Randy Johnston and Brian Tankersley, CPA, examine Microsoft Agent 365 as a control plane for deploying, monitoring, governing, and securing AI agents. They frame the shift as a move from per-seat software licensing toward an AI token economy, where tokens function like staff time, prompts replace checklists, and agents perform repeatable work at speed.
Practical examples include invoice extraction, email drafting, financial-statement analysis, and budgeting—tasks that may cost pennies in model usage while still requiring review and judgment. The hosts argue that Agent 365 gives accounting firms an alternative to “Wild West” experimentation by extending Microsoft 365 security, auditability, and governance over agents.
The Accounting Tech Lab (full YouTube series at link) is an ongoing series that explores the intersection of public accounting and technology.
View the video below:
Key Takeaways
- The economic unit of AI is shifting from a user license toward token consumption and task-level cost.
- Tokens can be managed like staff time, prompts like procedures, and agents like digital staff assignments.
- Agentic workflows depend on steps, loops, exception handling, context, permissions, and human review.
- Agent 365’s differentiator is governance: visibility, monitoring, security, auditability, and centralized control.
- Public AI tools create material concerns involving PII, PHI, client confidentiality, retention, e-discovery, and vendor training practices.
- Proprietary workflows may be valuable intellectual property and should not be surrendered casually to a model provider.
- Model selection should balance quality and cost with privacy, security, regulatory fit, and data ownership.
- Accounting firms should begin with controlled experiments and measurable use cases rather than enterprise-wide autonomous deployment.
- Human reviewers remain accountable for conclusions, professional judgment, client context, and exceptions.
- Firms need AI governance policies, token budgets, approved-tool lists, monitoring, and documented escalation procedures.
Transcript
(Note: Some errors may appear due to automated transcription.)
SPEAKERS: Brian F. Tankersley, CPA.CITP, CGMA, Randy Johnston
Brian F. Tankersley, CPA.CITP, CGMA 00:00
Welcome to the Accounting Technology Lab, brought to you by CPA Practice Advisor, with your hosts Randy Johnston and Brian Tankersley.
Randy Johnston 00:09
Welcome to the Accounting Technology Lab. I’m Randy Johnston with co-host Brian Tankersley. There is a topic that we probably should have covered earlier with you, but we were waiting for it to stabilize a little bit on May one. Microsoft released Agent 365. Now we’ve been digging into this over the last 60 days, and frankly before that, because we had a preview release at least 60 days in advance under non disclosures. So unfortunately, only four or five months of experience on the platform right now. So, but you need to be aware of what’s going on because we believe that we are transitioning into an AI token economy. Now we have another accounting technology lab where we’ve talked about MCPs, model context protocols, and the importance of agents. And over the next little while, as this AI vortex continues. We’re going to try to get you out of the vortex into practical things. But the fact of the matter is that tokens generally are going to cost somewhere between $2 to $5 per million for input tokens, and somewhere between eight and $15, maybe as low as four for output tokens. Now the cheapest tokens right now are from Google Gemini. The input tokens there are only 35 cents up to about 75 cents, and they’re well, they’re using the phrase Brian. I don’t like to use, but it’s become my catchphrase for the year. They’re good enough. You know, good enough sounds like I’m compromising, but it’s good enough to get the job done right, and so maybe I’m okay with good enough now.
Brian F. Tankersley, CPA.CITP, CGMA 01:48
Yeah, so it’s an F-150 and not a G wagon.
Randy Johnston 01:50
Yeah, exactly. And you know, as long as the work’s done right, I’m not sure I care. Now, what we think is going to happen, friends, is you’re going to wind up subscribing to enterprise level subscriptions, which are expensive, a couple of $100 a month for Anthropics Cloud and for ChatGPT, and then you’ll have a new license from Microsoft for Copilot that we’ll talk about, and you’ll probably have a Gemini Pro Ultra license, and that will get you token access from three of the four majors with Microsoft Copilot. Four of the majors to build an agent strategy. Now, tokens still sounds pretty confusing. It’s hard for me to get the right sizing for you to get token budgets. But just as a simple example, if you’re extracting an invoice today, it takes about 1,200 inbound tokens and about 400 outbounds, which in the Claude world today is about a penny, 0.096. And I’m not going to try to quote exact numbers to you, but I want you to get a sense. It’s about a penny to do it in Claude. It’s about seven tenths of a cent to do that work in OpenAI’s ChatGPT, and it’s about a half a cent to do it in Gemini, and the Gemini results just as good as the other two. Now, if you’re going to draft an email, you know it’s pretty clear it’s about a penny to do it in Claude. It’s about seven tenths of a cents to do it in ChatGPT, and it’s about six tenths of a cent to do it in Gemini, and the results are pretty similar. One could argue that maybe ChatGPT does a little bit better job. And by the way, Copilot can do all these jobs, and it’s included in the subscription, so you don’t really have a separate charge. But it usually doesn’t do quite as good a job. Now, if you’re going to analyze financial statements, and again, remember a caution: we don’t put client information, we don’t put PII, we don’t put PHI in any of these platforms. It takes about 6000 inbound tokens to do a financial statement analysis, and about 2500 outbound tokens. So that’s about five and a half cents to do that in Claude. It’s about four cents to do it in ChatGPT, and it’s about three cents to do it in Gemini. Now, Brian, I know you read fast, and you can look at a financial statement quick. You know, is it good enough for you that you know you can spend three or five cents to analyze a financial statement instead of how long would it take you to analyze a financial statement? Some of yeah,
Brian F. Tankersley, CPA.CITP, CGMA 04:14
I think it’s absolutely worth it. You know, the way I look at it is that I’m still going to look at it in detail, and I’m going to have my observations, but more than anything, this give this is like sending it to staff first, and the staff costs three cents. So you know it’s a no-brainer because it’ll catch, it’ll pick up the 60% of the comments that I would have made, and it will in for five cents it will enter it into the com into the analysis, and then I can add the 40% of other color that I have, as well as the things that are subjective from my coaching sessions with the client, that are going to be a little more different, and you know that you wouldn’t be able to pick up out of the financial statements by itself because the AI model would. Context, but it’s I think in general getting using you know staffing out a lot of things to AI is something that people need to do just from a personal effectiveness time.
Randy Johnston 05:13
Yeah, and that’s the point of the Microsoft Agent 365. And you know I should have given you full credit for your session at Scaling New Heights this year, where you actually provided a prompt sheet for financial activities, it was quite good, and you know. And I’ll
Brian F. Tankersley, CPA.CITP, CGMA 05:30
actually, I’ll, and if you, for those of you watching this on video, I’ll actually embed the QR code for that one if you want it. It’s just, it’s a fully AI generated, but it’s 100 prompts for accounting and bookkeeping. It
Randy Johnston 05:43
really was quite brilliant, as it turns out. And so, part of the trick here is we’re going forward. These prompts become the basis of the agents that we’re going to create, and that’s the end game that we’re trying to get you to. But lastly, and again, I probably should have much better token budgeting, but I don’t have it for you today because I’ve been thinking about how to help you understand the costs. Because we’re going to go from a per seat licensing, as I’ll explain in just a moment, to a token budget. But if you’re going to create a budget, 2000 tokens in, 3500 out, it’ll cost you about six cents to do it in Claude, about four cents to do it in ChatGPT, and about four cents to do that inside Gemini. So if you contrast that to the license costs that you have, most of you are using business class licenses, so you’re paying 20 to $30 a month for a protected set of code and the protection we believe is still not as strong as it should be, but the fact of the matter is there’s a higher end model of each of these platforms, and the going price right now seems to be about $200. Google’s AI Ultra is about $250 a year, but you know if you say I’m going to buy that license to get much stronger results. Good, you can do that, but we’re going to consider that for just a minute. The productivity platforms, and we don’t think you’ll need to license productivity platforms as much by user. So consider how many users you have where you’re currently paying 20 or $30 per user per month, and that you might be able to switch them over to agents that you just pay per usage for the tokens rather than a subscription. So again, if you’re a smaller firm, that’s less applicable. If you’re a bigger firm, it’s real applicable. So that said, this session is about agentic AI, where agentic AI allows us to customize. In this case, Microsoft Copilot, so it meets your specific needs, where the task can run on a fix schedule, and you know you may and you will probably have to ask your IT staff or contractor to enable the custom agents for you, and that can be done today inside the Microsoft 365 Admin Console. So when we’re talking about this, agentic AI allows autonomous decision making, where they can make independent decisions and act without constant human guidance or supervision. Now we’ll talk about human in the loop and monitoring and and the new IRS regulations that have really changed the game on AI related to taxes. I would see it, but these AI systems set up this way can pursue set objectives and adapt to new situations or changing environments very effectively, and they plan actions, evaluate results, and learn from experience. See the trick two agents are the steps and the loops. Because if you think about most accounting processes, they’re steps, and when something’s not right, we adjust and we go back and take the steps again. And so, steps and loops is the world that this is in, and agents can extend Copilot’s functionality directly, and it can do it in a safe environment. So, if you start thinking about digital assistants as agents as digital assistants, it’s like staff, like Brian just talked about in the financial statements. It’s a way to assign the work to an AI, and it can do it repetitively without a vacation and so forth, and be ready for you. For example, when you come to work, so you can have an agent analyze your calendar. Now,
Brian F. Tankersley, CPA.CITP, CGMA 09:33
so test me out here. So as I’m looking at this, I’m seeing tokens as staff time on the schedule. Okay, so in the old world we had staff time on the schedule to work on a project, and now we have tokens. Okay, we have we used to have checklists, and now we have prompts. Okay, and prompts have context that is a that are the I guess the procedures that we’re supposed to follow or the workflows we’re supposed to follow. We have. Steps and loops, which are which again are related to our workflows, and then basically the agents are equal to staff. But what’s different about these agents that’s that’s so fascinating is that they you know they they get work done so fast that the human is now the logjam in the process. You know the it takes the human much longer because we’re doing all the agent work at massively parallel processing, and so it’s going to be much more succinct. You know, I remember when we first started using outsourcing in the accounting profession, we had a lot of the folks offshore working on during their daytime, and so you would like throw everything over the wall before you left before you left work, and then then when you came back in the morning, they would have worked on it for a whole day, and then you attacked it again and kept going, and then you threw it back over the wall again. This is going to be much more like that, except when you go get a cup of coffee, it’s going to come back over the wall, and it’s there’s going to be a lot done. So I want to make sure everybody gets this about this because it’s it is still analogous. I mean, you still have to have procedures, you still have to have checklists, you’re going to still have to have you’re going to still have to lay out the way you want it to look and things like that. And that is now going to be documented in a prompt, and that’s how you’re going to tell the AI engine to do this.
Randy Johnston 11:21
Yeah, and I think the way you’ve set that up really helps, Brian. And the whole analogy to outsourcing is in place. One of the pieces of bad news, I guess I’ll say, is I don’t see that the outsourcing companies can survive in the long term as the agents get better. You know, people keep trying to convince me, no, you’re still going to have humans doing a lot of that work. Well, you might, but if you get your agents good enough, and you can get a process done in a few minutes your way, you know I’m having a harder time seeing where that’s economically going to make sense. Now the other piece is I don’t want you to think that you’re going to be sitting there feeding the prompts in. They’re going to be coded, and you’re going to have to adjust them occasionally, just like you adjust procedures occasionally. But the thing is, they will run the prompt. So if you’re analyzing financial statements, they’ll look for the things that you look for, regardless of the financial statements, and you’ll learn over time. Oh, I should ask it to do that. You add your prompt, and next time it always gets it just like you might go down to a staff, a manager, a senior, and say, “Hey, you know, whenever you do these, would you do this too, please? You know that type of thing. But you put it in the agent, and boom, you got it. So now in Microsoft Agent 365, there’s a new strategy to do this. Now up to this point, when we’ve been asking people to craft agents. We’ve used a tool out of Germany called N8N to be our agent repository, our launching system, our library, if you will. But you know, you had to really control that, and you could use any of the different platforms to get inside there. But in the Microsoft world, we believe that you can control all these agents inside the Microsoft 365 environment. So for many of you, that gives you protection of the data. There’s lots of advantages to being in Microsoft 365’s AI from a security perspective. Now, notice I didn’t say we have to quit using Claude or we have to quit using ChatGPT, but those platforms can be used to help you create the agents, and you can copy those prompts into Copilot, so you go from the Wild West of Claude, ChatGPT, or Gemini into the town with the sheriff, the Microsoft Copilot environment, and you’ll see that Microsoft is allowing the selection of ChatGPT or Cloud Xanthropic inside the platform, but they’re not quite as crisp because of some security issues. But here’s the deal: if you think about the Gen AI platforms that we’ve been talking about as being kind of employee empowerment and productivity, that’s the starting point. But what Microsoft Agent 365 adds is the agentic AI, including the authoring tools inside Microsoft Copilot, the much more extensive tool of Copilot Studio, the AI Foundry, and the Azure AI. But what I’m going to ask most of you to consider is use try using the new Vibe coding tool inside Microsoft 365 and the Agent Builder before you graduate up to the Copilot Studio. What you’ll discover is if you’ve got accounting knowledge, you can write these routines, these agents, and get a whole lot of work done inside the platform. Now, Brian really led the way from the two of us on Vibe Codin, and we’ve got another accounting technology lab on that. It may be time to revisit this, but the the key thing I want you to hear. Is this new Microsoft Agent 365 is a structured platform to control a lot of this, and there’s lots of relationships that we can talk about. How it empowers your workforce, how it can streamline your business processes with these agents, how you can improve client and customer experience, how you can make augmented decision makings with the intelligence insights and the integrations into Power BI and all these other things. It’s a real deal, as I see it. But it’s early, so again, Microsoft Agent 365 is the control plane for agents. So we now have an environment that is controlled. Now it’s early. Is it perfect yet? No. Is it pretty doggone good? Yes. And you know, I had an old rule that I used. It still works.
Randy Johnston 15:51
Microsoft Project products are never very good in version one or version two, but they’re pretty good by time to get to version three, and then they get better and better. Right now, we’re kind of in version three of this stuff, and so we’re just getting ready to take off. But what’s the control plane do for you? It allows you to observe and audit, to monitor and manage all your agents in real time from a single view. And what we believe is happening right now is you have team members using AI in any old which way, and you can’t control the results or the way they’re using it. You get some governance in here, and I believe that we’re going to talk more about AI governance. And your AI policy should have governance just like your WISP policy, your written information security plan has governance, but it gets you guardrails for the agents and the users, and it gets you security because you can protect the agents so they don’t get tampered with to give what I’d call false results or inconsistent results. So, Brian, you can see why this control plane inside this platform begins to make a lot of sense to me. Now, is it early and have I not quite got everything right? Yes, but we’re trying to help you navigate because the role that Brian and I are trying to play for you here is scout, and you know when you’re exploring the wild west, sometimes you know bad things happen out there, and you know accidents happen and you get attacked. But
Brian F. Tankersley, CPA.CITP, CGMA 17:19
yeah, and I will say that I think Mike, I think Microsoft’s approach to this is they’re really I think they’re really like the railroads coming through. And to your Wild West analogy, where we have the grown-ups in charge now, and we’re going to now systematize this in a way that we can regulate, and that we can do the right way, and that we don’t have to worry as much about train robberies and ambushes and things like that, you know it. And so, what I want you to what I want you to get about this is that yes, Copilot has his has historically not performed as well as some of the frontier models. Okay, but the problem with frontier models is that you don’t you. I’m not sure you can exactly trust them on the privacy side, or you know, again, there’s wishy-washy stuff there. I mean, just as an example, you know, we I’ve not, you know, as we’re recording this, as we’re recording this, the U.S. Department of Commerce has released the Kraken, as it were, with the Claude Mythos now as Fable, and you know, we don’t know exactly what limits they have, but I will tell you that one of the thing, one of the things, one of the solutions that Anthropic offered up was keeping all of your prompts and all of your history for 30 or 60 days, which is completely unacceptable in the world that you and I live in. And those, and again, the other thing is you don’t necessarily want the tools to learn from your prompts and then share that with other people because then it can memorize things off of here. So I want you to understand here that Microsoft is not out at the bleeding edge, okay? Because they’re having to be grown ups and be accountable for what they do, because they have a market cap in the trillions, and you know they’re going to make themselves very attractive to attorneys and and every ambulance chaser in the world that can create a grievance out of nothing if they’re not extremely careful and they don’t do things right. They’re also going to make themselves a an object. They’re also going to kill their brand if they’re not cautious enough to permit regulatory compliance to work appropriately. So you know, I think it’s good that they have that they’ve done this, and I want you to understand that this is the kind of approach that you want because you’re handling the most sensitive data that most people have with this, and so it’s important, I think, to be cautious rather than go out and doing bleeding edge. You know, just because you can put stuff into ChatGPT that’s sensitive doesn’t make it a good idea. And the nice thing about this is this kind of helps you with your team put guardrails around it so that you don’t. End up in a situation where now you’re having to deal with attorneys and cyber insurance and Microsoft and all these other things because you didn’t use the grown-up tools.
Randy Johnston 20:10
Yeah. So Brian, as I’m listening to you go on that, you know many of our listeners have not seen and had the benefit of your teaching of AI privacy and security, talking about how the AI engines are keeping this data permanently, which kills your records retention policy, and that under litigation, it’s already been established that content of the AI engines is discoverable in e-discovery, and you know I can just go on and on about why the grown-up approach probably is needed. I truly appreciate what these AI engines are doing, but let me give you one more insight that you and I have not even talked about yet. I have discovered that all of the big AI platforms are going to try to learn how to automate workflows with AI over the coming year. That’s the next big deal for their models. So what they’ve got to do, they’ve already overed up, gathered up all the public data. Now they’re after private data for workflows, so they can begin doing some of that. Okay, wait a minute. You think about the environment that accountants work in, and how sensitive is the data? Now it’s not the data they’re after; it’s the workflow. But think about if your workflow has a proprietary value, which my NMGI workflow has a proprietary value. It’s enabled us to do way more work with far fewer resources, and our competitors can’t figure out how we’re doing it. And I’m not going to teach them, right? So again, I share a lot of stuff, but there’s certain secret sauces. I mean, Coke doesn’t share their formula, and you know, Kentucky Fried Chicken, one of your favorites, doesn’t share the secret spices, right? Yeah, that there’s certain things that are intellectual property. Well, side rants, sorry, but I
Brian F. Tankersley, CPA.CITP, CGMA 21:56
will say, however, once you work inside the sausage factory, you can’t eat the hot dog again, and that’s kind of how I feel about Kentucky Fried. So you know, again, the good thing is that these AI engines aren’t seeing the things that can’t be unseen. So that’s good.
Randy Johnston 22:12
Yeah, it is good. So you know, then friends, what we believe you’ll want to consider: Microsoft on May one introduced a new licensing for 365 called E7. Now we’re not trying to do Microsoft licensing education today, but you know if you look at business premium, which we’ve recommended for many of you as being the right approach, it is clear to us that we’re going to have to come back and retrofit business premium at $22 with the $15 per user per month Microsoft Defender and Purview Suites for Microsoft 365 Business Premium, there’s a product name that you can repeat real fast. That’s $15 a month for Microsoft Defender and Purview Suites for Microsoft 365 Business Premium. Now, this platform, which includes Copilot and Microsoft Agent 365, is $99, and we had said 10 years ago that there would be a point in time when you know Microsoft got $100 a month from you for their platform. Well, it’s now, and it’s probably worth it. But there’s other pricing changes that also occurred on May one, so we want you to be aware that your next renewal, Microsoft 365, is likely to go up, because for the first time in a while, E3 increased in price up to $39, I believe, and Microsoft E5 increased from 57 to $60, but Microsoft Business Premium fortunately stayed the same at 22. So when you look at the pricing that became available on May one, Copilot for Microsoft $30 a user a month. You can select ChatGPT or Claude from inside the model today, and when I just throw a prompt at Copilot, I can tell whether they’re using ChatGPT or Claude to answer it. And I politely disagree with Microsoft’s announced strategy in mid June that they’re going to try to incorporate the Chinese deep seek model. They hopefully will get it done with the right protections. But there have been a few models we’ve asked you to stay away from because of privacy issues. So we’ll see what happens there. But copilot. But
Brian F. Tankersley, CPA.CITP, CGMA 24:30
I would suggest to you that that’s probably a move on Microsoft’s part to try to placate their Chinese overlords and try to keep their market share in China because the Chinese government is going to say that you have to use this if you’re going to use Microsoft Office. So yeah, I think the problem is they have too many they have too many overlords they’re having to please.
Randy Johnston 24:54
Well, and you know that’s an advantage in my mind, Brian, because the Microsoft overlords globally. You know, 160 jurisdictions that they’re compliant in, good for them. But I also just want to call out for our listeners that the copilot credit packs right now are priced at $200 per 25,000 credits, and that there are pay-as-you-go agents on the Azure platform and so forth. So all of this stuff is changing very rapidly, but if we’re going to a token-based economy, I believe you’ll start seeing token pricing from the bigs, as opposed to the, you know, let’s charge you $200 and like the old cell phone plans, cap your data when you hit so many tokens. Those of you who have been around cell phones for a while might remember when you got data bottlenecks once you hit your data maxes on your cell phones. That’s kind of what’s happening right now, and we are seeing organizations starting to limit token usage because a rule that they’ve been working on is that your token budget should be roughly double the employee’s salary. So, if you paid somebody 100,000 a year, the expectation is you would pay $200,000 in token fees to make that person more productive. Okay, I get that strategy. No, that’s probably not how we’re going to get it done because companies are wind, big companies are winding up with multi-million-dollar token bills. That’s as granddaughter from New York City would say, “Grandpa, that not be the plan. Well, that not be the plan. We’re going to try to teach you how to take advantage of the token economy and agents and get these workflows right as this technology evolves. So you can see why I wanted to make sure we covered Microsoft Agent 365, give it a little time to soak. So you know it’s been out for a bit. It will change a lot. We’ll probably do updates with you, but at least at this point, that’s the thinking behind it. My thinking may be wrong, but you can get started with Agent 365 today by going to admin.microsoft.com, and you can enroll in either the Frontier program if that’s what’s available to you. That’s probably going to be the case for our Canadian listeners, or you can just enroll directly in it. So it is really a fine environment to work from. So, again, we appreciate you listening in on the Accounting Technology Lab. But Brian, I need to give you a parting shot. Any other closing thoughts for our listeners today?
Brian F. Tankersley, CPA.CITP, CGMA 27:33
You know, I think that this whole thing is changing. And again, the intellectual property you have, you’re going to want to protect, like Randy said. So, this is something that it’s going to be hard sometimes to talk to your peers about. Like Randy mentioned, that he doesn’t, you know, they don’t always share. You don’t always share the super secret sauce, and so I think it’s, I think it’s going to be interesting as firms try to adopt this. I think it does give your your tech forward people a way to shine and a way to distinguish themselves by taking your checklists and turning them into prompts, and by by taking taking tasks that are on the schedule and pushing them into agents. But again, as we’re doing this, this is a new world, and we don’t know exactly what it’s going to cost. We don’t know how it’s going to work. But this is another area of your life where you really need to you need to do some experimentation. You need to try some things out, okay? Because this is a whole different world. Just like outsourcing was 1010, or 20 years ago, we now have agents, and perhaps the agents can displace some of the offshore stuff going on, you know, and I don’t know what the whole thing looks like, but it it’s a it’s a very interesting time, and it’s a time where you really need to push yourself to try some new things and to try some new approaches with AI, so that you don’t get left behind. Because like Randy was Randy and I were saying in our in another podcast we did on scaling new heights, one of the things that we’re observing is that it’s really almost like an inverted bell curve. We have the leaders that are out there way ahead of everybody else, and we have a lot of laggards that aren’t doing anything at all, and the people in the middle are the ones that are rare that are that are streaming from one place to the other, and so we want you. You know, we’re not saying you need to be bleeding edge, but we’re saying that you’re going to have to get off your laurels and you need to try something because the work. You know, the train is leaving the station, and if you don’t get started on this stuff, you know, you’re going to have to wait for another train to come along, and you’re going to be behind, and so it’s time to go.
Randy Johnston 29:43
Thank you, Brian. And that does, you know, trigger one final thought from my side. There are a lot of false prophets out there, and a lot of people that claim that they’ve got AI guidance for you. And I am listening to some of these consultants and shaking my head and thinking. Oh my, this is not pretty because people want to innovate, and it sounds good, and you have a hard time separating the wheat from the chaff. So you even then this year, bad wheat harvest, we need some food here. So Brian,
Brian F. Tankersley, CPA.CITP, CGMA 30:17
and I have to say that it is critical that you get your hands dirty with some of these things, because it’s the only way that you will be able to tell. You’ll be able to evaluate whether something really works or not. Okay, otherwise you’re going to be lost in a in a in a I don’t know what a sea of unfamiliar lingo and unfamiliar tools, and you’re just your context. You’re going to be too far. Your your context is going to be too far off. So it’s important that you try these things out so that you can adequately evaluate the offerings and you can figure out what’s real and what’s not. Because there’s there is there are going to be a lot of charlatans out there that try to sell you things that you don’t need, and it’s important that you have the knowledge you need to be able to manage this new expense that you’re going to have.
Randy Johnston 31:07
Yeah, amen. Thank you all for being with us, and we will talk to you again soon in another accounting technology lab. Good day.
Brian F. Tankersley, CPA.CITP, CGMA 31:15
Thank you for sharing your time with us. We’ll be back next Saturday with a new episode of the Technology Lab from CPA Practice Advisor. Have a great week.
= END =
Sign in to get access to this free resource, and all of our whitepapers and reports.
Download this content today!
Register Now Already registered? Click here to Log In