609 6 Blockers to Adopting AI In Accounting Firms [And how to build an AI native accounting firm]

My friends, we are in the thick of my workshop tour right now. I'm visiting 20 cities in the US, throwing half-day parties, continuing education, snacks, a dinner, an open bar. These are events that I have built with my team from scratch. We host them and we pull in 50 to 150 local accountants who all want to hang and hear a couple of talks from me and meet each other and just vibe, you know. These are unbelievably fun events. We are in the middle of running them right now. If you haven't checked out the tour schedule yet, please do head over to JasonOnFirms.com/tour. /events to see the list of all the cities. If you know somebody that's in one of those cities, give them the heads up. It's a super fun hang and I want to see you there. Okay, on to the show. Enough about your workshops. Let's get on to the actual episode. I think most of us can agree that AI is the biggest existential threat to accounting firms in the history of accountingdom. But for whatever reason, we're struggling to, I don't know, lean into this more. Like the right amount of investment in AI should probably feel uncomfortable, should feel expensive. After all, biggest ever change. But a lot of us are still stuck, and that's what we're talking about today is what are the 6 most common blockers to getting your firm to become AI native, something we all know needs to happen. Also, how do you get your staff to stop carelessly speeding through work? How do you pick a practice management system? How do you feel when a client keeps complaining about their price? And will AI kill after-the-fact accounting and tax work? Plenty of ground to cover today, friends. Come on in. Let's do it. The Dirty Dozen. How do you do, fellow accountants? My name is Jason Staats. Welcome along to the Jason on Firms podcast, where we talk all things accounting firms every Tuesday and Friday. Super fun hang, okay? People just coming out trying to figure out how to run better firms, whether you run one, whether you work at one. One of the, the only places in the world where it is cool to geek on this stuff. We are midway through the Dirty Dozen series. These are the 12 biggest issues ailing accounting firms, as sourced by the over 1,000 accountants that came out to my live events last year. Most of us, we are struggling with the same problems. On the food pyramid of things that ail accounting firms, these are the one on the bottom. What is that? Brains, breads, and grains. How could bread be the foundation of your diet? Anyways, today we're getting into the struggle that is how to get that firm going with AI. Easier said than done. And in 2026 feels higher stakes than ever because the AI is gotten good, man. In fact, if you do tax work, like If you went heads down in December into busy season and emerged in April, my friend, you are emerging in a completely different universe because the models have gotten so good. The model harnesses have gotten so good, and we now have AI that's capable of like doing production. Not this cute little assistant on the side, but AI agents that ought to be like the first place work gets done. Big mental shift here. Imagine AI being the first place work gets done instead of your humans being the first place work gets done. That is the big, like, seismic shift we are undergoing right now. But let's dig into the 6 reasons that firms are getting stuck and give you some practical tips on how to overcome these issues. Number 1, first and foremost, the "where do I start" trap. You know it all matters, but you're paralyzed by the number of tools, by not being able to find a clear entry point. Here's what this sounds like. Actual folks have told me this. We're trying to hone in on an AI plan to calm down all the hype and formalize our strategy. We're stuck on deciding what AI tools to use or develop ourselves. I really want to know what AI tools I can use to make my job easier, but I don't know where to start. I have good news for you, my friend. At this point in time, in the year of our Lord 2026, what are we? We're almost to June. The answer has never been clearer than it is today. The answer right now is Claude, C-L-A-U-D-E, Claude. It is the most performant AI agent tool that we have ever had. Also, Anthropic, the company behind it, 80% of their business is enterprise. So it is made with kind of security at its core. And so will the rest of your accounting tech stack implement AI in some way or another? Yes, 100%. And really building your accounting firm's tech stack can almost be thought of independently from how I build AI agents that work across my accounting firm tech stack. This is almost an entirely new layer of software. You have your legacy software stack, your practice management system, your tax software, your accounting ledger, all that stuff. Where Claude is novel, it is, it is a tool that works works across those apps, not unlike ourselves, Intrepid Accountants. We work across all these different apps. We are like human integrators. Claude is exciting because it too can work across all the apps that we use. Now, that is tech selection, but where to start? The hard thing about learning to use a new tool like this is you don't know what it can and can't do. Every other piece of software you've ever implemented and rolled out in your accounting firm, somebody gives you a big old list of features. It has all these buttons and the button does the same thing every time you click it. That is not the case with AI assistants. And because these tools, like these AI tools, know how to code and can use open source libraries and all these things that I can't even guess or understand, then it means I'm not able to guess what it is going to be capable of. So what do we need to do? We just need to spend more time with it. Need to bring like that thing you're working on to it and be like, here's what I'm doing. How can you help? And there will be times where it amazes you. There'll be times where it is pretty disappointing. But when you find a good use case, have it documented. Have Claude create a skill. A skill basically makes that workflow kind of reusable. Even share that skill with all the other Claude users in your organization. That is how you work out what, what can it do. And there is like the version of this where you are just working in the afternoon and you're like, ah, I don't really wanna do this thing over the next 40 minutes. Maybe Claude can help. But there's a whole other version of this where you sit down, you get the big brains together and you say, this status in our workflow or these collection, this collection of tasks, is there a world where we could get AI to do this stuff for us? And that is more thinking about AI as infrastructure, AI that just happens in your firm. It's in the walls. That is time worth spending and conversations worth having as well. But you won't be comfortable having those bigger, more ambitious conversations until You just spent some QT with it, some quality time. All right. Second thing that is blocking accounting firms here, the champion vacuum. The idea that somebody's got to take the lead on this and I don't have time. I already am in over my head. Maybe we just go out and find the right hire, the accountant unicorn that's really good with AI too. I'm sure they're out there. They're not. And this whole like waiting for the champion thing, you can't put all of your plans on pause until you find it. In fact, even if you did go out and find that champion, What about the rest of your team? Does that mean that they don't have to learn AI now? Of course not. This is like an entire team thing. Listen to a few of these things from actual firms. I need a dedicated staff member who can lead AI and automation. They could pull in a few members of the team to support them. This person/team would handle R&D and lead implementation. This makes AI sound like this machine that's going to sit there and it's got a couple of maintenance workers and everybody else still does their thing, but this machine is churning along. That's not how this works. This is something that everybody has to be using and learning. Another quote, "I'm looking for a champion since I do not have the time." Another one, "I need them to continue to flag processes that we can automate. I'm not necessarily in the weeds anymore, so I can't always identify areas we can streamline." This last one is a great callout because many of people's issues are unique to them. And historically, tech has always been framed as it will solve the problems of the masses, but not go further than that. After all, this is like off-the-shelf technology, right? It's not really being tailored to our needs. And so if even if you feel like you're not in the weeds of production in your firm, know that there is always a weedier weed. And even the people who are in the weeds, they're not in every weed. And this is why AI is an everybody responsibility. Now, there is a certain level of like who is bringing the enthusiasm to these conversations inside of the firm, and you need that. If you are the sole source of energy, that can deplete your batteries so fast. And I'm speaking from experience here, but all it took for me was like a few energetic folks that I could get into the mix to sort of distribute some of the load of forcing the excitement into everyone else. These are accountants after all that we're talking about. So could you find this hire? Potentially. Hopefully you've already got somebody with a level of this enthusiasm as it is and we can enable it. Not so that they can run off and tick the box of doing AI and so everybody else is like, oh good, it's that person's thing. It's not my thing. That isn't our goal here, but it is someone who can be the cheerleader and like share use cases and sort of drive some of the energy. If you have that champion, you need to get them connected with other champions. What I would be doing right now is I would have at least one of those people in Automation Town. Just Google Automation Town. It's an AI community run by Chad Davis. It's got like, I don't know, 300 or 400 people in there. And they are sort of folks that are really trying to build not only with AI, but technology in general, specifically inside of accounting firms. You throw a member of your team in there, they're going to get an abundance of inspiration that they can bring back to your team. But this inspiration needs to be built into the processes of the firm. Firm. Like, what is the feedback loop? How do we ensure that this is always happening? Otherwise it gets forgotten because we get busy with everything else, right? For the last couple years, I've been saying people need to meet 30 minutes a week with their team to talk about AI, like meet live and share what worked this week, what didn't work this week. Now I would argue 30 minutes a week, honestly, not enough. Like, when we've got this AI agent sitting on the sidelines that can literally do our jobs, is 30 minutes a week talking about it Enough? I don't think so. My take here, the expectation, honestly, I would be looking for something more along the lines of, hey, I need everybody sharing like what they used AI for in Slack and Microsoft Teams every single day. Like, what is one way that you used AI today? Because we're trying to build as much surface area as possible for the rest of the team to get an abundance of ideas of ways that they can use it. And the faster we're sharing this stuff, the faster you're kind of laddering up that skill set and the next use case becomes possible. It has to somehow be built into the weekly cadence of your firm. Otherwise, it just— it won't happen. Right? All right. Third blocker here is hitting the adoption wall. You bought the tools, but your team's not touching them. Some people take to this stuff so fast and it's easy for them. And the first time they saw ChatGPT, they were like, where's this thing been all my life? But then there's a whole other side to this where there's a lot of folks when they first saw ChatGPT, they were like, "Okay, tell me what to do. I need training. What's next?" If this came naturally to you like it did to me, you're like, "What do you mean what's next? Anything is next." But then that person is like, "Yeah, no, sorry, I'm still gonna need you to be more specific." We can't treat AI like any other software rollout because it's not that on rails. But I would argue the biggest reason why folks don't make use of this stuff once it is in their hands is that their incentives are not the same as yours. And this has always been the blocker to change management inside of accounting firm. This has always been the change management issue. You see something new and you're like, this is going to save us so much time. And then you share it with your team and they're like, meh. And you're like, well, how are you not as excited as me? Well, it's because they are not. If you're like a leader, an owner in a firm, it's because your team is not equally motivated to line your pockets as you are. And I experienced this in my firm when we roll out like a very basic production, a version of production-based pay, when they have skin in the game and it actually impacts them when they are more productive. I mean, everything changes real quick. But in most accounting firms, we are employing people to sit in a chair a certain number of hours a week. And, you know, number of hours does not really equate to output. There's not a 1-to-1 ratio there. And so as long as a team member is meeting this hypothetical bar of what the sort of general expectation is for what they will get done, then I think what we have to do is plan on that person spending the next 40 hours in whatever the most comfortable way is that they can spend that next 40 hours. And that is almost certainly not changing the way that they work and learning new tech. Now, you may not have been wired that way. I wasn't that wired that way. But when you're employing accountants, you almost have to assume that will be their mindset until you build a system around them that will drive a different sort of behavior. And if they are incentivized to get more work out the door, more correct work out the door, and we did a whole, a whole week on production-based pay on the podcast. Easiest way to find it is you go to the YouTube version of the podcast, go to playlists, and you'll see a whole week on production-based pay where we interviewed a bunch of firms that do this. But if you are excited about using AI to get more work out the door, consider what's in it for your team to get more work out the door, especially if a lot of the narrative around AI is that it is going to do your team's work. Not exactly an incentive for your team to get excited about it, right? As I read this ad, I am feasting my eyes upon talented accountants that you could literally, literally book an interview with right now. Like Ruru, a CPA, a tax pro, $3,500 a month. He's got experience with CCH Access, bless your heart, CCH Engagement, QuickBooks Online. Gang, he is one of many accounting pros that you can peruse at the Cloud Accountant Staffing portal. We'll link it down in the show notes. It is a veritable smorgasbord of accounting talent, folks that they have pre-vetted where you can go in, you can see, see their experience. You can literally see an intro video that they've recorded and you can book an interview with them without ever even having to hop on a call with Cloud Accountant Staffing. Amazing. We got tax pros, bookkeepers, auditors, executive assistants. Oh my gosh, I got a lot of people in here right now. 45 candidates. Because I know you need that next dream hire. You really do. What do you got to lose? Okay, go check out the portal. Book an interview. Feels like a win-win, right? You're only going to move forward if you find somebody you love. Cloud Accountant Staffing, run by folks that used to run an accounting firm. They make it as easy as possible for you. To learn more, check out the link down in the show notes. Hey, if you're done licking your wounds from tax season, if you run more than a few hundred 1040s a year, you got to look at this episode sponsor. Sorban, one of my leaders in tax workflow automation, can save you a tremendous amount of time right now, specifically on the intake part of 1040 workflows. But they're also working on connecting to your tax software automatically, like pushing data in there. They're working on automating the delivery of the tax return. But this is like the fastest growing category in tax tech. If you're a US tax firm, you absolutely need to have a good tax workflow solution nailed down. We did about 1,800 1040s a year in my accounting firm. This would have saved us so much chasing clients for documents and just tedious organizing of all these PDFs. Oh my gosh. Can't say enough about Sortman. We actually did a full demo of their platform on my YouTube channel not so long ago if you want to see it. But to learn more about Sortman, check out the link down in the show notes. 3 more blockers to adoption here. And number 4 is the training gap. Some quotes from actual firms. We're looking for formal training on AI literacy, prompt design, tool-specific workflows. A lot of us are like waiting. For just the perfect program to come through. And I can tell you, as someone that makes education content, which is ultimately what my stuff is and our videos are, I mean, it's just education content with me falling over. But I can tell you just how hard it is to make education content around AI. If I thought there was a linear path to people becoming experts with how to use Claude or ChatGPT or something like that, believe me, I would have made you that course years ago. I'd be sitting on a beach somewhere right now while everybody watches these paid access versions of my videos. In my opinion, AI is not so simple as to walk people through this just standard sort of progression. Now these days, there's actually a lot of folks building education content for accounting firms. Literally just go on LinkedIn and there are tens of people that now have this in their headline of AI educator for accountants. Hector Garcia has got one now. If you're looking for somebody reputable, Byron Patrick and Twyla Verhulst, who are both now at Karbonne, they've got some AI courses. We talked about Automation Town. There are lots of places out there to get it. Obviously my content, there's a bunch of that online. We're in the process of— in fact, I think it's coming out tomorrow. We're putting out a really big Lift Cloud video that is sort of like Cloud for beginners, the sort of thing you would run every member of a team through when they're getting started with it. But there is no replacement for simply time being spent with it. Like give people a very minimal, like a basic level of guidance, understand how to use it responsibly. But then after that, honestly, it is about sharing use cases and just spending time with it. And this is where it comes back to like, what is that feedback loop inside of your accounting firm? Are you awarding a cool use case of the week every week to your team? Do you have a dedicated space where people can discuss it? Because this is a tool that isn't like any of the other tech that we use where it has a finite number of options and things you can do with it. That flexibility is simultaneously what makes it really exciting and really hard. And some people will take to that. Other people won't. The solution to me to bring this back is still incentives. Why should they be excited for it? Is there anything in it for them to learn AI and get more productive? If there isn't, expect them to be scared. Expect them to not take to it the same way that you might. Next blocker, and this is still a big one for folks, fears around AI's impact on your pricing. The whole notion that AI could potentially come in and automate work. Is a lot of folks kind of staying on the sidelines because as if that their involvement is going to accelerate this happening. Like if I go out tomorrow and I automate the work that I do for my clients, will my clients then not value my work? There is a world where potentially over the next few years, the value of the work that we do changes because AI makes it easier to get that stuff done. That is absolutely a possibility, by no means a foregone conclusion. But right now, the inputs that go into getting a project out the door on a micro scale when it comes to just your firm do not impact the perceived value of the client. And to give you a really basic example, let's say a client drops off 1,000 receipts and you point Claude at a folder with that 1,000 receipts and you say, rename all these files a certain way so I don't have to drill into every single receipt. And it renames those, those 1,000 receipts or maybe organizes them into subfolders by payee. And it does that in 10 minutes. If that would've taken a member of your human team 2 hours to do previously, are you gonna charge them now for the 10 minutes it took the AI, or are you gonna charge them still as if a human did 2 hours of work? You're gonna charge them as if a human did 2 hours of work because the value of the work is the same. And I think we can all agree on this, but we have a harder time agreeing on, oh, what is the value of what I do? Like hourly folks, folks that still cling to hourly pricing, if I give them that example, they'd be like, well, I'm still gonna charge 2 hours. And so it's just worth being aware that how the work gets done isn't driving the price, the value that is perceived by, by the client, by the customer. And that has always been the case. People complain about the price because they don't value it. People are happy to pay the price and will pay big numbers because they value it. And it doesn't matter how many hours you spent on it. We're entering an era where we have the opportunity to literally trade human hours for machine hours. Whether you bill hourly or not, there is work that we're doing every week right now that AI can do for us. And who knows if we'll be able to keep making that trade for the next 10 years or if the, the dance slows down over the next 2 or 3 years because anybody can do it. But the only option here is to take any human hours that can be replaced by a machine hour and get the machine to do that work. It doesn't mean that there's not something else for the human to do at that time, but choosing to not pull in the machine to do that work, that is not the right bet to make. You or I don't have much control over big picture where this is all headed. But the one way that we can guarantee that we're not a part of that party in the future is by choosing to not take part in this whole progression. I mean, this is like the industrial revolution of knowledge work. Nobody could have told you what the world would look like after factories when everybody was living on a farm. But pretending it's not happening, that's not the move, brother. Which ties in really nicely to the last blocker here to adoption, the what's left for my team problem. Somebody framed this really, really well on the— so I'm on tour right now. We're going to 20 US cities. We are 7 cities in. If you haven't looked at the dates yet, Jason Staats on FirmzDoc events, come and hang in person. We've got like 50 to 150 people in 20 cities. It's been amazing. But somebody framed this question really well. They basically said, how do I enable my team to lean into AI and using it to help them to be more productive without the AI itself doing the job that I employ them to do? I can totally understand that. You got a team of people and you're like, I'm going to train you to do all these things, but what, isn't that complex when I'm going to put a tool in their hand that will literally do their job? The answer is yes, that is complex. And like the details of how you navigate that matter. But we got to meet that head on. Like we are doing nobody any favors by pretending that a tool cannot do that person's job. If you are a leader, a firm owner, and you are spooked right now, imagine how the like staff level folks feel right now. Team members that are tuned into this stuff. Imagine how spooked they are. Honestly, the only thing that I can tell them is a lot is changing right now, and almost certainly we will have tools that can do a lot of what your job is right now. But here's the thing. Is honestly, you should be so lucky to find yourself at a firm where I'm going to give you the time and the resources to learn this stuff because nobody is better suited to replace you than yourself. We are much better served plugging into this stuff and figuring it out and navigating it ourselves than pretending and acting like, you know, we're just going to wait it out like we're on some sort of death march or something like, no, plug into this stuff. You've got clients that are navigating the same really scary thing. How can you call yourself an advisor in 2026? To clients who are navigating AI if you yourself are not navigating AI. We should not run from that reality. In fact, I think it like if you employ people, it is your job to stick your nose, but also the noses of your team in that reality. But then also giving them the resources and the time and all of that to actually lean into this stuff and get comfortable with it. To me, it has never felt like more of a moral obligation than it does today to effectively be an employer of people, because there's a world where you don't lean into this stuff enough and 2 or 3 years down the road you are like, geez Louise, my whole team now doesn't know how to use AI and how am I ever going to get out from under this when it's like, well, it was your fault. Like, you got these people here. Whose responsibility is it? The skills of your team right now? I would say it's like 80/20. You, if you are the manager, to like give them the time and the resources and the space to be able to learn this. 20% them. Like, they have to still engage with it. But imagine one day coming upon the realization that like, oh my gosh, AI can do all of this stuff that I'm employing these humans to do. Now what? Are you really going to cut bait with the whole team when you're the one that brought them to the dance? Like you're going to send them off into professional exile because you didn't give them the resources to learn what they now need to know to do their job? That doesn't seem very fair, does it? And so yes, like there's going to be situations where you put the tool into a person's hand that will then do their job for them. But what is the alternative, honestly? And it's like, it is not as clear-cut as that because we do so many different things for so many different people. Like if you and I were working for Johnson Johnson and we are just like, we got an entire floor of people that are processing, you know, one type of purchase order or invoices or something like that. And that was the only job we did. Like, to be clear, those are the people that are at risk, the people that do one thing all day, every day. For most of us, that is not our reality because clients come to us with so many different things and so much of our job is still about the relationship management of it all with the client. And so we are doing, you know, a thousand things a day once, not one thing a thousand times. So I struggle to see a reality where you hand somebody a tool and it like does their whole job. We can't pretend that that is not a thing. We can't pretend that the tech is not in the process of stepping into doing some of these things. And it is up to us to figure out then what do we do next? Because if you're worried that the client won't value the work when AI can do the work, let you in on a secret. They've never cared about how the work got done. They really haven't. They care about the fact that you won't call them back. Right now, they care a lot about the fact that AI might be having an impact on their business. And while it's true that like over the larger arc, like maybe it does have impact on sort of the pricing in not only our space, but, but everything. But in the short term and kind of in more of the micro perspective of how we work with our clients, the inputs have never impacted the pricing. The inputs have never impacted the client's perceived value of what we're doing for them. And if we're using AI to do that work, I still don't think it will. Now, if you like eliminate your client communication with AI and you put AI between you and the client, that's That's tremendously risky in my opinion. Then you are inviting them to be like, "Well, they're just doing all this with a bot now. Like, what am I paying for here?" But that's the last thing we should be doing because that is sort of the most valuable thing that we can cling to is those relationships. Spring is here. It is spring cleaning season. The perfect time to clean out the dusty corners of your accounting firm and just get rid of the stuff that you don't want to be doing still. And for most of us, one of those cobweb-laden services that is just stuffed into the corner of the garage, it is bill pay. Well, good news, my friends, artificial intelligence is here to help. You might have heard of it, the platform that's crushing it right now, this video's sponsor, Maker's Hub. Listen, nobody likes bill pay, okay? It sucks, but almost every business has to do it, and oftentimes it's high-stakes It's messy. It impacts inventory. You're going after like savings. The details matter. You miss a, you miss a zero or a comma here or there. Oh my, oh my, oh my gosh. That's a big deal on bill pay, right? We are getting wildly good new AI models seemingly by the week right now. And honestly, MakerSub are the folks that I trust to like put this work. MakerSub have been like arguably making the best use of AI models. Over the last 3 years to solve a very, very specific problem. The problem of bill pay. Nobody likes it. So if you do this stuff for your clients or you have clients who deal with project tracking, receiving, construction people and manufacturing folk, HVACsmen, super recommend you check out Maker's Hub. A great example of AI done well. To learn more, check out the link down in the show notes. Hey, real quick, the most important question accounting firms are asking right now in 2026 is how do I use AI securely in my production workflows? I've said this many times, don't take security advice from a YouTuber. What you need is actual, like, proper IT help, like this episode sponsor, Farido. If I had to shoot from the hip, uh, with what I've seen from folks, Farido might be the fastest growing IT solution for small accounting firms. I say small, I mean that like 100 employees and under. You need an IT group that just works with accounting professionals, otherwise you are paying this IT group to learn accounting firm things. Like, I had a local IT guy that I once brought over and spent 2 days watching him learn CCH engagement, like how the, like, database worked on my server. Oh my gosh, please don't do that. Work with an IT group that only supports accounting firms, particularly right now as it is so important that we are beginning to implement AI effectively inside of our firms. You want to do that responsibly. Responsibly too, right? You need better help with this stuff? Look no further than Verito. They do hosting, they'll be your entire IT group, they'll do your hwisp. Can't recommend 'em enough. Check out the link down in the show notes to learn more. There are certain things for whatever reason that we're finding that people just don't want to work with AI over. Like even if it's perfectly capable of it, like do you want to, uh, go into therapy with an AI assistant? Like the fact that you could doesn't mean that you should or that you want to. And at least right now, certainly does feel like there are a lot of things that people still want a sort of human liaison for, even if they know that that human is using a lot of AI on the backside. What is the appropriate amount of money and attention to invest in your accounting firm leaning into AI right now? I would say the appropriate amount is the one that feels uncomfortable. Has there ever been anything bigger in how our work gets done? Maybe like the introduction of the personal computer, but that played out over a pretty long time horizon. This is the most disruptive thing to accounting firm And so if we're not investing at a level that feels uncomfortable, that is not a level unlike anything we've ever invested in before, then we're probably not pumping enough into it yet. And I don't think this has to be like this huge investment in training and all that stuff. Honestly, I think that kind of misses the point. I think it's about like time, spending time together, carving out dedicated time to do like little hackathon or just like making space for your team to be able to lean into this stuff more. All right. Got a bunch of questions on more kind of general accounting firm strategy, a little bit of AI stuff as well. Let's get to the mailbag. Mailbag, mailbag. The segment is called Mailbag. Can I just say, the number of people at these, the live events that I'm running right now, who are sharing how much they enjoy the transitions, who are sharing how much their children enjoy the transitions in the podcast, it absolutely, absolutely warms my heart. All right, first up, I'm working on finding ways for dopamine-needing staff to slow down and not get high on checking off the checklist. List just to say they're done. Each checkmark really means something that you need to think about and consider whether it was done right. Any tricks inside of apps to help with this? Brother, humans are just bad at doing checklists. I mean, get the iPad generation to sit down and go through a 35-item checklist. We have never been built to do this, and TikTok has made it worse. But the good news here, AI is tremendous at doing checklists. It really is. If somebody had told you 5 years ago, that 5 years in the future you're going to have this magical tool that can do checklists eagerly, earnestly until the end of time. Like, we literally have this technology now if we just figure out how to use it in our firm. Honestly, I would say we are post-checklist at this stage. I don't know that you need to be asking humans to do checklists when we have AIs that are perfectly capable of doing checklists and documenting their work. So I do think this is an opportunity to shift something that humans stink at to a computer. But also this points to maybe another issue inside of the firm that is giving your humans the space to do deep work. I think most accounting firm setups are tremendously distracting, as evidenced by Six Monitor Guy having his Outlook up and having his Slack or Microsoft Teams up, having all these different things that at any given point in time you're just like inviting distraction. Most accounting firms don't have these conversations around like, What does it look like to do 3 hours of focus work where my team can't bother me and my clients can't bother me? Like most people in public accounting, that's not even an option. And if that's not an option to have a literally fully silent, quiet, focused afternoon, then we got a problem. Like we got to be able to create an environment that is not so distracted. We've talked in the past on the podcast about the whole concept of the maker schedule versus the manager schedule. Another big problem here is you have managers that are trying to get time on everybody's calendars. But if these people's jobs are to do production work, all it takes is 1 or 2 meetings on, on the calendar and it kind of blows up their entire day. So build a firm that is built around deep work and not like avoiding distraction. This for me personally is why I'm such a fan of like asynchronous businesses. That's why I run my companies in an asynchronous way. Meetings are extremely rare. Everybody has their responsibilities and the flexibility to do that stuff when they can get to it. Taking into account, like, ultimately they can't get in the way of the rest of their team trying to get a project over the line because every time somebody needs something from somebody else, every time somebody has to hop on to a meeting, each message, each email, that is, that is stopping someone from being able to, to do deep work. On the subject of getting AI to complete your checklists for you, actually, next podcast episode, We're going to go through sort of the progression of how to get Claude to do tasks for you in the background. So if step one is having a conversation back and forth with Claude to get it to do a thing, we're going to go through the progression that is then making that task sort of just something that happens and you not having to drive that conversation back and forth. Checklists are a great example of where you're like, what if I have a 1,000-item checklist? Obviously that's way too long for a back and forth conversation. How would I even begin to build something like that? We're going to talk about that progression next podcast episode. Like, subscribe, auto-download, all the things, you know. Jason, I'm a small firm of 3. I'm trying to go paperless in a small town in West Virginia. I've looked at your Excel review cheat sheet. That is my app recommendations. It's very helpful. I'm looking at TaxDome, Canopy, Firm360, and Qount. I do a demo of Qount on Friday, but I've demoed the others already. Can you help nudge me toward a particular one? It is very difficult to pick. Brother, that is why the app recommendations exist. But also the way to choose any software in your firm is not demoing each of them. A demo can be like part of the input that goes into it, but all a demo does is award the company that can make the most razzle-dazzle demo, which is very rarely the company that has the best platform for you. This is 100% a situation where you need to be able to sit with other firms that are users of these platforms. Honestly, folks that have used this stuff for years. Do not make decisions based on demos from sales reps. That is where, oh, dreams go to die. But there are a lot of corners on the web where this stuff is happening. I run an online community, Realize, now has over 800 accounting firms. We have like bimonthly meetups for TaxDome, Canopy, and Firm360. Kount is a little less common, but like we have meetups where people are sharing their live setups and what's working and what's not working. That is how you make decisions like this. Is making friends that use these platforms, getting them to show you how they use it and what works about it and what doesn't work about it. That is vastly better information than a demo from a sales rep because especially practice management systems at their core, they have like work management and CRM and kind of the core features. But I'll let you in on a secret. All around that, they have bedazzled those core features with all of these other features that let them stand up a landing page to say, "Yes, we too do proposals. Yes, we too do tax intake. We do this or that." And most of those features are just paper thin. We are talking one-ply, use at your own risk type of features. But they're not gonna tell you that in the sales demo. Sales demo's gonna be built around the razzle-dazzle stuff instead of like, what does it actually look like to live with this tool for years? So please, please, please build out your peer network. You can do this a lot of ways, from in-person events to making friends on social media. Paid communities are kind of a shortcut for this. They're also a bit of a walled garden so that the quality of the conversations and the stakes of everyone there are a little bit more, as opposed to going to like, you know, TaxPros subreddit or something like that, which is a tremendous resource, but literally anyone can just rock in and drop a bomb and then be like, see you later. As opposed to like a network of people where there's some accountability and folks know each other and hang out at live events in person, like all that. All this gets so much better once you have kind of built a friend group that does this stuff. All right. Next up, this was a comment on a recent podcast. Appreciate the reminder to increase my prices, but I got one client who keeps complaining. How do I explain the need to increase prices to him? If you ever feel like you're in a situation where you have to convince someone of why your prices are what they are, you have already lost the battle, my friend. People talk a lot about like educating your clients so that they understand the value of your work. How do you think you would feel on the other side of that if the person you are buying something from educates you on the value of what they're doing for you in the context of them trying to justify a bigger fee? Either they value it or they don't. This is the magic of positioning your accounting firm, and finding the perfect marriage of what you do and what that person needs. And the two drivers that will make people happy to pay more is when you solve an expensive pain and when you solve a rare pain. An expensive pain looks like, "We don't do this right, it could cost me $10 million." And so they're willing to pay more than $300 for it. A rare pain looks like, wait a minute, you know about custom boat building done in international waters using rare reclaimed wood from an endangered forest? Yes, that's, that is, that's our expertise. When they can't go out and get that work done elsewhere, you have the power to set prices where you want them to be. And if the combination of those two things, if you are like the perceived as the only one who does it, but also you're solving a really expensive problem, you're able to command really premium fees. The opposite end of the spectrum there, is people coming in who need something done but who don't ultimately value what's being done. So they're looking for whoever will do it cheapest. And this looks like doing the work just like the guy across the street will do the work. And so you're literally competing on price. And then you end up in situations where you'll hear people say things like, yeah, the difference with us is we will do more for less. You get a higher level of service and it won't cost you as much. I had a guy I worked with who prided himself on winning every other client he sat down across from, many of these people coming from bigger firms. He would give them his cell phone number. I mean, he would talk to them at 8:00 PM at night and then charge them less than the other accounting firm. Don't do that. That is sabotage. Don't do both more and for less. Either do it for less and do less or do it for more and do more. To do both is just asking to fall flat on your face at your desk when you're 43. Next up, this is a nerdy AI one. How much are you budgeting for tokens each month? I'm worried about moving from flat rate pricing on ChatGPT to tokens on Claude. So tokens are what large language models generate. A token is like a word or a part of a word, and it is a yardstick for how much you're using the AI model. Bit of a misnomer here. Claude still has a base subscription price that comes with a certain amount of model usage. In fact, couple weeks ago, that base amount of usage just doubled thanks to their partnership with SpaceX. But if you have usage in addition to that, beyond that, you can optionally enable token-based spending, basically. So if you go beyond the usage limit that just comes with your monthly subscription, you can pay for all the tokens used beyond that, either pooled across your entire team, or you can actually allocate additional allowed tokens, like 2 different users in your team. And so if you're actually bumping up against these usage limits, there's a lot of things we can do to, to mitigate this from putting big tasks like setting them to run overnight so that by the time you are up and at the battle station, the usage limits have reset. But this, this question, how much are you budgeting for tokens each month? This is such a niche problem. I know very few accounting firms who are really up against the usage limits inside of a Claude plan. You may get there at some point. In fact, after next podcast episode, when we talk about how to get this stuff to just run in the background, that is where you potentially run into the issue of usage limits because AI is just happening and you're not lifting a finger, which is a good thing. But ultimately, like what budget makes sense depends on the type of work that it's doing. Like right now, I bought an 80-year-old tax practice right now. Claude, would have been my highest output employee and it would not even be close. I think the cheapest staff level tax person— this was 5 years ago— we're paying like mid-$50,000 a year. So that person's probably costing us $65,000 USD per year. That is a lot of AI tokens to get to $65 grand. But we're still talking about Claude being like across everything it can do, higher output than an entry-level tax preparer. So I don't have like a firm-wide token limit in mind or spend limit in mind. I think you probably have to look at the jobs that it's doing. Where is it worthwhile? Start with the basic version of this that is probably inefficient and kind of not optimized. If it really starts getting costly, that is where you start thinking about what parts of this process could be done by a lower-cost model. Because going from Opus, Claude's current biggest, most capable, most expensive model down to like Sonnet, which is the current generation model, but like the fast version that's not quite as big-brained, you have like an order of magnitude price difference. And so you can carve out the bits that don't require a bunch of brainpower. That can save you a lot of money. But to be honest, like accounting for most of us are so far away from this right now. Hey, I know you love Cloud Cowork as much as I do, but have you heard the news about Canopy? Coworker Subtle is a new AI assistant they launched inside of Canopy, one of this episode's sponsors. Canopy, a practice management system that is a feature nerd's dream. They're trying to cover as much of everything you need in an accounting firm in a single place. But as we talk about on this podcast often, it is much easier to roll out an AI agent inside of a closed system than it is to roll out an AI agent that works across like all of my systems, right? That's the trade-off. I want the AI that works across all my systems, but we're struggling now with like, what's the most secure way to do that? The AI agent that works inside my all-in-one system? Buddy, that is easy. Flip the switch, count me in, strap in, buckle up, I want it. Canopy is so convinced that you're gonna love what they got to show you, they're gonna cut you a check for 75 big ones. They'll give you a $75 gift card. Gift card to Bed Bath Beyond. What's that? Okay, I'm I'm not sure where the gift card is for, but if you book a demo with Canopy via getcanopy.com/jason, I'll put that link down in the show notes, they give you a $75 gift card just for coming and demoing. Do it. What do you got to lose? Hey, just real quick, if you're using Outlook or Gmail to manage email inside of your accounting firm, brother, those were tools that were built, uh, 20+ years ago and not for accounting firms like yours. Of course, if you haven't heard of them, let me introduce you to one of today's sponsors, Missive. Missive, M-I-S-S-I-V-E. It is exactly what I needed. I actually used a bunch of like off-the-shelf help desk sort of email tools in my accounting firm. Man, do I wish I had Missive. It's a vastly more efficient way to manage email across your entire team because email is not a hill that anybody should be dying on alone. We did a demo on the main channel of it. If you want to see it in action, just search my name and Missive. Really cool, doing some cool AI stuff as well. Worth checking out to learn more about just what you're missing by using Outlook or Gmail. Check out the missive info down in the show notes. I would say a good problem to have is to be bumping up against usage limits. That should almost be a goal. Big brain, like AI nerds call this token maxing. How do you make sure that you are using up all the tokens that you are allowed from Anthropic so that nothing is going to waste? While some of that can be just inherently wasteful, like just trying to crank on compute for the sake of cranking on it. I do think we need a little bit more of that energy in the accounting firm space because we are genuinely talking about trading human hours for machine hours. The AI is just sitting on the bench. In many cases, you're already paying for it. So let's put it to work. 3 more questions to go here. If you got questions, pop them in the comments, shoot me an email, shoot me a text at 1516-980-4968. A lot of these questions come in via text. Is AI going to kill after-the-fact work? AI is definitely going to change after-the-fact work. AI is also going to change proactive work, though. There's long been this framing of like, well, advisory, kind of the big brain stuff is forward-looking, whereas the more transactional staff is backward-looking. And there is some truth to that. But AI, they can do both. So it doesn't map perfectly with like traditional accounting firm thought leadership saying everybody's got to be an advisor and you got to be looking into the future. And that's what real advisors do. 'Cause AI can look into the future as well. More important than any of these things, generalizing to like, am I just preparing tax returns or am I being a tax coach? And same kind of on the CFO side. More important than what side of the spectrum you live on is are you solving an expensive pain and are you solving a rare pain? Because there are proactive pains that are expensive. I mean, you know, doing after-the-fact work right now, not all after-the-fact work is made the same. There's great after-the-fact work and there's really bad after-the-fact work. So I don't know that AI really skews us to being more proactive versus more after the fact. I think that's a bit of a fallacy because people talk about advisory always being proactive. Just plan on AI changing all types of work. But how can we frame this as, as an opportunity? Because most of your clients, they will still want to have that conversation with you. Next up, got any tips for a fella wanting to offer freelancing payroll services with no experience? Yikes. Should I just get job experience first? I should mention I'm a fresh graduate and have learned bookkeeping too. I probably wouldn't offer payroll services with no experience. How glib do I want to be here? I think the opportunity for you here is to find someone who will pay you to learn. And that, my friend, is what we call a job. People sleep on jobs. People sleep on working for other people. Here's the thing though. It is like a safe little environment where you can learn the skills that you may one day go on to wield by yourself but you're literally being paid to figure out how to do a thing that you don't know. You're surrounded by other people that know how to do this stuff. Don't sleep on jobs. Plenty of other perks. Health insurance. You get paid to go to the bathroom. Don't start selling payroll services with— without any experience. There's a balancing act here because at the other end of the spectrum, there's also people that have like mountains of experience. They're like, oh, I can't do this thing. And the most common example of this is people trying to specialize into an industry and they're like, I don't have any expertise in that industry. Yet oftentimes the people in that industry will walk into a generalist accounting firm and they're like, hey, have you ever done taxidermis before? And the firm's like, nah, but we can figure it out. It's all the same. And so that's literally what you are up against is generalists just doing one of everything. So if you have general experience in tax and accounting, bookkeeping, whatever that looks like, don't be afraid to go out and plant your flag in kind of a new domain and learn a new space. But if you have no experience at all, the best thing you can do is find a firm that is most similar to the type of firm that you would one day want to run. If it's a wildly different firm than you want to run, that's probably not the place to go and hang out. I remember one firm I worked for, they delivered printed payroll checks every week to the client's business and dead dropped those payroll checks in a hollowed out wooden mallard. That's, that's a true story. We printed the checks, the checks went into a duck, and from there I honestly don't know what happened to them. That never felt like a sustainable system to me. And so the firm that is best for you to go to work for probably ought to be aligned with like, what is the type of way that I would want to run my firm someday, even if I'm not there yet? You know what, though? Don't sleep on wooden mallard firms. There's like a full, like, prepper version of building an accounting firm. There's also probably a never-AI-er version of running an accounting firm where you work with the folks who are like, I hate AI, took my job. I don't want anybody I work with using AI. The problem is what we do in accounting firms, at least like the transactional work, tax prep, bookkeeping, everybody kind of hates. And it's why back in the day, OpenAI launches their new GPT-4 model and their last like cool one more thing razzle dazzle demo is we're going to use it to prepare Becky's taxes. The reason they can do that is people are like, oh yeah, no, no, AI should definitely do that. We should totally use AI for that. You see a lot of creative people doing this right now, people in the movies, artists and all that stuff. And they're like, AI, let me tell you my thoughts on AI. Now, medical stuff, health stuff, we should definitely use AI for that. But to make a movie, to make a painting, it's just wrong. Meanwhile, the folks that are working in the medical labs are like, dude, what did— that's, that's my job, bro. It's far from a perfect analogy because medical has a little more, a little more stakes to it. But preparing a tax return, doing some bookkeeping. Listen, I'm an industry influencer. So if anybody's entitled to feel self-important about the work that we do, I feel like, I feel like I am. But how many people are actually gonna be like, finally, I found the accounting firm that won't let AI touch the tax return because I respect the craft that is filling in those forms. Forms. And it might be hard to build a client list of, of people who can get into that, especially when it means they're going to pay $1,200 instead of $300, right? Last question: What was the original goal of your videos when you recorded them, and how have you evolved your content to adapt to the busy accountant's day-to-day? That's a great question. Um, okay, original goal when I started making videos was to be distracted from my accounting firm because it was a 40-person firm It took forever to change things. I had the Zoomies. I needed some sort of creative outlet. And so I started making videos, just sort of documenting what I was learning so that other people didn't make the same dumb mistakes that I did. That was, that was how the videos started. And they were so bad and they're all still on YouTube. If you want to go watch my very first videos, you'll be like, oh wow, he was just as bad as everybody else would be when they're just starting video. The difference now is between the podcast and the YouTube video, I've done over 1,000 videos in the last 4 years. And so I hope, I hope that I'm better than I was or above average. The problem is now people see that and they're like, wow, you're a natural. I can never do that. But the original goal was like, honestly, it was just a distraction from the frustrations of running the firm. But how do you evolve the content to adapt to a busy accountant's day to day? What kind of a strange wording was that written by AI? This is more advanced. But these days, in terms of content strategy, you got to make the sausage into whatever form factor a person is willing to consume that sausage. This podcast, this is the 60-minute version. The YouTube channel is the 16-minute version. We're actually right now just rolling out podcast clips. That's like the 6-minute version. So you take, you know, just one question and answer and post that as a podcast clip to YouTube. Shorts, little portrait videos. That's the 60-second version. I would argue my AI assistant that anybody can use online, they can see into every single podcast episode. That is like the 6-second version where you can just go ask it a question and it'll pull the answer from past episodes of the podcast. Every different form factor will have a different type of consumer. There are people who will happily, there are people who would do a 60-minute podcast every single day. There's other people who are like, brother, I got a job, frankly. I already see too much from you. And so you sort of, you chop it up and build the programming so that for any level of appetite, there is something there for people. Also, people's like appetite for how much they want changes over time too. Like, and we're all this way. I'll be super into a podcast for a while and then I'm like, eh, I'm kind of cooling on it. And maybe I'm just like watching Shorts from them and then maybe I get back into the podcast again. So these days, best practice for like a thought leadership business like mine, is you are filling all those channels with like different sizes because different folks will, will consume different sizes. Now that from a content creation standpoint, that's a very sort of advanced concept. Where you start is just with one channel that you enjoy and grow there. And you can like— the world is big enough where anybody can find success in that one channel. Reality is at this point, like we've got a team of 15 and are pumping out the accounting firm stuff. Pretty much anywhere that you can find it online to just kind of be unavoidable. Also for me, like my business is always inherently going to be constrained by me and sitting down and shooting and my ability to write the next video and all that stuff. Right. And so a way that I can employ other talented people to help the business is filling all those channels so I can sit down and I can shoot a 1-hour podcast. It takes me about 90 minutes to shoot it. I'll spend 2 or 3 hours writing it beforehand, but then that Buffalo, can be cut up into all sorts of different parts by other people that get published on other platforms with little to no, like, incremental effort from me. And so that is a way that, like, we can build more leverage, leverage for the business that isn't really reliant upon myself and me sitting down to shoot. But if you made it here, you're one of the real ones. You're one of the podcast listeners. I appreciate you so much. Whether you are sitting down working, whether you are on that, that stationary bicycle, walking the doggos. I appreciate you tuning in here each week. Got any questions? Pop that stuff down in the YouTube comments. Shoot me an email. Shoot me a text. Next episode, we're going to be talking about sort of that progression of how to build like Claude AI workflows that run themselves in the background. You don't even have to have the conversation. Going to try to not make that one too kind of big brain, too nerdy. But for me, when I'm looking at AI, what I would be thinking right now is, okay, how do I just carve out like one status? Status and a process and just make AI in charge of that status? Is that a thing we can do? The answer is yes. We're going to talk about that next time on the podcast, gang. I can't wait to see you there. Until next time.