Video transcript
Transcript from available captions; it may contain errors.
All right, guys. It's official. The new era of eCom has really just begun. My name is Elon Moose and we are doing over 1.7 million dollars per day with our core store. As you guys can see, this was yesterday alone. We did over 1.7 million dollars. And what I want to go into in this video is what is allowing for all of this to be possible today and where is the industry as a whole going. So what we're going to break down in this video is first off, guys, we have officially entered into the AI era of e-commerce. This has been progressing over the last one to two years, but we've reached a critical turning point where the models have gotten so good that everything is just going to incrementally accelerate from here on out to make things that weren't possible before possible, [music] which is already been happening really over the last year and every single month you can see a lot of accelerations, but I think the big things that have now changed is going to be things around agents, full automations, where the industry as a whole is going and a lot more. So that's what we're going to be covering in this video specifically. And I also want to make sure that this video caters to you no matter what stage you're at. So obviously I'm posting big numbers, but I want you guys to understand that whether you're trying to get your first 50k days, 100k days, 200k days, 500k days, million dollar plus days all the way up to pacing towards two million dollar days, it doesn't really matter because all this stuff is going to be relevant no matter what stage you're at that we're going to talk about in this video. And I'm going to make sure of that. All right, guys, with that being said, let's jump into the video. So hopefully I don't give you guys a seizure with my handwriting, but we're going to get a nice clean Miro whiteboard over here just to demonstrate kind of where I see things right now and where I see things going. So the first thing that I think you guys need to think through is that almost every single workflow that you're doing today can probably be already be automated and within the next 12 months that anything that you're doing on a computer will probably be able to get automated. And this is all coming through AI agents. So for those of you that are less familiar with agents, the way that I think about an agent is essentially the AI's ability to do workflows in a consecutive manner to finish off tasks fully. So the way I could this down is let's say you wanted to write a brief for creative. So, in order to write that brief for creatives, you might need to start with, you know, doing your research, which might, you know, pertain to doing competitor research, looking at your own ads data, you know, doing some Reddit scraping, then taking all of that and then prompting it into Claude to then write the script for you and then ultimately finishing off a script, right? So now, rather than needing to do each one of those steps individually through browser approach, if you're using Claude or ChatGPT still on your browser, what you can now do is pretty much have your agent do all of those workflows. And I think the big breakthrough that's actually happened with all this is that the models are now good enough to actively do this at the same extent or better than humans if they have the right context and training. That is really the big breakthrough that's happened really with Opus 5.5 is what I'm using to build out agents, but Astro on ChatGPT is really good and there's a lot of other models between Grok bot and, you know, other places where you're probably seeing stuff already and people talking about it. So, the models have got to a place right now where they are good enough to actively automate everything. Now, the thing that the agents actively need though to be able to do this is really just two things. One, it's going to be the actual workflows, and then two is just going to be context. So, the workflows, the way that you go about this and at least the way I've been going about it, right, is just you have to think about this agent like it is essentially a cyborg version of you that has no context on the thing that you're doing. So, you or someone on your team is really good at a specific function in the business. Let's just use the creative example again of writing briefs and that's something that you really know how to do. What you need to do is give as much detailed feedback as possible for exactly how you are actively going about writing those briefs to your agent. So, the exact step-by-step process of how you do it, look for the output that the agent gives you back, give it modifications based on where it's screwing up. And the beauty about when you're training an agent is that an agent will never forget. So, all you have to do is tell the agent how you're doing it one time or as long as you explain it correctly, your agent will never make the same mistake twice. And I think that is the biggest piece with actually training it on your workflow. So once it understands your workflow, now that becomes fully automated. The context standpoint is you need to give the AI agent as much context as you have being able to access anything on your computer. Well, the agent can also access anything on your computer. You just have to give it the ability to connect into those things, right? So if a big thing that you're doing is let's say, you know, you like to go into your Triple Whale and then you go into your ads manager and then you use this spy tool or that spy tool, right? You have all these different things that you're doing, but your AI agent needs to be able to do those same things. So if you give it the context and the ability to do it plus the workflow of how you're doing it, now your agent can do that same task just as well as you. And I've already seen this. We're actively using agents to do very, very big functions in our business. I'm not going to get into all the details of what we're already using it for, but we are actively using it today. And part of the $1.7 million per day in sales that we're doing is because of these agentic workflows. That is what I'll say on this, right? Just a quick plug before we jump into the heart of this video. If you're a brand owner and you're doing over a million dollars per month in sales and want to get a full comprehensive audit from me directly where I can break down your biggest bottlenecks in your business and give you really the playbook that's allowed us to get to over a million dollar plus days, just apply down below. I'm going to be doing one of these per week. And the way that this will work if you do get accepted is we'll start by gathering your data so that I can get all the numbers and metrics and have understanding of where your business stands today. From there, I'm going to make a 90-day road map for you of what I personally believe are the priorities just from my experience of what you should focus on and what you should execute on to help accelerate the growth as much as possible. And then we'll take a 1 to 2 hour long call to break down everything that I put together in this road map. So if that sounds like it would be helpful for you, applying the link down below, it's a 2 to 3 minute long form. For full transparency, the reason why I'm doing this right now is because we are looking to potentially execute on a deal over the next 6 to 12 months. So I want to just start working with some high potential brand owners that have potential to scale their brands from million to multi-million dollar months to million dollar and multi-million dollar days. So apply down below if this sounds like it would be a good fit for you. Right now, the the thing that you have to think about with agents from a workflow standpoint, at least the way that I see it, is it has to be things that are pretty straightforward things that humans are doing, right? So like the way I try to think about it from a workflow standpoint is where am I already using AI to do most of the functions, right? So like if I'm already, again, using the scripting standpoint, right? Like a lot of you guys are probably using AI to do your scripting as it is right now, right? Like you're probably prompting it, you have specific ways that you're doing prompting, specific research that you like to feed into doing the prompting, but you are doing things already with AI to get you the output. So those are going to be the things that are the easiest to automate because as long as you give it just the context that you're doing every time you sit down on your browser to give those explanations, you can just feed AI that context and that workflow and now it can do that super easily, right? Because it already you're already doing those things with AI. So things that are very straightforward, step by step, there's a process that's already very clear in terms of how you're doing it, those are the things that are going to get automated first. The thing with these agents right now is that they can't really free think, but as long as it has a workflow that's already proven to work, that is what is going to be super important. So what needs to happen with these workflows that you're giving it in those step by step processes is that one, it should be very straightforward, but two, it needs to be proven, right? Because this is really the most important part because if you don't have a proven workflow that you're automating into an agent, all you're essentially doing is automating right? Because if something doesn't work or let's say you're testing something and you have a lot of faith that this workflow could be super good and if you automate it, it would be insane. The thing is if it's not proven out and it doesn't work, you don't want to accelerate that into an automation. What you want to start with are what are those low hanging fruits of things that you're already using a bunch of AI for most of the functions or a lot of different parts of it. Of course, some of it might not be fully AI because you probably don't even realize you can do a lot of stuff through these agentic workflows because here's the thing, right? There's connectors in Cloud. So connectors are pretty much just the ability to easily add it into, for example, like Triple Whale, you could easily add it into there. But then there's also the ability to just use the browser. And the AI can use the browser as long as you give it context and, you know, you log it into those different things on the agent directly. So pretty much everything that you already can access on a computer, the agent can access as well. So really where you want to start is what's already proven and what is more of a straightforward path that you already know kind of the step-by-step process. And maybe you don't have this already documented into an SOP, but if you can explain it and articulate it in a very clear way, you can probably automate that already into an agent. Now, the crazy thing is, right, once you start having an agent that can do the full workflow for you, you're not just creating a a duplicate of yourself. You're actually now creating a agent that can work 24/7, right? So it doesn't need sleep. It can do tasks all the time, meaning just from this alone, the output of, you know, let's just use briefs as the example, if you yourself, let's just say, are a strategist and you write 20 briefs a week, with an agent, if it knows your workflow exactly how you're doing it and now it's perfected it because you've trained it enough with the workflow and with the context that it has, this can easily do 100 briefs a week or 1,000 briefs a week, right? There's really going to be no limitation. The limitation is really just going to be do you have enough editors to supply it. So the output now becomes pretty much unlimited because you've now, you know, trained it. It works 24/7. And also it doesn't fall into human biases, right? Because a lot of the times we'll use intuition or use our gut for different thoughts, but really a lot of the times the data will point us in a completely different direction. And if you follow the data depending on what tasks you're automating, sometimes again there is, you know, room for the gut, the intuition, creativity, things like that. But there's a lot of stuff that you just want it to be purely data-oriented on the way that it's making its decisions, right? So with things like that, it will always fall in line with that and not use human biases that come into making specific decisions. And then of course the cost is going to be way down, right? Because all you're paying for is the tokens of this. If you want to do this through a creative strategist to get them to write a 100 or 1,000 briefs, Like that would cost you a lot of money, depending on how many briefs you're you're actually putting out. And I'm just using that as one example. This extends way past just creative briefing, right? So this is really what you're thinking about. So once you get this automated, you now have someone that's doing this 24/7 that is only going to also continue to get better. And that is the other big thing is that the more that you train it, the more that you adapt the workflow, the more context that you can give it, the better it is going to get. The improvement rate is going to continue to compound incrementally, right? So that's the the beauty of this because again, it doesn't forget anything. So anytime you train it and give it more information and give it more memory and context, it will only continue to improve. It never will get worse as long as you continue to adapt the the core agent that you've been working on. So with this being said, right? Like there's pretty much just unlimited benefits when you can actually automate a specific function. Again, we're talking about one function, but what will start to happen is you're going to have an agent that can do your creatives, an agent that can then also do your editing, an agent that can do your funnels, right? So everything that you're doing, eventually you're going to be able to fully automate. And again, like a lot of this can already be automated right now. That's that's the crazy thing about everything that I'm saying is that we've already entered into this era of e-commerce where agents can do these functions. Again, we're actively using it and we're only going to continue to scale more and more agents to do these different workflows. So what are the actual limitations of all this, right? Is really going to be what I said before is do you have proven workflows? Because if you don't have proven workflows, there is no point of turning that into an agent because then you're just automating stuff that doesn't work, right? And why would you want to scale that? So where do proven workflows actively come from, right? So if we talk about a proven workflow, that's actually going to come from your best people, right? So people are still the most important thing in your business because this all kind of circles back to you're essentially downloading a person's workflow and context into an agent and you are making this person a cyborg, right? You're making a automated version of this person, except that person is now going to be able to work 24/7, do way more output. It's going to get better over time by just continuously being able to train it. And so, and the cost of it is going to be much lower than the person itself. So, that is what is happening when you are automating the people. But the thing is you can't automate everything because you are not the best at everything in your business, right? If you are already running a pretty large-scale business, you're going to have people that are better at specific functions than you are. The people that have the best people with the best workflows that are scaling the most right now are going to have the biggest advantage because those people can download themselves directly into agents and then just get way more output out of themselves because they now have this agent that can do a lot of the work for them. At least for where we're at right now, what does this mean for the actual people that you have in your company? It actually means that they're more valuable than they've ever been before if they are good people in your company. And I'll explain this a little bit further, right? Because what starts to happen is let just say, choose another example where you have a really talented video editor. And right now, we're not at the point where I think video editing can be fully automated, but we will get there in the next probably six or 12 months. That's that's the truth, right? And let's say you have a really good editor that has really good workflows for how they're doing their their editing, right? What can start to happen is they are going to do all the stuff that I just spoke about. So, that video editor will make an agent, right? So, that agent is now going to be able to do all the workflows and step-by-step that they're doing to get to an end-to-end edit, right? Now, the thing is here, the next part that you have to understand is that this can be local agents versus coded agents is what we could call it. And again, I am very non-proficient technically, so like the way I'm calling this in the terminology is probably completely wrong, but this is just the way that I see it. So, a coded agent is when you're using an actual infrastructure which we have our own software that we put a lot of our workflows into so that it becomes much more scalable for the entire company. And this is really where you have to do things because what can start to happen is now you can share this agent across all your other video editors. Differences just like different editors have different styles and different ways that they ultimately complete the edits, it can be the same thing with the agents, right? Because the agents are essentially copy and paste of that human and of that workflow that they're training the agent on. The agent isn't going to go out and learn more on its own or get creative. It's literally just going to take the function one-to-one and probably do it just as good with a lot more output, right? And that it can continue to learn over time. But what you can start to do is take another video editor that has a different workflow and once you put this into a scalable standard, right? So that agent is now on a coded platform rather than local. Local meaning that it's just on their individual cloud. Instead, it's on a a actual platform that is more scalable. They can now add on top of the existing agent that this first editor coded, change the workflow a little bit once they start to use it. And now that shared changes are going to go back into that agent so that the agent continues to get better, learn more, and gets more improved from other people in the company, not just that first editor that made the agent. So now the skills can start to be combined and it can continue to learn. So the other really important part that you have to understand about this and why humans are really really valuable with all this is because again it goes back to what I just said. Everything starts from the people and their workflows because what we're automating is the workflows and the context. So we don't have AGI yet. We're not at that stage yet where the intelligence can figure everything out on itself, but we do have the ability now to literally take things that are working and copy and paste those one-to-one. So because of that, what we need to still do is create net new workflows. So new workflows are really what will now be things that we can automate. So we need the humans to work on the new workflows, prove those out so that they actually work in, you know, whatever you're trying to automate and they're proven to work with data behind it that it's working well. And once we have this, then we turn those new workflows once they're proven and now that can become an agent, right? Because it's a proven workflow and it needs to be a proven workflow for it to actually work. So what the humans can start to do is work a lot on testing new stuff that can be better, more improved. And once it becomes more improved, it all gets collectively added into your company's brain or database or whatever you want to do, right? This is what the direction we're heading. Again, you can do this in different ways. I'm just explaining how we see it over here is that you want to have a collected place where all of those shared learnings and knowledge can compound over just from one individual local agent. story short, what the good news over here is, right? Because I think this can sound very daunting, right? It's like you're you're thinking about, okay, everything can pretty much be automated like almost today and for sure probably in the next like 12 months, right? Like if you think about it logically, you see where the direction of things are are going, right? But I think the good thing is at least right now for for the near term is that it it's not what people are saying where it's going to replace all of the jobs. Now, if you're a bad employee and you don't work hard and you can't innovate and create anything new, then yes, this is going to be a bit more dangerous. But if you're someone that is adapting with AI, you understand how to use AI and you are making newer and better workflows, we're in the best time ever for that because now that good person can be amplified 10 times, 100 times their output of what they're able to do before because of this agentic capabilities that are now being built into AI. So good people have now become more valuable than they ever have been before. So it's not about firing people, it's about you we probably don't need to scale our team as much as these things start to become more and more automated. You don't need to add more people, but good people you're going to want to continue to dump into and have them just adapt and merge with this technology rather than being scared of it. I'm looking over this and it it's super confusing with all of these abbreviations. So hopefully you guys are following with me so far. So let's talk about the industry as a whole. Like hopefully you guys have kind of kept up. Pretty much everything is becoming agentic. So your workflows that you're actively doing, you should start to think through what are things that are more straightforward step-by-step tasks that are already proven in my business. How can I now take that and put it into an agentic workflow and then how can we do this with more and more people, the best people that have the best workflows to do it and start to change the mindset of people once you get these agents working and these automations in place to be from people doing those same things over and over again in these repetitive tasks that AI can easily do for them to start innovating more and starting to think about what are the new things that they can add into the business that can compound things much more. Now, the industry as a whole is what I want to talk about next, which I actually think e-commerce is going to be one of the least disrupted industries, which is a pretty good thing as well, right? Because if we think about this, let's compare e-commerce to a SaaS, right? A SaaS, that is going to be way more disrupted because what is a SaaS? A SaaS is a digital product. And with AI and with automations and all these things that we're saying, people now have the ability to make their own digital products easier than it ever was before. So the consumption of what you're actually delivering as the product, most people now can just do themselves, right? Whereas e-commerce, fundamental of the business model is that we are taking a supplement, a physical product, actual real world products and delivering it to someone's doorstep for them to be able to actually consume or use. So because of the nature of our business model, the fundamental parts of the business model is that we are having humans consume the products. So and you can't automate a physical product or a sub, right? At least at least not yet or probably for for the near future, right? So who knows where it will go eventually, but at least for now. So with that kind of being considered, right? Like the things around the business model that make the model scalable, so like your marketing stuff, things pertaining to the back end, things pertaining to the operations, anything that you're doing on your computer to make the model work, which is a lot of the stuff that people have to scale, those workflows can definitely be automated like what we just spoke about, right? Like every part of the business will eventually probably be automated to some extent. But the actual fundamental of the business model is not going to be automated. So around the business model, what are the modes that we still have right now? So the the main things are going to be your brand, right? So how recognizable your brand is, what people associate with your brand, how do they feel when they see your brand. All of those things are going to be still super important. Let's take it to the natural extreme where creatives are fully automated, media buying's fully automated, everyone has access to, you know, pretty much full automations for those types of things that are at a very good level of scale. Well, really the modes that you have, like going back to what we just said on that is first off, all the stuff we spoke about over here, right? Which is what are the workflows that are being automated? And what is the context that you're giving for those automation? So, yes, we'll probably get to a place where everyone has, you know, very good levels of automations, but then you still need a way because it's ultimately people that are buying the products on the other side of the screen. So, if everyone is doing the same things, there are going to be outliers in different ways of approaching it. And that is where creating these net new workflows and new ways of approaching using ads as the example, right? Like figuring out newer ways and and better innovations that then you can automate, those are the things that will put you on the cutting edge of things just from like a workflow standpoint. But outside of the workflow, it's then also going to be the brand recognition, understanding one brand versus another person that's doing something very similar trying to copy. So, the brand is going to be more important than ever, capital overall, right? Because really the person that can afford to spend the most on customer acquisition is the people that will continue to be able to scale. So, having capital in place and saving your capital and being ready to reinvest your capital, that is going to be a big advantage that you still need physical capital to be able to do that. Good product, right? At the end of the day, like we said, the fundamental of the business model is that people are going to be consuming products. So, depending on how good your product actually is, this will be the difference of, you know, having the LTV, having the numbers that can compound over the long term versus not being able to have it. And the big difference that where this shows up in the data is going to be like I said, the LTV, right? Because a good product versus a bad product is going to be very different. And also in this era of e-commerce where we're going into, of this AI era, I think it's going to be much easier for the consumers to know what products are good versus what products are bad. The research that they have access to just by doing a quick search on Claude or Chat GPT, they can scrape the internet on all their reviews, understand every thing that customers have said about your product versus not. And if your product is not good, it's going to be harder for you to actually sell it over the long term. Good customer experience, right? So, like the things surrounding the product that they're getting, right? So, the box, the packaging, the speed of delivery, all those things that come into actually making the customer have a good experience in the physical world, not in the digital world, those are things that are to still be a motive getting this to be better and better and ultimately giving customers a better experience, right? With so much optionality and so much information, this becomes way way more important. Back end in logistics, at least for now, right? Where this is going to go also, I mean, once we start adding robots into the fulfillment centers and a lot more robotics just across the whole logistics process, which will probably happen in the next three to five years. It's already starting to happen, but it's probably going to become much more mainstream and scalable with these robots in the physical world. But at least for now, there's a big advantage for us. You know, we have our warehouse fully in house. We are doing a lot of our product development in house at this point. So all of those things give you a very big advantage when you can have the closer to that rather than just using the general 3PLs that are out there and things like that. Obviously, there's costs that are associated with that, but having a grip on that, that's where you can get advantage to get more profits ultimately, which allows you to scale further. And then lastly, retail, right? Because retail, although obviously e-commerce is increasing every single year of people buying their products online, still majority of transactions in, you know, commerce is still happening in retail, in stores, right? So people that are going into stores, they're they can't send at least right now an AI agent into a store to buy their products for them. They are going to be the ones still making the choices of what they're buying. So having a presence in retail and being able to do well in retail, I think will become pretty important. Again, like also, I mean, if we look at the trends of this, it's probably going to be retail becoming lower and lower of where people are doing their shopping with more agents that can do the shopping on Amazon, digitally, things like that. But still as far as we are right now, retail is still where majority of shopping is still happening as of today. So having a presence in the physical world in retail is becoming very important for this next era of e-commerce. So I think keeping all these modes in mind, and again, this is not something that happens overnight. We're going a little bit longer into the future, but these are things that I think will become increasingly more important to think through and optimize to the full extent because everything else is going to get much more leveled out on the playing field, which to be honest, already a lot of it is, right? It's not like the things that you're turning into agents you can't, most people can already do, right? Cuz like everything people can do with an agent, you could probably do with AI just more manual. So agents are essentially just taking those things that are more manual and turning it into something that's more automated and thus more scalable because of the automation. So there's not that much difference, but I think the difference is still going to be in the workflows and the context, right? So like that's why you have to have really good people that can innovate and that's why I'm actually very bullish on I don't think AI is replacing the best people because I think that until we get to the place where there's actual AGI, which the world is going to be a very different place and we don't know when that will be, we're pretty much just automating what humans can already do, right? And it it just it gives you more scalability and so what humans can already do, you need the best people to innovate and create those workflows because right now it's very hard, almost impossible for AI to invent stuff and to create stuff that is actually net new into the market. I'm thinking about this a lot through like ads and creatives and things like that. So that's where you really need the humans that can do those things right now to create those things and that's where the leverage is when everything is evened out and everyone has the same tools and capabilities and automations, that's what allows you to decipher between it from a workflow standpoint. But then outside of the workflows, it's the people that have all of these things super dialed in in this era that I think is going to be able to make the biggest push over the next five years if we start to go a little bit longer term. So hopefully this video gives you some context at least in my head where I see things going and what we're actually using today. So again, AI agents is something we're using for a lot of different functions in our business already. I see this only continuously growing and really hyper accelerating over the coming months and we're investing a lot into these types of things on on the back end side, right? Like this is really a big focus across the board for us. So this is just how I'm thinking about it right now. This is where actively using these different parts today and you know, this is where I see things going at least in my point of view. So hopefully this video was interesting and valuable to you. Let me know what you guys want to see next. So comment below some different content you want to see. I appreciate you guys tuning in as always, and thank you guys so much for watching.
Assuntos: E-commerce, Plataformas e marketplaces, Logística e pagamentos, IA e tendências. Publicado no YouTube em 2026-11-10.
