When it comes to e-commerce, staying ahead means understanding the impact of artificial intelligence on consumer behavior and product discovery. In this episode of The Seller’s Edge podcast, we explore these crucial topics with Vanessa Hung, a leading Amazon expert renowned for her deep understanding of backend optimization, AI tools, and e-commerce strategies. As the founder of Online Seller Solutions, Vanessa has a wealth of experience in helping brands negotiate the complexities of the Amazon marketplace. In this episode, Jonathan and Vanessa discuss Amazon’s newest AI tool, Rufus, and how it will impact product discovery, the evolution of search technology, and what it means for your business. Read the full episode transcript below.
Episode 30 of The Seller’s Edge – Vanessa and Jonathan talk about:
- [00:00] Opening and Introduction
- [00:57] What Impact Will AI Have on Sellers?
- [06:21] RufusS’s Consideration of New Sellers
- [12:21] AI Shopping Assistants
- [17:16] Importance of Backend Optimizations
- [24:58] Attributes that Rufus is Looking At
- [29:40] Sellers Need to Zoom In on Demographics
- [33:15] How to Optimize for Rufus
- [39:43] Having Answers for Customers is Essential
- [44:12] Prioritize Creating a Comprehensive Strategy
- [46:56] Recap and Closing Remarks
Key Takeaways:
- FORGET ABOUT KEYWORDS: Focus on audience-centric strategies rather than solely keyword-based approaches.
- OPTIMIZE BACKEND ATTRIBUTES: Details in Seller Central are essential to making your product stand out to Rufus. Utilize backend optimizations and flat files to improve discoverability in AI-driven searches.
- PUSH EXTERNAL TRAFFIC: Create content that resonates with specific audiences to drive traffic from social media platforms and explore partnerships with Amazon influencers to increase product visibility.
- HAVE A STRATEGIC ROADMAP: Consider your approach holistically and prepare for a more personalized shopping experience by diversifying product offerings to cater to different customer profiles.
Full Transcript of Episode:
JONATHAN: So Amazon wants to make it as efficient as possible for customers to find the product they’re looking for. However, I feel like there’s a critical mass where that becomes detrimental to Amazon because they’re going to start losing advertising dollars at some point if they, if they make it too good. So it’s like, how can we get customers closer to these products without actually hand feeding them the exact product? And then the other piece of it is it’s going to become harder for sellers, I imagine, who are just starting out because they’re not going to have you know, the sort of background context and like reviews that other products will have that Rufus would feed to customers. I’m, I’m just curious like, have you thought about like those things or heard anything about those kind of two sides of it of like Amazon trying to make it so efficient but not too efficient and also like, do they leave room for fresh new sellers?
VANESSA HUNG: Love this question. Because I, I think, I think there are three parts that I, I want to touch on. The first one being the fact that Rufus and that whole AI powered search engines and things like that are reshaping consumer behaviors in a way that we’re going to see this video maybe three years from now and it’s going to be like, like it’s so funny that we were talking about these little things that were, are going to be the norm in the future. So like, and I like to look at history to know maybe where we’re going. But if you remember back in or when the Internet started, like there were not search engines so you would go to, you know, a specific URL or website but nothing directed to that. And then Yahoo came and it was like a directory of a bunch of different links to jump like and to go from one place to another one. Then Google came with a search engine of put what you are looking for and then we will redirect you to this. So this new iteration of the same evolution, because we need to look at it in a perspective of evolution of the technology is now you don’t need to do the brain work of I guess pinpointing the things that you’re looking for in the sense of product or discovery of anything. Not necessarily it needs to be a product, but discovery of a solution will start with the question or the pain point that you’re having rather than the digested keyword or input that you need to give a search engine to get the answer. As an example, like I don’t know, my oven broke, right? And in the past I will go to Find, you know, on Google, for example, like oven repair near me, you know, that will be a search. But I already know that that happened, right? With AI, it would be more like a conversation. And let’s give the example from Rufus. It’s like, oh, you know, my oven stopped working. I don’t know, it’s not heating up. And Rufus will say things like, oh, did you check that it’s connected to power? Did you clean it? Did you do all of these things? If that’s true, then you might be needing a replacement for this thing. Like, I don’t know, the heater part of the oven or you need to hire somebody to clean it or something like that. That is the difference. That is where the magic of AI, it’s coming or it’s getting into our lives to solve that problem of discovery. It’s a conversation. That’s how you’re going to find things, by using these tools in a natural language that you don’t necessarily need to obsess with or know exactly what you need to get the answer that you need, if that makes sense. Which is again, feels like magic because it’s like, oh, you know, AI knows what I need more than what I, you know, more that I know that in that sense it will be, it is the future, it is the reality right now, and this will only get better. So I see that there is a potential opportunity for a disruption like that happening to the world of search engines. So the same way Google disrupted the Internet back in the days, AI will disrupt the way we interact with the Internet going like translating that same vision to Amazon and marketplaces. Now they have the opportunity to give solutions to questions or needs or help the person do the pro discovery. And instead of, you know, that customer looking through, you know, a catalog of hundreds and hundreds of different options, the best option will be available to them immediately and it will be ranked based on their intent and the information that there is in the listing. So in one sense, like those are two other things that you discuss in the question was first one, will this be worse for newer sellers? Like, will the competition be very fierce? Because the people that have been in the platform for a lot of time might have a prime position towards the people that are starting. I want to think that that’s not the case, but I also know that unfortunately in certain searches or inputs that you’re as a customer putting into Rufus is a zero sum game. Because Rufus right now is giving you three or four options to choose from when you have an issue. So let’s say that you want to buy a new coffee mug, that based on your query, based on what you like. Right. It will give you five options of the things that are more likely or that you are more likely to purchase based on your intent and based on the catalog information. So what is scary to me is that we still don’t know how to play this game of zero sum. When there is literally when Rufus speaks, the five winners or even less like three winners for this search. That’s it. So there are advertising options within the, you know, AI, Rufus and all of that. We will get to a point that that will be a norm too. The same way right now, you know, at, you know, ad positions are everywhere in the search. We will have that. We still don’t have it. And it’s not super clear how they’re going to do it, but trust me, Amazon knows how to monetize advertising, so they will. And that is the part where it’s like, yeah, maybe this will be harder on new sellers. But my hope is, and the reason why I think this is a blue ocean of opportunities is because within the world of AI, the knowledge that the bot has about the customer, it is very important. And I don’t know if you use in your day to day Jonathan, like any AI tool, I use ChatGPT and within ChatGPT you will see if you go to a settings section, you will see what the system has saved about myself and what knows that is true in my world. So ChatGPT knows that I’m an Amazon expert, that I, you know, travel for work, that I make presentations, that I create content. All of these things they like, gather. So the answers that ChatGPT is giving me are different from the answers that it will give another person, maybe somebody in the medical field that is looking for a somewhat similar prompt that is the exact same true in the E commerce space because results will be tailored to audiences and I am an audience. If I’m looking for a coffee mug, I am an audience. And based on what Amazon knows about me, which is a lot like insane amount of information about customers, they know that I like pink, that I like to drink coffee in the morning. That’s not true, but you know, a bunch of things. Therefore they’re going to offer me, you know, this pink bottle, right? Versus you, which you will do the exact same search and you will get something completely different. Why? Because we’re not the same, obviously. So that is to me where blue oceans start appearing. Because I might be the winner if I’m selling this bottle. This Pink bottle of water. If I’m selling this, I might be the winner of the audience for Vanessa, like Vanessa’s audience and maybe another product in the same category in the same niche, competing for the exact same keywords in the past now is winning in the world of Jonathan and in that audience. So that is where to me I’m hopeful that that is true for the newer people that are coming. And how do you get there? Well, you highly focus on audiences rather than keywords and that’s the prompt of that post that I did. Like keywords are dead. Well, yes and no. I mean, obviously that was a catchy hook. So you read the post. But I mean, keywords are not completely dead because at the end of the day that’s how we communicate. You need to use keywords to input that into whatever AI you’re using to get a result. So I mean, it is, it is a whole transition. I love, I don’t know, it’s kind of like I love that I am alive at this moment in time because it is so interesting where we’re going and a lot of what we thought it was true in the past will be challenged and it will change. So that’s where I think we are going. And I think I answered the three parts of the questions.
JONATHAN: Yeah, no, you did. And you brought up a really good point. Because I was thinking about this the other day of like a point before the Internet when someone wanted to buy a product and if there were no stores in the vicinity that sold that product, like they just couldn’t buy that product unless they drove like a state over and like called a bunch of stores on the telephone. So it’s really, I agree with you. It’s a really interesting time to be alive. It’s also having worked in SEO on the Google side and the Amazon side, it almost seems like they’re trying to, you know, round both sides of that square or square both sound both sides of that circle or the one side of the circle of being informational and solving problems while also selling products, which is great. But it also makes me wonder is like Amazon going to at some point try to become its own search engine and like take over the Google space like it tried to do with streaming with Netflix.
VANESSA HUNG: So that was in my bingo card for the predictions for 2025. It was the fact that Amazon will eventually come out with some sort of partnership. Well, they will say, by the way, when you look for products on chat GPT, we are integrated there and the results will show up similar to what perplexity did with their shopping assistant. So that I think will. I mean, there are probably already conversations about this somewhere in the world. I don’t think that’s not neither far or very, very crazy to think about. But I don’t believe that Amazon will do it natively. Like they need a partner and you know, especially with the whole FTC lawsuit and you know, monopoly complaint, blah, blah, blah, if they’re. Well, I don’t know. Sometimes logic doesn’t work in the Amazon world. So I mean, my gut feeling is like if you’re smart, you will try to do a partnership, something that doesn’t look like world domination, but they’re very ambitious and hungry. So I guess world domination sounds like a better option. Like, I don’t really know.
JONATHAN: I thought they were already there when they, when they purchased and merged with Whole Foods because I was like, oh well, this is the end.
VANESSA HUNG: And now it’s just the beginning.
JONATHAN: Yeah, it really is. Yeah, it is. And you brought up the shopping assistance and I feel like, I mean between that and AI, the retail experience as we know it is changing. Obviously Rufus is great when you are somebody who, when you use somebody from an older generation, for example, might like not know what they need and they’ll go to Amazon trying to find it and they’ll ask these informational questions and Rufus will troubleshoot for it and eventually get them to an answer. There’s also going to be the traditional shopping experience where people go on Amazon and I know many people who do this who will go on and have no idea what they want buy. They just want to buy something like, you know, they need some new form of entertainment, whatever it is.
VANESSA HUNG: Shopping therapy. That’s what we call it.
JONATHAN: Yeah, exactly. So maybe they should invent the Rufus like shopping therapist too.
VANESSA HUNG: Yeah, well, I think for that demographic, which I will shamelessly admit that I might be part of that demographic, that demographic is sensitive to social media. And that’s true for in my case, I buy a lot of things from ads that I see on Instagram, ads that I see on social. So for, for that I guess world there are solutions and, and that is the solution is driving them through content, which I think will be the priority for sellers or, or for brands that want to target those type of audiences is content, content, content, you know, content about. And it’s so funny because I’ve seen things, it’s, you know, ridiculous. It’s like things that videos or ads that I’m like, this is ridiculous, but I really need this product in my life. Like I really need it. So in that, in that sense, I think Amazon is already solving for that through social. So the creator connector, like, you know, I guess program that allows you, allows brands to connect with specific Amazon influencers so they can drive content to, to that, you know, doing external traffic like bringing them from Google, Pinterest, Facebook, meta, Instagram, you know, all of that. And you know, TikTok, which is kind of like a sensitive topic. I don’t know where the world is going with that. But you know, that’s a massive channel. And the crazy thing is the amount of revenue on eyeballs that they were able to take from those channels. It is probably bigger or it is bigger than the traffic that Amazon gets naturally to the search engine. Right? Like to, oh, let me look for whatever X you’re looking for. So that is where in our next reality or in the AI reality, those things are interconnected. And it gets even crazier when you start talking about creating the content with AI. So there will be a world where, you know, it’s not even a human, it’s not even a person even doing the prompt. It’s an agent directing another agent to do that thing to sell to the agent. So, you know, it is, it is possible. Like you can do it, you can do that.
JONATHAN: Right now I think that it’s definitely, I’ve noticed definitely seeing the younger generation going on TikTok and that being their primary point of sale for certain products versus Amazon, which is, it’s fascinating. I’m very curious to see how that plays out over time. And obviously E commerce sellers are going to hedge their bets and they’re going to try to be on both TikTok shop and Amazon. So good for them. And I’m just like curious, like discussing all these things with like shopping assistance and AI. I’m a seller. What are the things that I could possibly do to sort of start adapting to that? Are there any early indicators of what I should be looking at and what I should be going after? Obviously we talked about keywords and what I’ve learned about you. By the way, I just have to say, because I don’t think I gave you enough praise right up front is like I knew how important the backend was optimization. I didn’t know how important until I started listening to you talk about it. And I was like, oh, this is really important and something people should be paying more attention to. And I imagine that will also play into this at some point.
VANESSA HUNG: Yeah, 100%. I mean these tools, these AI tools are built on the information and the backend is one of the ways that you input information. You obviously have the front end, the Pro Detail page, but also you have the back, you know, the flat files, the attributes, the things that are not visible to a customer. So in that sense it plays a critical role because it is space where you are I guess, allowed to add information. And the information is specific, which that’s what the systems like. Meaning if you’re selling a specific, let’s say dress in the backend attributes you will see things like occasion type of dress, type of cleavage cut or something like that. That you could see a round neck or V neck or this neck or whatever or that pattern or this style, like all of these different things. So those attributes attributes signal the AI to understand the context of the product. So it is part of the equation. It’s a massive part of the equation because it is where you kind of set the foundation for your product. Then on top of that, on the Pro Detail page, on the front end, things you create content to solve solutions or, or kind of tie the product to the specific need for the customer. So they know that that dress, they will use it when they are in a vacation in the Caribbean, right? That’s not addressed for a cocktail party in per month. You know, like that’s the difference. But you put that in the copy, right? Like you say where you, you want to target, but in the back end you lay out the foundation. It’s like I want this dress to be shown to, you know, people are, are looking for a floral style versus somebody that is looking for polka dots. That like, that is very basic. You don’t even need to say that in your copy. But that is in the backend and the system knows that. So you know, one thing, and I’ve been, it’s so funny, I just had that conversation yesterday with a couple of friends that are in the industry as well. It’s like I really like to give examples of things that you can do right now. Like, you know, you should go to this dashboard and click here and do that. But I mean, I enjoy that, but I love to understand and see the bigger picture. Like I’m a thinker. I like to think and when you ask me like if I’m a seller, what should I be doing? Well, the first thing and, and this is really important and people some oftentimes overlook this. When changes this big are happening in the world, you really need to stop a little bit to understand where is that we’re moving doors. Because for many. And I wrote A piece about it in on LinkedIn is everybody’s so hungry all the time for the tactic, it’s like, what is next? What’s the hack, what’s the software, what’s the thing, what’s that? And then if you stretch it for enough period of time, they find themselves in a position where they have no clue how they got there and they have no clue where they’re going. I mean the goal is always growth, right? You always want more revenue. That’s the ultimate goal. Everybody wants that. Everybody wants to be happy and rich, like amazing, like everybody. But how you get there matters. So if you get there just by hacks and tactics and hacks and tactics and hacks, buying tactics, at some point you will get lost. So to me, for any seller that is out there or anybody that’s doing business, stop and try to understand the first principles of AI. And to me those first principles, and I spent a lot of time trying to figure that out, is how do they learn, how do the large language models learn? Where do they take information from? And based on those inputs, now you know, what is that you can influence? So there is a whole conversation, for example, oh, you want to show up for, you know, chatgpt answers, then you need to do this, this, that. It is the exact same game for things like Rufus. If you understand what factors played into its development, then you will be able, no matter what they had, you will be able to influence whatever you can to, to get the answer right. And I think we are sometimes lazy just because we are so busy with the day to day and everything that’s happening. Like, I understand it’s, it’s a lot to run a business on top of like sitting down and think about, you know, the future of the Internet, right? Like it is not easy but really stopping and say like, okay, let’s reset. This is what, this is the game I’m playing right now and it’s different from what we were playing before. This feels like multi dimensional, multimodal game. It’s everything at the same time. So it’s content, it’s backend, it’s ugc, it’s AI, it’s like driving traffic, it’s audiences, our insights. So that’s where I think the overwhelm will go away if you know the basic principles. So I like to think about that. And I don’t know if this necessarily answers the question, but I hope that if anybody listens to this and they stop at okay, this is where we’re going, this is what we’re Doing now, whatever comes next, whatever tactic, hack or whatever they see will make sense or not make sense. Because that’s the other thing. There are people in the industry still talking about how you rank in number one keyword by using the honeymoon period. It’s like, come on, dude, we’re not there anymore. Like, this completely changed. And I know it’s, it happens super fast. So, you know, give, give yourself some grace, but time to really change the mindset.
JONATHAN: Yeah, I think that was a phenomenal answer. And I think not only did you provide the information, but the insights of how do you or that you should be thinking of ways to create sustainable growth over time and how you maintain a cohesive marketing plan and strategy for selling, you know, not only today, but tomorrow and three months from now, is, is definitely great advice that people should be thinking about. And you brought up the honeymoon period, which I think is funny because I was just. Actually I made a similar joke the other day because there are a lot of snake oil salesmen being like this, just do these three things. And that’s never true because every single seller is selling a different product in a product category that has its own rules. And it’s a, it’s a whole other thing. But I’m curious, you brought up the machine learning and large language models. And I mean, to people who don’t really understand or don’t have the technical wherewithal for that, like, how do you think that might differ? Like what Rufus is looking for versus when we were in the world of honeymoon periods and keywords. Like, is there a huge shift? Is it basically looking for the same thing and we’re just reshaping it?
VANESSA HUNG: That’s a great question. Look, I think it is different in the sense of we cannot translate certain, I guess, frameworks and activities to the world of AI. For example, ranking, you know, ranking, organic rank, BSR category. Things are not transferable to the world of Rufus. Why? Because what I said at the beginning of Rufus will give answers based on customer intent, what they know about the catalog and what they ultimately know about the customer. And the customer behavior is extremely important in the, in its learning. Right. So if that’s true, then the other option of, you know, ranking for one keyword, it is not what’s going to happen. So I do right now a lot of Rufus optimization. So we’re offering that as a service. And you know, a lot of people come to us and ask like, oh, well, but how much is going to. Is my rank going to improve? And it’s like, I still don’t know how to explain, but it is not the same game. So it’s. It’s like, if you ask me how. So I say this, and I don’t know if this is a correct analogy, but this is the one that I like the most. If I ask you, Jonathan, how long does it take you to walk to the store? And you are like 15 minutes. And I’m like, cool, but what, what if I ask you, like, how long does it take to fly to the store? It’s like, what, what are you talking about? You don’t fly, right? Like, I don’t know. I don’t know how long it takes to fly there because I don’t fly. So same same thing like with Rufus is like, you do optimization for the answers that it’s giving at that moment for that client, for that audience that you’re reaching. This doesn’t really necessarily. It’s going to impact directly your ranking, even though the more you sell, the higher you rank. That’s one. One part of the formula of ranking is like the conversion. So that is where it feels very weird to me that we’re trying to use the same frameworks for a completely different game. Like, it’s like if we go to Mars, like, if we go to Mars, we cannot talk about, you know, going to the beach, like, but. Because there is no none there. And I don’t know, maybe they are, but I don’t know if they are in Mars. So I’m just trying to reframe it that way. So what are the better questions that you could be asking? Right. So in my perspective, as a seller, you should be asking yourself and really doing the work of understanding who your audience is. And sometimes I talk to sellers and they don’t even know what audience is because when they, they, they answer the question, they’re like, oh, yeah, there is. It is a male between 20 and 55 that likes soccer or football that live in a city and they make above 100k something like that. That is demographics. That is like a surface level information about your audience, but you don’t really know who the audience is. Is that person a Jonathan or a Joe or a Will or who is it? Like, what do they like, what do they taste like? How do they purchase? How do they get to you? So that is the work that in this AI world will be more important than what is your number one top keyword? Which. That won’t go away completely, but it will be secondary in the game. Right. It will be kind of the bonus point not the main. Not the main play.
JONATHAN: Yeah. It’s interesting because I’ve talked to a number of sellers who. Not a number of sellers. I’ve talked to enough sellers that I found alarming. Enough that how many were out there of people who just didn’t really have a strong sense of their audience? They did have the demographics, but they didn’t think about, like, hey, are these divorced dads? Are they divorced moms? Like, do they have a college degree? Like, what do they like to do in their spree? Free time? Do they like to hike? Like, whatever it is. And thinking about it at that level.
VANESSA HUNG: One insight, though, and I love that you brought that up, one insight that to me feels the most important from an audience perspective is what happened in the life of my audience that make them buy my product. And that is where I believe gray marketers know this. Like, phenomenal marketers know this. Why? Because you follow signals. And that is the thing with DSP and all of the information that you get from DSP and the Amazon marketing cloud. You can follow those things because let’s say that you’re selling, you know, kitchen appliances you want, and. And you decided that your demographic are, you know, people in the. Their mid-30s. Let’s say sometimes we could argue that that demographic at that age is getting married. And when they get married, they normally move into a new place together. And when that happens, they need new appliances for the kitchen because they’re not living with their roommates or their parents anymore. So that’s where you start following the signals, creating content to follow the signals to start selling to that audience versus the other option, which is right now, what most people do, create content based on keywords. Right. And. And maybe you’re selling that exact same appliance, but you’re never mentioning, like, this is the best thing for, you know, your new kitchen. Or with this, if you invest in this specific, I don’t know, garlic press, this will last you forever. And for some people, that’s important. It’s like, oh, I move. I moved to a new house with my partner. Now we want to buy things that will last us forever because this is my forever home. So that. That is what a good marketer does. And I feel that sellers have to shift hats all the time. And I think that’s a more important role. Right? Like the strategic thinking for marketing, where we’re going, what’s the audience, what you develop, rather than other things that right now, you know, I don’t know. I guess inventory management is one of the ones that comes to mind where it’s hard, it’s painful, like there are not a lot of options. So, you know, maybe you delegate that. But those are the things where I feel sellers fail dramatically. And this is everything that we’re talking about right now. It’s not new in E commerce. It’s new for Amazon sellers because D2C, like people that sell in their own website and that they use social media to drive traffic to their websites, have been doing this for decades, like years and years and years. And those are good marketers. So I think that, that people like that group of people will have an amazing time doing marketplace like selling moving forward thanks to the changes that are happening with AI.
JONATHAN: I love those examples and that you walking us through that because those are really great points, especially the DDC aspect because there’s so many sellers that I think you discover so much when you’re doing that. Not only just like through the, the Google side of SEO and figuring out what kind of content that your audience is looking for, but just when you’re pushing a lot of social media ads, you, I mean the algorithms are trained to find those demographics and like, do they have a new job, like, or did they just get a promotion, whatever it is? And I think that that’s a really good point. You brought up thinking about how to optimize for Rufus. I mean, what are some sort of unconventional things that you’re looking at or have thought about if there’s a specific product or use case that you can think of? I’d love to hear it just, I would love to hear like what was the before situation and then what was sort of the solution for it or way to optimize it.
VANESSA HUNG: Yeah. So I can give you some, some examples. This is always a question, where is how much of, you know, how much was the revenue increase and things like that since we started doing this in Q4 last year? I think trying to get credit for growth in Q4 is kind of, I don’t know, I feel, it’s something to me feels a little immoral. It’s like, yeah, I mean, I change your listing, but the 15% growth you saw, it’s maybe because it’s holidays. You know, it’s like, I don’t want to say that that’s because of Rufus, but we saw things like that. We saw growth. The one that, that I feel we have the most impact on is a brand that did the optimization for Rufus and they start seeing a massive, like, but I’m, but I’m saying like they are, they were having like 22% return rate, which is super high for their category. They were paying like the return, return fee. So with the optimization, we were able to really explain what the product was about, what were the main concerns in those returns. Because you also get, you know, when somebody makes a return or applies for a refund, they have to say why they’re doing it. So sometimes that gives you data. That data. We use it to create content and now to answer the questions to Rufus. So we saw that and it dramatically changed, like, I don’t know, we were seeing like 3% when their baseline for the category was like 7. So it’s like, wow, this is drastic. So you influence things like that. What do you do? Right now it’s like you don’t know anything about what Vanessa is talking about. This is just, you know, nice, like AI, futuristic thinking. Cool. But I will challenge you to do something. If you go to. It could be on your desktop, it could be on your phone. But go to your Amazon listing. And at the bottom, if you, if you are on your phone, at the bottom of it, you will see a bar that opens up and if you scroll a little bit, you will see at least five or three or five questions, three to five questions about your product. What are those questions? Those questions are information that Amazon have gathered from your specific niche, things that your customers are looking for or, you know, answers that they’re looking for, information they care about. So, for example, we do a lot with people in the supplements world. In supplements, you will see things like, is this product tested or lab test or certified? That’s an answer, right? Or it will this. Do they use allergens in the formulation? Like, what are the results? Will I see results in 30 days or 60 days or whatever? Like, what are the results? What. So that is the work that Amazon has done for you, which is gathering your audience questions to prompt it in the listing so you can compare whether you are or not answering the question. So if you go to a listing, we did this recently for a supplements brand. We weren’t a listing. And one of the prompted questions was, does this contain allergens? And when you click on the question Rufus was answering, the product description doesn’t state if this product has allergens. So that is a flag. So that is the type of answer that you never, never, ever want to see on your listing. That Rufus answers that like the fact that it’s saying we don’t have information. Or in the pro description, it doesn’t state this terrible, bad pr. Bad pr. So now what you do is find those gaps, right? Sometimes you are somewhat, slightly, maybe not really answering a question, but you can do a better job at answering that question. Then you will change the content to answer that question. In our case for the supplement brand, it was like, no, it is completely zero allergens, gluten free, vegan, paleo, certified by the fda, plus plus plus a bunch of different things, eco friendly, hypoallergenic, all of those things. So you’re like, okay, cool. Now when we optimize it. Now Rufus is answering. So the word that this bot is doing for us as sellers is that it is kind of our sales rep or our advocate in a mega store. So when you go to a mega store, right, and you walk in, there’ll be somebody asking you like, oh, hey, how can I help you? That’s Rufus. And then you will say all of the things that you need and Rufus will tell you what is the better option for you to buy. So if we are not on top of our answers or on top of our content to answer the questions that our customers or our audience are doing, then we’re not selling to AI. And if we don’t sell to AI, AI won’t sell for us. At the end of the day, you try to leverage this tool to get more sales, right? That is the ultimate goal. So that’s what you do. Like go to your listing, find the questions, click on the questions, get the answers and evaluate is if that’s a good answer or not. And it is scary sometimes when you see answers that are coming solely from reviews, that is scary. You don’t want that. You want your answers to be from your product description. Obviously, like, and this is what Rufus does, it takes broad description or information in the listing and adds a little bit of social proof with the reviews. But the answer shouldn’t be solely reviews because I’ve seen many listings where based on the reviews, customers are saying that this doesn’t have any allergens. You don’t want your customers to answer that question because the moment that answer is incorrect, that is a huge problem.
JONATHAN: Those are really great insights. I love that analogy of like the sales rep and a megastore because if they didn’t have that answer, I’m not going to want to buy that product. So that’s a really great point. And also looking for trends and customer reviews that you might not have, you know, sussed out yourself and put in your listing. Definitely keep an eye on that because it’s amazing how many times I will look at A seller’s reviews and find some sort of use case that’s just rampant and with customers that will use it, but it’s not, not mentioned once in their listing. There’s no featured image about it. It’s really fascinating and it’s a really great place to find those things. I had not thought about the questions though. And that’s a really great, really great point.
VANESSA HUNG: Yeah, yeah. And if you start getting creative, you will go and you know, look into your competitors questions. I mean your niche question, your category questions. Like, there is so much. And also that’s the other part where audiences and you know, groups really, you don’t, you don’t need to marry to one. And I was actually talking to one of the top sellers in the world last week and they were saying like, well, we have a bunch of different products that are number one in our category. In each category, different categories. And it felt like there is no work to do. Like, we’re, we’re good, right? We’re number one. Like, we’re not competing against anybody. Like we are. 80% of the market share is ours. I’m like, yes, let me look. And I did the same exercise and I found a bunch of gaps of information and you’re like, even the ones that are doing great have opportunity for this is. Does that mean that their listings suck? Not at all. It’s beautiful, amazing, like perfect. But can you do a better job at selling to AI? For sure. And I think everybody, everybody is in that position. I don’t think the native way that we used to build these things before answers really the questions that Rufus is asking. So there is a need for that optimization whether you do it or not. It’s the fad of like. And I get this question a lot. Like, why would I do that now? Like, people are not using this and. And I’ll again challenge that thought and say, well, according to what Amazon has put out, they get tens of millions of questions every month, right. And they’re only. Or they recently, just recently roll out Rufus for everybody. Before it was like a beta group of people. After that was just prime members. After that it was just a little bit of other people. So now supposedly in the US anybody can have access to it. So you know, it will get bigger fast. Very, very fast. So that’s where I advise you or anybody that is listening not to wait. Like, you don’t need to wait. This is not an either or like. And this is one thing that. But I feel at the beginning of me, my journey into this AI world. I was really like, keywords are there. Like you don’t need to do that. And I caught myself in this dichotomy of like either you do keywords or you do AI optimization. The reality is that you do both at the same time and because you have to. Because there are two games going on at the same time.
JONATHAN: Yeah, I think like looking at things holistically, I mean you brought up that point before about like the inventory management and all those pieces, like you need to be on top of them because they think it’s like a. What is it like a game of Jenga, you pull out the wrong piece and like game over.
VANESSA HUNG: Yeah, no, and, and imagine like Jenga in a multi dimensional way, like where it’s like, and it’s zero gravity. Like it is, it feels so complex sometimes where Amazon is going. And yeah, it, I know, I know as a fact that it’s not fun in a lot of ways, but yeah, I mean, I find joy in certain things. Like I always like to say, like we specialize in the unsexy. It’s like all of those things that kind of a headache and complex. That’s where I thrive. Like I love to, you know, make sense of this mess. But yeah, it is, it is, it is going to be challenging for a lot of people. And I think this is a front that we are, I guess playing in. But also the other front, which is equally or not even more important is the profitability front. With the results of the earnings in the Q4 for last year, Amazon, you know, the one that surprised me the most that is like this is insane, was their net income. So their net income year over year grew by a hundred percent. They double net net income. So, you know, as a shareholder of Amazon, this is the best thing ever. As a seller though, that means that that profit or that in. Or the net income comes from somewhere. And it for a lot of it comes from all of the changes that they have done in the marketplace to for them, for Amazon, be more profitable. And for sellers means that, you know, you pay more fees, you, you don’t have certain programs, you don’t have certain things. So you know, complementing your profit, profitability strategies with growth, it’s I guess the name of the game and my word for the year. So I always like to pick one word for the year. The word for 2025, if disruption, everything disrupted. Whatever you think you know about Amazon, disruption, that’s going to happen. So yeah, I’m just excited to see what’s next.
JONATHAN: I love that. And I can’t think of a better note to end on. Vanessa, this has been a really incredible conversation. I feel like I just learned a lot. I’m definitely going to have to have you back because I have so many more questions for you.
VANESSA HUNG: Yes.
JONATHAN: But I just want to thank you so much for sharing all of that with us.
VANESSA HUNG: Thank you for inviting me. It was a pleasure.
