613 Stop Writing SOPs for Your Accounting Firm. Do This Instead [The NEW way to build your SOPs]

Listen up, 'cause I hear that Jason on FIRMS is going on tour. Buckle up, it's the show you love now with snacks and an open bar. The tour's halfway through, but we still need you. For tickets and more, head to JasonOnFIRMS.com. JasonOnFIRMS.com. Now it's time we end this song. On to the show! Okay, my mind's kind of spinning from this. The way that we document work and processes in our accounting firms, the way we try to standardize what we do, I think it has completely changed. In the process, just for outlining this podcast episode, I was reviewing some conversations I had in the last year around best practices for building documentation. And you know what? A lot of it, like a big percentage of the outline that I built, I came to it the next morning and I'm like, this is pointless. In the age of good AI, like the type of AI that we've gotten in the last 6 months. You're never again going to have a human build like a step-by-step process guide. You're never going to put a big old checklist in front of a human again. Like, we have AI agents that can do this stuff. So what does this mean for documentation, for how we standardize work? How has AI changed this? And where do we, fellow humans, where do we fit into this process? Because standardization, getting work out the door, all that stuff matters. But what's the best way to approach that now in 2026? Plus, How do you build a client list around a new niche? We're going to talk about how to promote a sense of teamwork in a remote team that's working across all sorts of different time zones. Are CRMs necessary in the age of meeting recorders and AI that can kind of just see across my tools? And is QuickBooks Desktop— QuickBooks Desktop— is it the future? What? I'm going to give you a pretty compelling argument for why it might just be. Come on in. Let's do some accounting firm stuff. How do you do, fellow accountants? My name is Jason Staats. Welcome to the Jason on Firms podcast, where we talk accounting firm strategy, building better firms, whether you build one, whether you work at one. Pull up a chair alongside your fellow accounting firm maxers and let's get just a little bit more confident in what we do today. Now, that super cringe transition you just saw, we are in the, the heat of the Dirty Dozen, a series around the 12 biggest pains facing accounting firms as sourced by the over 1,000 accounting firms that came out to my live events last year. And today is a biggie. It is around SOP. It is around documentation. And as I shared up top, I have so many quotes from people last year, just, I mean, and this was last like October, November, December that these events happened. And I look at those quotes now from firm owners and I'm like, oh, you would actually never do that now because of AI. I actually wouldn't recommend this. And so today we're, we're going through what have kind of been the historical SOP documentation best practices and considering which of this, what part of this is still valid and what is not. What is the new kind of AI alternative way of getting this stuff done? It's actually, I think, like really, really interesting. And to give you just a really basic example of this, I'm in the middle of my 2026 tour right now. Uh, if you haven't come out to an event before, check out jasononfirms.com/events. Super fun, um, in-person events. I'm going to be hanging with 125 accountants in Austin tomorrow. There's a good chance of coming to a city near you. But over the course of this tour, we've been collecting use cases from firms. How are you already using AI in your firms? Some of this stuff has been amazing, but one of them that I kind of keep coming back to, and I've actually got a couple follow-up emails about since I shared it, was someone sharing that they had this big ugly process that they do inside of NetSuite, and they recorded a screen recording of going through every single step of that process in suite. And then they gave Claude the video, like just the raw video file. No, no narration, no like verbal explainer, just the screen recording. They gave that to Claude and Claude, bless its heart, put out like a flawless step-by-step process guide for everything that was clicked in the app and everything that was tippy-tapped in. And that kind of blew my mind because I'm like, how is Claude even doing that? Like it's probably analyzing the frames and looking at like what's happening there., but the fact that that worked was kind of shocking. And this opens sort of a whole Pandora's box of, well, you know, in the past I was a real big fan of Scribe. Scribe, 4 or 5 years ago that came out as a Chrome extension. I remember talking with one of the co-founders like weeks after they launched and they're now a humongous company. But Scribe is a Chrome extension. It's also a desktop app, but basically You start recording and it click, it records everything that you click and type in and all of that. And then when you're done recording, it creates this really nice sort of step-by-step documentation. That is an example of something now that Claude can just do like straight out of the box. But it wasn't that long ago that I, I mean, literally 6 months ago, 12 months ago, I was telling people like, this is probably the most efficient way to create documentation. But if AI can now create the documentation, then we can actually take this a step further because Claude is obviously much more than a documenter. We can also put it in charge of like maintaining the documentation and proposing changes to the documentation. So this rabbit hole, it goes much, much deeper than creating like the little screenshot guide of how to do a thing because Claude is capable of so much more than simply documentation. Now, many of the headaches here are the same. The age-old problems that we have had with documentation in our firm. Some of these things have not changed. Many of them have. Let me walk you through the 6 traps when it comes to documentation inside of an accounting firm, and we'll also look at how much AI has changed each of these things so that the old thought leadership has— it's kind of evolved. First trap around documentation is just the whole notion that I'll do it someday, right? You know, you need SOPs, but you've never started. Maybe it's because you can't find the right tool for it, or you're not sure how best to do it, and so you just— you just never do it. Some good advice here: your documentation at first will always stink. Like, there is an art form to documentation, but it is one of those situations where the whole point of doing the version 1 is so that you're closer to the version 2. Like, it will suck. I mean, there's so many things in our firms that are this way. Capacity planning is the same thing where it's like, well, I just haven't found the right framework for capacity planning yet. Brother, capacity planning at its core is not complicated. You know how much capacity you have, you got a ballpark idea of, of how much work the work is gonna be. You just gotta do the legwork of building that plan. And the first plan, it's gonna suck and it's not gonna be right. It's gonna be busted, but the next plan will be better. But we never get to the good one until we do the really bad version. Documentation is totally the same way. Now, I shared the story the other day of someone I met at an event that told me, we've pulled in an AI coach to basically use as a personal trainer to sit alongside each of us and give us some advice. But the, the bigger thing is to use them for accountability because that is, I mean, that's the most valuable aspect of a personal trainer is I have a standing appointment at this time, a thing that I have paid for, so I will show up for it even though I don't want to. I love that analogy as like using a coach as a forcing function to learn AI. Just imagine if we had the same level of accountability to our processes. Our process is ultimately how the sausage gets made every day. So if you can't find the time for it, like, I mean, for most of us, like, there will never be an afternoon where you will look out the window and be like, I don't have anything to do right now. That's just not how entrepreneurship works. You signed up for the wrong game if that's what you're looking for. And so making this happen probably requires a different level of accountability, some sort of forcing function to ensure that on some sort of calendar, be it monthly, be it quarterly, whenever that happens, this documentation is being reviewed or produced or somehow updated. Trap number 2 is you are the SOP. If somebody has a question, just come ask me because everything lives in your head. You are relying on judgment rather than process. And this is such a learned skill. This is one of those things, kinda like delegation where like nobody rolls out of bed one day just knowing how to delegate. Your ability to systematize what is your abstract expertise is totally a learned skill. And it's so easy to give into the idea that, whoa, it's, it's just experience. It's this cauldron of the things that I know and the things that I've seen, and it can't possibly be distilled into a framework. It's so easy to give into that line of thinking, but Like, this is the highest leverage application of your experience. It's the best thing that we can do. Take everything that we've done and seen and distill this into a step-by-step process so that anyone or a lower skilled person inside of your firm can walk a client through this transformation. Like, this is how you get from just a person being able to sit in a room and be impressive to how do I scale this across a team and create a repeatable transformation for my clients that other people can help me with. This is something that I've— it's not come natural to me. Like, how do I distill these ideas into more clear sort of step-by-step frameworks, right? You can think about this as like inside of your firm, how much of the process is reliant upon judgment and how much of the process is reliant upon a system. And most firms, we are overly reliant upon judgment. You hire a really expensive person, you say, oh, this has to be a senior level people with a bunch of experience because they can just figure it out. That is an overreliance upon judgment. As much as possible, We want to extract everything that can be systematized out of that. And you may never get to 100%. The goal is just to go from like 20% to 25%, especially now that we have AI agents where you can have 1,000 AI agents working in parallel behind the scenes, getting this job done for any number of clients. Like literally your scale is only limited by how well you can systematize. In the past, systematization has been like, How do I get more consistent results out of humans? Now it is, oh, if you can systematize it, you can scale it literally to infinity. That is what we're talking about in this AI age. That's really, really exciting, but it still requires the hard work of how do I create a system that will actually produce repeatable results? Trap number 3 when it comes to building processes, the idea that every client is a snowflake, a special, special snowflake. Uh, to be clear, they are, but the snowflakeness is more like the candy coating on the outside, that is the last mile where the opportunity is for systematization is the core stuff that is the same across projects. So if you do a 1040 individual tax work in the US, every single tax return will be different. But there are many things inside of all those tax returns that is the same, such as Social Security statements, such as the number of volunteer miles that they drove in a year, such as a W-2. Those are kind of the intersection points of the Venn diagram across a whole bunch of projects. Those are the places where where we start with systematization. Imagine you or your team never having to enter or review a W-2 again. Wouldn't that be nice? The fact that every single client is different because they are all different doesn't mean that a big percentage of what they do can't be standardized. Exact same thing on a month-end close. You're never going to produce a set of financial statements that is identical to another company's set of financial statements. But oftentimes we see that and we're like, oh no, you can never automate this work. Because every single one of them is different. Our job is to grow the percentage of those projects over time that is standardized. There's a real, like, fallacy here around the idea that you need to find clients that are the same in order for your systems to become the same. You know, like the whole problem of like, how do I standardize this stuff when every single one of my clients are different? Oh, I guess I just have to find a whole bunch of the same clients and then my processes can be the same. I would argue, and this is from experience doing this in an 80-year-old firm that had 2,000 clients and just a huge hot mess of different ways of doing the same work. I would argue that the opposite is true, that your clients will not start looking more similar until you build, like, the internal vernacular and the systems for speaking about the common things you do for your clients. Those clients don't start looking more similar until that language exists. My experience doing this inside of my firm was we started building language around, you know, let's define what Bookkeeping 1 is for us and Bookkeeping 2 and Bookkeeping 3 and Payroll 1 and Payroll 2 and Payroll 3 and Entity Tax Prep 1 and 2 and 3. We started creating this sort of arbitrary language around the idea that as much as there was overlap in these different projects, we would create sort of a naming and complexity system around this. And it was far from perfect. And so anything about that client that fell outside of that system, that would be what we documented. And it might have been 50% of the project. It might have been 30% of the project, might have been 70% of the project. But then once we had that internal language inside of our accounting firm to be able to say, oh, this is a Bookkeeping 2 project, then everybody in the firm knew what that meant. Another good example, benefits management from, you know, making simple IRA contributions after a pay run to double-checking the health insurance invoice against the deductions and employer contributions. Fiddly stuff like that. We built an actual system for that and we gave it sort of a naming convention so that inside our firm, everybody on my team understood if it is Benefits Management 2, here's the expectation of the work that we're supposed to do. And if there's anything we do for the client that falls outside of that or any deviations from that, it's literally just a note, a note to capture any exceptions. And it wasn't until we had that internal language around the different things that we did that projects started trending to being more similar over time because a member of the team, when a client raised a question and was like, how should we do this or how should we do that? They said basically what our internal system was. And the client's like, yeah, sure, I don't really care how it gets done as long as you think that's the best way. And then that engagement looks a little bit more like all of our other engagements. If we don't have that internal infrastructure, then when the client raises the question, every person on the team will give a different answer and those engagements will get more different over time. And so I think it's a fallacy, the notion that you have to find clients that are the same in order to be able to document more of their processes. I actually think building this documentation, this internal vernacular, leads to client engagements trending to be more similar. It's not to say you shouldn't go out and try to find more similar clients. Like, you know, if you've been around, I'm big into specialization and like finding a rare, expensive pain and just building an entire accounting firm around that. But my experience was really more of the opposite, taking a firm that was a big hot mess Creating that internal language, the internal systems. And at that point, all those engagements started to look more similar as a result. They were all special snowflakes. They were all very different. That was how I created a path to them becoming more similar. 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 Accounting and 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 and 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 Donuts. Hey, if you're done licking your wounds from tax season, if you run more than a few hundred 1040s a year, you gotta 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 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 link down in the show notes. All right, fourth headache around this stuff is out-of-date SOPs, and this is one of the areas where I think AI can actually be a big help. Many firms went down the path of building client SOPs, which is a tremendous amount of work documenting a client engagement from beginning to end. Best practice here is not client, uh, SOPs. It is system SOPs. It is SOPs around the standardized ways that we do things. And then the client-specific stuff is only like the last mile that is unique to that client. But here's the thing is many of the things that we can now build an SOP for, an AI agent can just do. Like obviously not 100% of things, but a growing percentage of things. If we can create a clear like set of steps, the AI agent can actually do that work for us, which makes then the ongoing management of the SOPs of the documentation sort of trivial because you can just go to the agents and say like, hey, for the last 70 times we've done this process, here was our original process doc. Does this still make sense? Update this, make this a more current version. But honestly, I like when it comes to the, the, the step when the AI will actually do that work. I don't know that we really care about that documentation anymore. And the best AI agent setup there that whole thing is actually self-managing. I don't know if I've shared this story on the podcast before, but I create these daily newspapers for my kids. They— it is like they absolutely love them. It's turned into this whole giant project. But every morning they get a newspaper that I generate with Claude. And part of that process of building that newsletter— newspaper every day is I do some back and forth with the AI around stuff I wanted to fix and things I didn't like about what it generated. But then those notes that I give it are automatically harvested into a feedback doc that the agent can see the next time it creates the newspaper. So all the feedback I'm giving it, all the notes that I'm supplying with it are automatically ingested into future runs of it creating that thing. And so if you bring this over to say a small process inside of AccountingPro, what's something like bite-sized and simple on the tax side? Maybe this is taking IRS transcripts and looking at the tax source docs and making sure that everything in the transcripts is reflected in the source doc. On the bookkeeping side, maybe this is looking at all the documentation that you had last month and the pile of docs you have for this month and making sure that everything is there and ready for the accounting pro. Those are both things that AI can, can totally do on their own right now. But for this agent to be doing this stuff every single day in production, there's probably gotta be some sort of feedback loop there where you're telling it like, hey, you did it this way this time. I actually wanted you to change it and, and kind of do it this way the next time. If this is now a process that the AI agent has taken over, in all likelihood, I mean, this is not something that any of our human accountants are ever going to touch again, right? Like once the agent can do it, the toothpaste is out of the tube. So the whole problem of an out-of-date SOP, I don't know that it really matters anymore because as soon as the agent takes over the process, there's a version of the SOP that I care about, like so that I can see how it's doing the work. And if a human has to take over at some point, then the human can see what the normal process is. But the whole notion of maintaining the SOP when it is like taking in daily feedback from everyone across the entire firm who are giving it essentially review notes, I think we have, we have outgrown sort of the traditional way of thinking around an SOP and modifying what a step might look like. I mean, this is basically an SOP with like, you know, after a number of months or a number of years, it's an SOP with hundreds or thousands of points of feedback from people all across the team that is much more robust than a traditional SOP. 2 more traps here around documentation. Number 5, nobody follows it. Uh, you wrote the SOP, it sits there, and the team just does the process however they think the process is best done. Or you've got like, oh my gosh, this was a nightmare in my firm. You've got a job that needs doing and everybody in the firm has a different way of doing it. They all get to an outcome that is okay, but they all think that a different way of going through the process is best because it is what they have always done. Now, this problem of people not following the process, this one might be going out of style as AI is simply doing more of these processes for us, right? Like, this only matters as long as the human is actually doing the process. And I think the bulk of what we're actually going to be documenting going forward, those things that can be done step by step, it's probably not the stuff that the humans are going to be doing for very long, right? I like, aren't we going to be like the ones kind of marrying everything together and getting something from one agent to another and sort of filling in the cracks? I mean, that's kind of what we already do right now for our software tools that can't talk to each other, getting data from system A to system B. That's like the last place that you build documentation. That's the hardest stuff to document. So to the degree that we can make step-by-step guides for things right now, aren't those kind of the things that agents are now coming into and handling for us more and more? And so this whole problem of like, like nobody following the process. How I'm kind of feeling right now is like, if there is a step-by-step process that can be outlined and the process is as simple as that, why is a human doing the work? Like, I mean, there's a lot of nuance to that, right? From like, well, how does the AI access this system or that system? But this whole problem, which I mean, I spent like a decade beating my head against the wall over the notion that we couldn't get people to agree on what the process should be. This whole problem in many ways goes away when the AI agent steps into that step-by-step process right now. In some ways, maybe it gets even more important because we have to get to like, what is the ultimate version of the process that an agent should follow? But there's another part of me that's like, I mean, you could crowdsource all the different ways to do this and the agent could literally do all of them. Like one of the most frustrating examples I had of this problem was I still remember and I still remember when OCR first became decent and there was this, this wave of OCR products coming into the market for all these different applications from bill pay to tax prep. And where we ran into this probably more than anywhere else was around personal tax prep, these OCR solutions that would extract the numbers from the forms and sometimes put them in the tax software. And there was all this marketing around like, oh, we've now achieved, you know, 94% accuracy, 96% accuracy. And every time you see that, you're like, well, yeah, but I mean, it has to be 100%, right? I mean, what if, what if the number they got wrong was tax withholdings and now client's getting a huge refund or they have a whole bunch of tax due? Like, we need 100%. And the whole 96% accuracy was exciting in theory, but in practice, we still had to do all the same review, all the same documentation. And so I would bring a 96% done project to a member of the team and they wouldn't do 4% of the work. They would do 100% of the work because they had to make sure that everything was correct. And there's a, there's a risk here that maybe we end up doing the same with AI agents because accountants are renowned for redoing work. Like if they have to take over a project from a colleague, they will literally rework the whole thing from scratch. Are we, are we going to end up doing the same thing with AI agents or can we truly treat this as, well, the agent got it to review? And now we just skipped a whole step. We are, I think we're definitely gonna be dealing with like an AI agent version of this. But, uh, to bring this back to like, you know, tax research, I think people talk around, talk about this problem around tax research a lot. Well, I, I can't just trust an answer from a black box. Well, the right way to get AI to do tax research or any sort of technical research for you is the same way that you would get an intern to do that research for you. 'Cause if an intern comes to you and they're like, hey boss, here's the answer, you'd be like, Well, I, who are you to tell me what the answer is? You know nothing. I don't believe you. Go document it. And basically what I want is the most, like the fastest provable path to an answer. And that is, you know, some process of documenting kind of all the relevant bits of literature and summarizing that for me so that I can go through that as efficiently as possible. That's what I actually want more than an answer. And so getting the output of AI to be in a format that actually yields time savings has to take some of that into consideration. What's the reviewable form of this? Not, not what is just like the black box complete version of this, because what I don't need is a really capable AI agent. And then I plop that work on a person's desk and they just rework the entire thing. Right. All right. Last trap, number 6, the following without thinking trap. You build an SOP, but yeah, well, here's a quote from somebody. I need more independent reasoning. Sometimes it feels like my team's so good at following my processes that when something is not written into the SOP, they quickly turn to me for help. Another similar quote: I really need my team to stretch beyond the work that AI will eventually replace. They spend so much time thinking about the right place to categorize a transaction when I need them to be thinking about the big picture for each client. It's very easy for a preparer to get in the rut of the process and avoid the friction of having to raise their hand and be like, should this be done differently? I mean, the historical version of this in accounting firms is simply following Sally. Just follow what last year was this because it's the closest thing we have to an SOP. But then the frustration is like, why are we just copying year over year and nobody's stopping and raising their hand and being like, isn't this an issue? Humans are like checklists, for example. We are terrible at doing checklists. Like we develop checklist blindness so quickly and we're trying to speed through this. And part of this is like the incentive structure of like just trying to get more work out the door doesn't lend itself to someone being the squeaky wheel, but the whole following without thinking trap, aren't— don't you feel a little bit like this maybe goes away a bit in the age of AI as well? Because we're asking the machines to think, not to say that we are not still thinking at all, but the grunt work, the heavy thinking ought to be placed on the machines and checklists are a great example. I mean, there's almost no reason to put a checklist in front of a human anymore because the machines can do the checklist really, really well. Not because they're more capable than the human, almost because they are willing to allocate more thinking resources to it. It's something that we are— humans are chronically— we are just bad at. Like, it just does not— like, get some, some iPad kid to sit down in front of a 30-item checklist. Good freaking luck, man. I love that that is what I call the youths entering this profession now is iPad kids. Pretty sure they are iPad kids though. But there's, there's just a new angle to this now where it's like, Is there a thing that humans just suck at doing? I would say checklists are a tremendous example. Is that a place that we can pull AI in? 100%. I mean, that is maybe one of the best parts about AI right now is you could point one very simple checklist item or point an AI to one very simple checklist item. You could literally have it work on that problem for 24 hours. I mean, don't, and that's going to be really expensive. But if you think about, you know, management leaders' frustrations with Staff, they need to use their brains. They need to stop going so quickly and think more about this stuff. And on the staff person side, they're like, really? Because that same person just told me I need to get more work done. So which of those two things do you want? AI, I mean, it literally has a dial where we can tell it how much do we want to work on this. The level of reasoning that goes into this. If you use the Claude desktop app, flip over to code, in the bottom right, there's a dropdown that is basically how much effort do you want me to put into this task? On Cowork, that effort level is actually overridden, but we literally have thinking machines and we can force a level of thinking. How is that not a solution to the problem that we're talking about here? The whole following without thinking. To the degree that we can pull AI agents in to help with these specific tasks, it feels like the whole following without thinking problem really is maybe not going away in full, but our evolving role in the process The only thing that is then left is like the thinking, what could have gone wrong here? How do I get this thing to the next step? Is everything passing the sniff test? What do I still need to review? The sort of rote following last year, that sort of stuff is kind of going out the window like shockingly fast. I do want to call out one resource here if you haven't gone into it before. I developed this like 2 years ago. It is called my Service Library Template. Basically, it is a solution to the age-old problem of how do I get my work to look more similar? How do we develop that internal vernacular around the stuff that we do? The entry point into learning more about this is a YouTube video, but on the back end of that is a humongous Excel workbook where I fully built this out for like the 16 most common services that accounting firms offer. It is maybe the most positive, positively received resource we've ever developed. It's completely free. If you haven't explored that yet, it's absolutely worth doing. I'll put a link to the video that's kind of the entry point to that down in the show notes. We have got some really interesting questions in the mailbag today and a kind of wild story I want to share with you. Roll that delightful transition. Mailbag, mailbag. This segment is called— Mailbag. Questions are sourced from my email, jason@jasononfirm.com, from social media DMs to my cellular telephone, 151-698-04968. Operators are standing by. First question: If you have a practice management system, is a separate CRM tool necessary? If so, what advantages does it offer compared to what your practice management system is already tracking? My initial reaction here is like, if you have to ask, then you're not missing anything. Like, there are no features to a CRM CRM that are so core and so important that rise to the level of like, hey, you might be missing out on this thing. Where most people pull an additional tool in is if they're doing like email marketing and like outbound and they need more of a proper sales funnel built around this. It is worth noting practice management systems notoriously have not been great about letting third-party tools access the data in the practice management system. The one most common area where they do give access is around contacts, around clients that are inside of that system. And this is so that you can build or you can pull in, you know, a third-party CRM or a, like an email newsletter platform or something like that and be able to push and pull contacts from your practice management system to that third-party tool. But I would say it's only worth going through that pain of adding yet another tool if it's solving a meaningful problem for you. An interesting development here recently is meeting recorders, which are incredibly important. The assistants that will sit in on your, your virtual meetings and capture transcripts of those tools. Those are kind of a light CRM that now is almost a mandatory part of doing accounting firm work. Like it is so, so important to build, uh, a, your sort of backlog of transcripts to be capturing these meetings. There are a lot of really cool sort of net new things that a CRM can do for you now that we have AI meeting transcript tools. So for example, let's say you're recording all your meetings with clients. And you can with a single AI prompt say, pull any exact quotations from clients here that could be used as a client testimonial. How cool is that? A client says something very kind and you're like, oh, this, we could put this on our landing page. Pull that stuff automatically, go ask the client for permission. But that thing almost rises to the level of like, oh, maybe we need an admin actually like running this every single month so that we're always in the process of harvesting these positive things clients are saying., kind of on the fly. We recently did a, a sponsored video on the channel for Ping, which is an AI meeting assistant. A recent thing they launched is an automated way to identify revenue opportunities. So it will see into not only your meeting transcripts, but also your client emails. You can even connect it to, you know, a proposal tool like Anchor so that the tool itself knows the scope of services that you're engaged to provide to them. But it will literally call out things from your conversations that are net new revenue opportunities, additional additional work that you could take on. This just, this like whole new group of things that are now possible once we begin capturing our meetings. But like, these are not things that are really solved by a CRM. In fact, I would argue there's an even maybe greater argument for a CRM because AI can automate harvesting from those meetings of useful things. Like, wouldn't it be nice if you actually kept an amazing CRM of all the conversations you have with your clients to be like, what's their kid's name? What, what college did they go to? What's their preferred tea? Like all these different details where if you brought them up the next time you met with that person, they would be amazed. AI can extract all of these details and put them into a system for you, like do the thing that we've always wanted in a totally automated way. But where do those details live? Can you put that in your practice management system? Does it require a whole other system? I really try to steer people as much as possible, try to contain all this stuff inside your practice management system because you don't want yet another platform, yet another source of truth. But oftentimes, like, the client list in our PM is just that— it's a client list. I do think we need to expand the thinking around how we can use this and what different types of information we could get in there that are actually extremely useful. 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 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, 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 is actual, like, proper IT help, like this episode's sponsor, Farodil. If I had to shoot from the hip, uh, with what I've seen from folks, Farodil 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 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— whoa. Can't recommend them enough. Check out the link down in the show notes to learn more. Next up, how do I attract clients for my new construction financial operations advisory service, given that I have one satisfied client and I'm developing a website, newsletter, and webinar? Brother, that is a lot of infrastructure for this early on. I would say at a single client, all you need is a landing page and a dream. I mean, if you're me and you're one client in, like, I don't know that I would even be totally convinced that construction is what I'm going after. Maybe you have a background in construction. That's how you end ended up here. If that's the case and you've got 7 years of hard-learned lessons in this space, for sure, start there. But how do you attract clients? You need a landing page that is a single-page website that is focused on the ROI that you can deliver to them. So this is not a list of my services. This is, uh, the average construction company is burning X% of their cash every single month on this. Here are the ways that we can help. I don't know if you saw the Claude Design video that I did on YouTube recently. I'm incredibly proud of it. Because Cloud Design is so, so good. And I had it make a website for my fictional accounting firm, Acruy. And this is actually great for this person specifically if you're getting into construction. The whole idea of Acruy was it was an accounting firm. What is stupid to say out loud is it is an accounting firm for construction companies. And the website that Cloud Design made was so ROI-based, like it called out all these different things of like sort of leaky areas of construction businesses, and then it roll all that stuff up into an ROI calculator of like, hey, here's probably the different things that we can enable for your business. Actually put in your company size and this detail and that detail. By the way, great. Now we've got those things. But the result is like this ROI live calculator on the landing page that puts the value of your services into real dollars and cents. That is the single most important thing that we have, because if we are taking in referrals, like any way that somebody finds you these days, they will first go to your website. And they need to be dazzled. And historically, for us to build a website that's truly different and truly impressive, honestly, that has not been a trivial thing. In fact, what most accounting firms do is they try to stand out on like aesthetic, like the look and feel of a website, a modern website design. But an aesthetic is not like cold hard cash. Like, I want that client to see if you pay me $6,000 a month, here is why that is a no-brainer for your company. A cute frilly design does not communicate that. Right now with AI tools between Claude and Claude Design, the ceiling for how cool of a website you can sit down and make yourself is so, so much higher. Where in the past it was like, uh, maybe I find a template that I like, maybe I hire a designer and we can make something that's cool looking. Now you can do very complex things on your own website just using AI to write the code. So don't sleep on just how good that landing page can be. But beyond this, I mean, you got to be really, really sure that this is the space you want to operate in if you're doing both a newsletter and a webinar, because those are, those are both like assuming you're doing a monthly webinar. Those are both pretty like big lift projects that pay like compounding dividends long term, but it takes a while to get there. The bigger thing to me is like, I think you need a landing page and you need to hit the pavement. Like you need to show up in person in front of these people, have a whole bunch of conversations. This also speeds up the feedback loop where you are learning about their companies and learning what problems you want to work on and really validating that this is where you want to set up camp. But the faster you can have a higher volume of those conversations, the better. So I would say keep it as simple as that landing page and showing up at events. Best practice here is always like, what is the choke point inside of an industry? Maybe this is some sort of licensing that all these people have to go through. Maybe it's the permit office. I don't know, like, where are all the places these people collect. You know, for dentists, it is the dental supply rep that is going in physically to all these locations. It's obviously the schools these dentists are coming out of. Try to go hang out on those choke points and have as many conversations as you can. How do I promote a sense of teamwork with remote team members working in different time zones all around the world? This is a tough one, man. I'm an old fart in that I really enjoy working in office. It's— and it's very uncool to say, But just working shoulder to shoulder with people is, is just a whole vibe. It's a lot of fun. Now, how do you recapture some of that in a remote team? Honestly, it is time spent together. And so like, it depends on your time zones. In a perfect world, you build a workforce that has at least a few hours of overlap every single day, and that is the expected time that they are turning up for work. In a perfect world, you've got a little bit of overlap there. To build a greater sense of camaraderie and like greater transparency between people and kind of understanding of what they're doing and their life circumstances and all that. It just takes time spent together organizationally. How you structure the teams also makes a difference here. If you have, for example, like running a hybrid accounting firm is very, very difficult. If you have an in-office team and a remote team, those kind of become like two different fiefdoms and they don't really overlap all that much. But if you just think about the structure of like a pod, for example, simply putting people inside of a pod together forces them to work together. So there is some sort of intentional design that can solve for some of this, creating more touchpoints between team members and then having shared goals. If everyone is like working in a vacuum and just has their own projects, very hard to build a sense of camaraderie around that. If they are working towards like a shared goal of here is sort of the output expectation of the entire team. Then people have more opinions about how all the work in the team is getting done and have greater stakes in getting involved in helping out however they can. The historical model of an accounting firm where it comes down to like every single individual's realization rate, that is inherently isolating. You do that in a remote environment and that is frankly kind of awful, even more awful than the in-office version of that. I've always loved like building pods of like junior level people and senior level people, like a mix of capability, a mix of onshore and offshore. And then when those folks are working together towards a common goal, some of this stuff just happens through the process of them trying to get there together. Next up, Intuit Enterprise Suite. I have a client with 6 entities on QuickBooks Desktop who wants to switch to Enterprise instead of QBO. He also uses QuickBooks Payroll for 3 of the entities. How are you using QuickBooks Payroll still? I thought that was dead on Desktop. Does anyone have an opinion on good, the bad, and the ugly? Uh, Intuit Enterprise Suite is a, like, it's a consolidator. Like, we've had platforms that do this for a number of years now. And so it is their own sort of like kind of half-baked entry into the same space. Look at Translucent. Translucent's a tool that's like a roll up a bunch of companies into one platform and they don't even just have to be on QuickBooks. They can be on other systems. Also, hot take here, look at Claude. Claude is incredibly good at consolidation stuff, like the stories of people chucking in 10+ companies. That don't even have the same chart of accounts and Claude just does it. Like it creates the grouping. It does all of this. This is one of those fringe things where it's like maybe actually AI out of the box is just plenty good enough for this now. The bigger problem here is you got 6 entities on QuickBooks Desktop and it's, I mean, that's a, that is a mother to switch that to a different accounting ledger. Maybe some of those entities are very simple. If that's like one complex entity and then 5 very simple entities, that's kind of a different ball of wax. Very simple entities, honestly, in my opinion. You can do accounting in Claude now. You really can. And it feels like heresy to say aloud, but the company with 100 transactions a year, oh my gosh, that is 5 minutes in Claude and you're done with it. Like no need for another monthly subscription for a company like that. But the hardest thing in what you're describing is everybody's still on desktop. That's tough. That is a gnarly migration. A lot of kicking and screaming, tearing of robes. Just getting 6 companies to QuickBooks Online for the first time. Godspeed. You'll do great. It's going to be great. It's going to be great. Speaking of great, we are great at nonprofits, condo associations, trades, and professional consulting firms. How do you declare a niche on your website without shutting out your existing client base or making them feel they're not valued? Did any, did any red flags go up when I was reading that question? Let me, let me read that again. We're great at nonprofits, condo associations, trades, and professional consulting firms. How do you declare a niche on your website without shutting out your existing client base or making them feel like you're not valued? I don't know that you've gotten to a niche yet. I mean, this is great that you found 4 potential areas. This is like 4 areas where we could potentially specialize. And that's hard because that's like probably too, too many, but it is maybe better than having zero. So how do you niche without like ticking off everybody else? I mean, you can't niche into 4 things. Like you can't build a brand around Hey, well, come on in. We are great at nonprofits, condo associations, trades, and professional accounting for professional consulting firms. You're not there yet, but the progression for most usually looks like you've got your generalist website and then from there on your generalist website, you have niche landing pages. So JasonStaatsAccountingFirm.com/trades. And then when you're ready to go all in on, on trades, for example, you have a specialized website sub-brand. So Jason, Jason on Trades. So I've got a bit of whooping cough going right now. Most people, like, after they have their sort of personal identity accounting firm brand, once they specialize in a specific domain, they will be, you know, the Fitness CPA or something like that. And then in the footer it'll say powered by James A. Johnson CPA LLC or whatever that looks like for you. So that's normally the progression is generalist site, generalist site with landing pages. Those are just slugs. /trades, /condoassociations. And then when you're ready to go all in, you go to a standalone brand that is Jason on Condos. Jason, that really— I mean, we could really use that in a lot of different contexts. Jason on Nonprofits. Jason on Trades. I like Jason on Trades. Something about that. That's nice. Okay, next. This question is ever so long. I'm going to give you the short version. It's really more of a story. So this person, they're using a very well-known platform in this space, but they built their own version of that platform and it was working just as well as the paid off-the-shelf thing. So they've basically replaced a pretty core, not cheap— this isn't a practice management system, it is a more workflow thing. They've replaced this not cheap platform that they were paying for with Claude, just Claude. They are raw dogging it. And they told this company they had a negative credit balance that they were going to leave and they wanted to settle up that credit balance, and they had a little conversation around why they were leaving. And the company asked, can our founder hop on a call with you? They assumed this was just an attempt to keep their business, but then they're like, this is actually a very, very tiny firm. Like, it was not a call to keep their business. And they ended up spending 20 or 30 minutes on a call with the founder of this platform, walking them through how they do this stuff in Claude. The founder said this is the first time they've seen somebody leave their platform for Claude. People talk about accounting firms like being afraid of AI and all that. Software companies. Man, software companies are shaking in their boots right now. So here's the end of the question from this person. I listen to the podcast. I'm also part of Chad Davis's Automation Town community, so I don't feel like I was doing anything groundbreaking. That is a nice way of saying they feel like they're behind because they're listening to a lot of the really, really progressive folks. But this person said I was the first to ever leave their platform because they're using Cowork instead. And that really surprised me. Am I in a bubble and 99.9% of firms aren't doing things like this? But also like what other apps are potentially at risk of being replaced in a similar fashion? Thanks for the work you do. No, I do. I think, I mean, 99.9% of accounting firms are doing nothing with AI right now. Unfortunately, a lot of the folks who are listening to this episode, it's a huge spectrum though. I mean, even we had a, a roast, a roast of Jason on firms on, on the Tax Pros subreddit last week where, you know, 80+ people share, share their thoughts on me. And the biggest, the biggest negative theme was all he does is talk about AI right now. Now. But then literally 2 comments later, you would have people that are like, I use Claude for like 100% of XYZ. And there's just like this weird stream of, of comments that are like, these people live in the same universe. Like there's a bunch of people that are like, stop yelling about AI to me. And then a whole other group of people that are like, yeah, this AI he's talking about is like doing all of my work for me. And it's, it's weird that all of this is happening sort of in the same universe, right? And so it's absolutely true that there's a huge spectrum of comfort level with this stuff right now. But what are the apps most at risk of being replaced in a similar fashion? Historically, I would have said anything outside the accounting ledger and the practice management system is up for grabs, like the, the more peripheral tools. And I would even include tax prep in that. But honestly, I think even the accounting ledger is up for grabs potentially. Like we've, and I've shared before, I just like a lot of the testing we've done and we've got a video coming on this. I'm so excited for the stuff that we've done around like, can Claude just be an accounting ledger? And if you bolt a couple things onto Claude, the answer The answer is yes, it becomes a very, very good accounting ledger. Not for the biggest, most hairy, most complex things, but that is not most of the accounting ledgers that we manage. Most of the accounting ledgers, like most of our clients, they have one big successful business and then 5 other random rinky-dink small businesses that are just a pain. Claude can do all but the biggest, most complex stuff. So I think there's a bunch of software-dom that is up for grabs that Claude could potentially just replace in our day to day. And you know, the, the hard thing in our ecosystem is such a massive percentage of thought leadership is put out by software companies that they're not gonna tell you this. And so I don't think it's talked about enough. Like, I don't think people are realizing that this is actively happening, but I do think we'll probably be paying for a smaller number of tools over the next few years because Claude is just kind of gobbling some of that up. I, I wanna do a roundup. I've got actually a whole bunch of versions of this question. The short version is, is for firms under 5, for solo firms, for small firms, what do you recommend to have team-level security using Claude? Thoughts on using the consumer Pro/Max accounts for production? Small firms, solo firm owners are very understandably bent out of shape about this. Like, this sucks. The only way you can buy the business commercial version of Claude is with a minimum of 5 users. And this is not just a Claude problem, this is an entire software ecosystem problem. There's a bunch of companies in our space that will, for example, you can only use our tax workflow tool if you do at least 400 1040s a year or something like that. And there's been so many versions of this problem in our space, and I've had this conversation with many founders that are like, here's the deal, is we can onboard a single-person firm and then they expect us to spend like 40 minutes setting this stuff up. Or maybe with the nature of our product, we have to spend a half day with them. From getting this stuff set up. Tax software, if there's any integration with the tax software, this is a huge headache, takes a bunch of setup to get going the first time. But at a single-person firm, they're like, they're only gonna be able to send us X number of clients, like they're only paying us $40 a month, or, you know, whatever it is for like a flat sort of recurring fee product. A lot more if it's like per client. But the software companies are then like, we literally can't make money on a company that size. And so they draw the line and they say any firms below this size, you're not for us. SafeSend has started doing this now. I think their threshold is like 500 1040s a year, I think I heard. And I wish I had a great answer for this. It is just a really tough spot to be in for small firms. Now some companies are acknowledging this. In fact, I don't know when this will go out on the podcast, I just did an ad read for Financial Sense where they are like beating the drum of we are for small firms and proud of it, tired of being, you know, disregarded by all the companies just going after big firms. We are for you. And so there ought to be like a counterbalance to this of hopefully somebody acknowledging this part of the market is sort of underserved. But to bring this back to the original question, is it worth paying for 5 seats? That is $150 a month when really maybe you're only solo and that's just you. In my opinion, Claude is still worth that because even the Pro and Max consumer plans on Claude don't come with a commercial agreement that says your prompts won't be trained into our model. It is a setting where you can turn it off, but you still don't get that commercial agreement, and you still don't get AICPA SOC 2 Type 2 certification on the consumer plans. So I mean, you got to make up your own mind here, because even on the most secure versions of Claude, I can't say, yeah, you're good to go do XYZ. Everybody's kind of got to find their own comfort level with this. I would tell you myself unfortunately, I would still be burning $150 a month on a 5-seat team plan, even though the only seat in the room is just me and I'm, I'm so, so alone. It is a bummer that the team plan starts at a minimum of 5 users. There's also a threshold above that where beyond like 10, maybe 15-ish users, I'm gonna spring for Enterprise because you actually get some cool stuff with Enterprise, like a million token context limits. There's actually some cool like performance upgrades you get there as well. But yes, for small firms, this sucks. Like, it leaves you in, in kind of spot. Hey, real quick, I want to invite you to my super excellent weekly newsletter. My newsletter was like the start of my thought leadership journey. It's where I try to collect what I think are the most important things for you to see every single week, sent to you every Wednesday morning. Make sure you don't miss any of the good stuff, right? Increasingly, we're also building into that some like of the best evergreen thought leadership I've ever done, kind of make that like the most action-packed, highest quality email you get to your inbox each week. Because listen, we're all busy. All right? You're not going to see every single little thing that I put out. Goodness sakes. There's just too much. If you're not on my newsletter yet, check that out. Jasononfirms.com, or I'll put a link to it down in the show notes. Hey, quick word from one of today's sponsors. Me. Me. Yes. I run a community, Realize, that is for accounting firm owners. And Realize is kind of the secret behind how I've done all of my content and been able to like financially turn thought leadership into a business. About 100 firm owners in there. Realize is a systematic way to build connections with other peer firm owners. It's not a framework for how to run a firm. It's a way to introduce you to other super cool firm owners. Right now, we're putting together small groups so you can have a group of like 5, 6, 7 other firm owners just like like you that you meet with every single month, run very similar firms. Having this for me totally transformed firm running. My north star, like my goal for you coming into this group is to make a new best friend, somebody that runs a firm, because that was what I needed in my journey. The whole community is built around that. It's called Realize, link down in the show notes if you wanna learn more. Two more to go. Is there an LLM that you use besides Claude? If so, which one and for what purpose? Right now, I don't. All I use is the Claude stuff because I believe you and I's time is important enough that when we are in the exploratory process of figuring out what can AI do for me, we should only be doing that with the very, very best models. This is why I've been telling accountants for years, stop wasting your time with Microsoft Copilot. Your time is more valuable than this. And they're like, well, but we get— it'll just be a little extra work and we can, we can figure it out. We can get there. Meanwhile, on the shelf, you have a vastly more powerful tool. Use the most powerful tool. The only situation when work for me gets pushed down to a more basic, more cheap model is when I have an AI workflow that is running itself and kind of running in the background. For you, that might be, you know, every time a project hits a certain status, this, maybe this workflow kicks off, that sort of thing. It is in production, you could say. At that point, there's an opportunity to delegate some of those tasks to lower-cost models just to keep the whole thing cheaper. But if it's me, a human sitting down talking with an AI assistant, I'm a special snowflake. My time is valuable. I will not waste my time with anything but the very best models that are out there. This is not too dissimilar to being like, hey, there's two people you could possibly hire. Do you want the really, really capable one or the less capable one? I'm going to be like, what? I don't— what's the question? Of course I want the more capable one. So I think your time is valuable. You should only be messing with the most capable stuff. Last question is a big one. You ready for this? Is QuickBooks Desktop the future? This is a reference to a post I made on Twitter about the QuickBooks Desktop SDK reference. Now, this is like a bit of local software that, that developers would use to build things that connect to the QuickBooks Desktop program. This was like before the cloud ecosystem was a thing. This is how you could make a desktop program that that could connect with QuickBooks Desktop. And I made a joke on Twitter saying, is QuickBooks Desktop SDK the future? The notion I could have an AI agent that just uses— I mean, QuickBooks Desktop, let's not beat around the bush, the most capable accounting application in history. And still, it is still the GOAT. If you want as much functionality as possible in a single program, you can't touch QuickBooks Desktop. I will fight you on this. So what if there was this like like automation layer on top of QuickBooks Desktop. That's what the SDK is. You can write code to automate stuff inside of QuickBooks Desktop. How is this not the future? I'm sort of joking, sort of not joking, but at this point everybody's like, ooh, how do I just like get AI to drive a really good accounting ledger for me? Because AI is very capable and I, and I, we're close to a bunch of versions of this. Like, uh, the ledger company Qik just launched an MCP so you can connect your cloud to Qik. Kik, but it's not only a read MCP, it is a write MCP. And so it can do things like import transactions, classify transactions, run reports, all these things. So I could literally go to Claude and give it a pile of bank statements and say, go do the accounting in Kik, in the accounting ledger software, and Claude will run off and do that. I shared the example of, um, the open source library BeanCount. I think I shared this a few episodes ago. BeanCount BeanCount is this open source library that's been around for a while where you can basically do like your programmatic accounting, like write code to like classify the transactions and all that stuff. But it stores the full accounting ledger in a text file. And I know that sounds like kitty, but like it is actually like a database that is stored in a text file and it is lightning fast up to about 50,000 transactions. And this video we have coming out is around using Claude and BeanCount because Claude I can give it a pile of bank statements and it can extract all the transactions and it can do pretty darn good classifications. The problem is where does it store all that stuff long-term? BeanCount. I can literally just say the word BeanCount to Claude and it will go out and pull the open source library and give me a text file that is like a persistent database for that accounting file. And I can then take that text file to some other conversation with Claude and say, hey, use BeanCount to make these updates or post these new transactions, blah, blah, blah. Blah, blah, blah. I think we all are hungry for like a maybe desktop, just like low friction backend for some of the stuff that we're doing on Claude to make it persistent. Like I'm used to having like these reporting views for my accounting software and all that, and I can see just how convenient and helpful Claude is, but it's still not an app. Like it's still not a proper fully developed app that gives me all the views and everything that I need. And we're like, is there just, is there some sort of shortcut cut, some like light backend that we could develop and Claude just sits on top of that. And that's, that's where the whole joke around is QuickBooks Desktop the future came from. And, and I, I, I think the answer is yes. I think the way accounting work gets done in the future is not these humongous bloated programs. And if it is, if those programs are still around, it is an AI agent doing the bulk of the work inside of those programs for us. But we're in this now entering into this sort of messy transition phase where it's like, do I go out and buy one more accounting ledger? Monthly license? Like, is that a thing that even makes sense right now when Claude— I mean, we've shown videos Claude doing 6,700 transaction bookkeeping files in less than 10 minutes, and the classifications are as good as anybody I would have on my team. The tech is there, but we're like, oh man, what, like, what do I do with that? Like, what's the right way to use that? Could I actually, like, ship a set of financials using just that? That's kind of the tipping point that we're all on right now, kind of trying to figure out what makes sense here. Imagine though, so like the consumer plans of Claude, they can control your computer and like move the mouse around and do all these things. Not available on the business plan because it is a tremendously risky thing to do. The notion that it has access to every tool that you have access to, that's pretty scary. But hear me out. Imagine you just had this on a single machine and the only application on that machine is the greatest application of all time, QuickBooks Desktop. Let's call, uh, let's call this machine Mike, and I could be like, hey Mike, got this new client in, spin up an accounting file, here's a pile of bank statements. Mike works away in QuickBooks Desktop. You can hear it on the other side of the office, boop boop boop, entering transactions in the register. Mike wraps that one up. Hey Mike, can you go update that file we're working on last week? I mean, that's, that's a thing. The technology is here today. Is that a dumb way to do it? Claude automating your mouse and keyboard and using QuickBooks Desktop? It might be. It might be the nostalgia that, that makes that sound fun to to me, but I do think the days are limited of us like buying traditional SaaS software in many ways. Not that all SaaS will go away, but I do think we're gonna be relying on cloud for more of this stuff. But you know what's not going away? This, you and me, what we have together. Okay? I'm very excited for a whole bunch of AI use case videos we have coming to the YouTube channel, just like showing how to do all this stuff, make it feel more accessible. It is a wild time that we are living in, but we're gonna go through it together, right? Everybody lock hands. Accounting on 3. 3, 2, 1. Good morning!