What We’re Working On in 2026 | Episode 211

Host: Marketing is no longer enough by itself. If that's all you're doing, it's not sticky enough to get to 2mil.

Co-host: We've looked at Angie's, we've looked at Meta, uh, we've looked at Nextdoor, we've looked at email marketing, looked at Organic Social. No matter what we pilot, it just doesn't really stick. The next iteration of the business isn't marketing, it's operations.

Host: Aimax is going to be, in the future, the default search campaign setting when setting up new campaigns. So far with what we've seen, it's bad.

Co-host: I could 100% see AI mode acting as an agent to contact multiple businesses using paid lead forms. It takes up way less resources and it's way easier.

Host: That's gross. But you're probably right. I don't like that. Because I don't want to think that you're low key competing with everybody now. So you better call them back the second this hits.

Co-host: Here, they already knew it. With LSA messaging, as soon as you leave them message, it says, hey, try leaving a message to these other businesses too. It just duplicates the customer's message and everybody gets charged for it.

Host: The top comment in the chat was, why are you guys pushing PMAX so hard when it sucks?

Co-host: I just think of the Mr. Krabs interview where he's like, money, money.

Host: My assumption is it's just Broad Match on crack. And whatever they decide, you're going to Broad match too. And people aren't biting or they're trying it out and it's not sticking because it's terrible. And then they go, well, AI Max will let us eat people's lunches. And if you do it badly, they'll take all your money.

Co-host: We can finally do the cold open. I've been waiting like a week and a half to do. And it's that the big event we put on at the University of Iowa.

Host: Oh, yeah, I forgot.

Co-host: Okay, so for some context, for some listeners. Actually, double context context. Uh, number one is I'm moving. So if you're watching this on Spotify, uh, or YouTube in the future, my office is like, completely empty. So, uh, moving. In the process of moving. Next episode, you'll see this again, maybe, but in two episodes I'll be in a different environment. Um, okay, update. Another number two. Uh, on. I think it was the ninth. The ninth. Uh, the University of Iowa. It's Big Ten stadium or Big Ten football team. Uh, they wanted to do a big paintball event inside of their stadium for like a recruiting day. So where they have like coaches and staff and recruits and family all play paintball inside the stadium. So for the better part of like the last two months, I've been organizing this, this whole event, uh, for them. I've been sourcing like paintball bunkers and equipment from like fields that are closing down to like, to get this all in conjunction with KW Paintball in Waterloo, Iowa, uh, which is kind of like my, my hometown kind of field where I grew up at. And my buddy owns it now. But uh, he kind of contracted me to help him put this event on. And Lauren and I basically organized the entire thing. She did like everything in the press booth which was like, uh, she had like the, the step and repeat big banner people could take pictures at. They had like, you know, a table with like gift cards and things like that, uh, super professional. And uh, we hired a media team and everything and we were kind of short staffed. So uh, I mean the people that could help us that actual like at KW were like two of the owners, me, my wife Lauren, and then we had one guy who was like, who kind of plays for the team there but isn't really part of the staff. And then uh, I thought like we were going to be okay, like what, we should be good with that. And you messaged me and you're like, can we come? Slash, do you need us? I was like, uh, I was like, yeah, sure. So, uh, Lauren got, got uh, you guys polos and name tags and we had your guys help. But like it turns out we were actually pretty short staffed because the window that we had to set up, this field, the air station, all the guns, everything was not long enough. I think we had like two and a half hours max to do it. And then like there were so many times that we had to pivot. It was probably the hardest setup of like an event I've ever had to do and probably will ever have to do. Um, but we got it done and it was amazing.

Host: Media Guy was cool. The stuff that he's cranked out. Oh yeah, some of the videos that you've sent, they've been good.

Co-host: Yeah, some of those videos have been banger. It was a super good idea to uh, record that. So I did this entire time thinking that this is the only time it's ever been done in a college stadium. It's the only time other than like earlier this year when the NXL hosted inside of the Tampa Bay Buccaneer Stadium that this was, you know, the only other time that this was in a paintball Or a football stadium. And I guess I was wrong because two years ago, or two or three years, three years ago, a company out of Illinois did it for the, for Illinois State. But there, there's no pictures. There was only evidence that they were there. And then they, they put the event on for them and they did like a smaller five man field. We waited a full scale ten man field which is basically almost the entire length of the football field. And uh, um, then I guess they did it again this year too. So. But they, they're only posting like the layout on their social media of like a two dimensional overhead of what it looks like. Whereas like we actually hired a media person to go in and take photos and video. And I think the reason why they didn't is because NCAA regulations state that you can't take photos or video or publish photos, video of recruits. And uh, I'm pretty certain this other company or this other paintball company that did it was like, yeah, okay, if we just won't do that, that sucks. But like, um, there's always a way, there's always a way to turn something into marketing. So, uh, we just got colored, um, jerseys and those scrimmage pennies and we put all the recruits in Hawkeye colors, so golden, uh, black. And then everybody else was in a different color. So our media guy knew who to take photos of and who not to. And that's what it was able to get us are the photos and media. And we just got approval from the university on those photos. So now we're going to blast that out to the press.

Host: That's awesome. I drove the Gator. I want to say that out loud. I want to say the Gator is uh, a four wheeler. How do you describe it? It's a UTV sorta that you can drive on a football field. Um, didn't know how, but by the end of it, man, I was Mr. Gator and I totally knew what I was doing. That was fun. I never thought that I could say that I did that. You know, I drove a Gator on the football field.

Co-host: Yeah, I mean, I had never stepped in inside of Connect Stadium until that day.

Host: You know, Max always said that it was really hot when he'd be playing. Like he'd get done, he'd be super sweaty. And I was like, yeah, that sounds bad. But now having been down there, I understand. Like it's a, it's a combination of a bowl and a magnifying glass. You have a bowl that like scoops all the heat to the bottom and It's a magnifying glass to just roast you, cuz. And then there's walls all around so there's no wind that can get to you. It's just, uh, baking, dude.

Co-host: I. We were there till like 8 o'. Clock.

Host: Yeah.

Co-host: Packing back. Packing up. It was crazy. But, uh, it was a really good time. They want us back next year, which is good. Um, we'll just. We're gonna need more money and more time to set up because that's crazy, the amount of work that we had to. To put into that. Was that your first paintball experience? Have you, have you played paintball before? I. I guess I don't know if I've asked you this before.

Host: That was my first. My first, uh, rodeo.

Co-host: You never played paintball before?

Host: Don't think so.

Co-host: Did you get shot on the sidelines?

Host: Nope.

Co-host: Cause you had a. You had a mask on and I. I brought you a nice mask to put on. But our contact at the stadium, like, kind of screwed up with, like screwed with the whole setup. And so we were scrambling right at the end and just trying to figure out like, left from right. So you had to wear a rental mask, but, um.

Host: Yeah, it's fine. Yeah, it doesn't. I mean, I. It's not my thing. I don't get the thrills from it, but it was, it was cool to. The whole thing was just cool mostly because. Yeah. Iowa. And being able to go back and be like, I went here and I can't relate to any of you guys because I studied nerd stuff and you guys are doing cool sports, bro things.

Co-host: The, the entire defensive, uh, line came out to KW and Waterloo, uh, on Sunday.

Host: Oh, really?

Co-host: Yeah. So they, they, after the event, they came out, they said they had a blast. They want to do it again.

Host: That's awesome.

Co-host: Yeah. This is great.

Host: It's cool. D line.

Co-host: Yeah.

Host: Yeah.

Co-host: Okay, well, let me, Let me double check. Let me, Let me double check. Let me make. Yeah, because I think so, because D line was wearing, like, if you follow me on Social, you'll probably see like, the videos and photos that I post, but the defensive line was wearing the green jerseys and they were undefeated the entire day at Kinnick Stadium. Like, they absolutely just flooded the other team and destroyed them. Yeah, it was the, the D line. The D line came out on, uh, Sunday.

Host: That's cool.

Co-host: Yeah, it makes me want to, like, buy more paintball fields and just operate them remote and just be, you know, just go in as a minority investor and just teach them how to do things right.

Host: Do it better. Yeah. Marcus Lemones, that stuff? Yeah, that's his name, right? The Prophet dude?

Co-host: Yeah, I'd be Marcus. Paintball loners. All right, business. Wait, what are we talking about today?

Host: All the stuff we've been. All the things lately and working on. Um, I don't know. I figured, like, we were trying to come up with stuff, and I was just saying let's just talk about what we've been exploring, because we've been checking out lots of stuff, more so by necessity than free will. I just feel like on my end, the stuff that I've been peeking at has just been forced. I'm not excited about it. I'm like, yay, pmax. I hate pmax. Um, no, but, yeah, there's been some Google things. Um, of course, High level continues to be a thing that I'm checking out, but I don't want to go super deep into that, mostly because, uh, it's done to death. I've learned enough to know that there's plenty of information out there about that. Um, it's more. So you can see Google scrambling hard right now about everything, like, in every realm. Okay, not every. Um, ads. Yes. 100% Gemini. 100%. Anything else? I don't even. I don't care. What else is, like, beyond that? What, YouTube? YouTube's fine. YouTube's gonna be fine. Nothing really competes with YouTube. Um, but in those two areas. Yeah. I remember there was one day that, um, Gemini was just down, and I was like, you guys can't afford to be doing this right now. Like, Claude's gonna go public, and you're gonna get knocked out. Like, that'll be end. That's.

Co-host: Is Claude going public?

Host: I thought.

Co-host: I know. OpenAI was right.

Host: Well, I don't know the full details of this. It's funny. This is actually my, um. Like, I feel bad even talking about this because of how much I don't care and don't know, and I'll totally get wrecked by people correcting me. But it was actually my. My Japanese instructor who was telling me, like, what's some of the stuff that's happening with Claude? Specifically Mythos. Um, and Claude's intention to go public with. I don't know. I don't know the full.

Co-host: Yeah, well, it says here that they do. It says, uh, a public listing could reportedly happen as early as fall of this year.

Host: So I didn't mean, like, publicly traded. I meant, like, taking their

Co-host: fable and Mythos thing.

Host: Yeah, I heard this function guy and then saying, anybody can use it. Go Ahead, do the thing. It's primarily been B2B up until this point.

Co-host: Right. Yeah. I heard they took it down because the government made them. I believe that, uh, anthropics. This is two days ago. Anthropics fight with the federal government has become accidental marketing for Claude's ipo. Yeah. Um, ah, here's what I know. I know that it wasn't just because it was so powerful. It's because people found a backdoor route to basically hack it or jailbreak, um, it. And, uh, so the government was like, no, turn that down. Plus, that's, uh, I guess it can be construed as a national security concern.

Host: Sure. Well, it's good news for Google, right?

Co-host: I think it's weird to live in a society or in an age where a private company's product can be jailbroken and that becomes a national security concern.

Host: I've never thought about this far. I just think, well, powerful. But my brain is not. All I've thought about was Gemini not being accessible and then going, oh, you guys can't do this. You really can't be doing this right now. Because, um, I know it's crazy that I. I'll. I'll say it. I haven't played with Claude much. And most people, when they hear that, they're like, whoa, that's crazy. That's insane. Like, I've done a decent amount with chat gbt, decent amount of Gemini. Haven't touched Claude yet. It's on my list. I'll get to it. Um, and I know people are like, well, if you don't get to it soon, you're gonna fall. I don't care. I really don't care. Um, um, anyway. But, yeah, so Google Marketing Live happened and they were releasing things.

Co-host: Let's do this. Because, like, we're exploring a bunch of different things right now. Um, and the whole reason why we're doing it is because we're looking for that next iteration of our business. I know we've said it before in this podcast, but in case you haven't caught up on every episode, uh, marketing, ah, isn't like, there seems to be, like, a ceiling with the amount of leads and marketing you can provide to a client. Like, our business model is so lean to where, like, our industry isn't necessarily the most profitable industry. And lead and sales attribution is way more important than a lot of other industries. And having said that, we can't really go out to market and charge $2,500 to $4,000 retainers, uh, and expect to build a massive client base with our type of business model. Now we can, if we were like more of a consultative agency kind of one on one. But again, that's not what we're set out to do. So we've always tried to find the next iteration of marketing services that we can to, to upsell and make that just as sticky as our core foundation service, which is SEO and Google Ads. Um, but anytime we turn the faucet on, our clients only focus on the bad quality of leads that come in. And when you go up from high intent driven marketing channels like Google or like Search, you go up the funnel, your leads like the overall lead quality generally gets worse. That's just the nature of things. And the fewer amount of leads are actually qualified. That's, that's the point of the funnel. And when you start with something super sticky at the bottom like Google Ads and SEO and you work your way up, everything else becomes less sticky and so everything else becomes a fight because everything becomes compared to what the client originally had at the very beginning. That's always been our challenge. And if you've stuck along with us for this, this amount of time since we started this podcast like four years ago, uh, we've always had the foundation and we've tested Yelp LSAs, uh, which is probably the number one thing that's working for us in addition to our foundational service. But there's still some problems with it that we're experiencing and I think that we'll dive into the high level stuff we're tackling with that. Um, and then, uh, um, we've looked at Angie's, we looked at Meta, we've looked at Nextdoor, we've looked at email marketing, looked at organic social. Like, it's just nothing. Like no matter what we pilot, it just doesn't really stick. And what we've kind of found out is that if we can find a way to kind of like gate the good leads and like filter out all the bad leads so our clients never even see them and they only get the good leads, the customer acquisition cost will probably be the same. No matter which platform we go to, they just don't have to worry about the bad leads anymore. So if we can create that kind of system, and not even just that too, but like also adding, adding in a layer of like sales consultation. Can we get something to like listen to these calls from these different lead sources and provide coaching instructions like, hey, this person asked for this. This is a great opportunity to upsell or talk about this. Um, and we have like really good examples of that, which I'll get into later if I don't forget. Um, and then we've kind of realized that the next iteration of the business isn't marketing, it's ops operations. It's creating something that's, we created something that was sticky for marketing. Now we need something that's sticky for the ops to where if they need to pause marketing for whatever reason, the ops can stay, or if they need to pause the ops, the marketing can stay. And that will also help with the seasonality of our industry. As we head into the winter, people just pause and go to website management or something like that, where it's like we need something, one, to help with the seasonality and then two to basically double our MRR per client. Um, yeah, that's the dream I want to talk about. Let's talk about the lowest level stuff first. So let's talk about Google. Uh, uh, cause we're exploring some stuff with Google. You mentioned Google Marketing Live.

Host: Yeah.

Co-host: And, uh, which I want to say.

Host: Sorry, go ahead. I got a, I have an exciting thing.

Co-host: Well, I, I, I did want to preface this. Cause I don't think we said, I don't think we really said it, but like this episode is really just going to be like the things that we're exploring right now. Everything I just previously said was just setting up for why we're exploring some of these things. Except for Google. Google is just because Google Marketing Live came out and they have a bunch of new tools that we have to explore because that's our bread and butterfly.

Host: Yeah, I'm not excited about Google at all. We just have to do it, uh, because they're changing things and some parts of it were just funny. I don't, I don't take Google Marketing Live seriously at all. I, uh, tell our team when they sign up they should use fake business titles like Space Cowboy or just whatever they want to throw off their system that they're definitely going to try to use to upsell us things in the future. Um, I won swag, though. I don't know if I told you that.

Co-host: Wait, in this last one?

Host: Yeah. So I've won it twice now. So this, like, if you're, if you're listening and you just want swag, go, uh, to Google Marketing Live and do whatever they say and you have a decent chance of winning their swag.

Co-host: You have a decent chance of winning their swag.

Host: I thought Eva didn't win. I felt bad, like, well then I can't just give it to her either. Because then that's like a pity.

Co-host: She didn't win last time either.

Host: Did she do all the stuff last time too?

Co-host: I don't remember. I don't ever watch Google Marketing Live.

Host: I don't watch it. I just like, I. Okay, that's not true. I watch the first, however long I can make it until I'm just disappointed. Um, and then it's on in the background and I mostly listen.

Co-host: I think Marcus watched it because they brought on Neil Patel. And I was like, all right, we're out. I'm out.

Host: I forgot about that. Okay. Um, I only. The only reason I say any of this, the top comment in the chat was, why are you guys pushing PMAX so hard when it sucks compared to Vanilla Standard ads? Right.

Co-host: I just think of the Mr. Krabs interview where he's like, money, money.

Host: Yeah. Um, they do have some recent changes. I don't think it's to pmax. I think it's to, um, demand gen, which are actually very cool. And if nobody's doing anything with that, we're not. We've done stuff similar in the past. Um, but I, I see a future in which we can do it easily now because they finally did the thing that everybody wanted. Allegedly. I haven't looked, but they've said that they've segmented out the properties. So if you want to run Gmail only campaigns again, or Discovery, um, only like Discovery feed, that sort of stuff, you can now do that, which is awesome. That's amazing. Um, and somehow that's now tied into pmax. I don't really know, don't care, but blah, blah, blah, blah. I'm saying a lot to get to the point, which is we're playing with AI Max now. Um, and I just saw Ms. Marvin. Jenny. Jenny. She's the liaison.

Co-host: Yeah, Jenny.

Host: She was saying how AI Max is going to be in the future. The default search campaign setting when setting

Co-host: up new campaigns is that basically saying it's search and display with partners network turned on. It's just all of it and you just gotta turn it off?

Host: I think so, yeah. I think that's the default. And then you have to tweak it from there. I don't think they're removing it yet. Um, but so far with what we've seen, it's not bad. So pmax is for lead gen, for small lead gen. Not great, right? You don't have the signals that it needs. You don't have the volume that it needs to really work and perform well. AI Max the only reason that we were really interested in testing it out in the first place was to make sure that we're showing up in AI mode.

Co-host: Right.

Host: Because we know that that's growing and that more and more people are getting into that. And so that's why we opted in. You're forced to use conversion based bidding. Um, you and I think Caesar was on, and then we talked about this very briefly with him and just said, like, if you don't have volume, it doesn't really work. The bidding strategies don't cooperate. Um, it's working, it's generally working. Uh, and it's not, it's working within the confines that it gives us. So for example, um, some people will segment out if you're, if you're doing anything at scale. Like if you're running enterprise accounts, you'll have brand campaigns separate from generic campaigns. So whenever your, your business is the agency growth podcast, and then that's a campaign where you're targeting those keywords and anything associated with that, and then you separate that outside of that too. And they've got that built in now where you can do that relatively easily. And if you don't do that, what we've seen is that Aimax will just lean on grabbing those branded conversions in order to hit your target cost per acquisition goals. It's not great. Um, but at least it's smart enough now to understand to do that previously just wouldn't do that anyway. Like, it didn't get it. So anyway, if that's where things go in the future, where we need to be over to AI max so that we're showing up in AI mode, which we know is, um, that's growing for Google. Right. And if that's where Google is going and that we have to do that to stay relevant on Google in the future. Okay, that's not the worst thing.

Co-host: Wait, so how does AI max work though? Because like, I'm looking at one of our clients that we're testing like 500 bucks for and uh, like we're still targeting keywords. So it's not like pmax, right?

Host: Yeah, um, we're still figuring that out. Apparently there's a difference between setting it up and then adding it to a campaign. Right now we've just modified existing campaigns and if you go to the search terms and you look at the match types, they have like AI mode match, which is them just deciding. I don't even know if they tell you, they're just like, this is how we decided it's in the AI max.

Co-host: Well, it's probably like, it's probably like a grounding query or something or. Because, okay, because I'm looking at this and it says for their brand and for our similar brand names, they're doing AI Max. That's the match type. And then there's other ones like yard cleanup, uh, and landscaping near me, yard work near me and then like yard. Regarding cost, I don't know what that means, but those are exact and phrase match or close variants. And so I mean, I'm guessing when it says exact and phrase match, I'm guessing that's a standard search. And then anything with AI Max is going to be an AI mode or. Yeah, yeah.

Host: The weird thing is usually like previously we would intentionally segment this. We'd be like, okay, well we're going to do negative audience routing so that anybody who is doing AI node stuff we know is over there, but we just, we're not allowed. It's not letting us do that. My guess and my assumption is it's just broad match on crack. And whatever they decide you're gonna broad match to, um, they're gonna go in and I mean, which again is if it's doing this, the CPA target goals, fine. I don't really care.

Co-host: Right.

Host: I think everybody kind of agrees on that.

Co-host: Fortunately, like everything in this account though is, is like the AI Max matches are like entities like, uh, the actual brand name of our client. Whoa. What the heck is. Ah. Oh, okay, sorry. I saw, I opened up our other client that's like putting a hundred bucks towards it. And I saw the campaign name was Angie's AI Max Exploration. And I just saw it in the column and I was like, whoa, did it match on Angie's.com or something like that?

Host: We set it up and I was like, call it what you want, I don't care. Call it your own AI Max thing.

Co-host: You gotta conversion. What do you get a conversion? Oh, uh, on the brand name.

Host: The other. So the other thing then beyond that and I, I, if what we're seeing is where things are going, we can work with that. And I genuinely think what happened if I was. They're gonna do it in way fancier words and while wearing suits or whatever, uh, they do in, in San Fran. And you know, but in the room it goes something like, hey, we're pushing pmax, but it sucks and people aren't biting or they're trying it out and it's not sticking because it's terrible, blah, blah, blah, blah, blah. And then they go, well, AI Max will let Us eat people's launches, but it also doesn't eat people's lunches if they pay attention. So it does what we're already doing, but gets us even more money. That's what regular search ads have done. Right? That's why people pay people to do it, is because if you do it well, it'll work and if you do it badly, they'll take all your money. And that's going to be AI max in the future too.

Co-host: Yeah, I mean, I think, I think everything is just going to be paid. Whether, like, I don't think it is even going to be like there's always. I saw something on LinkedIn today and it was like the, it was like the PPC conundrum or something like that where it was like, don't have, it was like, you know, don't have a brand campaign because you're wasting money on, uh, your own brand name when people will click on it organically. Then it was like, but if you don't, then competitors might bid on it and you lose clients. And it was like this like circular reasoning type thing. And um, I've always been the camp of like, I'm not going to bid on my own brand name unless like, I know people are like bidding on it. But then like, it reminded me of a conversation that I was having this morning with one of our clients because they, we own LSA accounts. We own all AD accounts, but for LSAs we have to give them standard access. Um, and this is probably good to kind of like segue into the next topic, which is why we're kind of exploring high level. But uh, we have to give them standard access to their LSA account because they need to manage their leads. Now obviously we can take that access away and make them just do like a Google sheet. And maybe that's why that one dude we were talking about in the past, why he did it that way, uh, because he didn't want to give the control over that account. Um, but so they had to log in. They have to manage their leads. Um, however, if they don't pay their bill because we operate on prepay, so they got to pay us. Everybody got the invoice yesterday on the 15th and they have to the first to pay or we don't turn ads on. Well, it was the 12th of the following month and they hadn't paid yet. Yet we go into their ad account and their ads are turned on because they have standard access. They have the ability to turn their ad account on even though it's off Even though it's our payment profile on, on the account. So, um, I was just emailing him yesterday and just said, hey, you guys either need to be on autopay or like, we can't do LSAs together because you're spending our money without paying us. And uh, he just kind of ignored what I said and just emailed m me back and said, uh, please, please invoice another thousand dollars for ad spend. Because he basically blew through the entire budget. We weren't able to pace it correctly because they just turned it on without, you know, daily pacing. And uh, then he emailed me back and said, um, thanks for turning those back on. Curious if we had a hot lead market or if we have a hot lead market in your expertise or is this normal? So it's basically implying like we're getting a lot of leads, like we're blowing through budgets and it's great. And so you and everybody, dude. So, yeah, ah, I said, um, I'd say it's pretty normal for the leads you generate because their search ads are less than a hundred dollars per lead and then their LSAs are right around $60 per lead, which is a little bit higher than I think we're seeing. Um, but at the same time, like LSA leads tend to be lower quality. And the reason why is because LSA leads are a mix between people who are searching for service and people who are searching who used to just click on the Google business profile. And those people are, uh, a mix between people who are serious, people who are not serious, and people who are existing customers. Um, but I said, I think the difference between whatever sales, I think the difference is whatever sales or follow up system you guys use. A lot of our clients complain that LSA leads aren't very good, which is usually why we stick to a budget as basically like jabbing him like, stop, stop turning these on. We got to stick to a budget. If you have a bigger budget than what you're saying you do at the beginning of the month and tell us, um, I said your search ads are averaging under a hundred dollars, which is right on target LSAs right now, one of the, one of those things where your budget can be almost unlimited because now you're paying for clicks and calls that people used to just go to the Map Pack and find the top three businesses there. But now all those organic clicks are turning into $30 leads. $30 leads sound great, but two years ago they were $0 because LSAs weren't taking off yet and everyone was just going to the Map pack. And so like, LSAS is like one of those weird things. Like if you're not doing it, you're not spending money on leads, but you're also losing like 30% of all of your call volume because people aren't clicking on the map pack anymore. Like, they're just going to the LSAs because that's how Google is choosing to display them. And so, um, now we have this like, really awkward situation where Cody and I have been talking about multiple times of like, okay, well, um, we need to find a way to like either give them control of their ad spend and we, and we bill just the management portion of it, like a lot of agencies do, or, um, we need to find a way to restrict their access so we can pace correctly. So here's the problem that I don't want to run into. And Caesar complains about this all the time too. Not all the time, but he has complained about it in chats with me, which is, um, clients will turn their ads on and off and they will mess up pacing, or if they have standard access to things, they will go in and change things. That messes up your progress. And so like, if you're going to hire an agency, I want to restrict that as much as possible. And that's one reason why owning the ad account is great, because we can restrict their access to do things, but still give them full access to view things. Um, uh, but the other reason we were exploring high level at the same time was because if we're looking at all these other lead sources like Yelp or Angie's or LSAs, everybody's getting like six different emails from six different platforms. So we just wanted to find that like aggregate lead dashboard. But then the other side of this too is like, okay, well what if we can have them log into their high level dashboard, manage the leads in high level as if they were managing them in local service ads, and then we just set up an API connection between high level and local service ads that automatically update the stuff in the LSA dashboard. That way they don't ever need direct access to the LSA dashboard. They can do their management within the platform and then we can control the budget and the spend like normal. Um, on the one more thing about the budget and spend too. Most agencies will charge a percentage of ads, a percentage of ad spend as additional management fee. Right, because it doesn't make any sense for you to charge $1,000 a month to manage Google Ads and then the client spend 1.5 million on Google Ads. That doesn't make any sense at all. It doesn't make any sense at all to charge the same management rate for $5,000 versus $100,000. So like you either have to have these like tiers of ad spend, um, with different management fees or you need to have uh, a percentage margin markup because it's going to take that much more, uh, I mean not you're going to take that much more attention, um, to detail and time to manage the account and pace appropriately. So if our client is in control of their budget and if we go the route of just giving them uh, control of their budget and that budget explodes because they pay something incorrectly and they come to us and blame us for that, we then need to be prepared to deal with it from a logistical standpoint or a financial standpoint. But if we just control the money then that would never happen. And if it did happen, that would be on us because we control that. So I don't necessarily want to go that route, which I think that's why like exploring this high level route is probably the safer route in my opinion.

Host: Yeah, the, some sort of lead dashboard is happening in the future for sure. Uh, I think, I don't know how well LSA is going to cooperate with it just because they're kind of doing a walled garden thing. Um, so we'll do as much as we can. But I know other software that specializes in LSA cooperation has said similar things. I think rating especially like you can't remote rate leads, you have to go and do that. And that's a very important part of LSA management. Um, but another reason why, another Google element of why this makes sense. I'm saying all these things out loud, it's new to me. Maybe this isn't new. I swear Google has developed some of this stuff recently. Like a lot of this stuff that they're developing at this point, they're not even talking about, they're just doing it. And I think it's always because ecom gets the focus. You saw that with the shift from old school analytics to GA4. Like GA4 is designed for e comm and lead gen. You're welcome to show up if you want to but like they don't care and that's, that's fine. But lead forms, Google lead forms recently has, there's been a ton of changes and I was looking back into this and I don't know at what point these things changed. Maybe this happened. I don't know if you're listening and you're a, uh, lead form pro, send us a message and say, cody, you're crazy. This happened a long time ago and you just weren't paying attention. And that'd be fine. But part of. So I'll talk in a second about like what has changed or what I've noticed that has changed. But they're managed by webhooks and webhooks are now accessible to people who previously didn't know anything about webhooks. Cause you can just go to Claude and say, help Webhook. What do I do? And it's gonna do it right? Um, you couldn't do that before. You needed a dev. So if, if that wasn't accessible to you, then that removes lead forms from your potential service offering anyway. Lead forms, now you can have them as an asset, so they can be an extension to your campaign. Um, or you can just straight up have lead form campaigns. And that's what blew my mind. I didn't know this was a thing. You can just, they don't make it clear how you set it up. And there's, there's restrictions to it. And I, uh, should say a lot of these things that we're now exploring are only accessible to us because of the position that we're in now. Like after getting invoice billing for our agency, we get so much cooler stuff now. Like we get access to things that we didn't have access to before. So lead forms are one of those things where they have restrictions anyway. You can make it so that headline clicks on your ad, pop up to the lead form and then that's it. So it's a headline in Google Ads. It never doesn't even go to your website ever. Never touches the client site. It's just, here's the lead form, ship it. And then that's why it matters for us to figure out a third party dashboard solution for that. So when we're getting leads from LSAs, Google Ads, the website slash lead forms that are never going to the website, Apple Maps, ChatGPT, Angie, Nextdoor, like, keep going, it doesn't matter. But all of those have to go into one concentrated place. And that's what we need. That's that thing that we have to work and build in the future.

Co-host: I wonder how many people are going to click on those headlines and expect, expect like or not be turned off by a lead form and expect to go to the website, but not go to the website.

Host: I'd be mad.

Co-host: I wouldn't like it. Yeah, I would be annoyed. But also I'm not a Normal consumer. So like, I don't know, you know, maybe normal consumers would find it, uh, more practical. The other thing I think is that I think the reason why they're overhauling this is a thought that popped into my head just now. But like, I think the reason why they're overhauling the lead form situation, making it more accessible with the dashboard, is because I can 100% see AI mode acting as an agent to contact multiple businesses using paid lead forms. And because it's native to Google, it takes up way less resources and it's way easier to, to reach those businesses through a very standardized lead form.

Host: That's gross, but you're probably right. Yeah, like I don't like that because I don't want to think that, hey guys, you're low key competing with everybody now, so you better call them back the second this hits here, they already

Co-host: knew it there to do it with, uh, LSA messaging, as soon as you leave a message, it says, hey, try leaving message to these other businesses too. And it just duplicates the customer's message and everybody gets charged for it. It's just Angie's.

Host: Angie's Google.

Co-host: Yeah, I mean, uh, they're just double dipping in leads. And that's part of the reason why too this aggregate lead dashboard is going to be important. But also going into the third thing that we're exploring because it's all ties together is because if everything is going to this response rate type situation where it's speed to lead or speed to close or whatever you want to call it, uh, a lot of our clients and a lot of service based businesses aren't fit to reach out to these leads the second they come in. They have real people answering their phones or on their computers and different platforms. And um, Yelp is a good example. Yelp and LSA is both. In order to respond to Yelp messages or local service has messages, it's best to have the platform open in a tab and wait till leak comes in and then respond to it. I've tested Yelp before and if you get the email notification, I've tested it. Like you can reply to it and then it would send into the Yelp messenger, but I think there's a delay or something. So it's like it's better if you're just in the platform, stay with local service ads and you can't reply to that one. I think in the emails you actually have to go into the LSA platform to reply to it or be the holder of the phone that that message is Going to um, and uh, so if you have all these tabs open, so if you're a service based business business and you have Yelp and Angie's and local service ads and maybe Thumbtack or whatever, you have all these, all these lead platforms open and people are responsible for answering that. Uh, you either have to be really big because you have to have the labor to facilitate all those platforms or like one person per platform maybe, or maybe one person for two platforms. That's still a lot of work. Um, so that enters in AI chat or AI voice to be able to facilitate those leads and send them, to qualify them on one hand and then also send them to their appropriate salesperson. And uh, I know I've talked about this a little bit because I explored it a bit last year, especially with AI voice and I know one of the biggest problems is the database. So there's a book I've been reading called um. Oh I forgot what it's called. The author of it is supposed to come out of the podcast. Um, but ah, it's like agent powered AI or something and it's like um, AI only operates well if data exists underneath. You can't just like there is no such thing as plug and play AI that's going to operate well. There is like plug and play AI that you can put into certain things but it's just going to be an LLM and it's going to source the information that you give it. But if you give it like really crappy um, disorganized information that's exactly how it's going to work and it's why a lot of voice AIs don't function that great. Um, it's why a lot of like you can break a lot of chat AI uh and it talks all about building a database that has access to and call tracking metrics is great. Like they have a knowledge bank that you can create assets in. The only thing they accept are PDFs right now and uh, web pages. But uh, I've noticed issues with the ability to crawl the webpage and, and uh, I don't know how much I trust PDFs. I've also, I've seen it like I've asked questions to it and it's made up information or it didn't scan all the information. It's like one of like the real big problems is in the service based industry is lining up zip codes and service areas with different services that are offered because some of our clients will offer landscaping to a wider service area while Doing maintenance to a much tighter service area and communicating that to the AI. So the AI understands if a caller calls and asks about this, knowing where that caller is from by asking and then verifying whether or not that's within that service area for that particular service. And then on top of that having access to a CRM or the calendar or things like that and being able to schedule all this stuff. None of that can exist without the data. So the data starts with the database. And that's what I've been working on while Cody's been in High Level World and Google World. But um, I know we talked to Josh, uh, from Bear north about this a little bit too, um, because I know he's building one or he has one, but essentially uh, we're building a, I'm building a giant database in uh, Supabase and it has like a client table with all of our clients in it. And as soon as you add a client to it, it will fire two make scenarios. So the make scenarios will then pull in the website. It'll um, it'll source all the pages on the website through the sitemap and then it'll parse out all of those pages in like 300 word chunks. But it ends on like specific thoughts and sentences. So some chunks are like a hundred words and other, other chunks are like 350 maybe um, just depending on if we can get the full thought out. And then uh, that all gets cataloged in the database and then so that's step one. Step two is gonna be like if they have like a pricing PDF or something, getting that into the database as well into a readable format. So like markdown or uh, JSON, um, so AI can read it. So then when we connect a voice AI solution to the database, we would just do it via webhook and say hey, all of your information, like you might like call tracking metrics for example, they might have like a proprietary knowledge base like they do right now how you add stuff in. But I would like to be able to then instead of using their knowledge base, connect it to ours and then say, and then see how powerful it becomes. Because then it can only grow from there like our database can. We can ingest it from literally anything. Whereas when you use these third party AI platforms, they build restrictions into what you can and cannot upload or use. So one of the things that we have an issue with in call tracking metrics is the ability to upload like CSVs. Our clients already fill out a, ah, client information document and it's like 130 lines long. And it has all their zip codes in it. It has everything like we need to know about the client. So the next question is how do we just ingest that into the database? So when a client fills that out, we know everything about their, their business or service areas, everything like that. And then also their website and everything else. Um, once we have that, now we can hook that up to a chatbot or voice AI that can then help facilitate all these leads, all these aggregate leads from all these dashboards over from Yelp and Angie's and uh, uh, LSAs. And then it can qualify or disqualify leads within their high level dashboard. And then our client only gets an email or a filtered view of the qualified leads that have been followed up with, that have been scheduled. And then uh, it's up to our client to then report on those in terms of profitability for that. So that's like the master picture of how everything is supposed to be coming together and what we've been working on.

Host: Yeah, I think um, Josh even mentioned what converts. Right. Is like ah, a. Well, that's just a simple enough solution for that if you, if you just want the lead dashboard alone. And it's like, well, Jake's describing all this other stuff that we're also cooking up and then these are just the things that we're saying. We also do want something that can do more. Like our clients are going to ask for more. Right. So, um, uh, one of the biggest learnings has been that marketing is no longer enough by itself.

Co-host: Right.

Host: If, if that's all you're doing is just the simple delivery of the core marketing service. Like it's, that's not sticky enough to, it's not sticky. Okay. It's not sticky enough to get to 2 mil. Okay. Uh, with our client, our clientele, if you're working with lawyers, right. Doctors, expensive doctors, probably fine. Um, but if you're not, then there's going to be other things too that also need to be added in, mostly administrative stuff. And there's stuff that we will be able to grow pretty easily into with high level, like email automations, text automations, things like that, like very simple add ons. So, um, that's why we're looking at that. I wrote this down. I wanted to tell you this. Um, I was driving last night. It's a Panda Express.

Co-host: Of course, of course. Where else would you go?

Host: Um, I was driving to Panda Express, but I went to the one in Altoona, uh, which is far away because I don't like going to the same ones too often because then they think that we're tight and it gives me anxiety. So I like to space out my, my rounds. I treat it like the mafia, where I'm like, oh, I'm going to check. Go check on my, you know, Ankeny Panda Express location. And then I'll.

Co-host: Anyway, the joke's for you because you doing this is going to make you make more people think that they're tight with you instead of just one location.

Host: Yeah, it could be. Well, we're going. We're almost to Bass Pro, um, because Jenny's with me and we look to the right and I kid you not, I see the Great Wall of Iowa and I'm like, what is that? I know we have all the data warehouses here. Like that's what we do in Des Moines. You'll just drive by and be like, oh, that's a new one. That's huge. This one. I could not see the end of it. I'm not kidding. Like, I'm looking and it just keeps going. And this is by. So on the one side of the highway you have a Amazon distribution, which is big, but it's not that crazy. Like, I think everybody's seen a, uh, normal Amazon distribution warehouse. But whatever that is that came up, if it's like a chat GPT, I don't know what. But it's. It is the gray wall of Iowa. And that's what I'm going to call it.

Co-host: Have you googled maps it to see if there was a label on it?

Host: No, I'll do it right now. I think it just, I don't, I can't remember the last time I.

Co-host: You do this. You do this. And I want to, um, shoot out a prediction about AI and data, um, centers. I think, uh, I'm going to take a contrarian approach to like what I, what I've always said about like AI and how I. Or LLMs and how they read websites. Because like I, we did my study, um, like it. It got published on the 6th or something like that and it showed that ChatGPT favored local business schema. And so like I'm trying to rack my brain on like why it's trying to like why it's favoring schema when like Google doesn't and search engines just don't in general or uh, don't even pay attention to it in terms of ranking. And what I. The only thing I think of is that, uh, LLMs prefer JSON markup and also markdown in order to read, uh, and contextualize content or websites. They favor it because it takes less processing power than to sort through all the CSS and JavaScript and stuff like that. And uh, I can see a future in which that becomes the norm because of all the pushback, uh, the public pushback that uh, data centers are getting in the environment and the economy and things like that. So um, I can see there being a huge, instead of the next iteration of AI, I can see the next push being how do we make AI more efficient to perform the same work before we scale that. And I think JSON, uh, and Markdown are going to play a role in that.

Host: I think that was my favorite part about starting down all of this with MAKE is because MAKE helped me to understand the usage of credits or tokens or whatever it is. Right. I think it's very easy to use any of these other ones and they don't really make it clear what all is happening or how much resources. But being in MAKE and then accidentally using too many credits and then going, oh, I didn't need to do all of that. Like I could have put limitations built in here that would help me to use less.

Co-host: So that's why, that's why my AI brain, my crawler, my website crawler is in two different scenarios because I ran into credit per minute. Um, yeah, credits per minute issues. And so what it did was like, I had one scenario that like crawled the website, mapped the URLs and as it was mapping the URLs, it was then chunking it into um, 300 word uh, JSON chunks. And uh, so what, it's split into a dispatcher and worker. So dispatcher goes to the site, it crawls all the URLs and then it basically generates a site map of what needs to be chunked. And then it sends that to the worker scenario and then the worker scenario just receives the URLs and it's no longer crawling, spending credits crawling the site. Cause it's just receiving them in a, in a big bucket and then it starts chunking them one by one instead of doing it all at the same time. Cause I was actually running into um, AI, Google AI Studio limitations.

Host: You sent me your scenario and I was looking at the modules and I was jealous. Like, man, these are new ones. I don't know. These, these look cool.

Co-host: Yeah, there's a JSON parser one, there's an iterator. Yeah, there was Erica. I still don't really know what they are. Like, I, I, I'm I'm lying to myself thinking that I did something cool because AI helped me do it.

Host: I looked up on the map where I think the Great Wall of Iowa is. Uh, and it's a cornfield. Which sounds.

Co-host: Right, conspiracy.

Host: So I think it's so new that. But it's just, it's hard, it's so hard to believe. Like, it was just so much concrete. Like, how did you do that that quickly? I don't think you, like, this is extreme Home makeover.

Co-host: But we did a data warehouse Enterprise Edition. Yeah.

Host: Yeah. I don't know. I'll take a picture next time that I pass through. Uh, but it was, it was wild. I remember thinking Facebook, because they don't like, if you guys never. There's no reason to come to Iowa, right? Unless you're just like, I'm going to go look at all the data warehouses that don't have names on them. Um, that's what I want to tell you. They don't have names on them. So you're just going to be going through the Des Moines Metro and then you'll see this giant building that it's not tall, it's just flat. It's like two stories, three stories max. Um, and it's gray concrete and no descriptor on the outside at all. So you have no idea which ones is Microsoft or, or Google or Facebook. But the one that people know as Facebook, um, I think was like one of the biggest. And that, that was like my frame of reference up until this point was just seeing that and then going, okay, yeah, that's, that's really big. And I'm sure that there's maybe 20 humans in there maximum. Um, and that's Facebook. But this, this new thing, I gotta take a pick. I'm not gonna go outta my way, but next time I do, I'll take a pick.

Co-host: I don't have a problem with data warehouses, really. Like, but what I, if they're taking up like that large of like plots, uh, I think there needs to be a way to farm the roofs of them, like plant crops on top of them or something. Or they need to be underground or something. Because I think as a society we never use, besides skyscrapers, we never use height. Everything we do is two dimensional. It's just like a plot of land and everything is two dimensional. But it's like there's so much sky above that that we're not utilizing.

Host: The problem here is people are going to be like, for what? Like, we already have all this space that nobody's doing. Absolutely anything with. I'm with you. I'd be like, put solar on top. Do something. But also, uh, the alternative was the corn field.

Co-host: Yeah. Which I mean, corn feeds the entire world. Like not even just corn, but like corn products, corn syrup, corn starch, corn like everything. Ethanol, like. Yeah. You know, so it's like the more you put up on and take away from the corn and soybean fields, like the less product we have everywhere. Yeah.

Host: Great Wall of Iowa. That's what I want to put it on. I'm going to put a GBP tourist attraction. Verify it because nobody else can get the gbps verified anymore. So I'll verify it for them and then people can come and see the new gray wall.

Co-host: Yeah. All right. That all? Yeah, that's all cool. Well, everybody, thanks for listening. We'll see you next time.

Host: See you.