Rethinking Scoping & Estimating: A Better Way to Understand Loss

Good afternoon, everyone. Welcome to Rethinking Scoping & Estimating. I’m Leah from Encircle. The reason for this webinar is, we’ve been hearing a lot from the industry about, once the job is wrapped, the an estimate doesn’t quite tell the whole story of the job, and we’re we hear more and more that it restorers are getting pushed back on their estimates. So we’re here with Jeff and David. I’ll intro them in a second to just rethink the the whole process of scoping and estimating and how we can think a little bit differently going forward and maybe see some better results, and we can untangle scoping from estimating a little bit. Without further ado, because I said I would be finished talking by 12:05 and I’m almost done, I wanna introduce our, our two panelists today. First is David Herring. He is, the founder of RiseDocs Inc and, WriteLoss Claim Write Inc. With over 3,000 clients, they’re the largest provider of property insurance estimate in North America. I’m super excited to learn from David today. He’s got 30 years of experience on both sides of complex claims, and he holds level three Xactimate certification as well as professional GC license. And he’s been retained in over a thousand litigation managers as an expert witness and loss consultant, so he knows his stuff. He’s also the author of an upcoming book called Peel the Onion, a methodology guide for negotiating property insurance claim. So thank you so much for being with us today, David, and we will have to have you come back on when your book is available and you can talk to talk to everyone about that. Also with us today is, my friend and our Chief Product Officer here at Encircle, Jeff McDowell. Jeff is an accomplished leader. Accomplished Jeff. That’s a good word for wise and old. Just kidding. Told you he’s my friend. Accomplished leader. He’s got over 25 experience in the high-tech 25 years of experience in the high-tech industry. He was the Chief Operating Officer of Primal.ai, which was an AI company before AI was was a thing, before it was on the tip of everyone’s tongue, before OpenAI, before Claude, before Gemini existed. So he spent five years developing products, IP, and commercializing AI solutions in the large language model space. And prior to that, he was a Senior Vice President at BlackBerry, which gave him this passion for mobile technology, which as as we know, that that’s that’s something we care care deeply about at Encircle is how our product can function on a mobile app and in the field where you live and breathe every day. So super excited for these guys to to take it away and share their knowledge and expertise. So I’m gonna disappear, and I’ll see you in 45 minutes. Thank you. Awesome. Thanks so much for the introduction, Leah. And, yes, as a friend, I forgive the I forgive you for the old comment, but, I guess, sometimes the truth hurts in my case. But it’s great to be here. It’s great to talk to everybody about a subject that’s, near and dear to my heart, near and dear to Encircle’s. I know it’s near and dear to to David’s heart. That’s why we have him here. Although, I will start off by saying this is the first time I learned about you writing a book. Usually, people join us on these webinars because they have a course or a piece of soft or or a new solution or a book to promote, and you didn’t even tell me. So, we will as Leah said, we’ll definitely have to have you back when you’re when you’re ready to, to publish. It’s a it’s a it’s a labor of love. And, you know, like anything you do, Jeff, you’re really, really self critical of your own work. And so I’m working on getting over that so I can release it. Well, I can’t wait to see it, and and I’m sure there’s lots of people with us today that would love to read it as well. So, yeah. So scoping. So interesting. We we we always like to do that poll at the beginning because, it just sort of sets us off on a on some on some context of our audience. And it does seem like a lot of our audience already agree with the the premise of what we’re gonna be talking today, and and that really is that scoping is important, and it needs to be seen as its own individual workflow and not something that’s smushed together with with estimating when just to step back for a second, Encircle started out with a hypothesis that field documentation was a really important thing that needed to be that needed its own attention, that needed its own thoughtful app and and and and way to do it so that, you know, you could get paid in full for for all the work that you were doing. And while it’s not a really con not controversial. It’s not really a mind blowing idea. I mean, everybody probably knows this. It was certainly a lot of work and and a lot of thought had to go into building a platform that assisted our customers in putting together, you know, good field documentation and producing all the reports and forms and everything else you need to to prove at a loss. But there’s been a recently there’s been sort of a a resurrection of a thought here around scoping than what I wanted to talk about today, and that is I think there’s some tools in the industry and, you know, everybody will know who I’m talking about that has sort of allowed scoping and estimating to get smushed into one, for lack of a better technical term, into one workflow that certainly saves time. You know, anytime a tool is built, it’s it’s usually in place, and and I’ll say Xactimate. I think that saved the industry tons and tons of time in in terms of the amount of manual, entry and everything else it would take to to build an estimate. But at the same time, it it may have accidentally created some bad habits around how scoping is done and taken our eye off the ball, if you will, around the diligence of collecting the data that needs to go into the scope, which then needs to go into the estimate. And that’s what I wanted to chat about today. I have seen others in the industry in various social media or blog posts talking about this very thing that, hey. The scope’s an important piece of work. It needs to be looked at separately. I know this is something that you’ve always in in our previous chats, know it’s something that you’ve always held near and dear. So let’s let’s get into it. Like, David, tell me your tell me your thoughts. What’s your opening statement around around how we should be able get scoping? I would say that when we’re talking about being in the field, we’re going out to that structure. I don’t think we often think about the fact that we’re actually creating a narrative. We’re creating the story of the loss, and that story is gonna feed everything that comes behind that, not just the estimate, but the communication with the with with the insured, communication with the carrier, any stakeholders. It’s gonna but what we do and, Jeff, you and I have talked about this before, is a lot of times we walk into a structure and we think about line items. We don’t think about actually damage. And people say, well, that that’s the same thing, but it’s not. It’s not the same thing at all. You know, if if I say if I if I walked in a structure and I say that I’m in the main hallway, four feet in from the entry, two feet out from the right right wall, there’s about two feet worth of where the water damage or appears to be water damage. I don’t presently know what the source is. That’s part of the narrative. It wasn’t scope. We all know what the scope might be behind that, but the rest of the narrative is going to inform what that is. Because depending on what it is, we might need containment. We might need to do more demo. We might need to figure out a drying plan because it’s because the the way the building is is put together. There may be a lot of things, but it’s not line items. It’s just a story. And I I’ll explain it to you like this. If you and I were walking through a loss and we were just talking about what we see and we never talked about 64 square feet of drywall, we need three air movers here, you know, if we never talked about any of that but we just talked about the loss and we recorded that or we videoed that, the net result of that would be substantially greater than if we had walked and talked line items. You know, Jeff, we spent years chasing the perfect scope sheet. I mean, people would call me for 20+ years and say, David, do you have a good scope sheet? And and the truth is there is no good scope sheet because there’s too many differences. Right? It’s, yeah, it’s too complicated. And what AI is doing for us is, to some degree, taking the burden of the bad scope sheet and removing it. And I think that’s the I think that I think that’s the most awesome thing in the world. And I don’t know. I’ll I’ll bet you there’s not one person in our business that’s been here more than 10 years who doesn’t who hasn’t worked with some stupid scope sheet that didn’t work. Okay. And and so I just think AI is changing that for us, and I and I and I’m excited. But more than anything, I just think that what you do behind the desk may be a result of what you do in the field, but what you do in the field has a greater deal of deal of importance not just for to understand the narrative of the loss and to write that narrative, make an informed better estimate, and make a better presentation to stakeholders that need to see that. It’s funny you mentioned the scope sheet. Like, it always a lot of times when you’re looking for ways to make things better faster, people think of you know, there’s always that, that old saying, and I don’t know if it’s true or not, but I always like to use it, which was, you know, if Henry Ford asked his customers what they wanted, they would have said they wanted a faster horse because it was that’s what they that’s what they knew. They knew getting around was, you know, on a horse, and they just really, what they need is a faster one. It wouldn’t even occur to them to think about an internal combustion engine driving four wheels that then got them, you know, got from from point a to point b in a in a in a in a multiple of horse power ways. And the scope sheet is is is is is something that people constantly come back to, which is and we hired an expensive consultant a long time ago to build the perfect scope sheet that we include in our and every customer that looks at it says, hey. This is great, but can you just change it? Can you just can you just make these these they’re just trying to make something that they’re grounded in, something that they understand, and they’re trying to make it better. But, you know, they there’s just the amount of variety on a loss. It’s just too great. So, really and and and you’ve already mentioned AI, and I definitely wanted to get into AI at some point, But we can talk about it now. And and I I would say that’s really been the fundamental basis for our hypothesis about the importance of scoping is what we realized. Because we started out in field documentation, let’s capture the ground truth of the loss. Let’s let’s just know everything that happened. And, we we’re we’re trying to make it better, but you mentioned video. So we we support video. We’re creating audio notes now, like, where you’re just basically using your phone as a tape recorder. So you could just walk around stream of consciousness, not worrying about what you’re saying. Like, that’s the great thing about AI. You could say anything you want. I even tell people because this would help me. You could even swear. Like, I don’t know if that’s just part of your if that’s just the way you think, and I kinda do sometimes, especially if I’m at a loss. AI will just completely filter it all out and and just, you know, get down to the salient points. But that that is what really sort of kicked off this line of thinking for us is we’ve got all of the you’ve got all if you if you do a good job collecting the collecting the data, you’ve got all the components you need to drive AI. And if you’re just jumping the line items, you know, you’re missing some of that. So, you know, know, I don’t know. Can you in your experience, what’s Well, here here’s here’s what we found, and this is what we found 25 years ago, k, was that the the talent set that walks the loss and the talent set that writes the loss are very different. We found that people that like to sit behind a desk for 50 hours a week are not the same people that like to always see something different walking into another loss and another loss and another loss. It’s exciting for that person that walks the loss and the person behind the desk that is utterly boring to them because they’ve, in fact, think in line items. So when we separated those, what we could actually do is we could have a less experienced inspector that had better social skills. Because of the And and I don’t think that and I don’t wanna discount that. Having somebody with great social skills so they’re really good at dealing with the insured. Right. That has good knowledge about how to assess damage could produce a set of videos and and and audio even before we were using AI that could better inform an estimate than we could with any scope sheet. And we’ve been doing that for decades, and only recently have we been able to take that. We used to have a person, and the person was called scopes. Send it to scopes so that every time this data came in, somebody would digest it and then write up that scope. But with the AI, we’ve basically been able to remove that and actually build a much better product going out because the the data is so much more robust. Jeff, I’d also tell you this, that as everybody’s moving towards this, whether they’re using your app or any other, okay, when you’re when you’re when you’re talking to when you’re talking a loss while you’re walking it, I want you to walk it and talk it like you’re talking to a 12 year old. And this is the reason why. You know, you can you can if you can do you can direct a 12 year old to do almost anything, but you have to direct them. They’re not very good at coming up with their own stuff. And so when you walk in a room, you say I’m walking in the master bedroom. On the west wall, I see this. You you you talk like like that as opposed to just just saying, oh, I see something up there. Okay? You you you you do you talk like you’re just having a conversation, but you’re having it very deliberately because that will generate something that’s so magical on the back end that it makes your product so much better. You know, Jeff, I’ll give you an example. If I took a bunch of line item estimators and I put them in a in a lot bunch of losses and they wrote a lot of losses, there would be very few that that interviewed interviewed the, the insured to find out that the HOA doesn’t let you start your work till nine AM and want you out at four. They wouldn’t realize that the dumpster has to live on the street, or we can’t have a dumpster here and we have to live load it once a day. We didn’t notice that we have a real hard time getting to this particular location because there’s no side yard. We have to go through the building and carry debris through the building. Those are those are things that that affect the cost of loss, and we can always explain that to to, like, a carrier, and a carrier understands that. But they don’t understand it if you don’t tell them. And when you tell the narrative at the front of front of this process and imagine just how magic this is. You’ve got you’ve you’ve pulled this data with video and audio. You’ve pulled it into being a good outline for an estimate, and you’ve created a narrative that helps you communicate with everybody and all because you did one thing really right. One thing. So it as it happens, that field wound up being the most important thing we ever did. Right. Because it literally drives the entire process, and its quality will be relative to what we do in the field. Right. You know, I was just thinking as you were talking there, we’ve we’re we’re talking about the the importance of separating scoping from estimating. Again, going back to the initial poll at the beginning of the the webinar, it sounds like a lot of people are already aligned to that. But but now that I think about it, even the scoping the the the scoping workflow, I think you’re suggesting there’s actually two parts to the scoping workflow. And now that I think about it, that’s the way it is the way I think about it as well. There’s a damage description. There’s a there’s a very deliberate description of what you’re seeing, an inspection, I guess, if you will, to use the language, separate from the work to be done, which is actually the scope. And and there’s even, I would say, probably in in in my experience when I like, you you almost can’t help yourself when you when you arrive on a on a property that you start thinking even even if you’re not thinking in line items, you still might be thinking in terms of tasks. I gotta drop some d hues here. Obviously, some fans are gonna we’re gonna have to vacuum up this water. Instead of sitting in the environmental variables and and and understanding environmental first, have I have I got this right? Yeah. You do. You you do. Look. Look. Let let me let me give you a let me give you an example of just one thing that that I did recently. There was a there was a water loss. We didn’t know exactly where the water was coming from at the time. Okay? And, I went in, I described it much like you said. I say that there’s water that’s under the floor in the, in the great room. There’s water in the master in the closet next to the great room. There appears to be some water in the hall here. The HVAC is leaking. Turn the HVAC off. This is the recording. The HVAC is off, and it was what was causing this loss. And the insureds have been gone for, like, four days. So we don’t know how long this has been going on. But I look on the on the on the wall, and I can see the the relative humidity here is 76%. I’m not sure where the back of this wall connects and what’s behind it. You see you see what I just said? I just told the a I just told the AI, I’m not sure what’s behind this, and it told the AI, you have a question to answer. Right. Okay? So as I went back around, I said, okay. The back of this closet backs into the right run of cabinets in the in the in the kitchen. I need to look under the cabinets and see if I see water standing there, and I’ll have to pull the kick plates to do that. Okay? I have just I have just issued major amounts of scope by talking about what I see in front of me, and that will turn into I arrived at the loss at this time. I found this, and I found this, and I found this, and I found this. We think we’re gonna have to obviously deal with these things were found damaged, we’ll have to address them. And that little bitty scope right there that I just said to you is $57,000 worth of scope, right. That was literally covered in about eight minutes and literally went up to an estimator and into a narrative within hours and turned into something real. And, Jeff, you and I talked about something a while back that I think is important. What do most mitigation companies do? They write a estimate down the road that’s actually an invoice. An invoice. Right? Yes. So how would you say this helps to avoid that? Yeah. It doesn’t. And that and that is something that you know, when I when I joined Encircle five years or almost five years ago now, I had to get my head around that estimates were actually invoices. And estimates are actually ROMs, which is a whole other thing that we won’t. I’m sure, we’ll talk about another time. But but, that is something I would love to I kinda have a mission to to to to figure out how we get the the the industry talking about estimates as actual estimates and not Yeah. Invoice that you’re about to submit that just isn’t approved yet. And and then I think that’s part of it because let let me let me explain to you what I see some people doing that’s really exciting. K? Some mitigation companies that are doing that’s really exciting is they’re doing this function where they collect data much the way I just said, where it’s very detailed. It’s very fast. It’s not too typing intensive. They they literally just have to punch some buttons, and that makes it good for anybody, even somebody who’s not good at writing. By the way, most people are crappy at writing. But, but but what happens is that gets into an estimator. And as we get started on this job, that estimate’s a real estimate that goes to the carrier because the data was so good coming in that the estimate could be could be written and could inform the crew that’s there what they’re doing and be an actual estimate so that we’re because one of our biggest problems with mitigation is we don’t tell the carrier that they’re spending money. Happening. And when we and when we give them an estimate that’s a coherent estimate with a good narrative, they’re more likely to just say go and not be shocked that you did something because you already documented what you did. And I just think the speed of that is different when we do it this way. Right. And I think I think that’s that may be the greatest takeaway is that speed is gonna change the way that we do this work. So I guess that is the the main unlock of AI is you know, again, maybe this was not I mean, this was probably known to everybody. If you do a good narrative, you do you capture a good narrative, a good understanding of what the cause of loss was, the resulting damage, that then led to a good scope, a good understanding of what needed to be done, the creation of task lists and and work orders and all of that stuff, and then, ultimately, the creation of an estimate. But to go through all of that in a serial kind of way just took too darn long. Right. Right? To be to be to be really diligent with the with the the field doc that that that justified all the work to be done. Like, in order to so so you you sorta learn to become a little bit hacky. You started thinking in line items. Oh, I gotta get an estimate off to the adjuster quickly so they can see it so I’m not, you know, doing anything that they’re gonna scrub or disagree with. But those hacks were almost necessary. The the time hacks were necessary because you wanted to get it done as soon as possible, but the time hacks at the time, know, previous to AI, meant that you you compromised something. You compromised the proof. You compromised a fully comprehensive narrative of what needs to be done. And any of those compromises could burn you somewhere along the line. And and I guess, really, that’s what led us to this hypothesis. Okay. Well, now with AI, you can do all of those things that are needed to to get paid in full and, you know, get all the stuff done for the policyholder that needs to be done in a short amount of time that is acceptable to whether it’s the carrier paying or the policyholder or so on and so forth. So I guess that’s what we’re saying here, and that’s why we’re sort of revisiting the scope and and having this conversation about scoping and estimating is because we’re really just getting back to a a situation that’s really the right way of doing the work in the first place. We’re getting back to our roots is what we are. Right. I really think we are. Now there’s another thing that you can do because the AI helps with this as well because this is something that often gets missed when we write it down. Okay? We might say in that very thing that I just talked about, we might say, I am not sure about the I’m not sure about the tile floor. It may have to be removed. We don’t know this yet. What I just said is that’s an open item. So my estimator is going to write that in the estimate and put a zero on it and say, this is a possible repair that may have to be done, but the field has not confirmed it yet. Right. What have I done when I do that? You’ve put in a zero dollar line item? And I’ve informed the stakeholders that there’s the possibility of greater work Right. To be done. Right. But I’m not jumping ahead of myself. I’m saying that we have to look at this further. And that way, we’ve noticed them that there’s a possibility. So let’s say we’ve all had a situation where Carrie didn’t want us to pull a floor or pull cabinets or something like that. Right? So this helps you to basically put those items that are harder to get in a in a zone which it says this is a this is this is likely or this is possible. We’re not sure yet. And it just it lets them because keep in mind, the the carriers still have to set reserves on all the things that we do. And so if we’re giving them that data, they can set better reserves. And that’s another great example of what somebody can do in the field, but it’s very difficult to do at the desk unless the field’s done that work to say that. And I and I and I love that. I think it’s a great aspect of it. Yeah. So a lot of folks joining us today, I’m sure. And and I think, again, going back to the poll, I remember there’s a number of calm owner operators or just smaller businesses where often you’re you own the business, but you’re also the one doing the inspection. You’re also one doing the estimate. What what advice can you provide just to to get people thinking about, you know, getting the right headspace for the inspection, getting the right headspace for the scope, getting the right headspace for them building the estimate in the office later or the truck wherever you’re doing it. I I would say I would say, first of all, that, what what happens in a lot of your smaller shops is that they develop some very dangerous behavior. I’ll I’ll give you an example. I’m I’m helping a contents company out right now that, had a fire working a fire, and they they lost the pictures for $60,000 worth of the contents, and the carrier refused to pay it. What they were doing was they were texting it to their estimator’s phone. And so these guys have a problem because they just didn’t recognize that that’s not a system. Like, natively? Like, native text SMS? Yes. That’s right. And that they didn’t realize that that’s not a system. I’m sure it works sometimes, but it doesn’t work all the time. Right. And and it’s and there’s no way to confirm that it worked. And so, I mean, they could have done something as simple as uploading that to Dropbox and been better off. Right. But they didn’t put any system in place except to use those those those devices, and it was it was it was just terrible. And I think a lot of times smaller operators do that. Look. We we’ve all been a small operator, we’ve all had to wear multiple hats. It’s why we’re good at what we do because we’ve worked every zone of every business of the business that we’ve been been in. Right? Incidentally, by the way, I I noticed in my own world that if, that I I could do everything, but if I gave it to somebody to do, they did a better job than me because they could focus on it. But but in that in that small shop, I think that you’ve gotta get in the mind that your first job is a narrative. Your second job is a scope. Your third job is an estimate. Your forced fourth job is presentation of that information to stakeholders, which is narrative and estimate. I think that’s that’s I think you have to get in those modes to understand that that’s the process. And more often than not, we just we take away the whole narrative aspect of what we’re doing. And Yeah. You know, we’re human. We all like a story. We all like to know we all like to have abiding understanding in what we’re dealing with. And when I could say the supply line on the second floor broke and ran for two days at 40 psi, which I think is about six gallons a minute, you could probably calculate how much water has has gone into this building. And it’s cascaded down all these areas, and it’s right down into the sill plate. That story right there is better than an estimate that says that any day of week because the estimate just says everything’s damaged. They’re trying to make they’re trying to just blow up the blow up the scope. Right? Yeah. Yeah. Yeah. And the narrative keeps you the narrative keeps you from having a situation where they’re trying to say that you’re you’re doing something that you’re not supposed to do. Yeah. I think I think the yeah. Totally. And I think the other you mentioned human behavior, and and I I’d mentioned, like, a hack earlier. And and humans just naturally look for ways to save time. And, and I think one of the things that, that really helps here is is just knowing that if you if you document properly and you and you provide the right narrative, don’t try to that’s not where you should be trying to save time. Like, again, all of these systems, whether it’s Encircle, and we’ve certainly launched a lot of stuff in the AI space lately, including scoping, but whether whether you use us or any other AI based tool, the the quality and the time saving is going to come in in a in a in a more comprehensive set of data, a more comprehensive narrative, and a a more complete narrative. So don’t skimp on that part. Just know that if you do that properly, there’s gonna be so many downstream workflows that AI can then use that data to infer what is the scope, what is the job to be done, what is the task list, what is the estimate going to be. We are I’m not announcing it here because I think we did talk about it at the last trade show, but we we will have an estimating product, of course, in the next couple months. But but, yeah, like, it all depends on collecting the data up front and and having a good set of data to work from. I I I I agree. And I think one thing, Jeff, that that that people probably don’t understand is that we you know, in the mitigation world especially, you’re basically asked to document things like you’re a grad student all the time. Right. You know, if we ask if we ask plumbers and roofers to do that, there’s no telling what it would cost. Neither one of them can write, so that’s a problem. But but the thing is that, you know, I had a guy I have a guy that works for me, and he is he’s the most prolific inspector that I’ve ever had. He was the first day he showed up. He just had a knack for it. But he didn’t know all the technical details when he got there. Just didn’t. Right. And so he would routinely say, well, at the end of this, there’s some doom of flachi. I don’t know what it’s called, but it’s definitely water damaged. And there are still some things he runs into that he doesn’t know, but he describes where they are and how they’re put together. And now the AI just handshakes that and just -yeah – doesn’t have a problem with it at all. And I think it’s important to recognize that to some degree, like I said earlier, you can, to some degree, change the people that have to to that walk the loss because they’re talking damage, not scope. And I think that’s – right – that’s the key. And I think I think in a lot of times, that’s a that’s a less expensive, personnel. I just I just – Yeah. Any and and and I I imagine there is a this is why I asked about Headspace earlier, even for myself. Like, obviously, in order to to do the job that I’m doing, I had to become quite knowledgeable and understanding of the restoration world and ASD, WRT, all of those things. But I’m not gonna pretend for a second that I’m a I’m an RC that’s been on the you know, that’s done hundreds and thousands of jobs like you you have. So I don’t have I don’t have anything to unlearn. So when I did my first when we were experimenting with the AI models and I first did a scope and my product managers and developers were saying, Jeff, just just do a damage description in the way you would know how to do it. I’m like, I don’t need to say, like, cat and class. And they’re like, don’t even get your head wrapped around that. Just describe the loss as you would have done four years ago. Actually, let’s go back 15 years when I actually had a loss at my house, and I had a foot of water in my basement. And I’m not gonna get into it, but just pretend take yourself back there and describe it. So I just did that. I just described what I was seeing. And the the professional output that was being was able that was that we were able to derive from a a a very uneducated narrative with swear words and a lot of ums and ahs. And I think what I’m looking at is this, and I think there’s moisture in the wall. Right? So it was a very uneducated, but the but the but the level of scope we were able to derive from it by you know, with all the research that we did with, you know, understanding how to build a proper scope without hallucinating that was, you know, something that we just said cannot happen. Models do like to do that. But, the amount of output that and the professional level of output that came out of it was kinda mind blowing because I didn’t know anything about, you know, I I took an approach where I didn’t know anything about, you know, the the the right approach to scope a house. And to be honest, I probably couldn’t do it in a way a professional could, but what I can do is describe what I’m seeing. That’s right. That’s right. And I and I think that’s the that’s that’s the whole that’s the whole nut. And I it seems like a simple concept. Okay? It really does. But if if you if you could just imagine the difference in the output when you do it like this, it’s it’s awesome. Like, I’ll give you an example. I had a guy just the other day, and it was just so cool. He he said he said, this is identical to what we did on the Smith loss last week. This is the same exact condition. Well, the AI knew to go look at the Smith loss and get get basically its inference of what that we were talking about. So the guy was able to say a lot of things that he didn’t say otherwise, and that is just absolutely positively cool. The other thing that I really like too is that, hey. I found this, this, and this, and we need to ask Mike what to do with this before we estimate it. Well, that’s gonna prompt the AI to say, Mike, we need your input here. Right. And and the problem is is if you don’t do it this way, then Mike’s not probably going to anything about it, is he? He’s not gonna know that he’s needed. He’s not gonna get that alert. And I think that’s that’s the that’s the power. But I I I wanna say one thing before we before we go into q and a that I I think is real important, Jeff. You can’t jump into Claude, and you can’t jump into ChatGPT. And just out of the box, expect that to have the coherence that’s needed to do this entire process. I’m not saying it can’t do it. Not at all. Okay? But it takes more than just pulling it pulling it out of the box to do that. And if you don’t believe me, I want you to just do that. Do this function. Say, go go get on, like, Otter or something like that. Record something, and then run it through ChatGPT and say, make me a scope. And then do it again and do it again and see if it’s the same scope all three times. And what you’ll find is it’s not. And until it’s the same scope every time, it’s not doing it correctly. It’s inferring it wrong. It’s making other decisions. And I think that’s that’s the that’s the takeaway you gotta learn that this is a tool that’s dangerous. I had to I I had a was in a lawsuit the other day where we there was an actual estimate somewhere that said that there were four layers of roofing. There were not four layers of roofing. And I did not look smart because there were four layers of roofing. And those four layer of rays of roofing came off somebody that ran it through ChatGPT for the scope. That’s what happened. And didn’t check. And didn’t and didn’t check. And so I think you need to move towards systems that allow you to do that that that allow you to to get the same result every time. I think that’s just just really important. Yeah. You’ve essentially described two years of research for us. This was certainly and this isn’t meant to be a a product pitch, but but, yeah, we we were we were doing this experimentation at least two years ago, if not more, where we we had had this this same moment that a lot of people are having today where if you put notes and photos through a through a, through a number of prompts, you can certainly come up with something that looks pretty amazing. But doing it at scale is where where it gets tricky. And, and what I mean by that is what you were saying, the more you do it when you’re when you’re using a probabilistic model, the chances of getting something wrong increase the more you use it. And getting something wrong and not reviewing it and not catching it could be could have catastrophic results. And Yeah. And that’s really what two years of research that we’ve we’ve spent doing is is making sure that anything that we infer is double checked, triple checked, triangulated. And and if we can make an inference, we surface it and say, we weren’t able to to come up with, something that we can be confident in here. We need more field data to – Leah’s back, so you have to be quiet. But I was just getting, we were just getting started with this conversation. It’s just like the the music playing at the Oscars. Your mic is now muted, Jeff. No, this has been this has been such a such a cool conversation. It’s interesting to see the the comments that are coming through the chat. I have a I have a few questions that have come through as well. But Laurie made a comment, and I wanna pull on pull on that comment a little bit. She made a comment in the chat that she spent 20 years as a unicorn, and I think that comes from something Andy said earlier is there’s very few unicorns the myth of the unicorn scope slash right slash run PM is a huge part of our current problems. And so Laurie spent 20 years as this unicorn, but the level of documentation and negotiation required at this point in time has made it impossible impossible to maintain that level of workload and and and also enjoy life. So I think it speaks to burn out a little bit. So, David, one of the things that you talked about was breaking breaking this new workflow down into four steps. And correct me if I got it wrong, but creating the narrative or the damage description, then scope, then estimate, and then package it all together to communicate. Right. How is that gonna help someone like Lori? Because on paper, it seems like it’s more right? It seems like it’s more steps. So which I think I understand that in in the end, it’s not more work, but I want you to can you just walk us through how it’s gonna help somebody like Lori go back to Well, I think first of all, you just said it. You just said it that it’s a much more technical game than it used to be. It’s much harder. You’ve gotta get it writer earlier than you used to. You used to could start the process and say, hey. We’re here and work it out over the next five days. And we can’t do that because everybody wants the answer today. And so it becomes very hard to do that, but if you’re able to speed things along by using the AI to collect that first part of it and to help write the second part of it, then you’ve got an estimate, and you’ve got the core functions for your for your narrative going out. So you’ve basically done one process that informs all the rest of it, and I think that’s where the power is. I’d also say, by the way, that one of the things we gotta really pay attention to is, there’s some research out that says that two out of five, insureds would alter a, loss image in, AI if it meant to get they get their claim paid. So the custody of our documents is gonna be important as well so that we take them and load them somewhere so that the custody is is is recognized. That’s a that’s a really good point. I wanna ask you too because I’ve had some of these conversations over the last couple of months as we’ve been, you know, launching scope into the world and talking about creating this damage description up upfront. We get a lot of questions about, well, what does this narrative what’s the what’s the ideal narrative format? So I’m gonna pose this question as is the the narrative script or the damage description script going to become the new scope sheet? It it really it really is the new scope sheet when you get right down to it because it’s allowing you a verbose level of communication that you could never get into the scope sheet. And it also keeps you from having to do all the calculations right there about what equipment you’re dropping, how much drywall you’re removing, how long the drying’s gonna take, on and on and on and on. It it it removes that because what you’re doing is just describing describing the loss. And, you know, you know, the carriers have been doing this forever. Like, they have a report they have to do on every one of their losses, which is literally the narrative of the loss. They’ve had to type this for years. And this just makes that easier for them because we’re giving them such more verbose information, and that results in better approvals and better cash flow. So my next question then is if if the narrative of the damage description is the new scope sheet, inevitably, we’re going to get questions about what does the best damage description look like? What is what’s the script? So can you walk and there and maybe there is one, and that’s the answer. But can you walk us through, like, what should people be ensuring that they are, you know, training their teams on including in that damage? It’s basic triage. Basic triage. It’s the same thing every time. I am at such and such and such and street, such and such, such in claim number. I am, inspecting the loss. And then you walk the loss, and you talk about every single room in the loss. And I mean every room, not just the rooms that you think are damaged, all the rooms. One thing that this really helps with is we have this problem in our industry. Sometimes we miss make mistakes in estimates, and we copy scope into multiple rooms that they shouldn’t be there. When we walk in a room and we say, master bedroom, no damage discovered, that allows that system to then say, don’t put anything here. And I think that’s all you do is you just walk the loss every single room, say where you are and what you’re and what you’re saying, where you are and what you’re seeing, where you are and what you’re seeing over and over and over again. And that may mean they pull out a moisture meter and say, look. I found this I found it this wet on the on this wall. That may mean that, but but instead of having to take know, they may take a video of that, but I’m saying since they’re really they’re what they’re saying audibly is the most important thing they’re gonna do in that mix. Yeah. I have a actual, it’s funny the way I I I got thinking about it is analogous to did anybody feel a sense of relief when you went from the the old Google search paradigm to the LLM ChatGPT paradigm where in Google, you were always trying to, like, formulate the exact query on what you know? And you and you’d be like, okay. I have to remove all of the, you know, connector words, the and and and things like that. And I’m just, like, trying to come up with a perfect query to Google information. Then when you went to ChatGPT, it’s just sort of stream of consciousness. Oh, I can just I can just say whatever I want in any way I want, and it’ll figure it out because it’s a language model, and it knows what I’m asking for. And to me, I think that’s the mindset that you know, I would love to promote certainly with our product is let go of trying to, like, think your way through the damage description. Just talk. Don’t worry about your format. Don’t worry about but just be very detailed and just talk naturally. I think you as you said, like, you’re like, you’re explaining it to a 12 year old. We have a question from Derek who is in his car. Hopefully, he’s he’s safely driving his car, but he’s really enjoyed the the conversation today, David. But he asked, what what systems for taking notes do you recommend or voice reporters? He does both and still handwrite notes. 16 years as an IA, 10 years as a general contractor. Paperwork and computer time kills me. I, I like Otter. I use something called Plaud as well, which is a hardware I just attach to the back of my phone, and I just hit a button and it records. That’s p l a u d. I I think any app that does that does dictation will get you much farther down the road than you realize. And and I think that I mean, I probably would start start with Otter because it’s just such a nice interface. It does a good job of summarizing what you have and giving you action items, and it does a good job of like, when you export it, when I drive it into a a large language model, you can you can get deeper information or ask it to work on it in different ways. But I think I think any of those tools are fine. I I like I I really like Otter. I use it every day for something. We we also, we we encourage video scopes or sorry. Video. We video narration. Sorry. We started calling it video scope early on, and I just can’t stop calling it that. But it’s not. It’s video narration where where you’re using video to walk around and and and do an inspection and and do the narrative at the same time, and we actually have an ability to extract the narrative and sort and summarize it in in in any way you want. Obviously, we’re big fans of Encircle, so taking notes in Encircle would be something that we would recommend. We’re coming up with a, what I would call audio note, which I guess is like video note, but it just allows you to do dictation without having to worry about pointing your phone at things. And we do feel some people like the video. In fact, we have a lot of customers who it’s written in their SOPs that their inspectors need to do a video. But we also have tons of other people that would rather just do a voice audio. And using that stupid little, like, voice to text built into the keyboard is not a good user experience, so we’re, you know, we’re creating a a much better way to use a voice note. And, and then, of course, make it available to the all the downstream AI for summarization, task note, task lists, things like that. So definitely lots of options. And I’ve actually started looking at Plaid myself just to see if there’s integration points where if people wanna use some third party hardware or software that they’re comfortable with, be able to get that directly into Encircle. Yeah. And the and the nice thing about Plot and and Hatter is both of those things have, great connectors to move them into other systems. And Yeah. I I matter of fact, I even like that, Otter, I can query with, with Claude today, which is nice, because it can give me some some some some and you said, look. Jeff, you said about video. Video is always gonna be great. K? You have to chunk the video because it’s usually not one contemporaneous video. And so that’s one of the reasons why audio is so important is because you can extemporaneously do 30 minutes of audio without much intervention other than using your voice, and that’s why it, by default, is so powerful. Yeah. Agree. Yeah. We would never again, I love it when people take videos because I think there’s just a ton of rich evidence that’s captured there. But the thing is nobody wants to watch the five, ten minute video. And that’s why I think, again, the AI piece is so important because now you can extract all of the salient information out of the video, write it into a note using AI, and then the video just stays stored as a piece of evidence that, hopefully, you never need again, but in the very case that you do because it’s either, you know, heaven forbid, a lawsuit or or some sort of adjuster pushback, at least you have that video evidence there that can be researched. But for the most part, you’re just gonna use the AI summary to extract what you need from the video. Yep. There’s one one question, Leah. I just wanted to jump on, which somebody was asking about when pricing I’ve lost it now. I’ve I’ve got it here, Jeff. It’s when will the AI scope feature, I’m assuming they meant in Encircle, start incorporating pricing now that Encircle has merged or acquired the others. So I was gonna I was just gonna mention because I did mention it earlier. We will have an estimating product. And when I say estimating, we’re starting with Xactimate. So, it’s taking the scope tool that we’ve already built and we’ve already released and then going one step further and creating a fully, a full estimate that will be then integrated into Xactimate and will be sent, into Xactimate to create that ESX for you. Other ways of estimating, whether it’s your own custom price list or t and m, are are future items that we will be doing as well. David, I’ve got two questions left that I wanna get answered before I do before we do the wrap up. One is from Chad. Who should be in charge of creating the scope? And I think the answer is not so so straightforward as that, but I’m gonna leave it. It says, who should be in charge of creating the scope, the tech, the PM, or the estimator? The estimator. Your estimator should because at the point you need to you need to speak in line items is where you need to make the scope, and that’s the best person that thinks in that in that way. I’m not saying that we don’t have to assist in some regard, but they should be able to do most of all of that. I’m gonna do it from evidence that was gathered. Yeah. Who should be in charge of the narrative? The inspector? K. I mean, they’re they’re the ones that are they’re the witness. Yep. And, there was a I think earlier there was a thread on, somebody talked about floor, moisture points and photos. We’re you’re not suggesting that the narrative replaces documentation. Correct? No. No. No. No. No. But but but understand the difference in how you could do this. If I’m picking up moisture points on the north wall in the great room and I say I picked up, you know, 59%, 46%, 81% as we move to the right closer to the source of the loss. If I say that into this to into that voice or into video, I’ve created I’ve created scope. And that’s why that’s why it’s probably you still gotta have the pictures. You still gotta have everything you have, but it allows you to do it much quicker. And the thing is it should allow the estimation process to happen much faster. That’s that’s the that’s the real goal. Thank you so much for joining. David, Jeff, thank you so much for for everything today. I think I think everyone can agree it was a really enlightening enlightening session, and, we have some unlearning to do and some some new habits to form. Thanks, Leah. Thanks, David. Thanks, everyone, for joining. Thank you. Good