. Versus we have to have lots of written descriptions about things and let’s, ruminate over decisions because we know how hard it is to go ultimately deliver the final product.

Now it’s allowed for a lot more iteration agility. Just get things built, then look at it and decide what you edit and what you keep. So it’s really allowed this development process to go significantly faster. We’ve released more code in the last two months. I can confidently say that than we did. In the last two years, and we’re gonna release more code in the next two months than we have, in the last six months combined.

So it is continuing to accelerate partly as the models continue to get better, but also partly as our workflow has evolved. So it’s made a significant impact in our business.

Rich: Yeah. And there are [00:36:00] lots of lessons and parallels in there for the MSPs in our audience in terms of rethinking and redefining existing roles and and workflows democratizing accessing use of those ai skills.

And in particular, you guys are enabling. I’ve had the exact same experiences as you, Evan. I am. If it’s even possible. I’m less technical than you are, and yet I’m building applications using ai. And for years, MSPs have been telling me we don’t develop solutions for our clients because we’re not developers and we don’t have that skillset.

You don’t have to have that skillset anymore. The AI has that for you.

Erick: That’s right.

Rich: So this might be a good pivot point into talking a little bit about what it means to own the agent layer for the end user from an MSP standpoint. And you folks at Rev io have said that MSPs are the natural owner for that layer.

Which, which implies I guess two things. One is that there are other people [00:37:00] who might be competing for that title. So who is an MSP in contention with to be the owner of that agentic layer? And then what are the advantages that an MSP has in that in that use case?

Even: Yeah, I think the the number one competitor probably doesn’t exist yet. But we are seeing some of these announcements. I read an article this morning, a company called Treeline that was just founded, and they’re out to really take over ag agentic workflows inside of companies. And, to, to them it’s really the e natural evolution of the MSP.

And, I think. There’s going to be more of these companies that pop up in particular. Just look at some of the factors that are happening, right? There are large amounts of tech talent. We’ve seen some of these big companies, do layoffs. I think AI’s been blamed for some of that, but I think that’s a lot of that is just kind economic cycles that we’re in.

You also see PV roll up of MSPs. We’re starting to see [00:38:00] mid-size MSPs get, rolled up under, PE conglomerates that are creating larger companies. Not every technician ends up being retained or some, don’t like that culture. So they spin off and wanna create their own thing. Net new startups of, single entrepreneurs who are creating these, solopreneur ventures.

Are not gonna necessarily create the same, MSP, right? They have a real opportunity to just go into, in particular the SMB segment, where I think it’s currently less than half of SMB owners have ever even used AI of any kind. Not to mention have they really integrated it into their workflows. So there’s real opportunity for companies, there’s a pent up demand for this.

And so a lot of people don’t even know they need it, right? They don’t even know they have the demand. But there is a revolution that’s gonna happen over the course of the next few years. And I think there are going to be new providers that pop up that [00:39:00] are willing to come in, create custom workflows, still leverage third party applications, right?

Whether they’re using the big oms, whether they’re using custom built that are, leveraging some of these models you mentioned, open claw extremely flexible. And if they come in and take over those functional business workflows. The other thing that we see in the industry is that, it’s not a 90% of SMBs are thrilled with the MSP that they use, right?

So the concept of, I’m gonna go to my IT guy for this. If someone else comes in and grabs a hold of your business workflows and automates that, what I think that creates is a risk for the traditional MSPs of that. If you let that go, then you know, potentially your traditional security workflows managing, core applications, whether that be Microsoft or the rest of the stack and say, can you take this over to, because you’ve been able to automate, key business functions for [00:40:00] me.

Can you become the one-stop shop for me? And I think it absolutely should go the other way around because MSPs of the security discipline, they have the infrastructure layer. They’re trusted with the most critical applications. They are the logical place where the business owners should be going to get these solutions.

But if there’s a gap there, there’s going to be new providers that pop up that get their foot in the door, and I think that’s risk to the traditional business.

Erick: Rich, do you wanna speak to the callback to treeline

Rich: it? It may or may not have been my treeline coverage that that just went up within the last 24 hours.

They’re, the portion has written about them, CRN has written about them, but yeah. I heard you said you were just reading an article that might have been mine.

Even: Distinctly possible.

Erick: Yeah. Think we even talked about ’em on a recent podcast as well. So Evan, you really were speaking you were leaning into kind of these additional revenue opportunities, I think that could be attractive to MSPs [00:41:00] leveraging some ai.

So to what extent does Agen AI expand existing revenue streams for MSPs? The easy stuff, the low hanging fruit stuff, and then versus creating new ones. And what might some of those new revenue streams look like for MSPs that are really on it?

Even: Yeah, I don’t know whether it’s completely new due to the fact that a lot of their revenue streams are margin on top of, resold or third party technology solutions that they deliver into their customer.

You have your management and service contracts and retainers to manage all of that stack. And I just think it creates new flavors of this. I definitely think there are. Packaged engagements that can be offered due to the fact that some of this will be more custom to their particular workflows.

I think this, then when you do those engagements, I think this creates opportunities for them to package and bundle and rinse and repeat, especially if you are a vertically specialized MSP. If you’re working [00:42:00] primarily with law firms or you’re working primarily with medical, their consistent workflows and you know the way the business operates, you don’t have to reinvent the wheel every time.

But really this is around how do I go in and create engagements for those customers? I’m gonna do the analysis. I’m gonna look at where you have, waste and where do you have limitations based on personnel? What that ultimately allows them to do is not just participate in a, 30% margin on a SaaS application.

I now get to go participate in the margin of efficiency that I’ve created in that business, wrapped around even a low level human employee. We’re looking at, five, $6,000 a month when you look at fully loaded expense of an FTE employee. And obviously it can go up to, triple that right inside of a business.

So if I can allow you to grow or I can, replace some of that [00:43:00] work by creating this automation I wanna participate in some of those economics, obviously at a rate that is extremely efficient to that business, but then also allows me to participate in that. And I think that’s very different than the model that we see today, right?

We’re charging to make sure that things stay healthy. We’re getting retainers and I’m solving tickets, and I’m working issues, and I’m delivering technology that gets used in the business, but am I really participating in those core workflows and am I creating value that is, FTE type equivalent that I can charge a portion of that’s a totally different revenue stream than they have today.

Erick: It’s almost like you’re push, you’re pushing into the, I want to become a profit center for you rather than a cost center. That’s right. Yeah. Sorry. Go ahead, rich.

Rich: It you just went straight into an area that fascinates me in any conversation about AI generally, agentic, AI specifically when you were talking about participating in economics, which is pricing AI [00:44:00] related services and all these different, I’ve written about this in my blog a few times.

I don’t really feel like I’ve definitively figured out the right way to price these kinds of services. You hear about people saying if I can automate a function that you pay somebody 50,000 or $75,000 a year to do, but I’m only gonna charge you two thirds of that, then everybody’s coming out ahead.

There’s a ton of talk, obviously, in the industry about outcome-based pricing.

What is your thinking about the right way or ways for MSPs to price AI and agent AI services?

Even: I think, this is something we’re talking about internally as well. We’re trying to figure this out Also, we’re seeing a, we’re definitely seeing a common pricing model as people are just charging, especially the big companies.

They’re charging for token usage, right? They’re charging for credits. You figure out some acceptable margin. I think that’s the, probably the lowest hanging fruit as far as you just make sure you’re not underwater, right? You don’t have to [00:45:00] worry about how much someone uses. You’re getting to participate the way we traditionally would on resold SaaS applications.

You’re getting an acceptable margin on that, and then you’re obviously charging management on top of that, right? Which is the, to me, the most traditional MSP model. If you don’t wanna reinvent the wheel, there is certainly the, Hey, I’m gonna show you how to use it. I’m gonna build workflows around how you use it.

I’m gonna be someone who’s there to support you as you use it, and we’re gonna manage this. And however much you use, you need to pay for what you use, right? And then I’m here to support you as a management layer. I think we’re gonna see that evolve, and I think there is opportunity to create more flat rate the way you would pay for a digital employee, right?

As you start to figure out what are the capabilities of this agent, what are the workflows that you want it to take over? How, what is the traditional value? Like you just said, rich Hey, it’s normally a $75,000 employee that does this. We can fully automate that. [00:46:00] It’s, we’re gonna need to be able to make sure that we’re confident as you implement those solutions in what your costs are.

Because obviously it can go nuts pretty quickly if someone’s banging on these things 24 hours a day. But I think there’s also the opportunity to. Put that in a managed wrapper where you know what that usage is. And it’s not just unlimited use. Again, our, the people that we have in our businesses are not unlimited use.

They don’t work 24 hours a day, right? So it’s Hey, what workflows am I actually delivering for you? What business outcomes does that deliver? I think there’s opportunity to price it that way as well. I think, that’s something we’re still figuring out internally as far as how does that flow through?

What are the value of those outcomes? How does that fit from a margin perspective? But I think the, there is a easy way to do it, which is just, Hey, let me just mark up what I’m paying in terms of my costs and, make sure that I pass that through appropriately and I still add this into my [00:47:00] management and my service contract that, I’m going to manage these applications for you.

And then it’s really. Making sure that we have ROI analysis. So our agents internally, including my own personal one every week, does an ROI dashboard. For me that shows, here’s the workflows I did here was the token utilization around those workflows. We know our cost model. Here’s what I spent on these different processes.

Here’s what we believe that the ROI to the business has been either additional closed one deals in Salesforce, whether that was, a retained account that we’re tracking inside of Rev io new leads that came in. How did we optimize SEO spend, we’re trying to make sure we’re looking at all of the token utilization.

How much does that cost and what are the ultimate outcomes of those workflows? And did we have ROI and we’re seeing significant ROI? So I think MSPs could do the same thing, right? It’s, I’m charging you for the gasoline that fuels these agents. This is really the. The salary that you’re giving it, [00:48:00] and the more you use it for, the more you’re paying.

But we’re gonna have, intelligence dashboards that we build, really, that the agent builds itself to make sure that it’s justifying its own existence, right? And that’s how we framed it to our agents. You’re an employee, you’re a valued member of the team, you’re salary and compensation is your token consumption.

Calculate your costs and, justify you continuing to do these things and show the value that you’re delivering. And they’ve done a pretty good job with it so far.

Erick: Evan, I’m just starting to, put my, my recovering MSP hat on for a second and thinking, boy if we were able to create dashboards like that, that could include all of the other things that the MSPs do to reflect the value.

And here’s your dashboard Mr. And Mrs. Client Now, boy. Now we’ve like it, it adds. Tremendously to the value of the MSB is already delivering, but in many cases, that is, hidden in the shadows, right? For a lot of business owners. Like why are we paying these guys? We never hear from ’em.

Here’s why you’re paying

Even: [00:49:00] that. That’s right. And it’s hey we don’t have to that’s the point. Yeah,

Erick: exactly.

Even: I think to your point there though, this creates something that is in the business owner’s face all the time. They’re using it, they’re feeling it, they’re touching it versus maintaining health and make sure you don’t get a breach and make sure you don’t have things that are broken.

That’s the silent value that ends up creating that conversation that you’re talking about. Like why am I doing this? I have to explain this to someone who’s not technical. If they’re interacting with a digital entity, or their phones are getting answered after hours and leads are showing up in their CRM automatically with context and follow-ups are happening.

That is very tangible and visible. And that now I think really changes the conversation around am I delivering value? Like it’s very tangible and it’s upfront versus behind the scenes.

Erick: Absolutely. I think you earn a seat in the boardroom rather than a stool or a ladder in the [00:50:00] server room when you’re That’s right.

When you’re delivering those kinds of services. So Rev io has said that MSPs can own the agentic layer if they move quickly enough. And I think you stating something like 18 months. Let’s visualize what that 18 month journey would look like, Evan, and give us what your thoughts are around what that, what the short term like priority stuff needs to be, the midterm and then the longer term steps that an MSP just starting to think about.

This should be considering,

Even: one of the things that I’ve experienced personally is the first place you start is not actually the AI layer itself. It’s what are the core tools that you’re using as a human and what is your own tech stack and what is your software stack look like? Obviously I’m a president of a software company, but we firmly believe, and I’ve seen this in practice, that the tools you are [00:51:00] using, they are what either make or break how your agentic workflows actually play out.

Certainly you could have the agents build you everything from scratch, but that’s where you’re spending all of your time is now you’re a custom software shop for yourself, and you are spending months and months, and you are burning that runway that you have to go deliver this value to your customers.

So you need to make sure you’re on modern infrastructure. You have to make sure your system of record is API compatible, that it’s designed to be used by agents, and that ultimately those are the tools that your agents are going to use. So this is how we’ve thought about rebuilding our PSA and our back office application is that right now we have then most of our features that you see coming out and the things that we’re putting in press releases and that we’re putting on community and on X and on LinkedIn are very much human based features, because that’s what our customers are asking for.

You don’t see underneath the example of the MSP that we talked about earlier. What’s that [00:52:00] silence stuff that’s happening in the background? Is making sure that everything is covered by APIs. Making sure that we have the ability to have, ag agentic interfaces. We are testing, having vy use our software.

When we do demos, we have vy actually through its voice agent come in and talk about, these are the new features that I’ve seen released inside of the API. This is how I can now use Rev io where I couldn’t use it before this release. We need to be thinking about the agent as a user of these applications.

And if you have areas in your own stack and your own business that aren’t compatible and aren’t conducive to that’s the first place that you have to start is again, make sure that you clean up your house. The second place is start to automate these workflows and start to use it internally right before you go put your name on it with one of your customers to say, Hey, let me go automate these workflows.

Then you realize something didn’t work out quite as well and that. There’s always gonna be problems, right? So we do need to be [00:53:00] mindful in terms of what kind of commitments do we make and make sure users know that this is still technology. Nothing’s flawless and nothing’s perfect, but that’s why you have us, to be here to help you.

But bill, pick the right stack. Make sure you have the right tools and particular your core systems of record data platforms, financial platforms. Then start layering these tools on yourself. Whatever stack you decide to choose, whether that’s Open Claw, rev, io, will vary in short order. Be a agentic platform as a service, right?

Use Rev internally in your business to help manage your own processes. But then we, like we’ve done for years with our billing platform, where we have our bill on behalf program, how do we allow our clients to extend our software into their customer so that they deliver additional value and they get new revenue streams?

We’re gonna do the same thing with our agent platform, but automate your own house. Look at what are those workflows, whether those are basic ticket solves. Knowledge sharing, ramping up new techs. I [00:54:00] definitely think this is opportunity for MSPs to expand due to the fact that they always have challenges getting enough talent to support the amount of customers that they want.

How do I offload pieces of my workflow and then have humans in the loop so that every one of my people is more effective? And once you get really expert discipline in-house, now you can start to take some of those exact same workflows that you set up for your own company. And now how do we go implement that with our customers?

Ultimately it’s what stack are you using? What stack are you recommending to your customers? And make sure that all fits with the ability for agents to use it and to automate.

Rich: I wrote a post two weeks ago now, I think about AI governance and how end users badly need it, how few MSPs are actually doing it right now.

You folks have touched on governance as one of those considerations, issues, blockers that end users have to be mindful of, and therefore, MSPs need to be focused on right now, along with, cybersecurity more [00:55:00] broadly and data readiness and so on. I would love to get your thoughts.

We can maybe use governance as an example of the larger issue around some of these basic kind of blocking and tackling sorts of issues around AI that are a good starting point for a longer term conversation and engagement and relationship with the customer around AI that the folks in our audience need to appreciate and understand and start educating themselves about now.

Even: You can certainly do some damage to yourself, in particular with a couple of the platforms that you mentioned, right? Like open claw is like a Formula One car. Like you tune it how you want, hook it up to everything. I think we’ve all seen the post of, watching someone’s agent delete their whole inbox.

Hey, clean it up, right? That’s right off the show. Silicon Valley from 10 years ago, right? Like I asked it to get rid of all the bugs, it deleted the entire code base. Wasn’t wrong. It did get rid of all the bugs, right? I think the average user could certainly inflict some damage on [00:56:00] themselves, right?

Which is why they need a trusted advisor who’s managing all of their other sensitive data, their security. This is why I think the MSP is really poised to be the right provider that comes in and says, we should be the one that teaches you to do this because. We’re gonna make sure that when we set these workflows up, that there’s security wrapped around it, that we’re taking compliance into consideration that may curb, some of the effectiveness.

We certainly don’t wanna cut the wings off AI and chop its legs off and then be like, now let’s see what it can do. Now that I’ve completely made sure everything is totally secure, to the point where, this thing can’t operate. So I think MSPs are probably going to need to adjust some of their risk tolerance, right?

But we still need to have core security, making sure that, we’re delivering sustainable outcomes that don’t do damage, right? Like we, we wanna take the Hippocratic Oath that we want [00:57:00] to create effectiveness, we wanna create optimization, but at the same time, we don’t wanna hurt these businesses and we don’t wanna expose them to unnecessary risk.

So I think that balance is gonna be really important. And I think it really just comes down to the MSPs have that expertise. I certainly didn’t I get, over the years have gotten to work with our, CTO and you hear about these concepts and I’m like, okay, that makes sense.

We probably do need to be very mindful of how our client’s data gets utilized. Can’t expose us. We’re gonna implement these workflows internally, but we have to do it thoughtfully and we have to do it the right way. And, I haven’t found that it’s limited our effectiveness. It just takes a little more thought and infrastructure upfront and, you can do it right, and you just need to be mindful what you give it.

Access to creating, separate environments. It’s not the best idea to, just turn over the keys to all your production systems to an agent. Give it its own access. Don’t give it your usernames and passwords. [00:58:00] There’s certain things that you can do that, really help this.

And again, I think MSPs have that expertise already and it’s very applicable.

Erick: I appreciate the callback to Silicon Valley and the battle between Pro Wolf and who, and it was such a great show. My goodness.

Even: Son of Anton, the son of Anton, was the first agent, I believe.

Erick: So you were unpacking a little bit a minute ago the steps that an MSP should consider and crawl, walking and running, internal really good advice there.

So let’s say that, okay, MSP, now I’m the MSP. That feels really good that I’ve got a handle on how, AI is working for my business. I now want to introduce it or introduce a version of this service, however I want to price it and monetize it to some clients. So what do you think the approach there ought to be?

Is it a pilot. A, an engagement to get some feedback from a small [00:59:00] cohort of clients. You deliver services in a pilot perspective. And if that all works, then you know, they pay you for a larger rollout and then you can expand more broadly, or is there a different approach that you would I

Even: think I think it probably makes sense to, the, most of these MSPs have customers, right?

And you have some that are, friendly and you have really good relationships with, and that they trust you and you trust them. I think this is where there’s opportunity and it’s is really critical in delivering valuable solutions to any customer is, make sure that you’re studying up on Socratic method.

Really have to be interested in their core business. And I think this may be. Where this does push some of our, MSPs out there to really have more for firsthand knowledge of what that small business actually does. How do they operate? Where does that owner spend their time, where their employees spend their time?

To me, that [01:00:00] first engagement is really around learning, and ultimately that’s just gonna create a stronger relationship with your best customers anyway. But I think what to, to your point, it’s, hey, what are the biggest pain points? Where are those workflows that are really inefficient, where also are tasks that they just really don’t like doing?

This is time that I have to spend with administrative overhead and it’s just not the part that I like. I really like doing plumbing. I really like being in court. I don’t like filing notes and I don’t like doing my taxes, and I don’t like reconciling my, ledger. Where are those areas by going in and talking to those customer, and it’s very interesting.

Oftentimes it is those activities that they don’t like doing that are real low hanging fruit for optimization. ’cause they’re not strategic, right? They’re rote, they’re administrative copying data. Any of these use cases that people would be thrilled to offload and not to have to do anymore. And we have seen [01:01:00] this inside of our own business where our agents are taking advantage of that.

Transcribing meetings, sharing notes, following up on action items. Now these are just some basic things that before we had to. Do extensively, and every person in the meeting was taking all their own notes and trying to track down follow ups. We have a dashboard that we run our company off of for the last 12 years that had 20 different tabs in an Excel file.

You had to copy data from System A into here. Then we have to give a quarterly report to the board that’s all the same information, put it all over somewhere else. That is all now completely automated. And I will tell you, it has not led to us doing less work. It just leads to us doing different work. So every one of our MSP’s customers could give you a nice list of those things that they would love to get rid of.

I’ve never had time. I don’t know how to solve that. I don’t have a better way for it. We have to do it, but it’s not something that I want [01:02:00] to do. So I think you could very quickly identify some opportunities inside of your most trusted, best customers. Start with the friendlies. Don’t go to your most difficult, highest support, account and start there.

Start with the one you have the best relationship with and really just start learning about what their day is like and where they spend their time. And I think some of these opportunities will jump out at you.

Erick: Yeah. I love that concept of get the list, I’m gonna call it the naughty list, so that it’s the stuff that nobody wants to do that’s noisy, that gets in the way and prevents folks from doing the things that only they can do.

Rather than all this other stuff.

Even: That’s exactly right. How do you focus on what your business is designed to do, right? Again, you have specialties that many of them are service providers themselves of, different specialty services. Focus on that, not on the things that every business has to do.

Those are real big opportunities for automation.

Rich: You pick one item on the naughty list, you fix [01:03:00] it, and the, their customer’s gonna be so grateful and excited about getting to everything else on that list, that there’s gonna be a lot more business headed your way. Absolutely right. Great way to get going and then build from there.

Evan, this has been a blast. I really thank you for making some time and sharing your thinking about these topics with us here. For folks in the audience who wanna get in touch with you, learn more about Rev io where would you point them?

Even: Rev.io is our website. Best place to learn about us.

We have contact information directly on there. There’s contact forms. You can find me on LinkedIn and Twitter, best place is rv.io and we’re very easy to get in touch with.

Rich: All right. Thank you very much. Evan Rice, president and COO of Rev io. Folks, Erick and I are gonna take a quick break now When we come back on the other side, we’re gonna share some final thoughts about this very interesting conversation with Evan ai agentic ai.

We’re gonna have a little fun. We’re gonna wrap up the show. Stick around. We will be right back[01:04:00]

and welcome back to part three of this episode of the MSP Chat podcast. One last thank you to Evan Rice from Rev io for sharing his thinking about agentic AI and its meaning for the MSP world. This is actually Erick, only the second time I’ve had a chance to speak with him, but it’s always easy to tell when you’re talking with him that he does a lot of reading a lot of thinking.

I really value his perspective on anything AI related because it’s coming from a particularly informed point of view. And that interview I think showed that, and there was something he said in there, there were a lot of things we could talk about. The I would say the thing he said that I’ve been stewing over most since we recorded this conversation just a short time ago, he was talking about this concept of what he called de specialization, where basically you really needed to have a certain degree of specialization [01:05:00] to use the tools to perform various business functions.

Whether that, is at the help desk or as a developer or as an accountant or what have you. There are tools you needed to Mastered. That means you need to specialize in those tools. And what AI is doing basically is making us all a specialist in all of these fields to some degree, just because we can do a lot of the things on our own, like developing in the obvious case of vibe coding that we couldn’t before unless we had some of those specialized skills.

And that has real implications for. How you structure a workforce. And, he was talking about the kind of person they hire for at rev io. It’s a, that you’re looking more for a more generalized set of of attitudes and aptitudes. It’s not about can I perform this function?

Can I use this tool? It’s, do I have a more generalized kind of intelligence that we can, throw at hard problems and get interesting answers [01:06:00] to. We heard a little bit about how they’re thinking about that at Rev io. I’m gonna need to think a little bit about that in terms of, how it affects MSPs, how it affects, channel mastered.

It, it’s an interesting point.

Erick: It is and I keyed in on the same thing and it’s more of a. You want someone that is, is open-minded, a problem solver, an outcome-based individual that can see a bigger picture, a bigger scope of things. So it’s a different individual because we don’t really I won’t say we don’t, we’re able now to rely less on a person’s actual expertise or skill in a certain thing.

When we give them the, when we give them the the objective, right? And they go to work with AG agentic ai, and they go to work to figure out what an approach might be and is able to test the output and confirm the deliverable and things like that. It does make everyone just a better, I think overall orchestrator of [01:07:00] looking at things from many different perspectives.

That’s the thing that, that happens with me, rich, when I’m using ai, when. I think I know exactly like I have to break some of those old bad habits myself. This muscle memory where I think I wanted to go do this way. And then it says, Hey, would you like to think about this? And I’m like tell me more.

And it goes down a rabbit hole, so you gotta stop it because I think secretly it wants to burn more tokens because keeps stringing me along. But all in all, I think it forces us to think differently and to be a little bit more open-minded about, collating and assessing different types of data to potentially improve our output or confirm what we already thought was the right way to do something or the right approach.

But I’m constantly learning things like all day, every day, just interacting with ai.

Rich: Folks, that leaves us with time for just one last thing. And Erick and I do a lot of travel and I’m good for a 80 a hundred [01:08:00] thousand miles a year easily, which means I’m on a lot of flights. And that means inevitably I, I encounter delays of various kinds over the course of the year,

And the years plural, right?

And yet I don’t think I have ever encountered this particular kind of delay. This concerns a delay of an American Airlines flight about to depart from the Charlotte Douglas International Airport in North Carolina. Delayed for about an hour due to a swarm of bees in one of the engines on the plane.

And I’m just imagining myself at the gate, and they’re trying to explain this to me. We’re gonna be about an hour late. Folks. There are bees in the engine and we’re working on getting them outta. That’s a new one on me. And it immediately reminded me of course of, this is bees on a plane.

Like we’ve just given Hollywood a new idea for a movie. It immediately reminded me of snakes on a plane and

Erick: yeah

Rich: yeah, I’m tired of these bleeping snakes on this bleeping plane. I’m tired of these [01:09:00] bleeping bees on my bleeping plane. I wanna go home.

Erick: Yeah, you took my line. I was gonna say, at least there weren’t snakes on the plane, so bees in the engines.

Yeah, I wonder what kind of an impact that has? ’cause I know, you can’t have birds flying to the engine. You’ll down a plane with that, but I wonder, if it was just an annoyance more than just, or were they just being extra careful because, bees are a living thing, invaluable, to the planet, trying to take care of them.

Humanity.

Rich: How do you get bees out of a jet engine too? I, there’s some video I haven’t actually looked at attached to this story. I’ll have to see if there’s a technique to that.

Erick: Yeah. AI might say maybe get some queen bee, royal jelly or hormone or something and follow them.

I think maybe a document or I saw one I was dealing was like, oh yeah, they follow the queen bee center hormone. Who knows?

Rich: I was wondering if there was like an emergency beekeeper who you call in and that situation. But I doubt it. I’ll research this. I’ll let you know later how you

Erick: Yeah.

It might be a. [01:10:00] I wouldn’t be surprised if it was just the handy old shop vac.

Rich: Yeah.

Folks that is all the time we’ve got for you this week. We thank you very much for joining us. We’re gonna be back in a week’s time with another episode for you. Until then, I will just remind you, this is both a video and an audio podcast, which means that if you’re listening to us right now, but you’d like to check us out on video, go to YouTube, look up MSP chat.

You’re watching us on YouTube right now, but you’re into audio podcasts. Go to Spotify, Google, apple, wherever you get your audio podcast. You’re probably gonna find us there too. And wherever you find us, please subscribe, rate, review. It’s gonna help other people discover and enjoy the show just like you do.

This show is produced by the great Riley Simpson, part of the team with us here at Channel Mastered, where we help vendors build, grow, and optimize thriving MSP channels. You can learn more about the many ways we do that. www.channel Mastered.com. Channel Mastered has a sister organization called MSP Mastered.

That’s Erick [01:11:00] and his team working with MSPs to help them grow and optimize their business. So you can learn more about that at www.mspMastered.com. So once again, we thank you for joining us. We’ll see you in a Week. Until then, we will remind you as we always do, you can’t spell channel without MSP.