
Be the first to curate this episode — add a title and quick summary.
Add title and summaryNo information listed yet. Be the first to add who benefits from this content.
Suggest who benefitsNo detailed summary yet. Suggest a summary to help the community.
Suggest summaryNo questions listed yet. Be the first to add a question for this topic.
Suggest questionThis week, Tracy Bech explains how and why she spent six months building, training, and testing her brand new 60 Minute Custom GPT, which is essentially an AI CFO that can perform all kinds of financial analysis on your business. You can interact with it as easily and conversationally as you do with ChatGPT, and it can help you figure out why your margins are off, or how you should expect to perform next quarter, or whether that new service you plan to offer will be profitable. And it’s free.
Transcript from YouTube captions. May contain errors.
[Music] Welcome to another 21 Hats dashboard. I'm Lauren Feldman and I'm here with Tracy Beck who is co-author of the 60inut CFO and CEO of Driven which facilitates CEO forums. Welcome back Tracy. >> Hi Lauren. Thanks for having me. >> Oh, it's great to have you here. Tracy, you you recently sent out a newsletter in which you pointed out what you called a painful truth, which is that sometimes driving more revenue can actually make your business less profitable. Uh you also offered up uh an interesting AI tool that owners can use to figure out which of their revenue streams are most profitable. But let's talk about that conundrum first. Uh h have you seen owners drive more revenue uh but make their businesses less profitable? And by any chance, did you learn learn this lesson the hard way? >> Well, yes. Actually, like all lessons that I learn and most that I speak about, I've personally walked the walk. So, you can count on me to go ahead and and make all the mistakes for you. Um, and then learn my lessons um and then teach and teach the fix for it. But, um, thank you very much. >> You're welcome. Happy to do it. Um, but yeah, no, there's a couple different ways that this works. And um you know, one of the one of the classic ones that I'm seeing a lot is that um as founders often times we think more revenue is better. So we sort of blindly go for it. And what we don't look at behind the scenes is what it's going to cost us to get that revenue. And we all we're all well, okay, I won't say all because that's a generalization, but you know, we're pretty familiar with how much money we're keeping at the end of the day because we can see it, right? like we can uh we we hope that we're pricing our our products and services appropriately so we have some money left over to um pay for the overhead of our company. But what happens as you grow is that often times you're still pricing your products and services appropriately to um to to make enough money. But what you forget is that you might have to stair step up a level in your overhead expenses. So that might look like hiring more people, maybe renting a larger facility. Um there might be some one-time expenses that you take on like you might decide you need to rebrand or invest in a new product um like software management tool or something like that that's going to either long-term or short-term really uh affect your profits. So so one of the things you have to do is as you're doing this is be really real about a couple of things. One is if we chase more revenue, what is that going to do uh to our overhead? Is it going to cause us to increase overhead? And if that if that increase um does need to happen, is it one time? In other words, that's just something we can sort of um take a hit on one time, but then get back to business as usual, or is it a long-term thing like these are going to be month after month, you know, increased rent, increased salaries, increased, you know, electrical or whatever it is going to be that we, you know, we need in order to run the business. So, in in a way, you're excited because remember, uh revenue is a vanity metric. We get very excited about it, but that revenue isn't necessarily going to drive our profits if it increases our overhead, right? So, it's a trap. I want to acknowledge you know f founders sometimes think as you say uh more revenue is better because there are a lot of people telling them to think that and you know I would include venture capitalists whose you know the the business model that they follow is uh investing in companies and trying to drive them to grow as quickly as possible get as big as possible without a lot of focus on profitability but we in the media uh meaning me in the media uh I've worked at companies that, you know, really emphasize that vanity metric uh that you're referring to and in part because it's the one that we know most companies are willing to share. But, you know, when you hold that up as the ultimate yard stick, as often happens in the media, it encourages founders to think that's the thing they should be chasing. Well, and that's okay because it's as long as we focus on the keyword invest because investing implies um putting money in and watching it grow over time. So, what happens is is your revenue, you know, on all investor models is always supposed to go up and to the right, right, on a graph over time. And what h and the way your overhead works is it stair steps up along, right? So there is a there's an inflection point at various revenue levels within your company that you will be required to increase your revenue. However, that investment in that or excuse me your overhead but that um revenue hopefully continues to go up and to the right and that's where you know private equity is bank that's what they're banking on. So they're saying keep growing revenue because what they're then looking for is the next the next inflection point before you need to increase overhead again. So if you can drive revenue on top of that increased overhead, the question just becomes how long will it take for us to uh regain our previous profitability levels? And that's where you want to get into some really good scenario planning and forecasting and you know some real strategy that's probably going to be pretty specific to your business and your industry. I think one of the strategies you mentioned in your newsletter was to make sure you fully understand what your fixed expenses are and what your variable expenses are so you can kind of plot that growth. Correct. >> Exactly. Yes. And that's and that's because those variable expenses they take care of themselves. So as revenue grows or as revenue declines, variable expenses by the definition, their very own definition, vary with respect to revenue. So they they bump along right alongside your revenue. And if if revenue goes down, so will they. And that's kind of the beauty of those things. It's the beauty of also knowing which one of your expenses do that, right? So a great example of a classic variable expense is uh commissioned people, right? people who work on commission make more money when you do and make less money when you do. So think of think of the other expenses that behave in your company like that. Um your cost of goods, right? The more more t-shirts you sell, uh the more cost of t-shirts that you had, right? And then the fewer t-shirts you sell, the fewer t-shirts you you have um as part of your cost of goods. So um your your fixed expenses however those are what we classically also refer to as and not all overhead expenses are fixed but classically you know generalizing um overhead expenses are fairly fixed right so your rent for example isn't is a classic overhead expense that just doesn't really change it doesn't really care how much revenue you did it just is what it is right it would be nice if it went down when revenue went down I guess it wouldn't be as nice if it went up when revenue went up but it just sits So you you think of it as um hey but the reason we pay rent is because we make a whole lot of money by being able to have this facility to work from, right? And in some people's case, maybe they even uh rent part of it out if you know if they're in my if they're in my moving company's niche um for storage. So um there's lots of reasons to have those things, but they do um they do typically stick there. So you want to know how much revenue can I drive on top of this amount of overhead because that's the cool thing is once you realize this is fixed um it is fixed until you need to grow it again. But how much can we grow before we have to raise overhead? >> It would be like somebody trying to sell something uh to be distributed through Walmart or Target or something where um you know the uh it it it looks great. It sounds great, but suddenly you have you need more warehouse space or you need more production lines or more employees, whatever it is. I guess employees wouldn't be fixed expense, but is is that what you're describing? >> Um, yes. And that's that's also a really interesting scenario when when outside pressures require you to grow at a much faster rate than you had previously been growing. So it goes from what we might consider kind of a bootstrapping organic growth rate to a surge of growth. And then we have to look for how are we going to fund this growth? And that's that's where we can actually say the company might be growing broke because it's going to take so much capital to maybe like fulfill that order um and so much manpower and like warehouse space and infrastructure to grow at that rapid of a rate that um you have to actually find a way to do it, right? Okay. And that that that that revenue, yes, well, we definitely want it is going to come at a pretty high price and risk to the company. So that can that kind of sort of teeters into the land of are we growing too fast as well. >> So one of the other things you suggest is looking doing an a margin analysis looking at uh what happens at different revenue levels. How do you think about that? >> Yeah. So that's kind of an interesting thing which you can again model it out, right? So that's going to help you see uh oh well at you can go backwards in time and you can go forwards in time. So if you go backward in time you can you can see at what different points in your business history you were the most profitable. Um I have a gentleman in my membership community right now who's a plumber and he has a 70% or something gross profit margin because he's doing all the work in his company and has no overhead. Right? And then you see, okay, so that that that revenue level, it's not a very high revenue level, but it certainly is a profitable one. It comes with all other kinds of, you know, catches and risks being, you know, a sole proprietor and service provider. You know, you you're doing a lot of work. It's a lot of stress and if you get sick, you don't get paid, you know. So, so then you look at growing and so he's literally modeling it out. What would it look like if I bring on more people, more trucks, more inventory so that I can scale to a new revenue level? I said, 'Y yeah, you're not going to have a 70% gross profit margin anymore. Now, will you maybe have a um maybe um a 48%? Yes. And is that in line with industry averages? Yes. So, is it wrong? No. It's just it's just the reality of the cost of doing the business at that revenue level. It's it's impossible for you to get to, you know, 500 or $800,000 in revenue as a single plumber. So, you're definitely going to lose some margin as you go. Um, and then of course like kind of um kind of intertwined in that question is also just really making sure that you're looking at the various margins of the various things that you do. So um there's going to be services and goods within your business mix that have higher margin than others. Right? So, um if we if we take the classic case for me of um someone who's in the moving and storage industry, um they have they can make more money on maybe like a a local move than they than they make on a long distance move just because of the the time and effort that gets put into something like that. And then their highest profit margin work is on um just packing and storage, right? Because they can they can pack something, they can get into storage, and then that that revenue becomes recurring revenue. and they don't do anything to that stuff while it's sitting in storage. So, it's really important to look at how how much you are doing of certain um services or you know and then same thing like if you're selling you know different things within your store, you probably have different margins on various products. What are your high margin services? What are your high margin goods? And if we're going to chase revenue, let's chase the revenue for those things, right? And be very strategic about that. >> Got it? So, here's what I'm really eager to ask you about, which is you have created a tool to help business owners figure these things out. Uh, I think you're calling it the 60-minute CFO custom GPT. Um, which, um, sounds a little scary to me, but also really interesting. Tell us what it is. >> Okay, so it's an AI tool, right? It's built on chat GPT. And one thing that you probably know about me uh by now, Lauren, is that I love big pictures. I love to look at the broader uh scope of things, but I get really tired if I have to do things that are down in the in the details of things or um do them over and over again. I don't have a very high tolerance for repeating tasks. And the very cool thing about business finance is that once you've ga you know garnered your education around business financial fluency um when you start doing analyses of income statements and balance sheets and as you know there are 14 key ratios that 60-minut SEO teaches. >> Of course I know that >> you know that like the back of your hand >> of course who doesn't know that and I know all your listeners know that. um it does not change and so I've saw a real opportunity to outsource a lot of this analysis to um AI and uh so the the opportunity to build custom um agents right within chat GPT has been around for I think a couple years now and there's a nice market or it's not a marketplace but it's a library of of GPTs they call them so I um I've been spending the last six months just working with and training my little custom robot that is this, you know, custom GPT. And the cool thing is is that we were able to um upload our book to it. So, it learned everything that's in our book and then it understands all of the the methodology, our our verbiage that we use that's very, you know, non-b businessiness finance jargony and it um so it does the a really good job and I would say, you know, certainly on these things that don't change with regards to business finance, it does an excellent job of of really combing through the details of your income statement and your balance. to look for the nuance. Um, I think it's a really great, you know, I wouldn't say it's totally black or white, but it does a really nice, very thorough job that, you know, my business brain I'd like to think is pretty darn good. But, like I said, it's it's I think it makes sense to outsort this outsource this kind of thing to a robot to help you along, right? And it gives you so many more details than and it's very consistent. And that's the other thing that's nice, right? It's not going to just like have a bias toward um looking for inflated costs. Like it's going to look at the whole picture for you and really give you um a couple different angles at which to see your business from. >> And do you interact with it the same way you interact with chat GPT? >> Yes, it is almost identical. So, the thing about a custom GPT that's interesting is that if you especially if you have a paid account through custom GPT or chat GPT, it knows you, right? Like it's going to remember things about you. It's it's forming a relationship with you. It knows your name. It probably for me in my case, it knows what I do for work. It knows about my family. It knows all kinds of things, right? It's my friend at this point. And um when you get over to a custom GPT, it doesn't know you anymore. And that's uh specifically for ours um by design because I don't want um users. This is your financial information. So I don't want it to retain information about you um in particular. I only want it to go off of what you're telling it in this specific chat. Um and that's because of this the the privacy issues around having financials in a tool. Right? So, um I think it's really important to note that that we have specifically told this this tool not to retain information or to share it with the broader LLM at at Chat GPT, right? Um and that being said, I still think that anybody using AI should know that you are still interacting with um a set of servers and computers that you don't own. So, um you know, that's not specific to my tool. I would just argue that you want to be careful whatever you put in to these tools. Um we don't have actual control and we might find out in a year that you know all of this was a bad idea. But >> sure, >> for now I'm taking my chances. Um but that's how we're we're um structuring it on our end. So if you ask this if it knows your name or your company's name, it's not going to know. Um it doesn't retain that information. >> And are you making this available to the public? Are you charging for it? How's this work? >> Yeah. No. So, it is available for free. I have um I think it's worth a lot of money, but I don't think um that at this point it's something that I feel like I I mean, I've created it and it has all of our um uh intellectual property in it, but I you're going to you're probably going to end up wanting to pay chat uh chat GBT for an account to use it. And I don't think monetizing on top of that makes sense. So, this is free. you might if you start to run out of credits with chatbt, it'll ask you for money to use it with them. Um, and otherwise it all it takes is um the I think the best way to interact with it is to have open um your your QuickBooks uh reports like your income statement or your balance sheet, export it into Excel, and then copying and pasting it from there. And if you're already a 60-minute CFO devote, you can copy and paste it from our business mastery spreadsheet, which is the template we teach from. Um, either one will work and then it will and you'll say analyze my financials, and it knows what you mean because we've taught it to run what we consider a full report. So, it will spit out all of your ratios that we teach, those 14 key ratios. It will, if you've given it more than one year, um, it will do a trends analysis. So, it'll say, you know, your company is liquid and getting more liquid or it's risky and getting riskier. Um, it will highlight three um areas of excellence and three areas of concern and give you suggestions for what to change. And then I've also, you know me, I like a good mom joke. It will infuse it'll infuse a little humor into it and say there's this it it kind of uh I asked it to kind of do like a superlative statement like you'd see on the Tonight Show where they're it's like you're your company's risky. It's risking it for the biscuit and the biscuit's getting burnt like it comes up with wacky things. Um but I think when we're dealing with finances it's good to keep a little bit of a light a light heart about it. It's hard enough. Um, but what gets really cool is let's say you're like, well, let's say it gives you your three challenges and it says, you know, your gross profit margin has um has been is is strong. And if you tell it what industry you're in, it will go out to the internet and grab what it knows about your industry averages, which is really cool. So, it might say, "Hey, your gross profit margin at 48% is strong and it's in line with industry averages. However, it has been declining by half a percent per year." and you're like, "Oh, can you?" So, you'll say, "What is driving that decline?" So, now you're chatting with it and it'll say, "Well, based on what the information you gave me says, it's your labor." You're like, "Oh, well, what should I do?" And it will it will say, "Well, there's two ways to, you know, fix this." One is to look to do your work more efficiently with fewer people. Or the other way would be to raise your prices. And so, you would say, "Oh, well, I don't want to raise my prices. So what are what are five ways I could reduce you know my labor force um and work more efficiently. So now it's doing scenario analysis with you. It's it's using what it knows. And remember this is getting information from the internet, right? So I always I always think um the best way to interact with AI is a sandwich approach, right? So first use your brain, then use its brain, then use your brain. Again, we can't we can't just end on the AI brain note. You got to like put your own your own thoughts and analysis into this. You know, I asked uh Chat GPT to help me prepare for an interview with an immigration lawyer that I did recently and the answers the the suggested questions that I got back were pretty good except it kept referring to the immigration crackdown being conducted right now by the Biden administration. And I I said, "Don't you mean the Trump administration?" And it said, "Oh, thank you for correcting me." Yes. That started under the Trump administration is now being continued under the Biden administration. >> Oh my gosh. Yeah. You're like, it's the year 2025. Yeah. Help me out here. >> Yeah. Yeah. You got to remember that you are um you are still the master of your destiny. We got to we got to like put our own common sense in here and and and make sure it's the thing I've learned about um AI is it doesn't like to not know the answer. So rather than saying, "Oh, I'm not really sure about that," it will just say whatever. it will make something up. So, you really got to look um and and then you you correct it and it goes, "Oh, you're absolutely right." And you're like, "Oh my god, >> good thing I good thing I said something." >> So, how does someone uh find this free tool that they can try out if they wish? >> Um, the easiest thing to do is if you go if you head over to 60inut CFO, we have a big bold uh banner at the top that says get access to our custom GPT. that'll take you right to a a page where you can uh trade me your email address and then I'll send you over the the link to it. Um and you can get going. You can and I always tell people I'm like please for your own time management decide ahead of time how much time you're going to spend on this. Set a timer on your phone and then be done because >> Are you suggesting you get lost in this and uh >> spend hours? >> Yeah. I just I just unveiled it to my community yesterday and we did a whole demo and they're like, "Thanks a lot, Tracy. I just I'm going to lose a whole day to this," you know, and I was like, "Hey, man, you got to you got to manage your own time. It is addictive." Um, and it's the thing is it's it's really user friendly. So, you know, spend some time with it. It's going to it's going to make you more financially fluent just by talking to it because you can ask it so many questions and it will answer you. So, um, I do recommend that. Are there particular use cases or questions or issues that you would suggest starting with? I definitely start um I think you should grab your income statement in whatever format you want like if it's year end last year or if it's year to date of this year and um preferably maybe even like two periods like if it's you know two two years of information start with your income statement cuz that's where everyone everyone's more financially fluent on profit margins right and let it um start there and get get some ideas and answers there and then um and then copy and paste in some balance sheet information to round out the picture, but um and and it will it it when it's as soon as it gets data from you, it runs our full report. So, you're going to have so much information and then it will just like any good agent ask you what three things for it to do next and you're just going to get carried away. So, don't even worry about dead air with with this tool. It will it will continue the conversation. >> Can it help you figure out whether uh future revenue is going to be profitable or not? Yep. Absolutely. You you um I mean give it some information to go off of, but that kind of scenario analysis it is so good at. So you can say um you know given what you know about my business and remember this is also all about um asking it very pointed questions. There's there's so much talk about being a good prompter, right? So sure >> um you know you don't want to it will otherwise play to your your happiness. it wants you to be happy, right? So, um you have to say, you know, given what you know about my business, if I bring on another million dollars in revenue in this business line and assuming I need to bring on these overhead expenses today in order to support that work. This would be a great prompt. Tell me um how much revenue more how much more revenue I need to get back to today's profit levels. Right? That would be like a really good prompt. So, it's like asking it, hey, I'm this profitable today. I know I'm going to increase my revenue, probably take a hit in profitability. Tell me where I need to get to in revenue um in order to re regain that level of profitability that I'm at today and start there because then it's going to it'll give you like 16 different things. It's so fun. >> That sounds fabulous. Yeah. >> Um what did you learn about creating your own uh custom GPT that you can share? Uh I think you said it took you 6 months. Um was it painful? Did you enjoy it? Uh what would you advise others who are thinking about doing something similar? >> Oh man, I would recommend doing it. Um it's no. So it took me one afternoon to build it, but it took me 6 months to gain confidence in it because I didn't want to share it with clients until we had all um you know, my analysts, my human brains that work with me here at 60-minut CEO, I wanted them to see what they thought of these answers. uh not just me who you know I think it I think it was like spot on. I wanted a group more of a group effort on that. So that's the six months. But um it literally here's the thing. It's a very meta exercise because I used my regular chat GPT to write the uh code that I then put into the custom GPT. And there are um there are podcasts, there are resources, there are actually multiple AI developer communities and I joined one of those. I learned how to do it there. And once you sort of listen and watch other people, you know, listen to the directions, watch other people have success with it. I I just kind of carved out an afternoon and just sat down and said, "I'm going to do it." And then it just was surprisingly shockingly simple. So, I would highly recommend if someone's if you're out there and you have those those busy work tasks or those repeatable um you know methodologies that you continually continually employ over and over again. See if you can't build that into something that AI can do for you and then um and then use it for a little while. And we we did it side by side. We would do our human analysis and then we would run it through chat GPT and we were just like wow this is incredible. So, it was um it was well worth that afternoon and and kind of um devoting my brain to the the learning of the process. So, yeah, highly recommend. >> That's amazing. Tracy Beck is co-author of the 60-minute CFO book and CEO of Driven. And if you want to try out her uh 60-minute CFO custom GPT, just go to 60minut CFO.com. >> That's it. >> All right, Tracy, thank you so much. This was great. Oh, it was a blast. Thanks for having me, Lauren. >> Have a great week, everybody. [Music]
About 21 Hats
21 Hats is an online community for business owners. Entrepreneurs have to wear a lot of hats to build a business—but some hats fit better than others, right? When you’re not sure where to turn, the 21 Hats community is here to help. The 21 Hats Morning Report scours the web every morning for the most important stories for business owners (https://21hats.substack.com/p/coming-soon). The 21 Hats Podcast has been tracking six businesses throughout the crisis in weekly conversations (https://21hats.com/).
People who have contributed edits to this page.