The Digital Revolution with Jim Kunkle

AI That Works For Accounting Teams w/Peter McCarroll

• Jim Kunkle • Season 3 • Episode 28

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 39:37

Send us Fan Mail

The fastest way to waste money on AI in an accounting firm is to buy a tool that looks great in a demo and quietly dies in the trenches. We’ve all seen it: licenses get activated, dashboards look shiny, and then busy season hits and everyone goes back to the old way because the workflow doesn’t fit, the rollout is fuzzy, and the team is exhausted by yet another “game-changing” system.

We talk with Peter McCarroll, CPA and AI advisor, about what actually makes AI adoption stick in bookkeeping, tax preparation, and client advisory services. Peter reframes AI as a thinking partner and makes a blunt point: buying software isn’t transformation. The real work starts with leadership, clear problems to solve, and pilots that prove we can work the software, not just that the software works. We dig into cultural resistance, staff fear, and the hidden ways “we’ve always done it this way” can signal a firm that’s about to fall behind.

From there we get tactical: how to avoid shelfware with small pilot groups, why “AI and the gaps” is the best place to start (especially when platforms like QuickBooks and desktop tax tools limit integration), and how to measure success beyond time saved by looking at whose time you’re saving. We also explore where tax automation is moving fastest right now around the tax binder and document intake, plus how AI can expand your advisory scope into profitability drivers like sales, marketing, and operations.

If you feel the clock ticking, you’re not imagining it. Peter argues firms have one busy season, maybe two, to get moving. Listen, share this with a firm owner who needs it, and subscribe so you don’t miss what’s next.

Peter McCarroll LinkedIn: https://www.linkedin.com/in/petermccarroll/ 

The AI Accountant: https://theaiaccountant.ai/ 

Fuel Accountants: https://fuelaccountants.com/ 

Download (PDF Ebook) "The Evolution Of Digital Transformation By Jim Kunkle" Here: https://drive.google.com/file/d/1z1NjoP7SMs3w7hwXVHT6mVc3--RNrD_1/view?usp=share_link  

Referral Links

StreamYard: https://streamyard.com/pal/c/5142511674195968  

Contact Digital Revolution 

  • Email: Jim@JimKunkle.com 

Follow Digital Revolution On:

If you found value from listening to this audio release, please add a rating and a review comment.  Ratings and review comments on all podcasting platforms helps me improve the quality and value of the content coming from Digital Revolution. 

I greatly appreciate your support and Viva la Revolution!

Why AI Tools Get Ignored

Jim

Accounting firms keep buying AI tools, and yet inside the office nothing really changes. The licenses get activated, the dashboards look shiny, the vendor demos feel promising. But when the dust settles, the staff quietly goes back to the old way of doing things. Not because they're stubborn, not because they're they don't get technology. It's because the rollout of AI and C inside accounting has collided with something deeper tech fatigue, workflow disruption, and growing skepticism from seasoned bookkeepers who've seen dozens of revolutionary tools come and go. You know the moment I'm talking about, that subtle eye roll when a partner announces a new AI initiative, it's the eye roll that says, great, another system I'll have to learn. Another tool that won't fit how we actually work, and another promise that won't survive busy season. And that eye roll matters because it's a signal that the firm's investment is about to become shelfware. Today's episode is about fixing that. It's about getting AI out of the demo environment and into the real trenches of bookkeeping, text preparation, and advisory work, where deadlines are tight, clients are demanding, and workflows are already stretching thin. And to help us break the cycle, we're bringing in someone who's not just talking about AI adoption, he's actually making it happen. Peter McCarroll, CPA, AI advisor at the AI Accountant and Few Accountants, has spent years helping firms integrate AI into daily operations without triggering the eye roll, the resistance or the quiet abandonment that kills most tech initiatives. Here's to uh talk about what works, what doesn't, and how firms can get finally get AI that actually gets used. Let's get

Meet Peter McCarroll

Jim

into it. Peter, welcome.

Peter McCarroll

Hey, thank you so much. Oh I'm not sure I can avoid the eye roll. I'll just be honest about that. I think that's human nature.

Jim

I appreciate uh having you for this episode. If you could do a favor, maybe provide a little additional background about yourself and also, too, about uh your roles and advisory roles as well.

Peter McCarroll

Yeah, so um I'm a CPA in Canada. I'm based just out of the Toronto area. I'm originally from New Zealand, so I'm also a chartered accountant in New Zealand as well. I've got clients in uh New Zealand and Canada, and we have a US office as well. Um been a CPA for most of my life as well, you know, uh since I was uh in my 20s. So it's quite a while now. Um, and have always been tech forwards. So when the AI, you know, I was always a leader in the cloud accounting movement. And when AI came along, I saw the trajectory, I knew exactly where it was going, and uh, you know, was able to understand it and gravitate it uh to it pretty well, which I think is one of my biggest strengths is that you know I understand the tech, even if I'm not a techie, uh, and I'm able to translate to the world of accounting.

AI As A Thinking Partner

Jim

Perfect, perfect. And let's I want to get some of your perspective, especially related to you know, looking at you know what AI really means for accounting workflows. And one thing that I've noticed and really picked up from doing research on you and the roles that you play is that you know, your view is that AI is a thinking partner for accountants, is it's not a shiny automation tool. You know, why is bookkeeping, tax preparation, and advisory, why are they so uniquely suited today for AI augmentation?

Peter McCarroll

Well, I think it's a case of not either or, but both. Um AI in the bookkeeping space in particular is is so uh you know influential to us today because a lot of what we do is very repetitive. We you know, bookkeepers in particular, uh, you know, I mean it's a reasonably boring, you know, repetitive job because you know it doesn't matter whether you've got one client or 50 clients, you're doing the same thing for every client. You know, the the context changes, but it's that repetitive nature. And this is where AI augmentation or AI replacement, I don't like using that word, but it is it is a reality, um, is is is able to make massive changes because a lot of what we do is just the repetition, the following of the rules. Yeah, we're doing the same thing over and over and over again. And that's where AI can you know can can really help. And yeah, it's a blessing and a curse. I always say AI is the thing that's gonna kill your business, it's also the thing that'll save your business. Um, because it's gonna take the monotony away and elevate us as advisors back to the position that we actually put away when things got too complicated of how do I actually help advise my clients rather than how do I just get through this tsunami of transactions so I could get some reports out and file that darn tax return? You know, the tax return is never the objective. It's a byproduct. It shouldn't be the objective. The real objective should be helping a client run a profitable business. Tax return just has to be done because you don't do it, you get in trouble.

Demos Versus Busy Season Reality

Jim

So when we're looking at the world of accounting, you know, what's the difference between AI that demos well and AI that survives the busy season?

Peter McCarroll

Yeah, well, there's a lot of that right now. AI that demos well. I've never seen a bad demo. Um two main issues. One is what problem are you trying to solve? All the the vendors will tell you that it solves every problem, but they don't solve every problem. They solve a very specific problem. And if you try and buy a product thinking it's going to transform your practice, you will end up with a you know a failed or a bad implementation. You'll get part way, but you won't get to where you thought you were going to get. And buying a tool isn't transformation. Buying a tool is a tactic, it's not a strategy. So that's probably the biggest reason is that people don't understand what are the problems we're trying to solve, and then what tools help us solve those problems. Most people go into it the other way. I've bought a tool. Now I've got to help, I've got to use it to transform my business. It's the wrong way around.

Jim

Yeah,

Staff Fear And Culture Pushback

Jim

it's definitely the wrong way around. So let's talk about some of the core barriers uh to adoption of AI inside accounting firms. And the first I want to talk about is the cultural resistance. And this, these really, these barriers kind of they they also bubble up to other uh other industries, other sectors as well. So the cultural resistance is typically takes the form of you know staff that fears that they're going to be replaced. And then the other aspect is it's like, well, we've always done accounting this way, and this way works forever. Why do we need to change? You know, how do you deal with that core barrier of cultural resistance in an accounting firm when it could be a good one?

Peter McCarroll

Yeah, I've actually written on both of these topics just in the last couple of weeks, funnily enough. Um, so the first one is staff fear. This is a very real issue. Your staff know the AI is coming, and they don't, and if they don't know what your approach to it, this goes for any business, not just accountants. If they don't know what your plan and approach is, they will be fearful. Now, the great news in the accounting space is we have such a staff shortage that I don't think any good bookkeeper or accountant should be worried that they are going to be out of a job. Okay, because we have such a shortage. AI is going to allow us to fill the gap without having to let people go. That's my belief. Um, yeah, can we do more with less once AI is fully implemented? Yes, of course we can. You may end up having to move somewhere else. But there are certainly firms that are growing tremendously, uh, even in the midst of this AI boom. So I'm not personally that worried that we're gonna see a whole bunch of accountants sitting around twiddling their thumbs and and you know, unwillingly drinking my ties on the back porch. Um, I think there's plenty of work out there. Uh the question is, how do I get you from where you are today to where you need to be in the AI world? That is actually the bigger challenge. And the reality is there will be some people that won't make that shift. That is the challenge of leadership and people management in the AI world. So some firms will drop headcount and they'll do that intentionally, or they'll do that because they don't want to grow revenue, or the revenue is shrinking, which is also a real problem. Um, but many other firms will be hiring staff as fast as they can. Okay. The key thing for any person is rather than see this as a threat, see it as an opportunity. How do I learn this new way of working, this new tool, whatever you want to call it? How do I take advantage of it? Even how do I solve problems with it? That's the mindset that you have to have. How do I solve problems with it? If you can do that, you will you have a job. There is no question in my mind, as an accountant or a bookkeeper or someone in the profession, that you will ever be out of work. You will have a job. The people that won't have jobs will be those that refuse to engage with it and want to just keep doing it the old way. And that brings us to your second point. We've always done it this way. Um, this is probably the hardest, and this is an indication of people that are not going to make the transition. The reality is we can no longer do it this way. So those that say, hey, we've always done it this way, uh, you know, it's it's like people saying, Well, we we've always fed the horses and we've always done this. The fact that we're no longer taking the buggy out for a ride anymore is is irrelevant to them. Okay. The buggies changed. That we have to acknowledge that. We're no longer in the horse and cart era. You know, this AI movement is going to change the work. And it has to change the work. And there's all sorts of pricing pressures uh that we could talk about that that tell us why this is now inevitable. Um, so it's not a question of we've always done it this way. Why do we have to change? It is we have to change. Um, are you joining us on that journey? And I told my staff right from day one, no one gets in trouble if we're using AI in our firm. And we want to bring you all with us. End of story. That's the commitment we've made.

Workflow Mismatch And Tool Churn

Jim

Excellent, excellent. And so the uh another barrier is uh workflow mismatch. And what I mean by that is that, you know, in right now, in the current stage of AI as a disruptive technology and its adoption, and really still its its evolution, you know, there are a lot of areas where you know AI tools and resources don't always fit everything on a daily for a daily basis. And also, too, there are steps that are more involved with AI to get AI to function. Uh, and so people need to learn prompt engineering, they need to learn a lot about it. You know, how do you deal with you know today's AI where we're at um when it comes to some of the mismatches that might happen with workflows in an accounting firm?

Peter McCarroll

Yeah, well, just briefly, the tools that don't fit the daily grind you mentioned, um, that's a selection issue. And that comes back to knowing what problem are we trying to solve. Um, and of course, what we have to remember is that we are still incredibly early in this AI journey. And there will be a lot of tools that don't make it. And that's just unfortunate. And the answer is not, let's wait and see what stabilizes. The answer is we may be churning through a few tools to find the models that work the right way for us. It's a cost of doing business these days. The world is getting faster. You can't slow down, or you get hit from behind. Okay. So when you have a tool that doesn't that doesn't match or doesn't fit, go back and refine your selection criteria. How do we choose that tool? There have been a number of tools uh that you know have left people with with disappointments and sour tastes in their mouth, particularly last year. A lot of people hyped the AI stuff but couldn't deliver. That world changed less than six months ago. So it's it's a different world now. And I think the capability of tools is so much better than they were a year ago. So hopefully that's less of a problem. But as long as you think about what you're trying to solve first rather than jump into buying a tool. Okay. Uh and the problem with the tools is many of them solve in uh you know one issue, but not other issues, or they sandbox you in. You know, in the accounting world, particularly in North America, QuickBooks is is the is the elephant in the room. I'm not sure QuickBooks will survive. But so many tools only work in that space. And I that those are the ones I'm worried about. I don't see them as being viable because we're going to be in a world where you can no longer be a QuickBooks-only accounting firm. Now, this doesn't translate to other businesses, but in the accounting space, this is the big risk. And I think you know, we're we've always we started with zero and then we we brought in QuickBooks. We are going to have to be adding in new products into our repertoire because you know these new products are going to eat the dust of uh or leave dust behind, I should say, you know, but for for QuickBooks and Zero, because they are just so AI native that they can move much faster than these legacy products can move. And so I think you know, as an accounting firm, we have to be willing to now open the doors to new products because we want to be in that space.

Jim

Yeah, you definitely want to be in that space, especially what the future of competition looks like. Yeah.

Leadership Drives Practice Transformation

Jim

And uh probably, you know, out of the four core barriers that I kind of want to focus on, I think the next one is the one that's the most critical because it can really make or break uh break a firm or a company, and that's leadership mistakes. You know, we've seen for too long that leadership buys into platforms without really proper rollout plan or execution of uh of uh of the uh of the the program and everything like that. So I think also, too, not having a mandate, um not properly training and not properly integrating uh the AI resources into it is a big mistake. You know, when it comes to the leadership side, you know, what are some of the things that uh uh you recommend or talk about uh regarding uh avoiding this barrier?

Peter McCarroll

Yeah. So first of all, it's really important to acknowledge this is not a technology issue, it's a leadership issue. This is about, and one of the things I often talk about is practice transformation. The issues that we're trying to solve now are not how do I bring a new tool into my practice in and of itself. The issue we're trying to solve, or we should be solving, is how do I survive? How does my practice survive and thrive in this new world? That's leadership, not technology. The technology is the tool, and in some cases, it's the thing that's forcing us to have this conversation, but the technology is not the thing we're trying to do. We're trying to transform our practice so that we can uh thrive in this new environment. Those people that don't do this, and this goes for any small business, any business, if you ignore this, you will be one of those businesses likely that closes. Because you can't you can't ignore it. So we have to start with that knowledge that this is first and foremost a leadership issue, not a technology issue. If we jump into the technology first without having the harder conversation on leadership, we will be in trouble. Okay, so we always start with that. I run a program called Practice Transformation. It's it's 12 hours, no, 16, yeah, 12 hours long. And we barely talk about AI in and of itself. It's all about leadership and transformation, pricing, HR, decision making. Yes, we've got some prompts and things in there, but it's all about that leadership aspect before we get into the how do I train my people on on using Claude or what tools should I select? We've got modules on that. But you know, so so that's first and foremost, that's what we do. Then the second thing is we still have time. I think that from an accounting perspective, we have one busy season, maybe two, before the world will have changed so fundamentally that we we can't afford to be late. That means that we don't have to solve the problem today, but we should be piloting and moving in that direction before the next busy season starts. So this gives us time to trial and implement things. We shouldn't try and jump all in on a new AI platform and try and move everybody over to it until we've done the diligence work of making sure that we know exactly again what problems are we solving, where are the barriers? Let's take a middle case, an easy case. People want to go to the hardest cases first. It's like, no, that's that's a recipe for disaster. Go to your your average case and pilot with that. Prove that you can, and and this is this is something I was just writing in my in my transformation uh course the other day. The point of a of a pilot is not to test that the software works. The point of the pilot is to test that we can work the software. If we can't work the software, if we pick the wrong trials, if we do the wrong stuff, it's not the fault of the software, it's our fault. So the first pilot is always to test can we do this? Then we evaluate the software. Did it do what we expected, didn't give us the results we wanted. But most of the time, 90% of the fault is with us, not with the software.

Jim

Yeah,

How To Avoid AI Shelfware

Jim

and that brings me to the final core barrier for adoption. Earlier I mentioned about the you know, AI resources and tools can become the shelfware. And what I meant by that is that you know, they're they're added, they're purchased, subscriptions, however they're acquired, and there's really no plan on how to adopt them. And then they come up for renewal, and there's regret about, well, why did we do this? And then they're not renewed. You know, I think that aspect of the potential for AA to become shelfware. Uh, you know, what recommendations do you have for firms to avoid this type of a of a barrier?

Peter McCarroll

Yeah, it's unfortunately it's actually quite difficult because a lot of software now requires a one-year commitment. You know, gone are the days where you can just pay monthly and decide to switch it off. Um, you know, I always advise people to start small, start with a pilot group, start with some pilot clients and grow into it. Don't try and dump everything into your new software all at once. That's sometimes difficult because if you're paying for the whole firm for you know on a 12-month contract, your inclination is to just move into it as quickly as possible. Especially as we're, you know, you come up to the next busy season, time is short. You know, in the accounting world, we really only have a few months of the year where we can make these kinds of changes. So it is very difficult. If you're not in the accounting space, God bless you, you've probably got a lot more flexibility here. Accounting rhythms are a little different. Um, but yeah, go a little slow. I mean, the inclination is to want to go fast. Uh, prove that it's working first. Uh, take the monthly pricing option if it's available, even if it's 20% more expensive, until you've proven that this is the right solution for you and that you can implement it. Once you've proven both of those things, then you can say, yeah, we're now going to roll this out to the whole team. Now you add more seats. Now you commit to the annual fee rather than the monthly fee. Uh, that's where you lock in the savings. But if you go too soon, what you end up with is missed opportunities uh and stress on your team.

Start With The Gaps

Jim

Now you perfectly led into the next focus I want to talk about is you know what actually works, you know, getting AI into real booking workflows. And you talked, you said, listen, you know, get those micro wins. Start small, don't go firm-wide in a total overhaul. Focus in certain areas, maybe some existing systems, look at certain aspects of the firm that you can add AI into it. Uh, obviously, you're gonna monitor it and to see its use and everything like that. Now, what are some of the uses? Let's say uses within the firm that uh uh a new adopter, they're adopting AI, and what are some of the different areas they might want to consider doing it when it comes to the accounting functions?

Peter McCarroll

So I'm gonna exclude third-party tools for a moment here. One of the challenges in accounting, this doesn't always apply to every industry, but it really does to us, is that we spend a lot of time working in our core tools, QuickBooks or Xero or Sage, whatever your accounting platform of choices, we spend a lot of time working in that tool. Now, unfortunately, those tools are trying to build stronger and stronger. Uh uh, I call it a walled garden. They are they are in implementing AI in their tools, most of them not very well. Um, and the the ability for us to integrate or interact with that from the outside is limited. They have APIs, they have MCPs, but they haven't designed them in ways that actually work for an accounting practice. I call this problem the AI and the gaps problem. We have very limited ability to use AI to automate or interrogate uh the data in the accounting system. We've got some scope, but it's it's harder work. The first thing I tell accounting firms to start with is look for the gaps between the systems. Don't try and go in and automate a monthly close in QuickBooks. It's just not going to work. Um, you're not gonna be able to automate everything the way you want to. Um, what you want to start with is looking for places where there is no interaction with the uh core accounting system, data transformation, messaging, preparing of documents. Do all of that first. Get used to using the AI tools without touching those core platforms. That's level one. Level two is where you have to carry data in or out of the accounting platform, generating reports and putting that into an AI tool. Uh, having an AI tool generate a CSV file that you could import into the accounting tool. That's level two. Now you're starting to cross that boundary. And then level three are tools or API MCP interactions that now allow you to take away that human interaction step where someone had to carry the file over the threshold. We can automate that, but that should be the last thing we try, not the first thing we try. And how do you measure success of that? Obviously, we're looking for time savings. That is the number one step. Now, the reality is most of this automation will get us 10 to 20 percent. What we really have to look for is 60 to 80 percent. And unfortunately, most of that's going to mean choosing different platforms. You're only going to be able to automate QuickBooks and Zero to a certain extent. Um, but time savings are the number one issue. However, there is a hidden trap. If all we're doing is counting hours, not the value of those hours, we could actually end up going backwards, not forwards. I'll give you an example.

Measuring Value Beyond Time Savings

Peter McCarroll

Uh, I built a uh script uh or a skill in Claude to do a file review for a year-end file. It cost my team more time. Because what it did is it pre-reviewed the file before it came to me and it asked them to do a whole bunch of work. They don't like it. But it reduces my workload. And I would give three, I would give away an hour of my time and have them take on three hours of more time any day of the week. Okay, because my time is is more valuable. So time savings in the absolute sense isn't necessarily our goal. It's the efficient use of time and whose time are we using? An AI tool that puts more work on your more expensive players and less on your less expensive players is a bad implementation.

Tax Automation Around The Binder

Peter McCarroll

So tax, particularly in the US, but also in Canada, is uh somewhat problematic. And again, this comes down to the AI and the gaps problem. Most of our tax software is desktop based. And because our tax systems are so complicated, uh, you know, firms, uh software firms have been very reluctant to migrate to cloud-based systems. There is almost no tax system with an API, period, in the in North America. So what we're seeing in the market is a lot of firms putting AI in place around that around the tax return. But what they're doing is they're putting AI around the document gathering and the document evaluation side of things. Uh, you know, the the phrase that most accountants will use is a tax binder. It's a checklist of all the things that you as a client need to tell us and all the forms you need to give us in order for us to do your tax return. Okay. So there's a lot of automation happening around the binder, the data gathering, the data ingesting. And then what it does is then it passes that off to tax software. Uh, some of them have some automation options. There's a lot of desktop control that's happening right now, um, you know, with apps and uh and some of its import files and things like that. But all the innovation right now is happening in that data gathering side, not in the actual tax calculation side. I'm waiting for someone to say it's time to rebuild tax software on the web. There's a product in Canada that does it, but again, no APIs, you know, there's no real way to integrate with it, um, which is unfortunate. But uh the net the first the first person that builds a really good tax software on the web will probably get a lot of business.

Jim

Yeah, definitely. Now, onto the aspect of accounting advisory workflows.

Advisory Wins Outside The Numbers

Jim

Uh AI obviously is going to be a strong strategic co-pilot. In what ways can it be a co-pilot on the advisory side?

Peter McCarroll

I think there's huge potential here. You know, the the standard answers are things like um, you know, monthly reports, analysis reports, cash flow forecasts. Okay. AI can do reasonably well at that. The key for that is making sure you've given a really thorough context file to the AI so you don't get this generic slope of your revenue went up and your expenses were did not change. You had a good month. Or, oh, your revenue went down last month. You know, when we know that in in June revenue always goes down because it's your slowest month. Okay. So having good context is what's going to turn uh mediocre AI results into the AI results that you might have set because you knew the client. So that's that's key number one. Here's the other thing that I don't think enough people are really thinking about. As an advisor, as accountants, we have historically limited ourselves to being able to advise and coach our clients on purely financial and accounting matters. The problem is our clients don't need coaching in just that area. They want more profitability. Well, what's involved in more profitability? Well, it's sales, it's marketing, it's operations, it's human resources. There's a whole bunch of areas that impact their profitability. And most accountants have so much experience that we are actually quite qualified to speak in those areas, but we often don't. And the great thing with AI is you no longer need to be an expert to be able to help a client walk through particular challenges in those spaces. I'll give you a really great example of this. I have a client who bought a locksmith practice. Now, he's not a locksmith, he's an entrepreneur. So he bought this business and revenue was starting to go down in year two. He didn't know why. So we started talking through it. And I asked him about his ideal client profile. Now that's not a question most accountants would think to ask. But I asked him, you know, tell me about your ideal client profile. Let me umd any odd. And so I said, let me give you a few prompts. I had the prompt library and I had an ideal client profile from the um uh from a couple of marketing specialists. And so I said, let's run one of these prompts and we'll pick a profile and we'll work on it. So he ended up working out three ideal client profiles. And when I met with him next, he had taken this is now all on him, not on me. So I can't take credit for other than the initial question. He now took those three ideal client profiles. He had worked out how he could train his staff to do a security audit tailored to that profile and come up with upsell ideas. And his revenue two months later was now up 20%. And his teams were full. All from one question that I asked. What's your ideal client profile? Let's get AI to help you with that.

Jim

The power of a question and AI. Exactly.

A Practical Implementation Playbook

Jim

Let's talk about implementation. Obviously, that is super critical. You have a playbook for real adoption of AI and on the accounting side. Could you go through some of the steps that you recommend to implement these uh the technology?

Peter McCarroll

Yeah. So again, like with everything an accountant ever says, the answer is always it depends. Um, you know, every practice is different. So, you know, what I tell one practice may be different from another practice. But but generally speaking, we want to look at the big picture questions first. Where are we going? What problems are we trying to solve? Are we a bookkeeping practice or a tax practice? That's gonna take us down different paths. If you're a tax-only practice, software is probably the right answer. If you're a more of a general purpose CAS practice like me, one software tool is not gonna solve all your problems. Okay, so we probably want to be thinking more about uh, you know, using more generic AI with our own context system in place. We call it a practice brain, the practice brain in place. So, first step is obviously identify some areas that you could have some key impacts in. Okay, pick one workflow, pick one area that, again, not your most complex, not your least complex. Uh, you want it to be something that you can run several reps on over a two-month period. So it can't be something that only happens once a year in January. That's not a good example. We want, you know, if you're an American practice, you know, year most year ends are December. So picking something in your year-end workflow isn't a good uh you know, pilot right now because you're not doing year-end workflow. So it's probably something in your monthly or quarterly cycle that you know that in the next two months we can have at least five or six of these happening. Okay. You obviously got to pick a client that you know well enough because one of the things I get people to do is we write a client context document. Okay, we've got a process for creating a client context document. Then you're gonna trial this on two or three clients. And every time you run this workflow, you are going to evaluate the workflow. And the first time you run it, it will take longer than it would have taken manually. We just state that up front. Don't expect us to solve problems the first time you do it. But the second time you do it, it should speed up. Hopefully, it's no worse than what it was doing it manually. And the third time you run it, and every time after that, we're looking to see is it actually getting faster? Is it getting better quality? And sometimes with AI, the goal is not always speed, sometimes the goal is quality. AI can help us improve quality or it can help us improve speed. Not always both at once. Sometimes, but not always. So we have to be willing to understand what is the problem we're solving with this activity. Is this to improve quality? That's a valid pilot, or is this to improve speed? Okay. Once you've done that workflow, and you can repeat many of these simultaneously, you just don't want to overdo it. Uh, you know, now you've got the you've got the the plan, you've got the executions, you've got the review, you've got a particular date, you're going to review this. Now you come back and you look at that and say, is this now ready for me to roll out to more than these pilot clients? And that's when you start thinking about implementing it across other staff. What dependencies do I have? Have I got those client profile documents, those client context documents? If not, that's your first step. Now we roll out the context document preparation step to the next rate tranche of clients. Then we start with these workflows and then we start implementing them and tailoring them into that client's environment as well.

From Rear View Mirror To GPS

Jim

Very good, very good. Now let's talk about the future. You know, what AI-enabled accounting firms, what were they what are they going to look like in the future?

Peter McCarroll

I talk a lot about uh motor vehicle analogies. Um historically, accountants have been uh uh rear view mirror operators. Okay, we're telling you what happened in the past. Sometimes a year late, sometimes a month late. Okay. Um, but that's been our focus. You know, monthly reports, month end close. It's always backwards looking. Um, you know, annual reports and tax returns, they're always backwards looking. AI is going to allow our tools to do almost all of that work in near real time. We have to move, and I believe even by the end of this year, we will be at a position where dashboard, not rear view mirror, will be our focus. What is happening today? Where is my business today? Not last year, not even last month. Where are things today? That is the first goal, particularly of a bookkeeping practice, is to say, how do I move from the rear view mirror to the dashboard? Zero day close or one day close. Okay, not 20-day close. How do we get this stuff done on a continuous basis? That's not the goal. The goal is to become the windscreen with GPS and heads-up display. You know, where are we going? Where is where are the roads? What is our destination? Where are the pitfalls? Where are the police on the side of the road with a radar gun? You know, where are the traffic delays? Okay. That's what, as accountants, I believe we should be focusing on. That is our goal. That's where we should be. We want to be the windscreen, not the dashboard, not the rear view mirror.

Jim

It just sets everything up for

One Busy Season Left To Start

Jim

success. So, in closing out this episode, Peter, you know, you've provided so much detail, really great advice, recommendations for firm owners out there. Is there any additional advice that you would that you give or you would, based on this episode, want to give to firm owners who are watching, uh listening uh to this episode?

Peter McCarroll

I'll come back to what I said earlier. It is my firm conviction that we have one busy season, maybe two. And for a North American accountant, that's January through April, right? Okay, one busy season, maybe two, before the world as we know it will have fundamentally changed. You do not have time to wait to start. You have time to wait to implement, but it's not long. Okay, if you don't start this year now, you will be behind. Um, the AI CPA have published their roadmap 2040. It's a 15-year plan. Um, I'm not an AI CPA member, so I guess I could speak a little more freely about them. That is crazy. They I was at the AI CPA engaged commerce, and they were talking about the firm of 2030. I think 2030 is too late. By 2030, we'll be into the next thing. Okay. Um, the firm of 2028 is what I'm focusing on. What do we look like in 2028? And if we're and if firms haven't started doing that hard work of thinking about what our firm looks like in 2028, they will not be able to be there. And I think there will be firms that will probably slowly go out of business because they waited too long or they were unable to make that transition. So, right now, the focus is on you've got to be thinking about your practice transformation. That's why I'm teaching practice transformation programs, not recommending software. Okay, that's what I'm doing. So that's my big piece of advice. Don't wait. Start working on this right now. Uh, you know, yes, you can start trialing some tools. Yes, you should be piloting things. Uh, next year, the heat goes up. And if you haven't started at that point, your feet are going to get burned.

Jim

Yeah, and when it comes to practice transformation, I highly recommend that. And I will have uh links uh to uh Peter McCarroll uh for everyone to be able to connect with it and with him, and also to uh you had mentioned you publish articles and then you have probably a lot of other things that you do as well for information and knowledge sharing and everything like that. So I would recommend everybody to highly get engaged with Peter uh and get connected with Peter. Research definitely into this uh topic. And the other thing I ask of everyone with this episode is that if you know someone who would get value out of listening to this episode, please forward it to them and to have them take a listen to it as well. Peter, thank you so much. Everyone, have a good rest of your day, your uh evening or morning, no matter where in the world you're listening to this episode. And we'll catch you again on the digital revolution. Thank you, everybody. Bye.

Head Shot

Jim Kunkle

Host