The Death of Ecommerce Jobs? AI vs. Humans Explained

[00:00:00] Richard Gaffin: Hey everyone. Welcome back to another episode of the Ecommerce Playbook Podcast. I'm your host, Richard Gaffin, Director of Digital Product Strategy here at Common Thread Collective. And I'm joined by a very special guest today. He, I believe he's been on the pod before, but not with me, is Jacob Posel, our Director of AI here at Common Thread Collective.

Jacob, what's going on, man?

[00:00:18] Jacob Posel: Hey, how's it going? Happy to be back.

[00:00:21] Richard Gaffin: Yeah. So I, I think part of the reason we wanted you on is of course, we've been rolling out our Compass tool which applies AI tools to creative strategy specifically as the first of, of kind of a long string of things that we're going to be building in order to make the creative strategy process a little bit easier. But what we wanted to have Jacob on the pod today to talk about. Is a more, a broader conversation about the future of AI and how it specifically relates to e commerce. Now, of course, there's been endless speculation about AI in the news, just generally publicly and how AI is going to apply to society at large, but we really wanted to kind of drill down into the specifics of what AI is going to look like for the e commerce space specifically.

So, and of course, Jacob has been building an AI tool for that. So he's probably the best qualified person I know of to talk about it. So let's actually, speaking of which Jacob, maybe tell us a little bit more about, Yourself, kind of your background with AI, how you got into this.

[00:01:15] Jacob Posel: Sure. So I, I started getting into AI in an academic background. I did a machine learning based degree and I was graduating right around the time. That like the chat GPT moment happened when they released GPT 3. 5 open AI did, and everyone realized that we now had access to this revolutionary technology.

At the time I was working for a logistics and fulfillment startup in the e commerce space. So that's where I kind of started. That's where I got my start in e commerce. And I immediately jumped into figuring out how, how can I apply AI to To these complex logistic problems. You have a lot of unstructured data, a lot of problems there.

And I was really early in figuring out how AI fits into, into those pieces. I then left that startup and I did some work in the ERP world, ERP for e commerce and so went deep into inventory management and AI. And then I met Taylor and kind of fell in love with the vision of CTC. I see it more than just as like a growth marketing agency.

I see it as like a holistic e commerce management business basically. And so like, there was no better place to get in to build AI applications for e commerce than here. And so, I went deep into mostly the marketing side, but also the financial side of e commerce. So I've worked across the entire e commerce stack, I would say.

From from inventory to creative. And so I've thought deeply and I spent basically all my time thinking about, Hey, how AI fits into all the different pieces of the business.

[00:02:52] Richard Gaffin: Okay. So let's, let's then let's just kind of dig into that then. So one thing that we're talking about before we hit record here was specifically how AI or there's some problems that are endemic or specific to e commerce that AI is maybe uniquely equipped to solve. So let's talk about some of those, like what, what issues does e commerce have that AI is well set up to be a solve for,

[00:03:13] Jacob Posel: So this is my mildly hot take in the e commerce world, but.

[00:03:17] Richard Gaffin: Hmm.

[00:03:19] Jacob Posel: and age, when there's capital constraints and e commerce is going through the multitude of reasons why it's a difficult business, there's low margins. Cash is tight. Cash is tough. I don't think that e commerce gets access to the highest quality talent.

I think there's two reasons for that. I think they can't compete on price. With consultancies, investment banking and software. And so that's incredibly difficult. And then when you do get super talented people in e commerce, there's this culture of like agency and services. And so a lot of people realize, Oh, I'm good.

I'm just going to go do it for five other people as well. And that's super common. So e commerce is already an incredibly difficult business. And it's kind of exemplified by the fact that you have this high operational cost. You have high operational expenses. And you don't have the highest quality talent.

And I think the world that AI brings is that it creates this foundation where all the work being done by people that is wrote or within a specific system uncomplex, which you still need to pay people fair amount of money to do is all going to be taken away. And only the best people are going to rise to the top.

Only the highest quality talent is going to rise to the top. So the best people in your business are now going to be. They're not going to have the ability to have even more impact to do even more high level strategic and impactful work And all all below that is going to be taken away abstracted away by ai So there's going to be no more looking at a spreadsheet to see To try and estimate when you're going to run out of inventory based on your sales like your last Last week's sales pacing, for example you're not going to have people building campaigns in a media account You're not going to have people going in and clicking into your facebook account and looking at how things are spending Right, like all of that work is going to go away and you're going to have Access to the highest quality talent.

And I think that's going to have less headaches. People are going to have less headaches when you're a, when you're a successful business owner, every problem becomes a people problem. You hear everyone talking about that. You're going to have more cash in your pocket and you're going to have better people and you're going to be able to pay them.

So, I think that's the world and that's the value that AI brings in e commerce.

[00:05:41] Richard Gaffin: So, so paint a picture then of like, what, what does the org chart look like in, I don't know, actually, maybe let's start here. What's the timeline on this? Do you think?

[00:05:49] Jacob Posel: I think it's close. I think it's really close. I think it just takes someone to build it. It's already happening in adjacent industries, like professional services and law. Like, I recommend people just go look at Harvey. ai. It's already being done in law and it's the exact same thing. They just, lawyers now don't need to do junior lawyer jobs.

All they need, everyone gets to do the client work early, everyone gets to do what they wanted to do really early. It's just the most talented people get there. I think the technology is there. I think that the workflows are well defined. I think it just takes someone to sit down and to grind it out for a year or so.

But it's coming. I think it's coming. I think the intelligence is already there. I think AI is already there.

[00:06:30] Richard Gaffin: Okay. So let's say within maybe even a year, year and a half or something like that. What do you think that the org chart then looks like in, let's say an e commerce business?

[00:06:39] Jacob Posel: I still think you're going to need specialists, In the main kind of in the main departments of your business. So you'll obviously have your CEO founder and then I think you'll have your most talented people who are great strategic thinkers and operators who maybe will manage specific departments like they'll manage, you'll manage your marketing, your office team and your finance team, for example.

But all of the layers below that, I think, think of it as a tree that starts to slowly get pruned. I think that you kind of, like, you'll have that high level, and I'm not sure exactly where it gets pruned, and it's not easy. But I think you have an opportunity to say to those people, We only now accept excellence.

That's all we accept because there's no excuse not to be excellent because you have tools that are excellent. And so we're going to help you move up, but those things that you were doing before, we're going to give you an AI tool to do that. And you need to see how you rise above what that AI tool can do.

And it's simply going to be a process of cut that level. Now you need to rise up. Cut that level. Now you need to rise up. Eventually you're going to get to a point where you have two or three people who are just being able to do the whole thing, but that's going to take a crazy person to believe in that.

[00:07:51] Richard Gaffin: Yeah. Okay, so then kind of walk me through what it looks like. You mentioned, like, cut a level, rise up, cut a level, rise up. So, like, in practical, practicality, like, what, what would that look like? What would it look maybe to rise above the AI tool then?

[00:08:07] Jacob Posel: you have to be able to,

in order to grow an e commerce business, I think there's a couple of decisions that matter, right? And it's a matter of strategy. It's a matter of taste. I think in a year from now, AI is going to be able to forecast your inventory and it's going to be able to buy your media, but is it going to be able to say or is it going to make the final decision to say I'm going to grow my business by expanding my product line?

By creating this by creating a new moment, by building a story around that, by building brand around that and doing that in a strategic way. I don't think so. I don't think AI needs to do that. I like maybe AI can do that, but is it entirely necessary? No. It doesn't provide tools to help you with that.

Yeah. So I think in the short term reality, people with that kind of strategy and that kind of taste are the ones that are going to do very well. And then on the operation side. I think it's the people it's the people who are willing to get on the plane, to go to China, to talk to your suppliers, to manage that relationship, to manage those negotiations and who are really savvy that way, are the ones who can have the ultimate success, but it's not going to be, but those people no longer need to be the ones counting numbers in a spreadsheet to place your next PO.

[00:09:24] Richard Gaffin: Right. So it's a sort of almost all execution will be done by AI. And then the decision, anywhere a decision needs to be made, That that's where the human will still exist or whatever, 

um, 

[00:09:39] Jacob Posel: That's, what I think the short term reality is, yeah.

[00:09:41] Richard Gaffin: yeah, because what it sounds like is, is basically that like everything that is computer like will be stripped away and the only sort of kind of task left to do will be anything that's fundamentally human, like physical, for instance, you're saying flying to China. That's something you might not do. If you were in that role currently or whatever, but if AI is performing all of the executions, then the difference maker will be to what extent are you willing to put boots on the ground and go above and beyond in these very specifically 

sort of face to face human ways, right? 

[00:10:11] Jacob Posel: I think that in the world of AI, what matters is taste. What matters is agency. What matters is human to human relationships and strategy. And like that, that next strat, that next level strategy that allows you to go beyond what everyone else is thinking. I think those are the things that really matter.

And there's definitely people that have that and those people don't need to be bogged down by all the work they're doing right now.

[00:10:36] Richard Gaffin: Yeah. I love that. You mentioned taste. That's a really interesting. Kind of, thing to think about that. I feel that people don't, don't mention a lot, but that idea that you have to make, you have to have some understanding of what people are going to like or respond to or whatever. And that's something AI can't necessarily provide.

[00:10:52] Jacob Posel: Yeah, it's like taste doesn't necessarily come from the same way that AI gets his intelligence, right? It's like, kind of like this, this this thing that exists outside the realm of what you can necessarily define. It's not necessarily based on previous training data. And it's something that certain people have.

Or it's something that people can convince you that they have and you believe them. And so you believe that it's tasteful. So like having, having those kinds of skills what I think can grow a business and define the business.

[00:11:28] Richard Gaffin: Okay. So actually that, that kind of. Brings up an interesting point or, or maybe a kind of a flip side way of thinking about this, which is what is it? So like, let's say in five years, even, you know, where there, there will be a significant amount of development in the AI space. What is it that AI will not be able to do? Like what, to what extent, what, where do you, what do you think is the limit? What's the ceiling for AI's capability? Yeah. Yeah.

[00:11:52] Jacob Posel: a five years from now could be an entirely different world. That's a really difficult question to ask to answer. What AI won't be able to do. I, I don't think it. There's a world where there's nothing that I can't do. There's a, there's a, there's a, that reality potentially exists. I think it's already smarter than 99 percent of people it's just a matter of getting it into different different mediums, right?

Like I was gonna say, AI can't make you feel good, can't make an, can't make a person feel good, but like, that's not necessarily true. Like maybe there's gonna be a version of AI where I'm talking to someone through a screen that looks just like you and sounds just like you. It makes me feel great, you know?

But if I were to guess I don't think that AI will be able to foster human to human relationships. I think that there will be a paradigm shift where human invented by human created by human like tastefully human is important. I think that will be a culture shift. And so just inherently AI won't be able to do that.

But any knowledge work, any execution work, probably any strategy work, it's going to be able, it's going to be able to do. So I, I think it's difficult to define it in like the current paradigm, but I think there's going to be a complete shift where that changes. And it's going to be like, not AI for the sake of not AI is what is considered valuable.

[00:13:21] Richard Gaffin: Ultimately, what I'm getting at is like, is the singularity is that coming up or do you have a thought there?

[00:13:27] Jacob Posel: Well, I don't, the reason why I say this is because I don't think that most people I think it's very, very rare for people to be exceptional. And so I don't think AI needs to be exceptional. I don't think it needs to be some extreme super intelligence to be able to do what most people do. I think there are exceptional people.

And so maybe they won't be completely replaced. Like there's something inherent that Elon does that allows him to like, That allows him to do what he does, right? And it's probably very personal. It's probably very in person that allows him to drive people to succeed but I think that's rare and I think like i'm thinking of most people who I interact with in the e commerce space who are incredibly incredibly intelligent and great operators, but like so is AI You know, like AI, it's not, it's not like to say that people aren't great, that they're very, very smart, but so is AI.

And so you have to be really, really exceptional to rise above what AI can do. So I don't know if there will be necessarily be the singularity where it like recursively self improves into oblivion. But I just think it will be very good and most people probably won't be as good at it as it is

[00:14:41] Richard Gaffin: that's interesting. And I, and I mean in, in certain aspects like ai, there's no, it's impossible for a human to be better than AI already in most things, I would say. Right? 

So it's like anything, it's the whole like we think of chess as like the highest level of human intelligence, right? If you're chess grandma, you're the smartest person in the world.

But that's actually one of the very first things that we were able to crack. Um. So anything that's like in a closed system or whatever, or it requires computation in like the most basic sense of the word will, AI will already. Be like light years better than, than anybody, even the smartest person in the black, highest IQ person in the world or whatever.

Um, so do you think that there's. what, what are maybe like the next layers to fall? So if at a certain point, like AI is probably not going to be able to replace things like taste. And of course, human connection. Is it talk to me what it looks like for AI to replace like strategy, for instance?

[00:15:38] Jacob Posel: So People will have started playing with the new The new open air product called deep research and I think it's the first glimpse into how ai can Go off define its own task, define its own goals, and then research until it completes that, right? And so it can also write its own code, so it can analyze its own data.

And if you think about the decisions that people make that are strategic or like the next steps to take in your business. If it's not something like the tasteful ones that we defined, if it's just like, what's the next best step for my business to take, like what's the next best product or what's the next best thing?

Like, am I pricing correctly? And so on imagine you have deep research, but you are able to create 500 of those. And they all are able to do that at a cost that's close to zero, and they're all able to give you the data driven and research driven best next steps, or at least a report to be able to give that for you to be able to make that decision.

That's probably the reality that's coming. So I think what you're going to have are these like agents, which are able to. Research every single possible outcome and consider every single possible outcome and then present you all of the facts and let you make the decision and it can probably make that decision like probably say based on what I've discovered.

This is what I would recommend and it's probably going to do a really good job. Another important part, another important shift, which you, which it's worth mentioning that is now these AI models have come up with these AI companies have come up with these reasoning models and reasoning models of basic, they basically just took regular models and told it to think harder before it gave you a decision and that showed to increase intelligence by like 20%.

And now they're just coming up with ways to scale up, think harder. And what that does is it's able to consider multiple different variables and it's able to come up with better decisions if it needs to reason and logic across a complex data set. So I think you have this world where you have every single possible outcome.

Considered and you have a super smart agent that can reason that can consider these things and it can go through multiple lines of logic and go back and reconsider and come back to you I just don't see how strategy isn't replaced that way, you know Like I don't see how I don't see what a person can do Besides for moments of human genius, which absolutely exist that will that will beat that, you know, so, the price of intelligence is going to zero.

I know that deep research currently is like 200 a month. It's not actually worth 200 a month. People are just willing to pay for it. It probably costs OpenAI significantly less than that. And so, and everyone's coming out with their competitors now. So the cost is going down, the intelligence is going up, and you can just see how this becomes like it's the best, like it's the best time to be a founder though, like now you have the knowledge work is free, like the best level, the best knowledge work is free.

So that's how that's how I think strategy kind of becomes abstracted away from people as well.

[00:18:51] Richard Gaffin: Yeah. So when you're talking about the, like the, so maybe help clarify for me. So the value of intelligence going going to zero, was that how you phrased it?

[00:19:00] Jacob Posel: Yeah, so I think the price of Intel, the cost of intelligence is going to zero, it's all converging. The cost of all of these models of AI is all converging to zero.

[00:19:10] Richard Gaffin: Right. So what, so what does the economics of that look like then in the near future? So it's basically, this is just almost like a public utility of some kind or like, what's, what does that look like?

[00:19:21] Jacob Posel: Yeah, I mean, that's the problem that everyone is facing right now. You know, you add this deep seek moment. Where everyone thought OpenAI and Anthropic and Google were leading and then a Chinese hedge fund came up with an equivalent an equivalent AI model to OpenAI for a tiny, tiny fraction of the cost.

And that just showed that either they were way more efficient or OpenAI was inflating their margins. And I think it was a combination of the two. So the costs are being brought down because everyone's competing the costs down. There is a consideration here, which is that as long as one model is consistently better than the rest, I think it's a zero sum game.

And I think that people will pay for that. So you may have deep seek, which is almost free. But for the most advanced work, everyone is still going to pay for the most expensive one. But yeah, it's a scary time to be a company that just makes this money selling, it's called selling tokens, selling the words that the AI outputs for you.

Because the cost is, it's going down and the cost is going down so much and they're getting better and better. So yeah, I do think it becomes a commodity. I do think, I do think that soon you're going to be able to have an equivalent of a state of the art open AI model living in your house, like a private, in a private cloud, living in your house, living in a computer that's able to work for you.

And it's very, very cheap to do so.

[00:20:45] Richard Gaffin: Okay. That's interesting. So, so talk to me then about the, the drawbacks of AI either for specifically for e commerce or maybe just generally for society or whatever. Like, what do you see is like, what is like the techno pessimistic side of you? If it exists, like, what, what do you see is like the potential drawbacks there?

[00:21:02] Jacob Posel: I mean, we spoke a lot about how AI improves talent for e commerce businesses, but like, what if you're not that talent, you know? And what if you don't have the, what if you don't have the motivation or the ambition to rise up using, utilizing these tools? Then, I think, I think work will become different for you.

I think that's the reality. I don't think this is stopping. I think that the world will look different, but I don't know if it will necessarily be worse. Like, we also no longer dig holes with shovels, right? We have, we have machinery and equipment. But those people who were using the shovels aren't any worse.

They're just doing better jobs now. So, but I do think that's going to be a short term. That might be a short term problem. I think that a lot of people I've seen start to use AI as a crutch and they lose the ability to think. So they start putting everything into chat GPT and like, I've interacted with people who I push people to use AI as much as possible, but.

There is a such thing as overusing it and they'll send me something. And I'll just be like, did AI make this? Because it makes literally no sense because

[00:22:16] Richard Gaffin: hmm.

[00:22:19] Jacob Posel: it doesn't know context. It doesn't, and sometimes people assume that it understands your context, but it's like, yeah. Right now, when most people use something like chat GPT, they're starting a conversation with a completely blank slate, but people don't really take that into account.

And so, they use it as a crutch. The work isn't as high quality because it's overused and they're forgetting to give context. And then it also it kind of means that you're not flexing that muscle that you need to do to do the work. So like super concrete example, I code all the time. I code every day.

[00:22:51] Richard Gaffin: That

[00:22:52] Jacob Posel: There's really good AI coding tools out there.

In the beginning I was using them a ton

and I realized that I was getting worse at coding because I wasn't flexing. I wasn't training that muscle every day. So

because it's not yet

perfect at that individual task, I realized that this is actually hurting me more than helping.

So I made a very I made, I made like a conscious decision to Use it less step back a little bit and make sure that I was still independently good at the task and didn't just get AI To do everything for me and then I realized that there was this nice hybrid happy medium in between where I was better and AI was helping me as well So like that's that's the problem when people use it for imperfect tasks right now because it's not perfect at everything yet by any means

[00:23:37] Richard Gaffin: That's interesting. Yeah. So I was, this is probably, I don't know, several months ago. I was, I gave a couple of like lectures at a college and this was in the English program, which is my background. So, like, liberal arts, you know, and was interesting is like, all the kids were using AI to like, summarize the things that I were saying.

And then Kind of regurgitate it back to me. Right. And it struck me specifically, like in the realm of English, where you're learning to write and communicate to other people, there's a potential, I mean, all those kids seem fine or whatever, but this is just a bit of me putting on like my old man hat of like, I wonder if there's a scenario where the type of talent that you're talking about can no longer be nurtured because they're so reliant on AI to do their thinking for them. And so it's like the ability to develop the type of talent that's going to rise into those positions actually may become. Sort of hindered by the tools themselves, unless we can figure out some way to create that balance that you're talking about. Yeah,

[00:24:33] Jacob Posel: the talent that i'm describing finds ways to rise above what everyone else is doing regardless, I think they would always do that. And so I would kind of hope that the, the more ambitious people would realize that this is hindering them rather than helping them or that at the very least they're doing the same thing as everyone else.

And so how do you rise above that? And how do you do that? So yeah, I, I, but it's possible, it's possible that it just becomes every single person who's entirely reliant on AI to do everything for them. Like, I don't know how to, I don't know how to read a map anymore because I use Google maps for everywhere I go, you know?

And so, but like, does that hurt me? Like, no, because I have my phone everywhere that I go. So we will definitely, definitely have that reality and some things for some people, like we're going to have people who don't know, like the next generation won't know how to do things that we know how to do now and.

Will think it's funny and silly, but they wouldn't need to anymore because they're gonna have AI with them everywhere they go.

[00:25:37] Richard Gaffin: right. Yeah, right. It's like, to what extent is that going to even be? A problem. I mean,

maybe in certain, certain instances, it will be, but yeah, my inability to read a map is will likely never be a problem in my life unless there's some kind of apocalypse or whatever. But,

um, okay. So then let's, let's get kind of down to the practical level again here for e commerce specifically.

So you're a library of cool AI tools. So what do you think are the most necessary? Five, 10 tools that every e com user should be either looking out for or using already.

[00:26:12] Jacob Posel: Okay. I'm going to give another kind of hot take. I think the best tools right now for e commerce live on people's computer in private code bases that only the best people are getting that only certain people are getting access to. Like, I think that

the things that I'm building for certain people at CTC are way cooler than anything else that exists in the public market.

And so I do feel like, and this is also something which I want to get across in general, which is that. The level of difference in quality and building a well defined workflow, like getting an actual solid engineer who works closely with domain experts to build a certain workflow or like build an agent and apply LLMs to that problem is so vastly better, so much vastly better than what you have access to with chat, GPC that I think a lot of people underestimate it because they haven't seen those yet.

So. Off the shelf tools right now are lagging behind in the frontier because the frontier exists in private tools, but that said, I do still have recommendations of cool tools. So, I would say that everyone needs everyone needs an open AI subscription. Everyone, you don't necessarily need the pro account if you can afford the 200 a month.

Do the 200 a month deep research is like really really good. It's really really powerful it's a step change level above just like a regular the regular search Function on chat gpt. So if you can afford that you should do it. It's worth it If you can afford to have multiple, also get a Claude one.

Claude is really good at certain things and Anthropic's about to release a new model, which will probably be better than any of the open AI ones. So it's worth just having access to those. I would say those are the base two. If you're a little bit more experimental, check out AI studio by Google.

They have some of the best multimodal capabilities. So Gemini can Gemini probably has the best image and video understanding. I'll give a concrete example of how I've done this in the API, but you can do it. using just their regular AI studio, which is a web app. I uploaded long form UGC contact content or interview content.

I asked it to analyze multiple different videos, extract key quotes that was said by the user create angles for those different quotes, and then create ad concepts for those quotes with the, with the With the timestamps of the video, which is something that people, a lot of people will pay offshore offshore agencies to do, to analyze their videos for them.

So that's something that Gemini can do for you. And it also has really cool streaming capabilities, meaning you can share your screen with it.

Like I can share my screen with Gemini

and get her to look at my problem, look at my screen and I can talk to it and I can ask it for help.

So if that's something that you're comfortable doing, I think it's definitely worth it.

And then

[00:29:04] Richard Gaffin: going to

[00:29:05] Jacob Posel: on the other side

you now have this ability, this like realm that's coming up of something called personal software, which is,

you no longer really need to hire an Upwork developer. You no longer really need to ask, like, I get less people asking me now to build them like silly little apps because you have these personal software tools.

So things like Repl. it. Things like lovable, things like bolt dot new. I think it's lovable dot dev. Those things will build you a prototype of an application

and let you deploy it, which is amazing. Let you push it to the internet for other people to use at a level. Honestly, like, if I'm being honest that I could do in the amount of like, or in a quicker time that I can do it.

It won't take you to like,

An app that you can sell to other people that requires more that requires more depth, but as just a prototyping tool, or just something that you want to personally use, it's incredible. So personal software is something that's great. And it might even replace a couple of the little SAS tools that you pay 50 a month for that you don't really need.

I would also check out workflow tools like lindy. ai. Another one is integral. Another one is word where those are basically the no code or low code agent building tools. So think Sapphire, but with LLMs built in, and it's kind of like native to AI that will take you to that next stage. So think like, read my emails, extract spreadsheet analyze, call Shopify, analyze my current.

Inventory. If I'm running low, like place a PO in this template that I have to find and email it to someone like stuff like that, you can actually start doing with these tools. So yeah, I would, those, those are probably the ones that I would get started with.

[00:30:59] Richard Gaffin: Yeah. So it sounds like, I mean, part of like the advantage of like the personal software or just like the, the kind of like custom developed tools that you're talking about is that one thing you had mentioned a little earlier is the issue with AI or one thing people don't understand about AI tools like open AI is that they. The tool itself doesn't understand context. And this actually feels like it would kind of get around that, which is you have a custom built tool, the tool, the context that you need for the specific tasks that you're doing, or just your life maybe is built into that tool. And so it can, it's, it's ability to answer questions for you or do work for you is sort of exponentially greater than just a general tool would be.

Is that 

some, 

[00:31:38] Jacob Posel: yeah, yeah, definitely. Definitely. It's definitely a lot better. I would say it's best for like, For defined workflow. So I feel like I know that there's people who will do the same custom workflow that they do for their job every day inside of chat GPT. So they have some kind of like, hacky process where they take a screenshot of one thing and then they copy another thing and they paste it in.

And that's giving the context, which is what I described. Imagine now you just have that like it's just built in it's just ready to go and it just goes for you, which is it accelerates the value of it a lot significantly.

[00:32:18] Richard Gaffin: And not to quickly plug Compass here, but of course, our copy generator tool within Compass is essentially that right. It's an LLM that has the information from your brand brings it in and allows you to write copy specific to your brand.

[00:32:30] Jacob Posel: Exactly. Yeah, it's like, If I was going to go to chat GPT, I would have to tell chat GPT. I'm writing copy for meta ads. These are some examples of copy that I've already written. This is my brand. This is what it does. This is the output format that I want. And these are the character limits that I need based on the constraints of the meta app platform.

For example, like those are all things that I need to just tell it every single time. And if you don't tell it that every single time is not going to do the job that you want. And so people often dismiss it. What Compass can do. Is it just has all of that inbuilt so it will pull all of that straight away It will pull your reviews straight away.

It will search your website. It has brand. Your brand guidelines and like your tone of voice and style and everything And then it just does it and it does it really quick. So that's like the example of The example of a tool that just does, it just does it better, you know, and it just does it consistently and reliably because a workflow has been built around it.

[00:33:27] Richard Gaffin: Okay, cool. So like, let's, let's, I want to leave everybody with. The question I like to ask people, which is if you can do one thing to take advantage of the upcoming revolution in AI and e com, like, what is the thing that you should do right now, if you're not doing it already to get ready for that or to kind of get ahead of the opportunity, I guess,

[00:33:49] Jacob Posel: Try and get as close to the frontier as possible. So you, you, by being close to new tools, you kind of become, you get this exponential value, like this multiply effect where, because because the increases in the tools are happening so rapidly. Just by being two weeks ahead, you can be double as good as someone behind you.

Right. So like if you stay as close as you possibly can to the news or to the new tools and you just like, you have the mental strength to just like, sit down and be like, I'm just going to try this new tool. I've tried all the old tools a million times, but I'm going to try this new one and see how it works for me and think deeply about how it can work in my business.

That will have a significant effect because as I said, it's like, because we're dealing with intelligence here, like you can get, you can get 10 percent better, but suddenly it has this ability to do something it wasn't able to do. AI has this thing called emergent properties where for whatever reason it reaches a certain threshold and now it has a whole new set of skills.

So that happens. And now you now have access to something no one else has access to. And if you can take advantage of that then, then you really, like, that's how you really Excel. And it also helps you develop this intuition of how AI works best for you. And like people, people joke about prompt engineering or like different things like that, but having a good intuition of how AI works, And what's happening under the hood or what it can do or what it can't do is really, really valuable.

It makes you feel less, less out of control when you're using it. So I would say stay as close to the news as you possibly can. I would honestly say Twitter is the best place for that. Everything kind of just ends up getting posted on Twitter first. Follow me, I'll repost the cool stuff. But yeah, I would say, I would say that's the highest leverage thing to do right now.

[00:35:41] Richard Gaffin: What's your, what's your Twitter handle?

[00:35:43] Jacob Posel: DTC Jacob. Yup.

[00:35:45] Richard Gaffin: DTC, Jacob, there you go. Short and sweet. So yeah, follow Jacob. Also I would be remiss of course, if I didn't mention here again, if you want to start using one of these custom tools, we've got one for you. It's Compass, our creative strategy, AI tool. And basically, so you can find us at adcompass.ai believe that's the URL. Check us out there, drop us a line. We would love to talk to you more about potentially putting that into use for your business. Jacob, thanks for joining us. Appreciate you coming on to chat AI and for all the rest of you, we'll talk to you next week. Take care.