#554 - 10 AI Prompts That Save Amazon Sellers Hours
Podcast

#554 - 10 AI Prompts That Save Amazon Sellers Hours

Summary

"Helium 10's new AI agent uncovers keyword gaps and competitor strategies without extra AI subscriptions. Use advanced prompts to find lost keyword rankings, analyze Brand Analytics, and fully optimize listings, including titles and descriptions. Transform your data into actionable strategies directly within Helium 10."

Transcript

How to use advanced AI prompts without even having to purchase Claude ChatgBT or any paid AI. Hello everybody and welcome to the AM podcast. My name is Bradley Sutton and I'll be your host and this is the show where we discuss all things Amazon Tik Tok Shop and Walmart private label and how to generate recurring revenue streams 24 hours a day during the AM and the PM. Hence the name of the show. Get it? am podcast. And as a matter of fact, this week I visited our teammates in Shanghai, Shenzhen, and Manila, Philippines. And even though I was traveling on airplanes most of the time, I was still making money. How cool is that? Pretty cool, I think. Hello everybody and welcome to our webinar/workshop. today is going to be great for anybody who's advanced, uh, anybody who's a beginner for AI. Um, obviously anybody who uses uh, Helium 10, but these prompts that I'm going to show you guys, you can use if you don't have Helium 10, maybe you have another tool or if you are using Claude or or Chat GBT, but uh, it's especially for those who haven't gotten Claude and Chat GP, you want to get their feet wet in AI. But again, all of these prompts you can use with the Helium 10 MCP as well if you're advanced and they're pretty cool. So, I'm going to be giving you guys some cool prompts and I'm gonna be inviting Andrew Bell on a little later and he's going to give some of his cool prompts. All right. Now, what are we talking about when we say the Helium 10 agent? It's an agent that we have in Helium 10. It's called Helium. If you go to the the screen, it's this uh kind of like little square sphere right here. This colorful sphere. We call this little guy helium. All right. It's just as if you're running Claude or Chat GBT inside of Helium 10 without having to pay any extra subscription. Pretty cool, right? So to activate it and this is for those who have Diamond, um you just hit it and then you can just start talking to it right here. All right, you would start prompting. So are you guys ready to copy some of these prompts that I did and to see the results? Let's hop into it. All right, here's the first prompt. You guys can go ahead and uh screenshot this and and copy and paste it in your own Helium 10 agent and you got to put your own ACE in there. Um or you can do this in the Helium 10 MCP and Claude. But here it goes. It says analyze the BSR chart of these competitors as and I put three competitor as and find for each of them the top selling month in the last 12 months and the worst selling month. Then use historical cerebro functionality. This is only available in Helium 10. To see which keywords were they in the top 10 organic spots during the high month but below rank 30 for the lower month for the same word, prepare a report of the top keywords where for that product losing or gaining rank was probably the reason for an increase or decrease in sales. For those keywords, run Cerebro on my own product and tell me where I currently rank for them organically. So, take a look at the results here in the hemp roll on category. It showed their best month and their worst month. Like, here's this medical miracle product. Best month, September 2026. Uh, worst month, August 2026, I should say. All right. So, here are the results. It says, here's the keywords where each competitor had a top 10 organic rank during their peak month, but fell below rank 30 or disappeared entirely during their worst month. And some of these have decent search volume like hemp oil for pain relief. All right, here's another one. Hemp roll on pain relief. Now, how is this beneficial? This is good because these are telling me the keywords that probably resulted in competitors losing sales. So, what does that mean for me? I don't want to lose my rank on these keywords. Here it gives me some key findings here, the Helium agent, and it says your organic rank check. To complete the report, please share your AS. All right. So, I'm going to have to put my as in here. Let me go ahead and do that. Copy and paste it. All right, let's paste my ASM and press send. And let's go ahead and see what the results are. And take a look at the results here. It says your position 13 dominant, seven opportunity, 14 gap. So the ones that I'm dominating, there's 13 keywords here. Okay, so I'm doing pretty good. But then here's the opportunity keywords. I'm visible but not dominant. Look at this. I'm in the like the middle of page one, maybe towards the bottom of of page one. And here's ones that I'm completely not even ranking for. All right. U so it gives me a list of those uh keywords. So it says I'm pretty strong and I would expect that uh this product is a number one seller in the niche. But what about you guys? If you're not the number one seller, I bet you'd have a a lot bigger list. All right. So again, make sure to use this prompt up here. And there's your first prompt that you can use with the AI agent or MCP. Let's go to the to the next one. All right. Here is the next prompt, guys. You can go ahead and copy it. It says here, analyze the keywords I have tracked for these two asens. Tell me what keywords were I consistently ranked in the top 10 six months ago, but in the last 30 days, I have been below position 15. All right, so here we go. Here is Penetrax Rollon. 6 months ago, I was number nine. Now I'm 17. Here's one where I was 10. CBD tropical rub. Topical rub. Used to be 10. Now I'm 18. Um, here are some keywords that had zero search volume that I dropped off. And on the other product, it says, "Hey, I'm not even tracking uh that keyword." All right, so that's not good. But again, this product is dominating. This is what you want to see, guys. You don't want to see a list of 300 keywords that you dropped off the top 10. All right? You want it to kind of be hopefully only 10 or less, kind of like mine. But you guys run it right now. literally copy this prompt, put it in your Helium 10 agent or MCP and run it and let me know how many keywords it found. All right, hopefully you guys have been using keyword tracker. You're not going to be able to run this if you weren't using keyword tracker, but this is another very valuable one where I might need to look into these keywords and see if I can get those ranks back up. Right, you guys ready for the next one? Here we go. I'm selling hemp pain creams and hemp rollons. I can produce hemp based products or other creams or topical roll-on products. Use blackbox keywords to find keywords that are related to this niche that have at least 1,000 search volume but title density of less than 10 and where at least five out of the top 10 results on the page have less than 200 reviews. All right, so this is next level product research here. And let's take a look at some of these keywords here. Flores migraine. Let me open up this right here. Look at this. Flores migraine relief rollon. That's probably a a um like a brand name I would I should have put in the prompt. Hey, exclude competitor brand names. But here's another one. Magnesium freeze cream for pain. Zero title density. 5,000 search point. Good grief. Uh B venom pain relief cream. What in the actual heck is B venom? It's got to be a brand name, right? Or is it actually an actual thing? B Venom. I don't even know. All right. Uh CBD rollon total relief magnesium cream for nerve pain. um capacian cream for nerve pain. I don't know what the heck that is, but a thousand people out there are searching for it. So, a thousand people know uh what it is. All right, let's see. Look at this. It is giving us a analysis of the opportunity. Magnesium pain creams are the biggest opening. Um let's see. Nerve and joint pain subniches are wide open. Rollon format has branded keyword spillover. All right, so there we go. There there's it realizes that those were branded keywords. Uh but pretty cool. Like this is just giving me quick product research ideas on how I can expand this like pain cream and rollon cream product line. You guys ready for another prompt? Here we go. Input my top two selling products into blackbox ABA stands for brand analytics. Tell me the keywords where from 12 months ago to 6 months ago I was one of the top three click products for at least 25% of the weeks. However, I have fallen off in the last 6 months. All right, let's take a look here. And the answer is none. That's great. I've actually, it says here, look at this. I have actually expanded the ABA top three presence. Again, this is what you guys want to see, but I bet you not all of you are seeing this. This product is just like going bonkers. We just had a a $1,500 sale day uh last week. It's going crazy. It's really good. Um let's take a look. So, it says here, there are no keywords to rescue from a visibility standpoint. So, on this this product, I am just crushing it. But I guarantee if you guys have newer products and you're not the number one in your niche, it's going to have examples. So, here's the other homework, guys. Copy this one. Copy and paste it. Guys, having fun yet? Let's look at the next one. Run my competitor as and I threw in an AS here in Cerebro and tell me the keywords that they have sponsored brand video ads running on. then analyze my sponsored brand video campaigns and tell me which of the keywords my competitor is targeting that I'm not. This should be self-explanatory how valuable this could be, right? Let's take a look. Uh, Cerebro found 22 keywords where this competitor as running sponsored brand video ads, but I'm only advertising my videos on six of those keywords. All right, so the fact that this guy is only having 22 keywords, what does that tell me? That tells me I probably have his exact manual strategy. He's probably not running like a broad campaign or a phrase campaign, like if he only has 22 keywords in one whole month. So, I'm literally uncovering with Helium 10 his exact video strategy and I can see, hey, what are the the the keywords here that he's advertising video for, but I'm not. All right. Hemp oil for pain relief, hemp roll, hemp and pain relief. There's a few keywords here. So, I've got some things that are uh actionable right here. And if I wanted to take it a step further, I could literally just say add these keywords uh to my video campaign now. And and uh you know, MCP would be able to do that. Uh Helium 10 agent is about to be able to do that as well. So guys, there are five prompts that you anybody can use whether you use the Helium 10 agent or you have the Helium 10 MCP. Pretty cool. All right guys, now let me pass the baton to Andrew who is going to, you know, Andrew is one of the most known people in the world uh for like LinkedIn and in the Amazon world for AI and he has been heavily using the Helium 10 MCP as well as the Helium 10 agent and he's got his uh five prompts that he wants to show you guys the results of. >> Thanks Bradley. So we're going to go off to do some prompts and Helium is a fantastic agent. I've been messing with it for quite a while now and am very impressed with what it can do and you know I think just like with anything um how you prompt makes a big difference. So there's lots of different things you can do as far as like optimize your listing, build a keyword pack for your product. U Bradley kind of probably already you know showed you that sort of thing. So one thing that happens a lot on you know when you you go to any system whether it be claude or uh codeex j they'll typically ask precursor questions to get more context um to make sure that your prompt is good as it can be right so it knows what to use same thing here with helium um I did when building a uh comprehensive keyword bank for wall decor um so I'm asking it to use every appropriate helium 10 source available including brand analytics top search term magnet cerebral keyword database historical keywords, top performing organic and sponsored keywords, opportunity keywords um and so on and then identify the leading wall to core asens too and use them for additional reverse as research. Continue through the available results page rather than stopping at first batch and then merge dduplicate and normalize all terms preserving search volume, ranking, conversion, etc. And then cluster the final bank by product type, material, style, room size, color, audience, use case and shopper mission. Separate broad category terms from specific subcategory and longtail terms and remove clearly irrelevant searches and rank the final keyword bank by demand relevance competitive importance and conversion evidence. Um, and so here I'm going to ask it not to approve intermediate steps because usually it would do that and to complete the research here uh return the strongest possible wall to core keyword here. So one thing you know it's a good response. We're used to like immediiacy with chatbt. uh claude. I think if you're a user, you've probably um come used to like waiting much longer, but with chatbt, I know it's been much faster before, but now as we all begin to use um clawed uh code and codecs, if you've heard of those before, um you'll see, you know, that you know to get a better answer, right, you have to wait for it to reason through everything. So, I'm going to let this here uh run in the background as I take you through an example that I had already done. Um, so here's that the core keyword bank I was talking about. Um, so 8,000 raw keyword rows here. Uh, gives you the seed and all of that. And then here's the bank overview of the total amount and the count. And then the ones with greater equal to 500 search volume and and so on and so forth. And then you have the type of keyword distribution. So there's broad keywords, subcategory, and more longtail. Uh, ones that are, you know, five plus words are the long tail. And then it shows the total search volume across all of that of course. And then when you go down here you have the top 50 keywords by composite score. Um so you have 40% demand, 20% source bread, 15% weekly sales, 15% CPS plus 10% trend. Um so now you have wall decor search volume here sales a week CPS score. Uh you have the 30-day trend here and then it shows you the exact sources actually um that it come from as well. um the keywords were found and brought to um so I have a lot of listenings actually that have keywords you can tell here that it must have you know brought from listening builder as well and then if you ever want to check the work of it too go to the thinking part and then you can see where it started it completed um and then yeah now you have your raw your generated answer so here's the product type clusters the top 10 per cluster I was talking about for general wall decor um I tend to focus obviously on like search volume and sales a PE I'm a big guy on keyword sales per um search term that's telling you a lot more than just merely search volume and then shelves um you know you have this number of keywords so like loading shelves shelves for wall then wall panels is a different cluster sconces and lighting is another cluster box plants greenery is another cluster um and then you'll know your own IC you know ICP whatever that might be uh bring those together and then you have style distribution here room distribution color distribution and then key findings too that will go along with that. And then I ran it again. It's like, oh, this is really cool. Is it going to do it again? And then it did this the same thing. Um, and it can do that specifically for for any keyword bank. Um, so you'll have the prompt available to you. All you have to do is take the prompt and plug in whatever search term you want and then you can use this prompt specifically uh if you haven't already screenshotted and then use AI OCR to grab the text and move it over to your own computer. Uh we can send the prompt too. Uh something that's I think has been very useful so far for me. Uh and then also too for for clients as well when you're building really big keyword you know universes. All right. Now we're going to look at another prompt. I asked it to fully optimize an Amazon as here uh for the US marketplace. And the reason I have all this text in there is because previously, you know, it would ask follow-up questions, which it rightly should. But just to show you like the full capability of what's happening, I asked it to fully optimize this AS for the US marketplace. First, identify and select the top five most relevant competitor as based on product similarity, shopper substitution ability, substitutionability, I don't even know if that's a word. uh sales, reviews, and keyword overlap. Uh optimize the complete listing including item name, highlights, bullet points, description. Um analyze all available catalog images and use all of them, all the images in there. I have no additional product notes or corrections. Uh use the number one selected competitor as the brand as reference and use the top two relevant competitors for the Alexa for shopping Q&A. Use all available and appropriate keyword sources to build the keyword bank using my AS and the five selected competitors. Determine the strongest keyword allocation across the title. um that is the item name, highlights, bullets, and description and make reasonable decisions at every stage and complete the fully optimized listing uh without asking me to approve or confirm intermediate steps. So, what Helium will do very consistently usually is um is ask you questions like to approve or confirm like, hey, do you want this to happen? And you can actually guide it to say what you want it to ask and what you don't want it to ask by the prompt that you give it. So, like I kind of got tired of asking like, oh, is it in the US? Like, well, that's just, you know, another question. And I totally understand that, right? But it's really cool to be able to work with all your data right here within um Helium. And so, uh you can definitely tell it's also an agentic. It's not just merely a chatbot. It's taken from a variety of data sources across the whole site, aggregating them, putting them together into unique insights. That's a very agentic capability. You know, it it begins it uses skills. So, here is fully optimized listing. You have the competitor selection here for the Stanley Quenter Tumblr. And here's the product why it was give you the reason why it was selected. And then it'll tell you which ones it chose for the Alexa for shopping Q&A. And then it'll give you your keyword bank and allocation. So 300 keywords sourced from six sources. ABA converting keywords, top keywords, opportunity keywords, historical top keywords, top performing organic keywords, and top performing sponsored keywords. All 300 phrases are were passed to this generator and 21 allocated keywords were placed in specific slots. So it basically based on a keyword performance score took the best 21 allocated keywords to be able to put those keywords in plan to put those keywords in for you in your bullets, right? And this is a part where you can ask it to edit. So if you need to edit like any of the keywords you want in there, but the point of this is to be able to write naturally while having as much keyword rich language because Amazon very much so operates on that still to this day as well as for AI visibility. So yeah, make sure you know you edit, you look through, but here's your item name, item highlights, bullet points, your description, and then it has uh missing keyword check. All were pure, their assigned slots, and then yeah, so this is another one. This is probably one of my favorites. I've been continuing to test this. There's a lot of gold to send. You got to keep in mind, too, we're the only ones on the market um as well that has helium. Helium is a very powerful powerful agent. And what's really cool is like you can This is the mode that you have here. And then you have switch to classic mode. I remember first you know conce you know conceiving of this really is like how can we you know distinguish ourselves from the rest of the market and just give something that you know directly works within and you know this is something that's Helium 10 has been on the forefront of um from the very beginning. It's really exciting to be a part of that and to contribute prompting as well um and to provide prompts too for you guys. But yeah, reach out to me if you're if you have any questions about Helium on LinkedIn if you'd like to and we can talk more about everything there. So, all right everyone, look, while we were waiting, right, just like five or 10 minutes later, you know, this is like a lunch break or something or if you're doing another task, you can just have it run in the background while you're doing a new chat, right? And just and just keep going this way. And this is where you can you can really chat with your data, right? So you can ask it like okay can you break this down by intent type now it's starting analysis the run started and then we can come back to it when we want to then maybe let's ask a question in here and ask for these keywords on this one let's ask can you sort these by branded non-branded this thing too is like asking follow-up questions to all this can be extremely powerful and can yield useful insights It's for, you know, your business now, um, but then also for the things that go on later. So, I got two going at once here. Okay, what we got? Sweet. Awesome. Check this out. Look at this. All the cord by in keyword bank intent breakdown. So, here's all the keywords, 10 shopper intents, ranked by total search volume. Wow. So, now you have, you know, general category here. And you can always open this up too to see uh more specifically. But you have general category browse, room targeted browse, specific product search, attribute specific, styled driven browsing, what's a solution problem solving. It says for solution problem solving like keywords. There's a total of 246,000 total search volume for that. Wow, this is really cool. I already know there's people thinking yes, it'd be a great idea. We'll probably have a a skill like a Cosmo skill if you will that breaks up this into keywords for Cosmo, right? For the 15 things people talk about. out. I think that's, you know, a good pretty good strategy, too. Brand navigational. So, once you get the bank here, the keyword bank here, you'll be able to, you know, ask it a lot more questions of what you want it to do. So, you know, if I were to say, oh, can you write down into more sub intent types? Okay. Now, you can break it down by the Cosmo intent types. Okay. So once that's done, I'm going to go into here. Looks like this one finished. And let's see what I asked it again. Said, can you sort by branded, non-branded? Yep. And it'll sort by branded, non-branded. Now I want you to can you organize by Cosmo 10 as defined here. I think what I love is just keeping the conversations going. But if you're not familiar with Cosmo, there's a lot of, you know, really neat things you can play around with it. Like for example, asking the questions of like, okay, what is it used for? or what event is it sufficient for? Who is the audience? Um what is it capable of? Uh you know, really getting down to how it's used, uh concepts, um you know, what time or season is this used? Um what location, possible location? Uh you know, is it used on your your body parts somehow? Like where use a cream of some sort, things like that. Okay, now see what we got here. Okay, so here it broke it down. um 13 intent categories broken into 93 sub intents specific to your product. So of all these you have the general brows room targeted. So let's see specifically for room targeted. Okay, this is great. Looks the search volume of the keywords that they were brought up your bathroom. They're using the bedroom, office, workplate, living room, kitchen, dining, dorm, college, laundry, staircase, um hallway, entryway, commercial, professional. Hey, that's something really I haven't thought too much about before is having wall decor. I'm always thinking home. I'm always thinking office, worksp space. I'm never really thinking like, oh, do you want to have it in a gym? You know, but that that makes sense. And then, of course, coffee shops and such. So, you can see how it sparks ideas like this. Um, outdoor patio. Yes, I love outdoor middle wall art and cave gamer, of course. Nursery, kids, teen. Uh, yes. And then different materials, too. Wood. Yeah, wood's a great one. Um, the color black. Glass and mirror. Yes, people love wall decor with like glass edges or a mix and match of both mirror and wall decorum in it as well. Love gold mirrors for example. Um exact dimensions. Okay, large um canvas here. And then it shows here the keywords next to it as well. So this is important. So yeah, this is this is very cool. Okay, so this is the chat where I asked it to break down. Yep. By the Cosmo. Okay, so Cosmo intent classification wall decor keyword bank. But most people don't know is like they talk about the 15 questions but there's actually 11 Cosmo relation types and and this is just one knowledge graph among so many at Amazon. So Amazon search A9 algorithm right does not uses Cosmo right but Cosmo does not replace A9. A9 is still the algorithm for for Amazon search. Okay cool. So summary by relation type used as functional role number of keyword search volume for each of these. Wow. And then it breaks it down why that's the case. So, here's the product types and here's the search terms that goes with that. Here's the used in location. So, very similar to what we looked at a little bit before. Used as, interested in, used for, used for here, used with, shelves, bed, fireplace, TV, couch. Yep. Desired activity. Decorate a room. Create gallery wall. Yep. You can see that here. Wall storage. Very cool. Anyways, yeah, if you're um only request I have million 10 is to let me build skills now, right? To put those in. But thank you so much and I hope you enjoyed these. We will get these prompts over to you as well. Um and also been building a uh Helium 10 prompt generator too. U been really excited to to get that going. So that I'm hoping to release that at at some point soon to reveal to you guys. Anyways, thank you. Appreciate it. >> All right. Thank you very much, Andrew, for that. So, what do you guys think? Is this doable for anybody, regardless of what level of AI you're at? I hope you can agree with me there. Whether you're advanced, whether you're brand new in AI, whether you already use Claude, whether you don't use Claude at all or chatbt, you can use these 10 prompts right now in the Helium 10 agent. Uh, let us know guys what you think of these of these prompts. Which one was your favorite? Which which one of mine was your favorite? Which one of Andrews was your favorite? Is there something that you guys uh want to see that we don't have yet for this agent? Like like did you run into any obstacles? Don't forget, go into your Helium 10 dashboard and at the very top there is this send feedback button. It's like a a little quotation box, right? Hit that button and then hit give feature ideas. If there's anything that our MCP or agent can't do yet that you want it to do, let us know. We look at this every week. Uh, all right guys. I hope you enjoyed this episode. Uh, comment below what you thought, which one was your favorite prompts, which ones you used, and some of those numbers maybe that came out. And we'll see you in the next episode.

This transcript page is part of the Billion Dollar Sellers Content Hub. Explore more content →

Stay Updated

Subscribe to our newsletter to receive updates on new insights and Amazon selling strategies.