Get Rid of Your Crap: Clean Up Your Business Data Before a Sale
Every business piles up digital clutter: old files, duplicate documents, outdated customer records, forgotten software subscriptions, and multiple versions of the same document scattered across shared drives, CRMs, and cloud storage. It rarely feels urgent. But that clutter slows your people down, raises technology costs, adds cybersecurity and compliance risk, and now limits the quality of the answers you get from AI.
It's also an ExitReadiness® issue. When a buyer begins due diligence, they'll examine your systems and information as carefully as your financial statements. Organized, current, trustworthy information builds buyer confidence. Outdated or forgotten information creates questions, delays, and doubt.
In this episode of our AI & Exit Planning Series, Pat Ennis and Corby Megorden talk with Dale Tuttle, Partner at FutureBridge, about what "get rid of your crap" really means inside a business, where to start, how AI can help with the cleanup, and why human judgment still matters every step of the way.
Key Takeaways
Run a practice deal room. Even if an exit is years away, ask: "If a buyer asked tomorrow, what would we put in the deal room?" The drill shows what's missing and surfaces what should go.
Embargo first, delete later. Move suspected clutter to a clearly labeled area of your network that your AI tools (and SharePoint search) can't see. Validate it before anything is permanently erased.
Use your domain experts, plus an arbiter. Your proposal manager, CFO, or HR lead knows what's current and what's obsolete. Have the owner or a senior leader make the final call. Don't do this in isolation.
AI doesn't know what's crap. It will confidently use outdated information to form an answer and tell you it's a great one. Garbage in, garbage out still applies.
More data isn't always better. The "lost in the middle" problem means large language models give less weight to information buried in the middle of a large set of documents. A smaller, cleaner set produces more accurate results.
Old isn't the only test. Relevance is. Date is a useful filter, but a well-written proposal from ten years ago may describe services you no longer offer. Ask whether it's germane to the business you're building today.
Keep what you're required to keep, and no longer. Records past their regulatory retention period cost money to store and create liability, especially anything with personal information. A simple retention policy keeps clutter from coming back.
Give buyers only what's in scope. Old budgets and outdated financials that slip into a deal room lead to questions, lost time, and reduced confidence.
Where to Start: Dale's Three Steps
Assemble your domain experts (proposals, finance, HR, operations) and explain why you're doing it.
Give them clear rules: don't erase anything yet; flag and embargo what they believe should go.
Review together, then act. They show what they recommend removing and why; leadership decides. Then put a retention policy in place so it stays clean.
About Dale Tuttle
Dale Tuttle is a Partner at FutureBridge, which provides operationally focused exit planning for business owners. He holds a PhD and works at the intersection of AI innovation and business operations, including hands-on work building AI-powered applications. Dale brings more than 20 years of experience leading technology consulting firms, organizational transformations, and teams of more than 50 people across government, commercial, and nonprofit sectors. He has held full P&L responsibility for practices from $5 million to $20 million in revenue, guided three M&A processes, and is a Certified Exit Planning Advisor.
Connect with Dale: LinkedIn · futurebridgeinc.net
Ready to Assess Your Own ExitReadiness®?
Building a more valuable, transferable business isn't only about growing revenue and profit. It's also about reducing risk, improving systems, organizing information, and creating a business someone else can understand and run.
Prefer to start on your own? Start free with ExitReadiness® DIY™: proven frameworks, practical tools, and assessments, with Certified Exit Planning Professionals available when judgment and experience matter most.
Want guidance? Contact us at email@ennislp.com or learn more about ENNIS Legacy Partners.
Disclaimer: The information presented in this podcast is provided for educational purposes only. Neither the presenters nor ENNIS Legacy Partners are engaged to render legal, accounting, tax, or other professional services. Consult a qualified professional regarding your specific circumstances. ENNIS Legacy Partners assumes no legal liability for any loss related to the information contained in this presentation.
Full Episode Transcript
This transcript was generated automatically and lightly edited for readability. It may contain minor errors.
Digital clutter has a cost. It slows employees down. It increases technology expenses. It creates unnecessary risk and in today's world it also limits the effectiveness of AI that you're starting to use more and more. After all, AI is only as good as the information it's given.
And there's there's another reason that this matters and it's one that we care very deeply about here at the ExitReadiness® Podcast and at ENNIS Legacy Partners. Someday if you decide to sell your business or transfer it, a buyer's, excuse me, due diligence team is going to look through your systems just as carefully as they well more than likely just as carefully as they review your financial statements. The more organized, current, trustworthy that information is, the easier it is to build a to build buyer confidence and reduce unnecessary friction during that transaction process. In other words, cleaning up your digital environment isn't just good housekeeping, it's good business. It improves operational efficiency today while helping build a more valuable and sellable business for tomorrow.
So today we're going to discuss why every business owner should get rid of their crap, what really what that really means, where to start, how AI can help with the cleanup, and why human judgment still matters every step of the way. And our guest is is our friend and strategic partner Dale Tuttle, partner at FutureBridge. Dale holds a PhD and works at the intersection of AI innovation and business operations building AI-powered applications with Claude Code and bringing that hands-on experience into into his advisory work. As a partner at FutureBridge, he provides operationally focused exit planning for business owners. That's why he's a strategic partner with us.
Drawing on 20 plus years leading tech consulting firms, organizational transformations, teams of 50 plus across government, commercial, and nonprofit sectors. He's held full P&L responsibility for practices from 5 million to 20 million revenue. He's guided three M&A processes. Like us, he's a Certified Exit Planning Advisor. His PhD is in international relations.
Well, what is how can that help with exit planning? Well, it's trained him to think rigorously and we've experienced this and conversations about social dynamics, incentives, change management skills he applies directly to helping owners build businesses ready to grow and sell. So, Dale, welcome to the podcast. It's great to have you here with us for the first time. Thanks very much.
I really appreciate it. And you know, maybe we can get into there is a connection between the PhD and what I do now. It's maybe a little ambiguous, but somehow it got me there's a link there that got me into this sort of stuff, but it's a pleasure to be here. You guys are doing great work. I'm happy to be a part of the show.
Okay. Fantastic. And Corby, hey, thanks for popping in here. I'll get us I'll get us our first qu get the first question. Get the conversation going, which Walter usually does, but I'm gonna do it today.
What made you want to bring this? Well, we were having a conversation on my porch, smoking cigars and having a a beverage or two, and we talk about this idea of getting rid of your crap. That's a quote. And so we thought that's a great topic then for a podcast conversation. So what made you want to bring that idea to exit planning specifically?
So it's it's kind of an interesting topic because it's been around for a long time. It predates AI for sure. In fact, in my previous lives, I've always had a knowledge management organization as part of the technology consulting organizations that I've led or been a part of over the last, you know, 20, 25 years or so. And that's always been an issue. So, you know, organizations have always struggled with crap and what to do about it.
And I think for a little while there, people thought that it was solved when AI started to make its appearance a few years ago. But In many respects, it's they can make things worse by not getting rid of your crap. So, that's where it comes into exit planning. So, there's that operational impacts, if you will, right? It's harder to find things.
You might you don't know what good information might be. All that sort of stuff just operationally makes things the friction, higher friction. That's really what I'm aiming at. But when the advent of AI comes around, you thought we would fix it, but it doesn't necessarily do it, which we'll talk about. But for exit readiness, it's the same problem.
When you get ready for a transaction, let's say you're, you know, you're you're going through due diligence, a deal room is going to be set up. Well, there's going to be mass panic in getting data together. You know, your contracts, your teaming agreements, you know, your financial data, all that stuff, good stuff needs to be put into the deal room. Well, if you don't have your house in order, that process will be a, you know, a complete chaos producing event within the organization. Even if you're a small organization, right?
Even maybe you're only five or 10 people, it will be chaos because you'll go through the exercise that we'll talk about today, but you'll do it in oh, a hurried fashion. Okay. Hey, Dale, say that last piece one more time because for some reason or another, you cut out. Oh, okay. The So, what can happen is in in the deal room is that there could be if you're have too much clutter, too much crap, the creation of the deal room will be very very difficult to do, cause a lot of stress in the organization and you may introduce information into the deal room that you really don't want to put in there.
Mhm. So, I think from an exit planning perspective, if you have things in order, more or less in order on the back end, when the time comes for that exit, you'll be ready and you'll actually demonstrate to the potential buyer or the investor, whatever you're really up to in your exit, that you have your house in order. And that means a lot. So, they'll have confidence that you have things at your fingertip and you'll be ready to answer questions with good solid information, data that includes financial stuff, contracts, teaming agreements, that sort of thing. Yeah.
So, hey Dale, we you're not an investment banker just like us. You're not an investment banker. You're not a business broker. We've had investment bankers of course join us here and they've said pretty much the same thing you just said. So you're coming at it from someone who who who doesn't do the transactions but you've you've you well you were involved with three M&A transactions on the sales side right?
Yep. And so you've experienced this firsthand as as u being on the team the internal team of the company that was being acquired. Correct. Mhm. Yeah.
And I went through that stress where you get the the message that we need X Y and Z and we need it tomorrow morning and of course it's the end of the day and you're not sure where those current teaming agreements are. You don't know where the last 10 proposals that you won are. Or they may be comingled. That's the big problem. Stuff is comingled because someone in your organization will know by looking at file names.
Oh, that's the good one. That one I wouldn't use again. That sort of stuff happens with the financial data too, by the way. Budgets, you have your old budget, you know, and then of course you're tracking against a new one. So a spreadsheet that may be too old, that works its way into the deal room.
That happened to me. Where you know, just stuff that should not have been in the deal room ended up in the deal room. Yeah. Yeah. Yeah.
Hey D comes to mind is you know and everybody knows this when you go to sell a house you finally get around to cleaning up all the crap but you know this there this woman named Marie Kondo did this book a few years ago about simplifying and getting rid of your clutter now. Are there benefits and I'm assuming there are to even if exit is like out there to getting rid of your crap now? Yeah, I mean that's a great question. There's every impetus to do it now and because a lot of the a lot of the benefits that there intuitively you understand what the benefits are but first of all there's cost retaining all that information. And we talked a little bit about it.
There's inefficiency that's introducing cost. If it takes someone three hours to find something that's costing you in a certain way. It might be a cost that shows up later, but also cost can be measured in a different way and in a it's not just inefficiency. First of all, incorrect information will worm its way into whatever you're doing. So if you have really old stuff, someone may grab it and it may be factually incorrect now or it may be it may be a problem with liability if something is disclosed.
That sort of stuff happens all the time as we all know. So if you get rid of stuff, you're reducing the probability that that will happen in the future and then impose a future cost. So I do think the the operational inefficiencies are you know important to consider but then there's just the cost of storing all that stuff. You know storage costs are going up. Now having said that you know one drive and a lot of these things are very generous in how much data you can store but when you start getting a lot of stuff if your business is around 10 years old you probably are exceeding your storage quota and that starts adding up especially if you're in the medical a medical practice and you're storing images.
I have a radiology client. They have their their radiology images. They store them longer than they need to and that costs big dollars. And they're actually now after speaking with them recently, they're they're starting to get rid of that stuff. Yeah.
Yeah. And and when we were talking the other week, you know, we talked around just, you know, AI impact on process, etc. And can you speak to like working with AI and the need for AI to have the right data to help you actually make good business decisions. You know the old garbage in garbage out concept in decision making and in management. So, I think it's really it might even be more important now because I don't think I don't think it's to your benefit to have the AI aware of your crap because it doesn't know it's crap necessarily.
And what'll happen is that it can start using this stuff as it it starts, you know, making its formulating its answer, applying its AI judgment. You know, you ask a question, it's going to give you its judgment. If it's using bad data, the crap to formulate an answer, stuff will can worm its way in. It can and will worm its way in. So really, this is probably one of the bigger things to think about now is just from an operational efficiency, you don't want your AI to be aware of that stuff as much as possible.
Now, and I know this is something you might want to dig into a little bit later, there are tools, AI tools out there that will help you reduce your crap, okay? Or they claim they can filter out the crap, but especially for a small medium-sized business. A, you probably can't afford those tools. B, I'm not 100% convinced that they're really effective. So, why not just old-fashioned get rid of the stuff that it's aware of, right?
Only make it aware of stuff that you think is good. Now, some some stuff might not be necessarily 100% current in there, but if you at least reduce what it's seeing, it's less likely then to formulate an answer with the bad information in it. However, I still think for now anyway, for now, you still need people in the loop because a person, especially in a small business, they'll know what's going on. They'll know what's good and bad with with the AI results coming through, but you want to reduce what they have to sift through. Then a problem I know I'm I'm talking a lot about this but the other problem is remember the AI will confidently give you an answer and it'll be obsequious too.
Hey great question a very insightful question and it will give you an answer that really sounds great but it might be wrong and that can happen actually even with good information by the way less and less now because the models are becoming extremely powerful. But you so I for now you still need a person who can evaluate the returns you know the stuff coming out of the AI and go you know what that's good stuff and but you want especially for someone that may not have that experience within the firm let's just reduce the probability that you're going to get a lot of garbage back and you do that by getting cleaning up your house like you're going to move and get stuff ready for it to ingest. Yeah. Hey Dale, when you you mentioned a few things, but is there anything else that you would say in regard to when you say get rid of your crap? You know, I think you've mentioned a few things that can be.
Is there anything else that you would say what that actually looks like in a real business that about getting rid of the clutter? Yeah. Yeah. Yeah. What what they actually should think about getting rid of.
Yeah. I I would So, you know, there's different ways to think about it. It a lot of it depends on the kind of business. So, if you're a if you're maybe you're a proposal, you know, you're helping people write proposals or, you know, or your business is, you know, federal based, you're probably accumulating a ton of RFP responses, a ton of teaming agreements, right? All this a lot of paper, okay?
Contracts, these things accumulate and grow quickly over time. And so that's really in that scenario, a couple things really need to happen. I would your proposal manager can be tasked with all right, what what ones are just are crap are garbage. We just got to get rid of them. Maybe you lost.
Okay. Which doesn't necessarily mean it's crap, by the way. You might have put together a great response, but you happen to lose, but the information in the proposal is still good. But when you really get old proposals, the services that you provided are probably out of date. Why would you want to like if if a proposal from 20 years ago about solving an information problem came came across and it started talking about setting up solutions with tools that are barely exist anymore, that stuff can worm its way in because remember the AI doesn't necessarily know.
But that would look like someone with some do domain knowledge knows what good content is. Have them start sorting through things. And it's not necessarily by date, but date is a great indicator. It's a great thing to look at. But if you're a small business, too, I do think owners can can be the person to to go, you know what?
I know that's no good. I know that's no good. But I default to people like a domain expert, like my proposal manager in this case, the CFO. He or she will know what's good and what's bad. I don't want old budgets in there.
You know especially from a long time ago I just don't so anyone with domain expertise should know and those things are typically stored in a couple of places just your network drives okay so your network drives are probably a great source but your financial system and any of your systems right your HR system okay your you know your one drive which is a network drive essentially that sort of stuff box. So, but the thing to keep in mind, Pat and Corby, it's it's content, right? It's, you know, it's document stuff, but it's also hard data, your spreadsheets, your financial data, all that sort of stuff. Yeah. So, it it clearly your clients hire you to lead that project and to coordinate it and get it all done.
Yes. If someone's listening right now and you're thinking, "Yeah, I've been this has been keeping me up at night for whatever reason and I need to get it done." and they want to take a first step internally. What kind of u person what what role on the org chart or what type of individual would that owner look to to kind of oversee this project and and move it forward internally? Yeah, I I think in a small business but even larger business has to be you do need someone who has ultimate decision to make a call on what to keep and what not to keep. So that that's that's key.
So while I you know I talked about like having your proposal manager you know that person or the CFO have they'll know what their good content and bad content probably look like looks like. However, you need to make you have to check their work, so to speak. So, someone a decision, you know, someone with a little bit more understanding about the about where you're heading as a business, the mission, or even if you're getting close to an exit, someone who might know who or whom is interested in the business. That will influence what you think is crap and what isn't crap. Right now, if you're way far away from exit, it can be pretty straightforward, but you do need an arbiter.
Someone that says, "Look, it we're going to keep that. We're not going to keep that." Your domain expert may have a different opinion on what's good and bad. You can have that conversation, but you you need to have that conversation. I know I'm stuck on a proposal manager scenario. He or she should show this other person, this is what I think is no good and this is what I think we should keep.
You need checks and balances. That would be the best best thing. Don't do this in isolation because you can lose stuff. Now the good news on getting rid of stuff is there's probably a backup somewhere you know like especially in one drive but certainly in Google and box you first of all you erase stuff it's there in the background for a little while before it's completely gone but your backups will probably also have it if you make a mistake but to prevent the mistake from happening you can again have like this committee view of it like proposal manager or CFO tell me what you think is garbage then you have a meeting people meeting this and they said this is what I think is garbage. Then you would need to move it.
Okay, oldfashioned move it. Now there are tools that'll move it. You can actually go tell Claude, go mo go move this stuff from here to there and you can name something you know clearly name it as as as junk or crap. So that is oldfashioned knowledge management work that hasn't changed in 20 years. Some of the tools have changed, but that work I I hate to say it, but it's grunt work, right?
You need to get in there, look, decide, and do it. It's it's really that simple, but you have to do it. Yeah. And Dale, there's this, you know, idea again, it's come up in our conversations, you alluded to it in passing, just the the context, i.e. What is the authoritative document versus the the iterations of the document etc.
Are there essential like if small business owner that's listening right now? What are the what are the top almost contextual documents that they've got to focus in on to start with? Because there's probably a ton of crap out there, but you know, where would where would they start get their act in order in these buckets? And do you mean in in terms of process or the types of things they're looking for? Types of things.
It's you know it's I I think we depends a little bit on the business right but like certainly around here we have a lot of federal contractors you right so it doesn't necessarily but they they will be dealing mostly with documents so your focus would be on documents then you think about types of documents what kind of documents are we worried about contracts is a biggie proposals is a biggie teaming agreements are huge. But then marketing material. Don't forget about marketing material. Your business u what you do and whom you've done it for that changes over time. So you need to make sure that that stuff is current.
It really doesn't matter what you did 10 years ago. It it doesn't you know and that stuff will survive over time. And especially in the age of AI, the AI will read that Now, it might pick up that the date was 1999, but it might not necessarily tell you when it says, "Hey, that we did this," but it doesn't might not necessarily associate what you did with 1999. If you ask it, it might tell you, "Oh, yeah, that what I told you was from 1999." But you need to ask it. So, look, a lot of a lot of times you can by your prompts, you can kind of get to where you want to be.
But remember it it doesn't know what good and bad is. I know l you know we all think that GenAI is you know going to solve every problem and I maybe someday it will but it doesn't make a value judgment. You know the CEO does the owner can make a value judgment. One of your trusted senior adviserss can make a value judgment. The LLM is simply creating a response based on an algorithm and it doesn't know if something is good or bad.
People do. Now, you can try to trick it and influence what it thinks is good and bad, but why do that when you can just make it a lot easier? Just give it stuff that is more or less good. It won't all be accurate and you can start, you know, being clever with your prompting, but your likelihood of getting a bad answer if you clean this stuff up and your data first. But, you know, you start looking at your network drive, start looking at your financial systems.
And I do and and I know it's a long-winded answer. I was thinking about this since we had our conversation out on the porch. I was thinking as a thought exercise, I'm a small mediumsized business. I'm starting to buy into the idea that we have a lot of junk and I don't want it out there, right? Because it's slowing us down.
What even if I'm not even close to an exit, go through an exercise about we've been asked to put a deal room together. What would you put in a deal room? And that will influence every focus everybody's mind about what you want somebody else to see. And that that little scenario, it's almost a drill if you will. That will I think it'd be a great exercise for a small medium-sized business to do.
Because you will then a be ready if something like that happens in the near future, but b you'll clean out a lot of your crap. Maybe one thing I should mention is when you do isolate your your your junk, don't erase it, right? Then just put it in the part of the network that is not going to be seen by the AI. Okay. And so that way if you've made a mistake, you do have access to it.
Of course, you know, we just talked about having backups and all that sort of stuff. So, I would delay, especially in a scenario of doing a pretend deal room. Embargo the stuff, isolate your your what you think is crap, but don't get rid of it, right? You could even then use AI to compare and contrast the content over there and over here. So you can do all kinds of great things we could never do before Gen AI showed up.
So I think it's easier to do it now than ever, but the need to do it is bigger than ever because AI is out there. Yeah, that's it's fascinating because as I've been as I've been wrestling with this this tool that we talked about that I'm putting together, a lot of the thought process is where do I want this data to wind up three steps down the road. But if we get that clear, AI is actually pretty good about saying, "Oh, you want this to go there? I can get it there." and and it it's it's actually very useful, but without the end destination defined and clear, it just it's going to pick a course. It'll pick and you don't know if it was correct or not.
It'll tell you, I got the I got this. Yeah. Yeah. And it'll sound great, but you won't really know. That's why, you know, I think the role people play in these processes is still very important, but it's changing.
Yeah. It's it's changing. Yeah. So, hey listeners, you just heard Dale give you a just a practical step to take and that's just move ahead and set up a data room and start to, you know, fill it up as a process to help you identify gaps here. And who knows, a buyer could come along or a potential buyer can come along at any point in time.
And it would be good to already have that that in place so that u you don't have to scramble. Dale, what's the worst thing you've seen happen when a buyer finds some something an owner forgot was even in their systems? Does that question make sense? Yeah. Well, you know, a lot of it is is Well, there's two parts.
What have I seen? I've just seen massive embarrassment in the M&A rooms that I've had, you know, the M&A processes I've been a part of. I've seen bad data go in. In fact, I think I mentioned earlier some old financial data went in. Somebody thought it was current, it was not.
And so then you start getting quizzed by the buyer. Well, what about this? What about that? And so that that takes cycles. It can reduce confidence.
So that that that's a bad thing. But I don't I also think the other thing is we talked a little bit about it is the confidence thing. If you can quickly get a deal room together or even have one ready, so to speak, that just shows to them we have our house in order. We know what's going on here. So but bad data leaking in is is bad form for sure, especially financial data.
But the other thing that can be bad and this is more of a for regulated industries, financial, medical, this has not happened to me but but it's it's a risk is if if stuff flows in that really you should have gotten rid of, right? And that you're only you're supposed to you're supposed to keep stuff for a certain period of time. Well, if after that period of time expires, it should be gone. And why? Well, most of it's liability.
If something if you're only like a medical record, I think the the mammography team I'm working with, it's either five or seven years. It doesn't matter what the time is. They have to keep their images. There's no reason to keep them after five or seven years or whatever it is. There's just no reason.
Because it's got a patient's name, a bunch of other PI, you know, personally identifiable information in it. Why keep it? It's costing you to keep it and it's a liability to keep it. So, that that is like a really bad thing. But, you know, I I think a lot of it comes down to why keep stuff that just might be even factually still true, but might tell a different story than you're trying to tell a potential buyer.
Or maybe have some bad financial results from eight years ago. Well, if they're that's beyond the period of their scope. Usually they'll give you a scope and what you know three years of information. Well, don't have five years of information in there. Because you just don't know how that will be interpreted.
So then there's one more thing we haven't talked about Pat and maybe you were going to ask me about this because we were talk about the intersection of AI and getting rid of your crap. There's a technical reason to do all this too. And that's when if you feed AI a lot of information, there's this interesting phenomena called lost in the middle. You may have heard me talk about this and maybe you've seen it out there. Large language models if you feed it a ton of information.
U what happens is that there may be really great information in the middle. Maybe in the middle of, you know, you gave it a 100 documents and those documents are all multimegabyte documents. There may be some great stuff in the middle. You know, the the order that it has ingested information, but it doesn't see it. For some reason, large language models tend to weight stuff that it finds in the beginning and stuff it finds in the end.
Maybe the first third and the last third look great to it. It retains that information. The stuff in the middle, that middle third, it tends to weigh it less. So, it will not necessarily come up in a in a in a search or be taken as a as a a factor to provide into an answer. Yeah.
So, so is a solution there to to make sure that your prompts u you know prompt AI to to also look at that lost in the middle data. Well, the the there there's a couple ways to deal with lost in the middle. First of all is reduce how much stuff you're showing to the AI. So, you're going to reduce the problem, so to speak. It's called a context window.
It's how much information that the that the AI can hold at any one time. If you just reduce how much is there, you're going to reduce you're going to reduce the probability that a lost in the middle phenomena will re rear its ugly head. The other thing you can do actually, and this is very technical, we don't have to dump into this. You can chunk your documents, right? There are tools out there that will take your giant document and and create chunks.
So then it's not dealing with one big piece of information. It creates a vector database. There's some other things in there that will allow it to associate chunks with what you're asking it for. And so that's it's called RAG, retrieval-augmented generation, and it's been around since LLM. This it's been around since LLM showed up because they the the scientists the you know the engineers figured oh my gosh this is happening we got to do something about it.
But for business. Yeah. So along these, you know, chunks and the other thing you were Oh. Okay. He's back.
Dale. So some people would say just point I'm just going to point AI at everything. All of what we've been talking about here. Yeah. I Yeah, I'm on board.
So, what I'm going to do is I'm going to point AI at the problem and have AI sort it all out. Is there something wrong with that? That's that is I would say the most common thing I hear with people that are that I know that are business owners, but even just people I talk to, like, I gave it all everything I have and it's doing a great job. Well, it sounds like it's doing a great job. It'll tell you it's doing a great job.
How could it not be? And it'll tell you you're doing a great job. Yeah. It tells you you're brilliant, too. Thanks for asking that great question, right?
But yes, that is I've seen this personally. We have a client right now that has tried this and we're weaning them off from this this idea that this is a good idea because it gets into that lost in the middle problem, right? If you're you're giving it all this information, it's going to lose some of it. It's just not going to weigh everything the same. And you may be thinking it's going to give you a great answer, but it completely missed some great information, right?
But you won't know. You won't know that it's missed it. It'll tell you you've asked a great question. It'll tell you it's provided a great answer. So, that will happen.
That will definitely happen because we're talking about, you know, most businesses been around for over five years have terabytes of information and it's just wrong to just feed all that into your large language model. Yeah. So, how does a human like you come in there and help them with that? What we do is we work with the domain expert. So, for instance, as this one scenario, it's a federal contractor and we're right now doing two things.
We're helping them identify the crap. We're working with a proposal manager. She knows what's good and bad. So, we we we ask her questions about all right, well, what what's older than five years, right? So you can start with that temporal that timebased thing but then I like to I like to ask la ask people what what's a good proposal and what's interesting will happen you'll get great proposals from 10 years ago but that's just not relevant anymore because the services that you were providing back there are probably not germane so I would my retort is I this is well written you guys.
One, I think you did a good job on that contract 10 years ago. But why is that important? People want to retain things naturally that they had a part in creating that were good. I'm like that. I reuse pieces of proposals that I won from the past.
But you got to just think, is this really germane to what you're trying to answer today? So get rid of the stuff. But you know, and I should be more clear when I say get rid of. Yeah, you can erase it, but don't necessarily erase it out of the gate. Keep it around, but embargo it, isolate it in your network and don't let the AI see that stuff.
Don't let SharePoint see it if you can help it. Put it on an old place on the network drive, but everyone knows that it's crap. Label it as crap. And so no one will mistakenly use it. So that that process is really working with people Pat and Corby still to this day.
Now we may then suggest once you get rid of the crap using some AI tools, pardon, right? And then have the AI start helping you find junk. And then what you've done though is because you've limited the stuff that it's going to see, it's going to ha have higher accuracy answers. It'll be better at helping you get rid of more crap. But don't let it make its decision.
You know, Claude Cowork came out what, last year, earlier this year. When Claude Cowork came out, remember it'll go in and do stuff. Do not let let a tool like that go out and do stuff without human oversight. And don't just click allow, you know, click allow it. Just it'll go do it and you'll be out there trying to recover stuff for the rest of your day or week.
All right. Okay, guys. We got to wrap up. I've got another meeting here in a few minutes. What here?
Last question, Dale. If an owner is convinced, and I'm sure there's at least one or two that are convinced listening, where do they start? Can you give them one, two, three steps where they should start to address this problem? I would assemble my domain experts. That means proposal managers, financial people, HR people, you know, we that's another component.
Assemble them and and give them clear instructions. We're trying to get rid of the crap. Speaking technically, and they'll intuitively know what you're talking about, but give them some rules, right? Don't erase things. Embargo things, okay?
So, but get your people together. Let them know why you're doing it. Okay? And and then that process could be them then making suggestions. Here's what we want to get rid of.
Have another meeting. Have you know this does not have to take long. They can show you. Here's what we think should go and why. And then let let people do it.
It it does not have to take a long time, especially in a a business. But if people if you tell people who know what they're doing that this is now an important part of their job, they'll go do it and then let them and and but just to be clear as we're introduced now, just embargo this stuff for now. Validate if it's crap. I'll make one caveat on this though. If you if you have a retention if you have to regulatory, is that a word?
If you have to retain information for five or seven whatever the heck years it could be obviously you have to keep it but after that there's no reason to keep it. So use that by the way as a guide about getting rid of information right if you don't have to keep it don't. I worked for a large accounting firm as a partner on advisory side and they had very aggressive retention policy probably too aggressive. Maybe we'll talk about on your back porch someday. But non customer data had to be purged after one year, which is a little bit too aggressive, but we could actually introduce some I would think about a retention policy.
That would be the last thing. What's our retention policy? If it's three years, after three years it gets flagged, reviewed, maybe it goes, maybe it doesn't, but someone's going to look. All right. Fantastic.
Okay, Dale. If listeners want to contact you for Oh, first off, is there anything you want to promote today? Well, you know, FutureBridge, we we take a little bit of a different view of exit planning. We're more in the operational side. So, we're talking about the stuff we've talked about today.
So, people can definitely visit us and and see how we can help them, hey, get rid of their crap, but also improve operations. We're heavily into AI like a lot of folks are, but you know it's not for everybody, but we can we can help solve these kind of operational issues around exit planning and that's what you know the owners out there listening today keep that in mind. All right. If they want to get in touch, how do they do that? Yeah, I'm at dale.tuttle@futurebridgeinc.net and happy to chat with anyone and of course it's been a pleasure chatting with you guys again.
And I hope to see you both soon. I'm sure you will. And listeners, if you'd like help building that more valuable, transferable, sellable business or getting ready for that eventual exit, you can reach Corby and I and Walter at email@ennislp.com. And if you if you're someone who you just have a DIY kind of mentality, particularly when it comes to exit planning, and you'd prefer to begin your planning process that way, check out our ExitReadiness® DIY™ platform at exitreadiness.com. And we appreciate you listening.
Thank you. If you found today's conversation helpful, please consider liking it, sharing it, and then subscribing as well. And until next time on the ExitReadiness® Podcast, this is Pat Ennis and Corby Megorden, without Walter Deyhle today, signing off.
