Meeting

Web Process Consulting | Recurring

Mar 24 · 5:30 AM
Transcript 166 turns

**Chris LaFay** *[00:00]*: Sa.

**Gavin Taylor** *[00:51]*: It. Good morning. Good morning.

**Chris LaFay** *[01:31]*: Good morning. We had. Oh, it's. It's Gavin from a different angle. What is this?

**Gavin Taylor** *[01:38]*: Yeah, I'm trying to fix my camera. My. The son was playing with it.

**Chris LaFay** *[01:44]*: I first thought my brain processed that as the sun was playing with it, because all I saw was, like, a ball of light outside of your window. I'm like, how did the sun mess with your camera?

**Gavin Taylor** *[01:56]*: Oh, I need a. Oh, yeah, no, I know. I'm.

**Chris LaFay** *[01:59]*: I'm an idiot. I, I. I am so sarcastic. Which I realized you weren't being sarcastic. I am so sarcastic. I crack jokes. Yet for some reason, I am the slowest human in the world when it comes to listening and hearing sentences. My brain latches on to, like, the first thing that it hears, and it doesn't hear anything else.

**Gavin Taylor** *[02:26]*: So I have the opposite problem. My brain attaches to what you're about to say, and, like, I. I'm sitting here just predicting every last word that you're about to say, and I sit in conversations just, like, let's just move on. Like, I know what's coming from. This is go.

**Chris LaFay** *[02:48]*: So basically what I'm hearing is you're a lot of fun to go to dinner with and hang out with, where basically the conversation is just somebody starting sentences and then you just being like, all right, next topic. Move on.

**Gavin Taylor** *[02:58]*: My wife hates it. She absolutely hates it because I cannot stay on subject to save my life. Is time. I. It takes you to, like, talk one sentence. I'm three subjects down the road. Now it all makes sense if you, like, put it together. My thought process. I, like, sat there and try to map it out. Like, how do I go from cars to, I don't know, astronauts or whatever? Like, I don't know. And I'm like, vehicles this, that, and, yeah, my brain just weird.

**Chris LaFay** *[03:33]*: Yay for being unique and wonderful.

**Gavin Taylor** *[03:36]*: Right.

**Chris LaFay** *[03:39]*: So before we get started, I sent you videos last week.

**Gavin Taylor** *[03:43]*: Yes. And I did not watch them. I'm gonna be completely honest. I need it.

**Chris LaFay** *[03:49]*: I dropped. So actually. Okay. It will actually. Okay. This is gonna sound really odd, but I recorded videos of the process that I went through last week.

**Gavin Taylor** *[04:01]*: Okay.

**Chris LaFay** *[04:02]*: The thing is, it takes long because there's processing time in there. I edited the videos down.

**Gavin Taylor** *[04:09]*: Okay.

**Chris LaFay** *[04:11]*: So I'm gonna make a really odd suggestion, and I honestly. I recommend that you go ahead and watch them now, because if I try to show them to you, it'll actually literally take longer to show them to you.

**Gavin Taylor** *[04:25]*: Okay.

**Chris LaFay** *[04:26]*: So now.

**Gavin Taylor** *[04:26]*: Now, like, on this call.

**Chris LaFay** *[04:28]*: Yeah. I would actually. I would recommend because it's gonna. You can watch or do you listen to stuff at, like, 2x speed? I talk pretty slow. Nvidia.

**Gavin Taylor** *[04:36]*: So, like, when I predict people's stuff, like, it's 2.3x speed. Yeah.

**Chris LaFay** *[04:42]*: So they're only. I think it's a total of, like, 12 minutes regular speed. So if you two exit, it'll be like six or seven minutes. So are you game to do that? Because I think that'll help guide the rest of the conversation versus me trying to explain it all over again. This is not a grumpy statement. This is just a. I want to be a good steward of time is all that.

**Gavin Taylor** *[05:07]*: No, that's completely on me. I went really far down the rabbit hole with sops for the past week, so I haven't really focused on anything else. Yeah.

**Chris LaFay** *[05:23]*: Do you need me to resend them to you? They were.

**Gavin Taylor** *[05:25]*: No, I got them. I saw. I saw the email. I started. I'm like, yeah, I'm gonna get back to that.

**Chris LaFay** *[05:32]*: All right, I will. So there's actually three videos. The last one is the simplest one. The first. Honestly, if you watch the first two that I sent you back last Wednesday, that's going to cover 90% of it. The third one, I can. I can tell you about the third one. The third one's easy.

**Gavin Taylor** *[05:50]*: Okay, I got the first one up.

**Chris LaFay** *[05:52]*: Okay, I'm gonna mute myself. So I. Shut up.

**Gavin Taylor** *[05:54]*: Talk. Question. On the first video, you're going through this. Have you ever used a tool like streaming Fox?

**Chris LaFay** *[07:23]*: Yes, I have. We used to have it back in the day, but we don't do a lot. We don't really do a lot of SEO stuff anymore, so we canceled our subscription.

**Gavin Taylor** *[07:33]*: Got. Okay, I got through the first one, and I think I missed the most important part of that. Are you trying to have it create the launch document, the fulfillment launch document? Is that what the goal was?

**Chris LaFay** *[11:49]*: Yeah. So basically, I wanted to create. In the process of creating the sop, I was like, there's got to be a way to create basically a prompt structure to where you could just type everything in, copy and paste it all in, have it organize it all for you, and then create, like, an actual final deliverable. And so the second video is actually walking through. When you put all that stuff in. Here is the packet. It is everything. And then the third video is. I realized afterwards I forgot the automated image migration moving around. And so I put in. I gave it in its own documentation. I have the list of core subfolders that you have with little descriptions about each of Them and then said, here's the main folder.

**Chris LaFay** *[12:44]*: I did not run it with 200 images because I did not want to kill my computer on a test run. But I tested it with about 60 and it finished it in about two or three minutes.

**Gavin Taylor** *[12:56]*: Other than that, I'm gonna be, I'm going to just be very blunt about this. I'm going to be a lot more interested in that specifically more than the launch document. Here's why. Because by the time I'm creating the launch document, we have all that information. So we don't need to crawl additional information. We don't need to like restructure it.

**Chris LaFay** *[13:23]*: But aren't you creating the sitemap on that Excel document from scratch each time?

**Gavin Taylor** *[13:31]*: No.

**Chris LaFay** *[13:35]*: Basically I wanted to create. If you had, if you had this information, getting it into a. Getting it into a structure so that you don't have to do copy and paste. Monotonous work.

**Gavin Taylor** *[13:50]*: Yes. And that's. I think we're on the same page where there is a piece missing there. And I'm gonna, I keep saying I'm gonna finish this launch packet. Let me show you what I've done. This is documents getting out of control. Oh, what was it called? Let's just go here. Okay, so all these I've already done. We have these will be done. So URL structure. I'm done later today. Side note, just remind me later, I want you to look at this for me, make sure I'm on the right track because I don't fully understand robots. Txt okay, so the fulfillment launch packet, the back on point. So the fulfillment launch packet, I'm already going to have like the UL structure. This is the, Is this the document you're talking about for like the sitemap?

**Chris LaFay** *[15:08]*: Yes.

**Gavin Taylor** *[15:09]*: Okay, so this document, when I run the screaming fog report, this is literally a copy and paste of the HTML pages. The service pages is. Okay, let's go to this one that I'm likely going to do next. The service pages. This site already has a bunch of service pages. So that would be an existing service pages. This document is not necessarily for the sitemap but for the developers say, hey, here are the pages that we need. These are existing pages that we need to have on the site. And then the title and description, just make sure that it matches here, additional service pages here, additional vehicle pages here, additional PPC pages. Now for the PPC pages, the likelihood of the PPC pages is without template.

**Gavin Taylor** *[16:24]*: We're going to have this pre built and this is, we've obviously pushed this off just a little bit, but we're going to work on the SOPs. Then right after the SOPs we'll go back and fix all the current clients. But during that time I'm going to be working with the developers to get the template fixed because now that we have all the SOPs in place, everything like that, I can actually work on the templates. The PPC pages, let's say we have three, 30 different ones that we use. We don't care if the PPC pages are completely templatized because the no index, there's going to be no penalty on that side. So we're going to have all these pre built in the template itself. And then so this page won't be very useful after that point.

**Gavin Taylor** *[17:24]*: So this is just telling the developers, hey, here are the new pages that we need. I know this is a dumb way of doing it, but in my mind it's just like, hey, here's the visual, here's existing, here's new, and I just categorized it by services and vehicles. I should probably do others as well. Now the launch packet itself, what I'm doing today, I tried to start it last night, what I'm doing today and this is going to be done today and I'm going to send it over to you today, I'm gonna have, I'm gonna work with Claude and recharge with you, whichever, and I'm gonna create the actual template document itself and that manual pullover is gonna suck.

**Chris LaFay** *[18:14]*: I already know what the template document itself. Which, which template are you talking about?

**Gavin Taylor** *[18:20]*: The fulfillment, sorry, the fulfillment launch packet. Okay, so I'm gonna have it laid out in a Google Doc, I've kind of started it and. What am I doing? Sorry, I had a really late night last night. Okay, so this is basically the idea is I'm gonna have. Okay, first it's gonna be all the business info, shop info, addresses, whatnot. We're gonna have a master document that we've already filled out through sales, vetting, onboarding, everything like that. And the only reason I don't have the developers that document is that is our master list. That is our, the sensitive information on there that we don't want out basically. And just to keep that secure as possible, we don't want to give them access. So with that said, that master documents would be a spreadsheet, give or take.

**Gavin Taylor** *[19:40]*: And what I will probably do is take that and then take the template that I have here for the fulfillment launch packet, upload it to Tribes of Tier Cloud say, hey, this is probably out of. It's too much. It's probably going to be a cloud thing because I think this is going to be too much for ChatGPT. But I'm like, hey, here's all the client information. I need this launch packet filled out. Go. And that was my idea in the first place. And on the master sheet it will have like the links to the Google Drive, it will have the links to the different spreadsheets. It will have everything that it will need to be in this packet. Does that make sense? That was my idea for the fulfillment packet. So.

**Chris LaFay** *[20:44]*: What all, what all is in the. Like go back to the main, your main sop like master sheet. So it's going to pull. The fulfillment packet is going to pull from different parts and pieces from each of these documents.

**Gavin Taylor** *[21:04]*: No, not necessarily. Our. Let me see if I can find Dan, who is in charge of client success right now. Has no one knows how to work in this. Hold on.

**Chris LaFay** *[21:25]*: Dan is the sales guy, right?

**Gavin Taylor** *[21:26]*: Like he's the guy that's doing client success. Yeah, yeah.

**Chris LaFay** *[21:30]*: Okay.

**Gavin Taylor** *[21:33]*: He has, I don't know if he's uploaded it in here yet. I don't think he has. Nope. I want to show you this master sheet and I don't know exactly where it is actually I do. There's gonna be an existing one here. Okay. This is an outdated one. And does not have all that information. This is why we need to do this because nothing is right. What is one that he's looking on? Of course it's on here. All right, we'll just. Here's a better one. Okay, so here is the master sheet. Obviously cleaning this up. Fixing it out, fixing it. But this is the important part now when I take information, put it in the fulfillment launch packet, it would be like, okay, here's the shop info, right?

**Gavin Taylor** *[23:26]*: Here's the social links, here's the, like they won't need the Google and the app, they're going to need the registrar, the cms, the technical onboarding stuff. They need the SEO stuff, they're going to need services and all this stuff. So that's just like in one category. And then we have sales onboarding, all this. So it's just gonna be cherry picked information from here that's actually relevant when you're building the site or when you're running the apps. A lot of this has to do with strategy. Like, hey, where were you at last year? Where do you want to go? How many lifts do you have? How many technicians do you have? Just for us to strategize with them. So we have one of the most extensive onboardings that I've seen anyone have.

**Gavin Taylor** *[24:44]*: We know so much about the business, but it needs to be organized better on this master sheet, and then it needs to be communicated better to fulfillment. That right there is the goal of the fulfillment launch packet is communicating all the information web dev needs, media buyers need, SEO analysts need all that.

**Chris LaFay** *[25:14]*: Got you. What you're wanting to do is basically take this document, feed it into ChatGPT Claude and have it output automatically a fulfillment packet to send to dev that.

**Gavin Taylor** *[25:27]*: And if we can't do that, it would literally be my assistant coming in copy and pasta. There's going to take about 10 minutes to do it.

**Chris LaFay** *[25:36]*: Gotcha. Okay.

**Gavin Taylor** *[25:37]*: And I'm not all too worried about it because she has enough time, to be frank with you right now. Like, when I'm working on all these sops and stuff. She has plenty of time. But I know when we get busy. Can be a lot. So I mean, if we could automate it, great. If not, the. The biggest thing and like the step one in this, the zero to one or whatever you want to call it, is getting the template created. So what I want to do is just sidetrack a little bit from what you. You sent me. Now, the photo side of things I am extremely interested in because that's a pain in the ass. And that's.

**Chris LaFay** *[26:28]*: It's pretty good too. Like, out of the 50 or 60 that it did, it felt right. Obviously, there's some photos that have a truck in them and also a toolbox in them. So there were some things that got duplicated. Yeah, there are some that are going to be mis. Not miscategorized, but just generically categorized. But if it gets you to 80%.

**Gavin Taylor** *[26:49]*: Of the way there, it's a lot better than what we got now.

**Chris LaFay** *[26:53]*: That's better than 100% in five hours. So.

**Gavin Taylor** *[26:56]*: So here's what I want to do from here. I'm gonna watch that video, and I want to work on this week. I really just need to get this out of the way.

**Chris LaFay** *[27:11]*: That's basically done. Like, I can basically give you a Claude code. I can give you a Claude code script that you could shove into a full. Now, the thing is, you will have to have your files locally, so you will have to download them, but then you can have one directory on your computer called photo management. There'll be a readme file inside the root of that folder, and then there will be subfolders for each client. So you would create a Grayson Automotive folder, dump all the unorganized photos in there and then spin up Claude code and that root directory and say go into the Grayson Automotive folder and organize these for me.

**Gavin Taylor** *[27:56]*: So how does that work with privacy? And go. Okay, it will be a local folder. So that makes local.

**Chris LaFay** *[28:05]*: Yeah, you don't. Yeah, it doesn't. Having having any AI access your Google like Google Drive and do management inside of it is lunky at best. And so trust I've tried. I do. I host a DND campaign and had all my DND notes for my DM sessions and all that jazz and a very organized Google Doc. And I was like will you read this and edit it for me? And wasn't very happy about it. So I then pulled it all down and completely redid things.

**Gavin Taylor** *[28:42]*: You know, I think that makes the most sense out of anything that you've ever said is that you play D and D. You have that animated personality. This is no way shape or form a bad thing, but you have that animated personality that I think would be really good for deity.

**Chris LaFay** *[29:02]*: It's actually so it's. I. I've noticed this recently about myself. I. It's a confidence thing. So in what you and I are talking about right now, web AI like all this stuff, I am uber confident and I'm a lot more animated. If you were to watch me play D and D, I am such a quiet person really. I it because I so what I'm not as comfortable with and part. And part of the reason I like playing it trying to get better is I, I don't think well on my feet. And that's all D D is thinking well on your feet. And so I'm trying to teach myself to make split second decisions based on what a character would do, not what Chris would do. And trying to get over that like people pleasing hump. But this, what you and I are doing today.

**Chris LaFay** *[29:54]*: I know everything about all of this. I'm a smart person and therefore I have confidence. And so it's really odd like one one. I'm a. This is, this is my normal personality one one. But you put me in a group setting. I am like hey, how's it going?

**Gavin Taylor** *[30:17]*: Do you play video games?

**Chris LaFay** *[30:20]*: Yeah. Yes. Mostly one player games because my wife and I both play one player games. So I think she's played Baldur's Gate three about four times now. We played Ghost of Tsushima when it came out. We yote in the backlog we played not I Am intrigued by World of Warcraft. I realize that's not the only mmo. Don't get me wrong.

**Gavin Taylor** *[30:45]*: Yeah.

**Chris LaFay** *[30:45]*: And I'm pretty sure when we chatted last time, I thought I saw a runescape icon on your dashboard.

**Gavin Taylor** *[30:51]*: You did it for. Yeah. So I. I've been playing Runescape for 24 years.

**Chris LaFay** *[31:01]*: One of my agency buddies has also played Runescape for 20 plus years.

**Gavin Taylor** *[31:07]*: It's so bad. But that's my version of deity. That's like, I will go into that. I have tens of thousands of hours into that game. And I was doing the math on it. I was like, I have spent literal years inside a MMO game and have nothing to show for. Except.

**Chris LaFay** *[31:32]*: It's like that whole Second Life video game that was out that was really popular like in the early two that like, it was like people were having jobs and they were like, maybe I think some of them were making money and stuff. But it was like people had a second life in this thing and you have literally nothing to show for it. It's such an odd.

**Gavin Taylor** *[31:55]*: There's one thing, and I don't tell anybody this, but I'm telling you this because you mentioned D and you won't judge the hell out of me. I learned marketing in runescape when I was a kid. So they have the whole economy and they have everything like that. I've had businesses in Runescape. I've bought and sold gold in Runescape. I've done it. All that I learned. I learned marketing inside of Runescape, which is hilarious that's now my job.

**Chris LaFay** *[32:32]*: So the. The guy that I was talking about, I have an agency buddy of mine who. Part of his thing that he talks a lot about is Runescape and he's. He's becoming one of those like founders that has like a thought leadership personality on YouTube and blah, blah. Not a. I'm not like a huge fan of it, but he has actually released a couple of videos on his YouTube channel where he talks about how playing Runescape was the found. Like how he learned like business fundamentals and stuff along those lines. And so I've never played runescape before. My wife played Runescape back when it was popular. Like when it first was popular.

**Gavin Taylor** *[33:19]*: It's. Yeah.

**Chris LaFay** *[33:20]*: You know, taking back up again. Exactly. And so I have been considering whether I want to bite the bullet and do it. I don't know.

**Gavin Taylor** *[33:34]*: I like the idea versions of the game. There's the new version. Yep.

**Chris LaFay** *[33:40]*: Classic. What I've heard.

**Gavin Taylor** *[33:44]*: It's. It can be for the players who've been playing it for 20 years. Right. Because it is very much a 20 old game. It basically in 2012, long story short, 2012 they made some decisions, players got all pissed off about it and a couple months later they went to the web archive, said hey, back in 2007 this was a game code and they split the game from there. Now everything that they've done since has been around player votes but the graphics hasn't changed, the core gameplay hasn't changed. So for us that player played back then, yes, it can better now the newer game, there's a lot more content, a lot more modern and overall just better for newer players I would highly suggest because they are doing this huge thing. If you do want to play runescape, try out the newer version first.

**Gavin Taylor** *[34:56]*: And the reason I say that is it just got rid of all the mtx.

**Chris LaFay** *[35:01]*: It just got what all the MPX mtx.

**Gavin Taylor** *[35:04]*: So microtransactions. It got rid of them all. It was just like entire big thing. They're redoing and simplifying combat, they're doing all sorts of stuff. But the reason I'm saying try it out is because all these big changes that they're doing, it's a huge marketing strategy, is saying hey, we're going to take this huge hit up front to take it back to the core vision of what Runescape was and you don't have to sit there. And one bad thing about the older game, it's called old school Runescape is all the XP rates are so incredibly slow, it takes six times longer to progress your account. So if you go try it out, try the new one, it's a lot faster early game, there's a lot more content.

**Gavin Taylor** *[35:58]*: But I would suggest it because if you like look at it on like a, there's a couple different ways to do it but like if you look at it like a business perspective, there's so many cool opportunities in that game. But yeah, I anyway that was down a rabbit hole. I, I really enjoy RuneScape. That's what I put so many hours.

**Chris LaFay** *[36:25]*: Into micro transactions I feel like ruined long term games like when you could. I read this on a forum about Runescape actually it might have been Runescape, it might have been World of Warcraft. I'm not sure which one of the two but it's like when back in the day in the early 2000s, if you wanted the big mega sword that does a bazillion damage, you had to go and do the quest. Like you had to dump the time to do the quest and to do like and grind it out to get there. Whereas now there isn't as much of a reason to do any of these quests and these other stuff when you can just go off and buy the thing for two bucks. It's like, oh, it's just two dollars. It's just two bucks.

**Gavin Taylor** *[37:13]*: Who cares?

**Chris LaFay** *[37:14]*: So I find it interesting that I think we're about to hit a shift that, you know, companies like Fortnite started with micro, like free to play, microtransaction supported and we might be seeing a shift back to story and game actually playing the game and investing time and maybe games become a little bit more expensive to offset the lack of microtransactions,

**Gavin Taylor** *[37:45]*: Which is exactly what they're doing is they just had a price increase. Everyone's bitching because now it's like 14 bucks a month or whatever and they're like, oh my God, it's so.

**Chris LaFay** *[37:57]*: But no other game out there constantly improves and constantly adds stuff. So you're basically buying one really good PS5 game every year now we'll call it two. We'll call it two. Like they cost.

**Gavin Taylor** *[38:11]*: They cost a old school. And so with one membership you get both.

**Chris LaFay** *[38:19]*: Yeah. And they're constantly updating stuff. Unlike Baldur's gate that I paid 70 bucks for. Now granted, we've dumped like 400 hours into that one video game. Played it multiple times through. Like we got our money is. We got our money for that event, the first one gameplay. But like no games just constantly update. So like if you're continually providing value, then you should continually be providing cash.

**Gavin Taylor** *[38:49]*: Absolutely. Like, man, I fully agree. Interesting. Yeah, it's not. I even try to get my son into and he's. He plays this game called Brawl Stars and it's just a little mobile game that it's 30 second matches, it's two minute matches. I don't know. I would much rather play a game like Minecraft or something that takes thought and strategy and everything like that.

**Chris LaFay** *[39:32]*: I'm actually due to get a new computer here soon because now that my business shifted at the beginning of the year and so I'm doing a lot more like actual development work and design work. I need a heavier. The last few years where I was only a sales guy and I just did Google Docs, I didn't need anything super powerful. But I have this Mac Mini sitting over here now, which is a good computer. It's just not good when you're running 50 different Google Chrome tabs and development servers and all that jazz. So part of what I'm thinking over the next couple of months, I'm going to get myself a new computer. And then what am I going to do with this guy? And so it's probably going to turn into something to do with D and visualization and stuff, like in our.

**Chris LaFay** *[40:16]*: In our house somewhere. Or maybe I download Runescape on it. And maybe it's not an. Or maybe it's an. And actually it's a yes and situation.

**Gavin Taylor** *[40:28]*: The last thing on this, because I know I could get way farther about it all. I've seen these YouTube shorts of people showing off the DND spaces and some of them just go all out. And I saw this guy. It was last night, actually. Yeah. He programmed an entire table where he set up lights underneath it so it would like, light up different parts of the map and all sorts of cool stuff. It was like a big old tv, essentially. Yeah, it's cool. I never got into dnd. I'm a nerd, but I get very, like, cringy and I don't know if I could like, actually act that out.

**Chris LaFay** *[41:16]*: Oh, yeah, it's. I don't do a lot of acting when I play. I do a lot of. I almost treat my character like you're reading a novel.

**Gavin Taylor** *[41:27]*: Yeah.

**Chris LaFay** *[41:27]*: Of like. You see my character and he looks like this. And because I don't improv lines well, I actually preface. Like, his goal in saying these next things is X. Chris might not be able to verbalize this, but this is the through line. And then I like. I almost like. I preface it like you're reading a novel. And that's. That helps me get over the hump. We should pause and just. Not just pump the brakes, but slam the brakes.

**Gavin Taylor** *[41:58]*: Yeah.

**Chris LaFay** *[42:01]*: We were talking about image manipulation and the fact that I had. So basically what. What you would do is you have cloud code on your computer. You'd have a single folder called. A single folder called Local Jungle. Local Jungle Image Organization. Inside that folder, you would have a markdown file that I would give you that basically has all the instructions for Claude. I don't. I didn't actually give it to you yet, but I will.

**Gavin Taylor** *[42:38]*: Is there still one that you were going through it?

**Chris LaFay** *[42:41]*: Yeah, but it's not. It's not. It's not the. It's not the code aspect of it. It's more just like, look at me, type in the prompt and look at how it does the organization. It's all the visual stuff. It's. There's nothing to learn there other than the Fact that it worked.

**Gavin Taylor** *[42:57]*: Okay.

**Chris LaFay** *[42:58]*: And then you would then just down, you would create subfolders for each client inside of it and you would then point claude code to that folder and say, go do organization. Let me know when you're done.

**Gavin Taylor** *[43:12]*: Okay, so that's 90% that we're looking for. Okay. So what I've been doing is I just tagged you in this document. It's the fulfillment sops. I think I showed it with you last. Yeah. Shared it with you last week. If you can for me drop a link or code, whatever in here, just so I keep it kind of somewhat organized because I've been hyper fixating on this document specifically. So like a video or whatever, a document for the code of prompt. And then when you take a look at the robots Txt file again, I know what it does. It allows things to call the site. That's the simplest understanding that I have of it. Now with best practices, the strategy, the actual implementation of it is what I want to make sure that we are focused on.

**Gavin Taylor** *[44:26]*: Now the idea behind this and I, I use strategy for this. I think the strategy side makes sense. But again, strategy procedure said, hey, this is what you should do. Basically I want to know when is it okay for AI to call your site. When it's not okay is like for us, like in general, is it just a good thing to allow them to do it? How to separate the robots txu like crawling crawlability instead of the indexing. Basically I have all this here. If you want to just kind of skim through it and give me your thoughts on that, I would very much appreciate that. And then what I'm gonna do, I'm gonna force myself to do it this week. It's absolutely gonna be done this week. The launch packet, I am going to get a template, templatize and send to Chris.

**Gavin Taylor** *[45:40]*: I'm gonna template this week and then I'm gonna send this to you. So that way we know, hey, this is kind of the handoff that I want to make to the fulfillment team. And then once we dial that in and you say, hey, yes, this is everything on as a developer that I would need to build out a site properly. Then we can talk about automating it and making it easier and all that stuff.

**Chris LaFay** *[46:13]*: Okay, that sounds good. Just a heads up this week, I'm actually leaving this afternoon and I'll be out of town the rest of the week. What I will do is before I leave, I will send you, I will send you over this markdown file. That way you can play with Claude code on your computer and having it go through and organize files and stuff. Now, the way that CLAUDE interpret claude is an AI when it reads the MD file is probably going to ask you questions and stuff like that might be different questions than it asked me because it always interprets it slightly differently. I'm adding in a lot of caveats into this script. So I'm trying to.

**Chris LaFay** *[46:58]*: What I'm trying to do is I'm trying to pigeonhole it into a very specific structure so that the questions that it asks are the same every single time. So that I can 100% do in terms of the robots. Txt and all that kind of fun stuff. I can get started on that. I might not have time to do that before I head out of town, but that is something that I can at least take a look at. And I might have an idea before we meet next Tuesday, but my only other working day is going to be on next Monday. So I might not have it all done by then, but I can have some stuff of it started.

**Gavin Taylor** *[47:38]*: And just so you can kind of see where I'm at really quick, the biggest one is this one. If you want to, like, boil this down all into, like an applicable script, this is it.

**Chris LaFay** *[47:56]*: So what are you. So if you. If you have. If you have this right here. What. What are you. What are you looking.

**Gavin Taylor** *[48:03]*: I want to make sure that this is best practice and this is going to make the best impact because I want to standard the biggest goal right now.

**Chris LaFay** *[48:13]*: You want to standardize the robots. TXT file across the board so they're all the same. Yeah, got. Okay, got it. Understood.

**Gavin Taylor** *[48:20]*: I just want to make sure that this is the correct one.

**Chris LaFay** *[48:24]*: Got it. Okay, cool. What I'm going to do is I will actually send that over to one of my SEO friends who knows SEO 10. X more than I do, and I'll basically just ask them, like, hey, I'm looking at like standardizing my robots file. Is this what you and your team recommend as default?

**Gavin Taylor** *[48:42]*: Okay, I'll.

**Chris LaFay** *[48:43]*: I'll do that. I'll send. I'll send that over to her and see what she says.

**Gavin Taylor** *[48:46]*: I very much appreciate it.

**Chris LaFay** *[48:48]*: Awesome. That sounds good. Yes. And I can. I can send that over to her. I can send over. I can send that over to her today.

**Gavin Taylor** *[49:01]*: Yeah.

**Chris LaFay** *[49:03]*: Have you shared this document with me?

**Gavin Taylor** *[49:05]*: It's on the. You won't have that. Dang it. And then I want. Probably want to share this. I'm gonna show all three of them with you. A lot of this is chatgpt fluff.

**Chris LaFay** *[49:36]*: But yeah, but honestly this stuff is very standard anyway. Like this is. There's some things that I wouldn't trust technicalities too. But robots Txt is a fairly standard procedure at this point. So yeah, I have a feeling what it.

**Gavin Taylor** *[49:55]*: If I do get a junior developer back in, I want them to be able to see like hey, this is exactly what we're doing. But yeah, the whole purpose of this is just standardizing absolutely everything and then what comes after this is going to suck because we got to go through all of our clients and make sure that all this is done correctly.

**Chris LaFay** *[50:24]*: Would it be, I'm curious, would it be helpful if rather than building all of this off of like spreadsheets and Google Docs, what if you just had a, like a local junk, A local jungle dashboard app to where your sales people like when they create a new client sales had a tab where they filled out everything. You had a tab where you filled out everything and all just lived in literally one spot.

**Gavin Taylor** *[51:02]*: It's something that I'm working on with them. The best way I can explain it is some people in this industry are very much technically like knowledgeable, but they're also very ignorant about application of things. So what I mean by that is I try to set up something very simple with high level. Right That's I have such a knowledge on high level. Well like oh hey, we could do XYZ and just make this process so much easier.

**Gavin Taylor** *[51:48]*: The problem is going from idea to implementation and for them just to fill out a form instead of filling out the form, they try to do things manually, broke everything and I need to get them in route of let's apply this as a standard and then work on that and which I already tried to pull them off of spreadsheets once which just need to standardize that spreadsheet first and then we can go down the rabbit hole of hey, let's just pull out a form instead and then it will pull in a dashboard. And I've built those dashboards. I, I think they would be absolutely phenomenal. But we just need it done right first.

**Chris LaFay** *[52:41]*: That's the biggest thing because one of the things that I would be more than willing to work on next month is if you have all these documents all nice and interlinked together. I've been building out stuff like this for my own company. So if you have all the documentation organized and structured well might be something to look into. And I have a general idea of, I've been playing around with this Stuff for the last three or four months. It's possible, let's just put it that way. So.

**Gavin Taylor** *[53:15]*: Okay.

**Chris LaFay** *[53:16]*: And then it's also exportable as well too. So if you needed to then take certain data sets out of that and then send it to a developer, you could.

**Gavin Taylor** *[53:29]*: The big thing I want to do upcoming, before we run out of time is. Is after is photo. That's going to be important. I've already figured that out. Swimming fog. I have done today.

**Chris LaFay** *[53:45]*: Do you have Claude code?

**Gavin Taylor** *[53:47]*: No, but we're getting the company one.

**Chris LaFay** *[53:50]*: Okay, great. So the one thing you will need for the. Actually, do you have. Do you have a hard stop at 10:30 or at the bottom of the hour?

**Gavin Taylor** *[54:01]*: Yes, I do. I have another meeting.

**Chris LaFay** *[54:04]*: Okay. Because I was gonna go ahead and send this over to you. And get this set up. I was going to show you how to get it set up because it's not. It's not hard. And once you have it set up, it's done.

**Gavin Taylor** *[54:24]*: I. I'll be able to figure out. If you give me just kind of basic instructions, I should be able to figure that out.

**Chris LaFay** *[54:30]*: You're from the. You. You know how to get API keys and stuff like that?

**Gavin Taylor** *[54:33]*: Yeah.

**Chris LaFay** *[54:34]*: Okay. Awesome. Awesome. I'm going to go ahead and send this over to. Perfect. I will write up the rest of this email with a couple of instructions and then you'll have access to it. You'll need to get. When you get cloud code, you will need to go into their API console and get an API key. So you'll have to set up API billing. It's stupid cheap.

**Gavin Taylor** *[55:03]*: Okay.

**Chris LaFay** *[55:04]*: Relative to the number of hours that any of your humans work on organizing photos, the API calls to Claude to read the photo and to understand it and then organize it is pennies on the dollar.

**Gavin Taylor** *[55:16]*: Okay.

**Chris LaFay** *[55:18]*: So yeah. And if you wanted to go faster. Just one thing to note. Most of the files in the folder that I looked at were all like 20 or 30 megs. It seemed like they were like raw photos from a digital camera. You don't need them to be that size. So if you wanted to go faster, which it went, really, I think it did like 50 or 60 images in like two or three minutes before. I could be wrong on that. I don't remember. But it was fast. If you wanted to go faster and hit less context windows, if you en masse resize them all ahead of time, that'll make it snappier.

**Gavin Taylor** *[55:58]*: Yep.

**Chris LaFay** *[55:59]*: So.

**Gavin Taylor** *[56:00]*: So appreciate it. Last question for that, since it'll likely be My assistant doing this and she's in the Philippines obviously Internet's not super stable there. It's going to be done on local files. Would it make sense to get her an external hard drive to do this? Slow down her computer?

**Chris LaFay** *[56:35]*: Possibly. You'd have to get her. You have to make sure that it's like I I. So I have an external hard drive attached to my Mac Mini and I paid about 120 bucks for is not as fast as I hoped it would be. So like there's read write time that happens with it. So it's good for now. I may have ended up buying the wrong type of USB connect like I might have on USB A 3.0 rather than USBC like I could have. Hardware is not my personal forte but if you're going to be doing in mass file management relatively quickly with right. It needs to be usbc. It needs to be solid state so that it goes much faster. Because this one, I don't think this is solid state. It's fast but it's not solid state.

**Chris LaFay** *[57:32]*: And I was migrating like 10,000 files to it and it was very sad when I did it.

**Gavin Taylor** *[57:37]*: I might have to pick up like a work laptop or something. I can see this adding a ton of files to a computer really quickly.

**Chris LaFay** *[57:52]*: She can delete them as soon as they're organized and uploaded to Google Drive and folders. She can then go through and delete them off of her computer. So just keep that in mind.

**Gavin Taylor** *[57:58]*: And I'm almost wondering if that's going to cause problems just simply because the download upload times. I'll figure that part out. That's not a problem. If I need to do it, I can do it. I got super faster than that. I got ton of him. Yeah. So we're good. Cool. Okay. Awesome. So I think that's where we leave off then. I We are working on the pre launch photo review and the approval stuff. We are working on the Roblox Txt and then next week we'll touch phase. Oh, that's what I was going to say next week. This is where I want our focus to be is the QA process. Because I don't. We're missing things and I don't want to miss things.

**Gavin Taylor** *[58:56]*: So it makes me look stupid and I want to make sure that the QA process is in depth and it makes sure that those stupid little mistakes don't happen.

**Chris LaFay** *[59:14]*: We can talk about how do we automate some of that too.

**Gavin Taylor** *[59:17]*: Yep.

**Chris LaFay** *[59:18]*: So that all sounds good. Perfect. You're about to get an email that's what I was typing up. I'm typing out the line by line instructions for.

**Gavin Taylor** *[59:26]*: For.

**Chris LaFay** *[59:27]*: For what to do here. So you'll get that email from me here momentarily. Does that sound good?

**Gavin Taylor** *[59:32]*: Yep, that sounds great. Thank you so much.

**Chris LaFay** *[59:35]*: Adios, man. Catch you later, sir.

**Gavin Taylor** *[59:37]*: Bye. Bye.

Full text stored in meeting_transcripts. Vault export: Brain/Chief of Stuff/Classic City/Meeting Transcriptions/2026-03-24 - Local Jungle - Intro.md
Highlight-to-quote lands with the Library shelves. Quote sources are polymorphic: articles, podcasts, and meeting transcripts all qualify, attributed to the speaking attendee.
Jump to
Projectsbrowse
Peoplebrowse
Companiesbrowse
Saved
Editor
Autosaves as you type