- Platform: meet
- Source: Read.ai 01KYW73BZN6K985HP5GG8S7ZVQ (via Ben Ornstein’s account, imported 2026-08-06)
Transcript
Michael Houck: What is, I mean, I saw your updates, but what’s top of mind right now? Ben Ornstein: Yeah, I mean, so the agents, I’m fairly confident the Jolt agents are in a pretty good spot. And I’m glad Hermes thought the architecture was not terrible. So that’s good. UNKNOWN_SPEAKER: Yeah, I liked it. Ben Ornstein: That was good to hear, which I felt more confident just building them in general this time. So that’s good. Yeah, I kind of figured based on the conversation yesterday that like the whole acquisition decision had been made. Honestly, I bet the margins are looking insane post-writers. So significantly better, I should say. Michael Houck: Yeah, it’s just good for growth because we want to, you know, we want to finish up the new funnel and then really invest aggressively in these ads. So we just, you know, it’s just better for the business, I think. Ben Ornstein: Yep, for sure. So, okay, yeah, Emil, you already talked about, yeah, so for the next, so my priorities now, I guess, I’m officially off all studio accounts. We finally got Jeffrey’s access to Charlene, who’s actually now taking over Jeffrey. Replies. I’ll continue to look over those for the next week or so before she like puts them in. But first day looked pretty good. Michael Houck: Sorry, just on the replies, like if we get this working with your account and we are able to successfully grow your account and they’re not terrible and you don’t flag them as being weird and AIE, I don’t know. We’re probably going to need to iterate on it a little bit. But like at that point, I feel like the process with the VAs probably isn’t worth it, right? Ben Ornstein: If we can get this to a point where it’s fully automated or very close to, yeah, I mean, we could probably consolidate like a review with just Venus and like maybe, maybe at a maximum. I guess I have to see how this all works and like how successful it is. You might be right. You might be right that we don’t even need them at all. Maybe. Michael Houck: I’d like to not have a review. Like, I’d like to get it to a point where it can just go. Now that we’re using Fable for, I don’t know what model we were using before, but like now that we’re using Fable and just spending a little bit more on that, I think I’m optimistic. I don’t know if I’m naively optimistic, but I am optimistic. So, I do think it’s good to have Charlene either way, just to have sort of more hands on deck as we need them. But the process, if we can get it fully automated, will be much cheaper. So that will be a positive. So let’s just monitor that and see how it goes. And you let me know each week, just like, hey, here’s, and obviously, if you find any that are like really terrible, you know, just like flag those. You can even just DM Hermes and kind of let it know. What the deal is for sure. Ben Ornstein: Yeah, so I mean, when building Casper and the other agents, it was, I mean, it was like, so there’s Haiku, which I was using for most of the lower-level tasks because it would be really cheap. Then there was Sonnet for like the main task. I never even touched Opus because I thought it was too expensive. So Babel, I mean, look, I don’t know how expensive it is for the tasks that we’re running. And I don’t know if, again, it was my architecture, but look, if Fable’s not too expensive, then hell yeah. I mean, that’s great. Michael Houck: Yeah, just for the comments. Like, we’ll see what it plays out in practice, right? But I ran a pretty, I think, sophisticated scenario estimate for it on a per client basis per month. It’ll be. I think, under 100 bucks to do 100 comments a week, which is more comments than we’re doing and cheaper than we’re paying the VAs. So, again, as long as we don’t sacrifice quality, I think it’ll be a good move. Definitely. Ben Ornstein: Yeah, it feels like this is a very interesting time with the business. It’s like it’s transforming into something much different than what it’s been for the past long time. Michael Houck: Yeah, honestly, it is. It’s like, I feel like I’m, I feel like I’m doing the things that I’m best at. Again, I’m not on the content yet, but that’s next. Like, I’ve felt for a while that most of my day has been like orchestration, management, coordination, responding to Slack threads, not like building things, you know? So in that sense, it feels good. I can be, I can have a little bit more of my Mark Lou energy. Ben Ornstein: Yeah, no, it’s great. I mean, you built this stuff really quick, and it seems like it’s going well. But okay, so yeah, just in general with Charlene, great start. Give her a tracking sheet. She’s up and running, handling a client. So I agree, just in general, like having these VAs, because there are tasks where I’m like, okay, could I have Venus or Rena do this? And then sometimes I do offload it. And so, like, just having another hand, as you said, that’s great. So, that was a lot of my time, like, making sure she was all set and just getting her onboarded. And so, that’s kind of delayed me on a few other things, but I think it was worth it to spend the time there. Michael Houck: Yeah, I agree. She needs, it’s her first week. She needs context, So, definitely. Ben Ornstein: So, on Jolt, yeah, the agents. Yeah, I’m excited. I want to see Tatum start using them. I hope she. Starts using them. I removed her DM access because I only wanted her to use that one chat for now to see the agent channel. Michael Houck: Yeah, that’s smart. That’s smart. That way we can see it. UNKNOWN_SPEAKER: Exactly. Michael Houck: And also, Hermes can see it too because he’s only in channels. He’s not at the right. Ben Ornstein: So that’s good too. And then, yeah, just that sales list. I’m just as if I have like a free few minutes, I go on TikTok and scroll for a little bit more. So I’ll just continue adding there. As I was mentioning, so I said that, yeah, Charlene and that all the other tasks that come up through the week have kind of delayed my anything on the ad side. I wanted to spend like a lot like ad creative creation. So I want to get on that today. I have not done that yet. Michael Houck: So that’s fine. That’s not, that’s not urgent. We still have two more weeks with the feed, and they’re presenting new copy, new, new creative to us next week anyway. So that’s not urgent. They’ve got another batch to do first. Ben Ornstein: So that’s what I thought. That’s maybe why you mentioned it to me. I thought, and I should have responded when you had messaged me that. I was like, oh, does that mean I probably shouldn’t do the next batch of creative? Michael Houck: But I guess you honestly like hold, you can hold off until we get that final batch from them, which I, again, I think is going to be next week. We’ll see. Because then we’ll get those live. And by that point, I should have the funnels set up with the VSLs recorded and have our things all set there. So we’ll start to see if that’s working. And then as that works, we will scale up spend as we scale up spend. We’ll need more varied creative. So we’ll want to do more frequent passes and new stuff. And we won’t have that many more to create either. So cool. Ben Ornstein: Okay. That works. Yeah. So for on the viral examples front, so I know you said we’re going to be combining the Monday and Thursday, like current X version into one per week for each the X and LinkedIn. UNKNOWN_SPEAKER: yeah. Michael Houck: Like, I don’t know if we’re gonna do a little micro essay at the top or not. I gotta see if Blake has time to do that. But the main thing that we’re gonna be pitching and you know, therefore, writing about in the newsletter is: here are three templates, they went viral this week. We analyze why, um, and um, we break it down in like a templatable version and example that you can use. Um, so want to do that for X, we want to do that for LinkedIn, and we want those to be separate emails because we’re going to run ads for separate funnels, one for X and one for LinkedIn. And then the people will get tagged automatically when they sign up or grab the lead magnet or whatever in Beehive, they’ll get tagged with the right one. And that’s the only email that they’ll get. Michael Houck: Got it. Ben Ornstein: I’m doing a test right now where I connected the TapLio MCP to Claude. It’s a lot of words there. I connect to those and I’m having it run two separate things where it’s looking for viral posts to see if there’s templatable ones, as well as I’m seeing if I can, because Rina’s only got about gotten about 50 so far. And she’s like, how do I find these things faster? And so I’m kind of saying, you know what, see if I can do this and just pull. I’m having Claude run it right now. It’s been going. It’s using the Taplio MCP to see if it can find a large list of creators. So I’ll get back to you on that. I might just be able to do it myself. Michael Houck: Okay, I mean, we don’t want to like Taplio is like, if we have to pay a subscription for it, we can. But also, doesn’t Appify do this too? Ben Ornstein: I have a free trial right now, so I’m not paying anything. And I can actually, I had a plan in my head where. Every like two weeks or so, you can click cancel subscription and it gives you an extra like two weeks. So, um, yeah, I have a whole system like a patched together solution. So I haven’t paid a dollar for it. Um, and you get the MCP for free. So I’ll just continue cycling through different emails if I have to. Um, but yeah, we’ll see, we’ll see what the output of that looks like, Michael Houck: Okay, yeah, we can make a call there as we as we see because we need the LinkedIn one. Um, I also have Hermes. Trying to figure something out, but I haven’t spent a lot of time on it. So I don’t know. Probably get back to that. You know, we have like three more issues or two more issues. So it’s not an urgent problem. But obviously, we’ll want to get it set up before, I think it’s like the 19th is that Wednesday. So we’ll figure something out, but not urgent there. Ben Ornstein: From the work that I had Charlene doing, and then Rina, when she found. When I had her do like the initial pass on X templates, I have like one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve. I have 13 templatable viral posts that I can put into upcoming newsletters. So I do have them. I just haven’t done the newsletters yet. So at least on the X side, not for LinkedIn, but so I could put those in, but I know we’re a little bit ahead. So. Michael Houck: honestly, if you want to do that, that’s fine. Obviously, not urgent because we are a couple of weeks ahead, but just so I can give Hermes clarity on when it should start doing this automatically versus us doing a manual process for it. If you want to get them drafted in Beehive, then I’ll just do the why it works section or whatever and do that next week. Ben Ornstein: Okay, sounds good. I’ll do that next week. There’s all that going on. And then I think there were a couple other things. Well, one, I’m going to review the the Hermes drafts today. I got through the first script of the one, the script that he created with the AI Tatum. That one looked pretty good, honestly. Like I could envision it. I could see it, even though it’s just text on a, but yeah. The blogs have been pretty high quality in general, like also. Michael Houck: So I think that automating lore fully is work, except for the Reddit stuff, which we can’t automate because we need the aged Reddit account, but that’s fine. That’s an upsell. I think for the standard offer, automating it fully and doubling our margins there essentially is work. And so. My plan there, I don’t know if I mentioned this to you. I just came up with it like yesterday or the day before. My plan there is to get this working for Jolt and then expand it to other brands and some of our clients. And then, once we have positive feedback from clients on it too, or they don’t notice or whatever, just we know it’s working, then I’m to Emma and I’m going to be like, hey, we can double your margins on all of your clients that aren’t related to Megaphone as well if you use our system because it works for us. And instead, why don’t we just acquire your business, give you equity in Allure, you make the same amount like net as a person and have more equity in this new thing, which is going to be growing super fast. We have the funnel set up, we have the AI system, et cetera. I think that’s a pretty attractive offer. But obviously, we need to prove it out first. But that’s my plan right now. I want her involved, I want her expertise, and I want her to lead that side of the business. But I want to approach her from like a position of strength with a fully baked system. Ben Ornstein: Makes sense. Okay, other than that. I guess I’m still, yeah, now, so now we’re in a like a very interesting spot where I’m now fully off studio. Um, and I have, I wouldn’t say I have time on my hands because I tell you, when I think about my tasks, I’m still busy all day, every day, because little things keep coming up, but like, I don’t have like a defined tasks per se, um, which honestly I kind of like because it’s uh. It’s fun. Every day is different. But I guess I’m thinking: okay, where we are right now with the, I mean, the business keeps changing every day. I can’t even keep up. Where would you like me being? Is it reviewing the content that Hermes is creating? Is that like the highest leverage thing I can be doing? I obviously want to be doing whatever is highest leverage. I want to be able to justify and depend on me, obviously. Michael Houck: Yeah. Look, I know I can lean on you. And I think, honestly, having seen what you’ve done over the past few months with the AI stuff is really like you, when we started, I think you even didn’t know if you could do it, right? But it’s very clear, especially from the way you built the Jolt ones quickly and that they’re high quality, like that you can do it. So I think that’s good. I think that you now have a lot of experience with these systems. And so. Being the sort of like human, like review layer of like before we send things to a client or out into the world, even for our own brands from these systems, I think there is a lot of value there. Cause, like, when I think about where I’m going to add the most value and what the highest leverage thing for me to do, it is sort of being this like builder creator person, having that kind of energy again, like coming back to that after like my and management, basically. So I think that’s how we can orient the business the best way because it’s going to let us create a lot of content and just build a lot of like authenticity behind what we’re doing. So I think that review layer on your side makes a lot of sense. I also think there might be some stuff on the AI side as well. I don’t know what exactly that’s going to be yet, but I think that like the, and honestly, this is a testament to you, like you clearly have this like founder mentality now that you didn’t have, two years ago or whatever it was. And so it just like, it has put into me the idea of like, hey, I need to be more in the arena myself, right? And so I think for a lot of reasons, like on the AI stuff, I need to sort of be running that part of the business because we can move faster, it’s less coordination back and forth. And if I’m going to be talking about it, I should also be the one who’s sort of like doing it. It’s just more authentic. It’s not to say, again, that I don’t think you can do it because you’ve done it and you’ve done it well. It’s just, I think about the business, what’s best for it. I think it’s content and building, not managing people and Slack combos. So, look, I’m totally candid with you. I don’t have a specific new thing where I’m like, we need to throw Ben at this beyond the sort of like interface layer between you’re the arbiter now, basically, between humans and the AI. But I thought about this quite a bit over the past couple of weeks. And like, if I’m going to be doing more of the AI stuff and sort of ever since I did the design stuff, I was like, I should just start be doing this stuff, right? I just don’t have like that next thing yet that might change when we launch like the creator partnerships. Agency. It might be good to have you run that. It might change when we run the context farming agency. It might be good to have you run that. But off the top of my head, like mostly research that I can’t get from AI. And so I need to find out where that line is. But I guess what I’m trying to say is like there probably is going to be less in like the immediate future, like immediate to near term, than there has been over the past few months. Just because I’m making a concerted effort to like automate some things and take on more myself as well. So I’ve been stressing about it, honestly, because it’s like, I don’t want to be in a position where you, for totally valid and legit reasons, like need to go, you know, find something else that’s going take up full time and be more steady and things like that. I totally get it if that’s where you’re going to land. To, like, hey, if I’m not, if I’m not sending you enough, you know, like, I get it, but I don’t want that to happen. So I just don’t have what that next thing is yet right now. And so I just want to be candid with you about that as sort of for August and beyond. Like maybe the ads thing becomes bigger, but I don’t know. So yeah, just want to flag that. Ben Ornstein: That’s kind of what I thought. Just like, I feel like I’m pretty in tune with how you’re thinking about things. Michael Houck: You’re very perceptive and you know me very well. Ben Ornstein: That’s kind of how I felt. So, yeah, I mean, I don’t, I don’t, that makes sense. I get it. Um, I don’t, yeah, I don’t know what that means, like, exactly right now. Yeah, and you don’t either. I know neither of us do. So, but I, I, that’s kind of what I was feeling. Um. At least, like getting past this, like sprint. Yeah. Michael Houck: Like, honestly, man, like something I keep coming back to in my head is like, you could literally just build something like with part of your time, you know, like you could just like you, you’ve learned, you’ve you’ve learned so much, I feel like from where we started before that, like, you should have the confidence to do whatever you want to do, right? And I want you to be part of what we’re doing, but like, I recognize that if I’m not feeding. I’m not feeding you, obviously you’re gonna need to eat somewhere else. So I don’t know what that is, but I want to have like open dialogue between us on both sides. And I’m gonna keep throwing things at you as they come up. And as like, oh, like, I need someone else to be doing this. But like, yeah, I just wanted to be kind of open and candid about it. Ben Ornstein: Yep. No, that makes sense. Okay. I guess I’ll have to think about like what that means in terms of the next step, like on my end. I don’t know. I don’t know. But I kind of figured with how the business was kind of evolving that like a chief of staff is like useful, but not like, you know, the top priority in terms of like things to have. Michael Houck: I there’s a lot of things you do that are just like, super high leverage and super valuable. And when I saw that, I was like, if I’m able to free up my time on other things, like those are things I should probably be doing or that I should cut to some degree. Right. And so, but I do think that the sort of like interface layer thing is real. I think that is a durable. Thing that is going to last as part of this business. I think it’ll grow as part of this business as we run agents for lore, agents for Jolt, agents for megaphone, agents for hockey stick for clients that we bring on. And even if we automate like the hard calls and my social content creation, like that’s the next project that I want to do this month is like automate that stuff too. So it’s like we’re at the trough right now. I feel like, you know? And I don’t know what the next shape of the hill looks like, And I know that that’s not fair to you. So I just wanted to be upfront about it. Ben Ornstein: No, that’s that’s literally exactly what I thought was happening. Like, I, that’s how I envisioned in my head. I was like, okay, we hit this like peak. We don’t know, or I guess the opposite, whatever. Michael Houck: Right? Ben Ornstein: We just don’t know what we don’t know, sort of UNKNOWN_SPEAKER: Right. Michael Houck: right. It’s like, it’s like we’re changing a lot of things, therefore, work expands to fit the space you give it. So there will be new things that come up, but I just don’t know what those are. And I know that, like, you would ask me, like, we should think about, okay, like, what does a new hourly rate look like, Right. And I’m like, you know, we can talk about that too, but like, it’s not, it’s not like a financial thing. It’s just like, I don’t know. Know what those next things are going to be, and that’s not fair to someone who’s put so much into this, you know, over a while. So, I just again wanted to just have that conversation. Ben Ornstein: Yep, understood. Um, yeah, I’ll figure that out. I’m not sure, okay. Michael Houck: Yeah, I know. I wanted to give it to you, like, before I didn’t want to give it to you, like on a Monday, because then it would be like I knew you’d just be like in your head kind of all week thinking about it, trying to figure it out. So I hope it’s all right. I know, I hope it doesn’t dampen the end of the week too much, but I figured just good to take some space and kind of think about what you need for him. You know, that you guys are married and you’re getting onto like next phase of life and things like that. So just wanted to, I don’t know. You let me know where you land. Doesn’t need to be on Monday, doesn’t need to be urgent or anything like that. But we’ll do 100% of whatever needed to support, facilitate anything else. But I also might just start sending you something too. So, right. Ben Ornstein: So, that’s the because like I could see that I see two different futures where it’s like, and I’ll be just as candid. I can see where it’s like, look, Hop’s automating literally everything. He doesn’t need me. That’s one future. I see it. I do see it. I can like, I see the beginnings of it. I see another future where it’s like, look, he’s automating a ton of stuff, but there’s still extreme value in having Ben do the research, like the VSL stuff. Like, I thought that was pretty high value. Michael Houck: But I would not have gotten that level of, like, I used your research on the VSL thing to train Hermes on how to help me with a draft of a script that I wrote for the VSL. Right. You know, and I don’t think that’s going to change. So, either pre-research or post-kind of verification review layer, I think both of those still exist and will continue to exist at a higher volume. But I just, again, like, because of how much you’ve put into this, I don’t want you to be relying on sort of the unknown from me, right? So, do I expect it will go up? Yes. Do I know that it will go up in some time frame? No, I don’t. Ben Ornstein: Yeah. Okay. Well, I appreciate like the transparency. I’m gonna, I’ll stay as transparent on my end too. Michael Houck: So please. Ben Ornstein: Yeah. Ben Ornstein: But yeah, in terms of like my like tasks, because there isn’t like a, like a set thing I’m doing, I’m kind of just like floating. Michael Houck: Yeah. No, I know. And like, just like as you’re you’ll probably find things too, and just like flag them to me. And then I’ll be like, Yeah, like, let’s do that, or you know, not urgent right now. But yeah, I again, like, just said we’re at that, we’re at that nadir right now, whatever that is. So, um, and that’s not super fair to you. Uh, so just wanted to. To be open about it, even though it makes me nervous and stressed to think about you fixing something else. But yeah. Ben Ornstein: I appreciate it. UNKNOWN_SPEAKER: Go. Michael Houck: The funny thing is, man, like I’ve been thinking about this too. We’re like rebranding the newsletter as like hard calls, right? These are the fucking hard calls you make when you run something. Like, this is not a conversation I wanted to even. Have, right? But like, obviously, we have to work through it and find what’s best for both sides, and we have to make the hard calls on both ends. So, yeah, I think just for now, just keep me in the loop with the things you’re identifying to take on. And I’ll be like, oh, yeah, like, go for it. Or, like, you know, either we don’t, we don’t want to do that right now or I’ll do it or something like that. But yeah, I guess just like think on it through the weekend and we can catch up on Tuesday and go from there. There. Cool. Ben Ornstein: Uh, where are you going, by the way, later today? Michael Houck: Oh, so Carly and I are going to Franklin, Tennessee, to scope it out as a potential uh spot to build a little homestead when we have kids. Ben Ornstein: Oh, oh, so so you would you keep this house? Michael Houck: No, so the plan, the plan was always Charleston to meet somebody somewhere else to raise the family. Um, so I met somebody way quicker than I expected to. Um, so that was good, and now we’re sort of we made a little list of like these different towns and cities across mostly the south that we’re going to be checking out over the next like nine six to nine months or so. Um, and then at some point next year, we’ll probably buy some land and start building something there. Ben Ornstein: Very, very cool, nice, that’s awesome, yeah. Michael Houck: Um, but all right, man. Well, let me know what your thoughts are next week and we’ll talk through there. But either way, yeah, let’s just keep it fluent, keep it open. And yeah, I hope, I think you know this, but I hope you know how much I appreciate everything that you do. So, you know, it’s, we would not be where we are without you. So I appreciate it. All right, brother, let me know. But otherwise, have a good weekend and talk to you soon. UNKNOWN_SPEAKER: Awesome. Thanks. See ya. Later, man.