Amazon marketing strategies: AI & beyond

Mordy Oberstein:

It's the new wave of marketing podcasting. Welcome to SERP's Up. Aloha! Mahalo for joining the SERP's Up+ podcast. We're pushing out some groovy new insights around what's happening in digital marketing. I'm Mordy Oberstein, the head of SEO Branding here at Wix, and I'm joined by the very marketing minded head of SEO Communications here at Wix, Crystal Carter. Hey, Crystal.

Crystal Carter:

Hello everybody. Hello, internet marketers, digital Marketers, PPC executives, social media managers, people who are doing CMO stuff and whatnot.

Mordy Oberstein:

CMO stuff.

Crystal Carter:

They do stuff.

Mordy Oberstein:

They have a little room they sit in and they all like the CMO stuff. That's where they smoke a cigar.

Crystal Carter:

They do stuff. They show up and they're like, hey, what's this? Do this better? It's good. It's important.

Mordy Oberstein:

More digital. We're do more digital.

Crystal Carter:

But better. It's very important.

Mordy Oberstein:

Exactly. The SERP's Up+ podcast is brought to you by Wix, where our Amazon apps help you showcase your products on your site, connect to an affiliate account and automate fulfillment of orders on your site using our Amazon multichannel fulfillment app. Best off, we built these both directly with Amazon, so it works out all of the jankiness, as Crystal likes to call it. I stole your word, Crystal.

Crystal Carter:

Do you know what that word is? Something that is from Northern California, and then I saw it in the page speed Insights report, and it was like, use this tactic to reduce jank. And I was like, jank is a tech term now.

Mordy Oberstein:

Wow, that's amazing. Not like janky. Jank. I've never heard of that in singular.

Crystal Carter:

Oh, yeah, it's a verb.

Mordy Oberstein:

Oh I like that.

Crystal Carter:

Yeah, reduce jank. It'll be janky.

Mordy Oberstein:

Thank Google. A little bit about this podcast. You might've heard of the SERP's Up podcast. We go deep into SEO, but SEO, as our head of SEO, Nati Elimelech, likes to say, touches everything. So we felt that our podcast should also touch everything everywhere all the time, which is why we decided to run a monthly bonus episode that focuses on the wider digital marketing world strategy, tactics, and whatnot. So if you're a digital marketing pro, this podcast is for you. If you're an SEO, this podcast will expand your mind, because we're not just talking about just SEO, but the wider internet that exists beyond or part of or connected to or that touches upon SEO. And that's kind of important. For our inaugural episode, we're kind of covering rankings, but from a different point of view and not just rankings, but strategy around Amazon and e-commerce and how it all kind of works together. It's a deep dive into Amazon and e-commerce. We cut through the thicket of the Amazon on this, our first ever episode of SERP's Up+. Crystal, you can't see her, but she's dying laughing at that pun.

Crystal Carter:

No, I'm just dying at that pun. I mean, honestly, cutting through the thicket. Wow, okay.

Mordy Oberstein:

That's fair. No.

Crystal Carter:

Okay. I don't know. I feel like now I want to do references to FernGully or something.

Mordy Oberstein:

Oh, great movie. Good poll.

Crystal Carter:

As soon as I saw Avatar, the first avatar, I was like, that was just FernGully. And people were like, what? I was like, have you all not seen FernGully? That was literally FernGully, even with the Twisty Tree. I was not impressed.

Mordy Oberstein:

I remember seeing it in the theater. That's how strong of an impression FernGully left with me.

Crystal Carter:

Yeah, FernGully is a good movie.

Mordy Oberstein:

So for an inaugural episode, we're taking an Amazon point of view as we deep dive into Amazon and e-comm as we look at how Amazon works hand-in-hand with your website and wider marketing strategy. And we're going to take a look at AI and Amazon to help us wade through the waters of the Amazon, sorry, Crystal, I did it again. We'll be joined by Tinuiti Vice president of Commerce Strategic Services, Elizabeth Marsten, and we're also going to hear from the Senior Director of Amazon Joe O'Connor, who's going to tell us where AI fits into succeeding on Amazon. So get ready to sell online like you've got somebody to buy the Brooklyn Bridge, as the SERP's Up podcast is primed and ready to go.

Crystal Carter:

Man. Honestly.

Mordy Oberstein:

How many puns?

Crystal Carter:

I mean, how many?

Mordy Oberstein:

How many puns? I don't know. I don't know. I'm sorry. Okay, so, a little bit of background. What made me think about this episode was actually Amazon released a few AI updates, and I'm like, okay, it'd be great to talk about that, but wait, let's not do the whole AI thing and focus solely on the AI. Let's take a step back and let's take a look at where Amazon fits into a wider e-commerce strategy. So I figured who better to help us? And thank you, George Nguyen, who's the Head of SEO Editorial at Wix, for making this connection. We figured, hey, let's have Elizabeth and Joe to help us figure out the Amazon. Here now, live, is Elizabeth Marsten. How are you?

Elizabeth Marsten:

Hello. Thanks for having me, and thanks for all the prime jokes. That's pretty much what we're living and breathing.

Mordy Oberstein:

Yeah.

Crystal Carter:

I mean, they were delivered well. And on time.

Elizabeth Marsten:

Thank you. Thank you. Thank you for bringing that home.

Mordy Oberstein:

For Crystal, it's every day. By the way, yeah, that was a good pun, but it's every day. The same kind of jokes endlessly. So I feel you, Crystal. I'm sorry.

Crystal Carter:

It's all good. It's all good. It's what we're here for. It's what we're here for. It's very important. It's all part of the process, but thank you, Elizabeth, for joining us and helping us to decipher this fantastic topic.

Mordy Oberstein:

So before we dive into the topic, per se, maybe you could share a little bit what Tinuiti does and a little bit of your backstory.

Elizabeth Marsten:

Absolutely. So Tinuiti is the largest independent digital marketing agency in, I think, it's North America, United States in particular. So that means that right behind the whole codes, you'll find us. And what makes that interesting is we get to do all of the things. So whether or not that's SEO or paid search or social or shopping, e-commerce, like marketplaces, I also work on, besides Amazon, Walmart, target, Instacart, Kroger, and about 20 other retailers. But we have all of the things, but it all allows me to do a lot of the flexibility as well. So I started in 2006 in paid search. I thought it was called paperclip Marketing. Absolutely not. I had to Google it. I got that correction. I was like, cool, figured it out. I read the descriptions. In 2006, you'd look at anything like that. You're like, yeah, I'll probably figure this out.

And I started at Portland Interactive under Ian Lurie. And so I worked there for eight years before moving on to Commerce Hub, which is one of the largest product feed management, catalog management, syndication, catalog feeds, everything that there is. And who recently bought Channel Advisor last year, to give you an idea how big they are. So that's just the demand side, but they also did order stream management. So they worked with about 65 different retailers across 8,000 partners across the US doing a drop-ship network. So I got to learn all of that piece, fulfillment marketplaces, plus what I already knew about Google and Microsoft. And then I come to Tinuiti and I go into what they call the Amazon group at the time. But that was because that's what there was, and that's how I got handed the first Walmart client and said, what should we do with this? So I built out the Walmart, Instacart, Target, and Kroger business lines.

Crystal Carter:

Amazing.

Elizabeth Marsten:

That's literally what I do now. I find ways to sell things online.

Crystal Carter:

I mean, that's so fantastic, and it's such a fantastic example of some of the variations that you can have in a digital marketing career and the iterations. I think one of the things that's great about digital marketing is you can build on your skills. So they're all sort of transferable, but you can build wherever you go. You'll take whatever you knew before and add to it.

Elizabeth Marsten:

And we're not particularly reinventing the wheel, either. Each time I took a pivot, it was like, oh, you already did this. So when Amazon, in 2015 came to our offices, at Commerce Hub, and we're showing off this thing called sponsored products, I looked at it and was like, what is this? What are they trying to do? They just ripped off Google from 2010. That's really what it looked like. In fact, they borrowed one of our guys.

Mordy Oberstein:

And Google rips them off because now Google writes sponsored and not ads.

Elizabeth Marsten:

I mean, yeah, we're not particularly inventive from one to each other. So when the first Amazon sponsored product came out, I remember one Black Friday, I was talking to the team, and they said, oh, we put our bids in. I'm like, when did you put your bids in, they're like, two days ago. I'm like, wait, what? It takes two days to proliferate to see how that went. And they're like, yeah. I went, well, geez. And then here we are today. So if you'd have told me in 2015 that we'd be sitting here with Amazon at how many billions of dollars in advertising revenues.

Mordy Oberstein:

Insane.

Elizabeth Marsten:

Maybe they followed a certain playbook. I don't know.

Mordy Oberstein:

I think it seems to have worked. I think.

Elizabeth Marsten:

That and AWS, man, that's it.

Crystal Carter:

Yeah. I mean, the recent stats that I found, I think, people think like, oh, Google's the king of the syrup. Google runs everything like Google, Google, Google. But actually, if you look at the stats for where consumers start shopping online, 61% of Americans start on Amazon before they go to Google.

Elizabeth Marsten:

And break that down into generations too, right?

Crystal Carter:

Right.

Elizabeth Marsten:

The rise of the TikTok search.

Mordy Oberstein:

Oh, no, I don't even want to know.

Crystal Carter:

I think also these things will vary, but by geo as well, because I know that there are people who are in New York who are like, oh yeah, I put something in Prime and I can get it the same day by the end of the day sort of thing. Whereas if I put something in Prime, it'll come to me tomorrow, probably in a hedge.

Mordy Oberstein:

It'll take me two weeks. They still do it.

Crystal Carter:

When the service varies, you're going to get different activity on it as well and stuff. But yeah, I think it's also really interesting. You talked about your agency, how you do all of the things, which I think is one of the greatest things about working with an agency is that if a business wants to work with you, then they want to add something to it. If they say, oh, we want to do some social now, you can say, oh yeah, we've got a social person. Or, oh, we want to do some different kinds of ads. You can say, oh yeah, we've got somebody who does that as well. Is that something that you find that customers are looking for more these days?

Elizabeth Marsten:

Yeah, it is. Well, it's a holistic thing. So what they're trying to discern right now, which makes total sense, is this idea of media mix modeling and not just media mix modeling among how much did Google search spend and how much did Amazon search spend. But if I activate something on social, and I push that coupon code via BOTA through Walmart, what did that do? Is there actual incrementality that happened? Were these new to the brand? Were these customers I was going to get? Anyway, so that's the bigger question that a lot of brands are trying to answer right now. It's gnarly and it's difficult. And there's a lot of back of the napkin math right now, because I question at times if we should be trying that hard to pinpoint shopper A, because shopper A is also the Amazon customers, also the target customer is also the Best Buy customer. Because you don't just go to one store, and you go to commercials, and you see bus stops, and you have social, and you're like, okay, well how do you really give weight to all of those touch points?

Crystal Carter:

Sure, sure.

Mordy Oberstein:

Yeah. I do a lot of brand marketing. It's the same sort of problem, because you're trying to build sentiment, you're trying to build trust, and you're trying to build a certain reputation, and it's a very thorough process. You can't say, this is what did that, it's all kind of together. But I was wondering if we could take a step back first, because I mean, listen, we're Wix. So one of the things that I wanted to focus on is your website. And what I've seen on the SEO side we're hyper focused on your website, and in a lot of Amazon marketers who are focused on that particular website, which is just Amazon. But I'm wondering how do you sync it all together? And it doesn't have to be a particular Amazon. If you're selling on eBay, whatever it is you're selling, how do you sort of combine Amazon or whatever other platform that it is with what you're doing overall into your wider marketing strategy? Which, again, touches on your initial point of it's all kind of together.

Elizabeth Marsten:

It's all kind of together. I would say, what are your differentiators? So the number one thing I always roll it back to is why are you selling on Amazon? Why are you selling on eBay? Is it a different user you're trying to reach? Are you reaching them with the same assortment? And then I would also flip that to why should they buy from your.com? What is it that you're particularly offering there that's any better or worse. If it's just simply to collect user data, because if they buy it through Amazon, they become Amazon's customer, and the big differentiator and then Amazon's cheaper, it usually is because that's kind of their thing. Why do they buy from you? What have you provided either from a.com experience is a brand or hype essentially that you couldn't do on Amazon? So I always start kind of there, and then from there you'll know whether or not you're supposed to be proliferating across other marketplaces, or you're supposed to be staying to your.com.

Because there are some brands that just say, you know what? I'm going to be very tightly controlled and my Amazon strategy is to not be on Amazon. And that's okay, but you've got to say that's what it is, and then you've got to stick to it.

Crystal Carter:

Now, the thing that I-

Elizabeth Marsten:

You got to say, that's what it is, and then you got to stick to it.

Crystal Carter:

Now, the thing that I've seen when people talk about variations between Amazon searches and Google searches is that Google searches tend to lean more towards branded searches, and the Amazon searches tend to lean more towards generic searches. So someone would say, "I would like a Starter baseball cap." They would tend to go to Google and they'd find a Starter baseball cap from the Starter website, for instance. And if somebody was like, I just want a baseball cap, I don't care what color, I just want something to cover my head, they're more likely to go to Amazon.

Do you find with clients who are choosing about their websites, how to position their website and how to position their Amazon, do you find that brands that are more established brands are more likely to lean into a less Amazon, more owned commerce sort of space? Or do you find that it's sort of fairly fluid?

Elizabeth Marsten:

It's fairly fluid. So I would take the example of Nike most recently. So Nike was the one that cut back, in favor of their D2C and providing a user experience, stores, all that kind of stuff, cut back on a lot of their wholesale relationships two years ago, let's say. They've now started to roll some of those back on. So whether or not that was through Foot Locker or DSW, it's a little more selective than it was before.

But it goes to tell you in terms of from a brand perspective, what your reach is. You assume the dot-com reaches everyone, and it does, but think about the level of effort that it takes to reach them, whether or not it's through social, there's newsletters. That constant, continuous expense, honestly.

But you also have to have a brand that goes with that. And not every brand can be a Nike, let's face it. I always look at it this way. If they got tired, in a sense, of that exercise or that capability or that cost, and they became... It just makes sense to me. Sometimes you just need to be where people are shopping. To your point about the baseball cap, sometimes they're just there and they're in the Foot Locker and they go, "I know Nike, I like Nike. I buy this Nike. Because I'm here now, and this lovely person in a striped shirt is helping me."

Mordy Oberstein:

First off, the striped shirt is an amazing branding thing. I love the striped shirt of Foot Locker.

The SEO in me looks at it like, okay, I want to be everywhere. I want to be on the informational search. Someone's searching for, "What are the best sneakers?" And I'm going to show up and funnel them from my blog post into my product page.

The brand marketer in me is looking at it like Nike will look at it. If I'm going to have a strong brand presence, then if someone's going to Foot Locker and they're faced with Reebok or Nike, they're going to know my brand, which is Nike, and decide, you know what? These shoes are exactly the same. They're made out of the exact same thing. They look pretty much exactly the same except I like the Nike brand way better, so I'm going to buy the Nike. Is that kind of what it is?

Elizabeth Marsten:

A little bit. And then here's the other thing you got to add in that I don't think a lot of SEOs think of from an e-commerce perspective, there's always the bonuses to it. So what if Foot Locker has a loyalty program where I've already bought two pairs of shoes and I add a third and I get another one half off? What if it's a buy one get one kind of thing? What if I am gathering points for my next whatever? What if I've got five kids? There's some-

Mordy Oberstein:

Now you're speaking my language.

Elizabeth Marsten:

Yeah. The dot-com experience is a planned experience in a lot of ways, and the retailer experience is a little less. So to your point around the SEO exposure, when we talk about those digital touch points, highly, highly important. You got to be there. That inception. I call it inception. I want you to have experienced the brand and have a positive outlook and have a, "Yeah, okay." Wherever that is. Whether or not that's on amazon.com, whether or not it's in the Foot Locker store, or whether or not it is on eBay, which has a fantastic sneaker outlet.

Again, those pieces all connect, but it has to do more with, what are your differentiators? What are you providing from an experience that you want? Do you care? Do you care where they really buy it from in the end?

Crystal Carter:

So I'm wondering what your opinion is on this. My hypothesis is that in a space like this, if you're a brand like that where there's a lot of high volume and Amazon is absolutely a player in your marketing strategy, my hypothesis is that one of the things that's most important about your website is the trust factor. And Google leans a lot into E-E-A-T. And that when you have spaces that are like Wish, or this new one's Temu that I keep seeing everywhere. Right?

Elizabeth Marsten:

I have many questions.

Crystal Carter:

Right. Exactly. And these people have big budgets and they're right in your face. And if you think about something like Pink Sauce that came out. Or there's all these folks who are just... Anybody can spin up a whatever and start marketing on TikTok or marketing on wherever and stick it on Amazon. On Amazon, everyone pretty much has the same page. Some people have a little bit more branding, some people might have their own channel or something like that, but the Amazon listing's fairly the same.

When I'm looking at these things, if I'm on Wish or something, or even if I'm on eBay, if a person has a website, I'm clicking on the website before I purchase. I'm clicking on the website before I give them my details. And also because I don't want to have to wait for this thing because it's not going to show up. I don't want it to be broken. I don't want it to be the wrong thing, that sort of stuff. So I'm going to go to their website and I'm going to get that sort of thing.

I mean, what important elements do you think a website needs in order to provide that kind of trust? That kind of-

Elizabeth Marsten:

You just hit it on the head. It's that customer service thing. What was it, three-ish years ago I want to say, there was a survey, and I can't remember if it was eMarketer. It was a reputable numerator, somebody who was big enough. And they did a consumer satisfaction survey. And Amazon came in number two. Number one was L.L.Bean.

And it makes total sense, because when you think about L.L.Bean and you think about customer satisfaction, end to end, I call them and I'm going to get help. My problem will be solved. That trust over the last 100 years, whatever it is. Fun story, I think it was Leon himself there, the first hundred pairs of the Bean Boot that he made, I think 80 of them or something came apart, essentially. He just took them all back, fixed them, and sent them back. Customer service from the get go. That's their go-to.

So when it comes to the website, so it says trust factors is like, can you call us and somebody picks it up? Am I going to get a response? Are there multiple ways to contact you? Review aggregation is a big one, because if you can use your dot-com to do the review aggregation, one, obviously it gets crawled and indexed in those other places. But also you can take those and use a third party vendor like a Bazaarvoice, and syndicate those out to your Walmarts and your Targets and whatnot.

Now, you can't do that necessarily on Amazon. They want the unique collection. It's theirs, hold it in their box. But your dot-com as it relates to other retailers outside of Amazon can be incredibly impactful because you can reuse a lot of that content or a lot of that brand sentiment.

Mordy Oberstein:

So let me ask you the reverse then. So on Amazon for example, I am not going to look at the website. It's Amazon. Whoever the seller is, I'm just buying it. I'm really looking more at reviews.

Elizabeth Marsten:

It's coming out of an Amazon FBA warehouse anyway.

Mordy Oberstein:

Right. It doesn't really matter. So at what point does the site say, "You know what? Let's give up on all of the sentiment that we're going to build by having the person on the website, and let's be on Amazon." When do you make that decision and how do you make that decision?

Elizabeth Marsten:

I think it really comes down to what you sell. So if the items that you sell are under say $10 or something like that, it's shipping. It's all about shipping and logistics at that point, because you have a math problem. Can you afford to ship it yourself? If you were selling cases of water, probably not. You need that to come out of a pickup from a store or from an Amazon warehouse. Bags of dog food. So things that you really have to... It's the business math of logistics. Can I conceptually even get it there? That would be like point one.

Then point two has to do with categories. So if your item requires refrigeration, what they call the chill chain. If it's chocolate, is it going to melt? Are there any kind of chemicals? Does it have to be wrapped seven times? So sometimes the brand is actually better off shipping those things because of those unique things, because they just know how to do the thing. And Amazon has a lot of rules when it comes to those things because they don't like all the returns. So they have a lot a lot of rules that you have to follow, and if you don't follow it, they charge you. They punish you monetarily for being wrong.

Crystal Carter:

They also want it easy to ship. Because I always get these things that are in these giant boxes with twelve pieces of paper.

Mordy Oberstein:

Giant boxes in another giant box with giant bubble wrap.

Crystal Carter:

And then they just throw them in.

Elizabeth Marsten:

It's well known within the marketplace industry, but not outside, they have a well-known, it's not factually a metric, but it's called CRAP. It's called can't realize any profit. So those are items that are under five bucks or weigh 10 pounds and are under five bucks or something like that and it's hard to see. They're not realizing any profit by maintaining it, and so that's obviously where the membership fees come into play, but also total basket build, getting out of the same warehouse.

But if it's this weird item that you only have in three warehouses throughout the country, but for some reason, I don't know, Oklahoma really wants it and they have to keep shipping inside it, the margins just melt off. And so you get punished, algorithmically really, on Amazon for those items. So they just kind of don't go anywhere because they don't want to sell them, because they can't realize any profit.

Crystal Carter:

Well, I think also that business math thing is something that comes into play for everyone. Working with clients, I've had it before where they were like, "Let's spend PPC money." And I was like, "Cool. Okay." We were doing this campaign and we had... I think cost per conversion or something was £5 or something. And that was okay for the stuff that was, I don't know, £100 or something. We were working in the UK. So the things that cost over £100, that's not a big deal, or maybe doing good volume.

But for the socks that were costing £2.35, I was like, "We're not putting those on the ad." They were like, "Oh, blah, blah." I'm like, "Nope. Nope. The juice ain't worth the squeeze."

And I think it's one of those things that I think a lot of people overlook when they're getting started with e-commerce and with dropshipping and all of that sort of stuff, is doing all of the math that goes into the entire supply chain, the entire from start to finish. And I'm sure that must come into play when you're devising marketing strategies.

Elizabeth Marsten:

I mean, the number one question from a D2C website typically is versus Amazon free shipping. Because we all know the cost, quote unquote, of free shipping. And so it's like, okay, well in order to make it free, you have to bake it in, which means that you have to either eat some of the margin or increase the price. And then it's like, okay, so versus Amazon, what's the differential? And then why buy it again? Why buy it from the dot-com versus an Amazon? Because okay, Amazon's going to get to me faster probably, depending on my assortment and where I've put it. And then, again, is it a unique product or not?

Especially right now in the economic times that we're in, you're going to comparison shop. Do I have to have it now? If I don't have to have it now, is the dot-com cheaper or close enough? Or do I just say, "Screw it, go with Amazon"?

Crystal Carter:

Then there's also the in-between where you have... Because I've had it before where I went to the main website and I've clicked on the thing and went through everything, and when then I got the package it came from Amazon.

And we have a multichannel fulfillment app on Wish. So you have your website and you can do all of the branding stuff, but you don't deal with the fulfillment. They handle the fulfillment. And so that gives you the best of both worlds. Have you seen that work well for clients?

Elizabeth Marsten:

Yeah. Yes and no. From a marketplaces standpoint, it's... I look at it this way, again, it's that business math. If the product is already sitting somewhere and it makes sense to let the customer buy it via Prime, and from a fees perspective, how does that shake out? Because you're not paying the listing fee necessarily on that so much as you're paying the FBA costs and the pass through costs. So the payment processing essentially.

So again, it really depends on what it is and what you put in on. What we've counseled is, be selective. You can test it. So they really wanted it to work, Amazon really wanted it to work. And when you think about it, it makes total sense. It's like, why let PayPal get all of the fees that they could have?

So test it out, and we did. It's okay. I think, again, it comes down to... I looked at it this way, it's a little bit desperate from a D2C perspective if you do. It's like, "I don't care, I just want to sell it." That's what I interpreted the feature as. Just for the love of God, let's go, let's go, let's go.

Crystal Carter:

Throw everything at it and see what sticks.

Elizabeth Marsten:

It doesn't hurt. I mean, when-

Mordy Oberstein:

Sometimes that's just what it is.

Elizabeth Marsten:

When Buy with Google came out... Because they just announced that they're going to subset that marketplace, right?

Mordy Oberstein:

Well, they just got rid of it.

Elizabeth Marsten:

So I was there for the 2013, was that when that... '15? '15 when they first came out with the buy button. And they didn't want us to call it that, but that's what it was. And then we said, "Is it the Google marketplace?" And they said, "Don't call it that. It's Buy with Google." Whatever. It went through four name iterations.

I thought it was interesting because it was kind of like Google's attempt to collect e-commerce information. Because you always wondered, so you would do paid search, it would go to the website, or you do SEO, go to the website. And unless they had Google Analytics or something, you don't necessarily know what they searched for is what they bought. Especially with paid search. Or which variation did they... Maybe they clicked on the black one, but they buy the red one. Or maybe they did and then they leave for 30 days and they come back. Whatever it was. That was Google's way of actually definitive...

Elizabeth Marsten:

... days and the come back, whatever it was. That was Google's way of actually definitively knowing what you searched for and what you bought. If they were the payment processor, then they knew 100% what it was, how many, what color, where it went, when you bought it, all that kind of information that Amazon has so much of. Then yes, you should be jelly of, but that was what I anticipated the buy button or that marketplace to be. We had a client that took part in it. One of the promos the first week it came out, they had some stellar promos with it. This client beat their Amazon sales for two days. They were thrilled. I was disappointed to see where it went as a product overall because it did have potential, but it wasn't sustainable, I guess.

Mordy Oberstein:

It is interesting now. I'm curious what's going to happen with this whole thing because now what Google is going to do with their testing and merchant centers sending you right to the checkout on the website. It is an interesting little balance, because as Crystal mentioned before, the website offers you context except the checkout page. This offers you nothing.

Elizabeth Marsten:

Yeah, that's nothing.

Mordy Oberstein:

It's a weird little bounce out.

Elizabeth Marsten:

I wondered when they started it, if they were like, because at the time there were so many sites that were still pretty janky, to use that word, or the checkout experience. Are you required to check out as a... You had to register for an account. Well, people love doing that, right?

Crystal Carter:

Also, sometimes when they do that, they're just like, "Oh, can we also have your middle name and your mother's maiden name and-"

Mordy Oberstein:

Oh, I love my mother's maiden name.

Crystal Carter:

... and the name of your dog?" I'm just like, oh my days.

Elizabeth Marsten:

Then they would make you make a password and then you didn't do it right the first two times and you're like, Jesus Christ, just let me give you money.

Crystal Carter:

And the password is all like, oh, it has to have a caps and a number and a symbol. And I'm just like, I can't.

Mordy Oberstein:

Unbelievable.

Elizabeth Marsten:

It should have frictionless, right? So that's what Amazon does incredibly well. They do better than anyone else. Frictionless.

Mordy Oberstein:

So as time sort of ebbs away here, and I hate to be asked this generic question, but I'm going to do I going to say two generic questions. One is, where do you feel Amazon is heading going forward? And a year from now, what's going to be different? And we like to offer our audience on the SEO podcast that we do the regular service up, I guess. So it's not regular. Some other people they can follow across social media to get more insights. So one, where's Amazon heading in a year from now? And two, who are some people that our audience might want to check out to learn more about e-comm?

Elizabeth Marsten:

Okay, so Amazon heading, they are headed for, you're going to love this, lead gen. That's where I'm putting them, non-endemic. So it's outside of e-commerce and it should be because they're so dominant in e-commerce. Where do you go next? So they have a DSP, which has... Think about how much information they have about you alone over the last 10 years, and that's all first party audience data. What if they were to take that and sell that to Geico, Liberty Mutual, T-Mobile, whatever.

So conceptually, and this is where your D2C comes in a good way. So not necessarily an SEO perspective, but ideally over these last few years, you've curated an amazing experience or checkout or lead generation process or landing page testing or whatever. You've got decades of information, you know how people go through your process, how many days it might take, all that kind of stuff. So if I knew that kind of persona information or that kind of audience information and I were to go to Amazon's DSP and start looking for audiences that could match up, show them the advertisement for your D2C website. So let's say it's Legion. Let's say it's real estate attorneys for whatever reason, take that information, run it through Amazon's DSP with your beautiful ad and drive it back to your D2C. What kind of quality lead gen do you think you might get?

Crystal Carter:

Very interesting. Very interesting. I think also-

Elizabeth Marsten:

You could do with e-comm too, but yeah.

Crystal Carter:

Yeah. I think the other thing that's interesting is that Amazon is very happy to couple their tech together. So it could easily be something that was coupled with say Amazon Echo, so that you were like, hey Alexa, my refrigerator is broken. Who can help me with this, for instance? And then Alexa gives you recommendations for, I don't know, refrigerator repair people.

Elizabeth Marsten:

Yeah, that's a service they have, right? The biz dev directory.

Crystal Carter:

Exactly. So that could be linked with that as well. So that's very interesting.

Elizabeth Marsten:

Yeah. If I know that for example, you've purchased lawn care subscription or lawn care items from amazon.com and a certain amount of time has passed, or I know that there's a drought or something like that, I can find you and sell you my lawn care services for fixing whatever it is that you broke or didn't have time to do, and you were totally pretending that you're going to get around to, but you didn't.

Crystal Carter:

I decided to concrete my patio and concrete is actually being really difficult and it took us months and it's still terrible.

Mordy Oberstein:

I feel like you should probably hire a professional concrete.

Crystal Carter:

Yeah. Yeah. But it's like sunk cost fallacy. We put so much effort into it. I'm like-

Mordy Oberstein:

That's it. You can't give up all the way. So basically what you're telling me is the way that Facebook retargets me when I even talk in WhatsApp about, I don't know, buying a new chair is going to be like a million times more powerful.

Elizabeth Marsten:

But less creepy and probably cheaper. One of the things with the Amazon DSPs, I don't think most people realize what a bargain it is in terms of CPMs and stuff versus some of the other DSPs that are out there. And then relative to their knowledge, it's their first party data. So with Facebook, it's captured in other ways. With Amazon, you straight up told them because you bought things and you sent them certain places.

Mordy Oberstein:

That is less creepy. You're right.

Elizabeth Marsten:

So they don't need to listen to you the same way.

Mordy Oberstein:

Oh, I see. So it's not like they're not against being creepy, they just don't need to be creepy.

Elizabeth Marsten:

Exactly. So when you're going to stand in front of the Department of Justice someday, you got to make sure that your house is in order.

Mordy Oberstein:

I think that's a good note to say, hey, who should our audience follow on social media to check out some more e-commerce insights before we go down that worm hole?

Elizabeth Marsten:

One thing in particular, if you are interested more in retail media, I would follow Andrew Lipson on either LinkedIn. I don't think he tweets as much, but puts a lot of good research out there from an analyst perspective, forecasting that kind of thing. So that's both Amazon and anything retail media. There's also Russ Derringer who is on LinkedIn and Twitter just a little little bit. His company's called Stratably and he writes on all things Amazon and retail media these days. A pretty kick ass. You could follow me. I write about this stuff once in a while.

Mordy Oberstein:

I'm going to get that next. I was going to get that.

Elizabeth Marsten:

I have a week article that is pending on endemic audiences. My next one is going to be on what I just talked about where Amazon's head of next year around non-endemic. So I tend to get out there and be pointy as I've been told. I'm trying to think if there was another human I would follow. In particular, Kiri Masters, of Bobsled or Arcadia. So she does a lot. It's mostly around Amazon. She does write for Forbes as well, but they have a lot of Instacart information as well. So she and another colleague of hers wrote a book on Instacart.

Mordy Oberstein:

Amazing. And where on social can people follow you?

Elizabeth Marsten:

I am at EB Kendo, so I'm on Twitter and I tend to cross post pretty much that exactly on LinkedIn and better at maintaining LinkedIn than I am at Twitter.

Mordy Oberstein:

Amazing. And we'll make sure to link to all of those social media profiles in the show notes. Elizabeth, thank you so much for coming on. This is a lot of fun and really insightful.

Elizabeth Marsten:

Feel free to reach out. I love questions. I love nerding out. I love my job. I love what I do. So yeah, please.

Crystal Carter:

It's really, really clear that you love what you do and you're really passionate about it. Thank you so much for sharing today.

Elizabeth Marsten:

Oh, thanks for having me. It was a blast.

Mordy Oberstein:

As I mentioned in the beginning of this episode, one of the reasons why I got stuck on doing an episode with Amazon and e-comm was I saw Amazon do a couple of new things around AI. So Elizabeth said, "Hey, why don't you talk to Joe O'Connor, who is the senior director of Amazon at Annuity because he actually used to work at Amazon, knows all about this." So that's exactly what we did. We sent a few questions over to Joe all about Amazon and AI. So let's dive in.

So one of the things that Amazon has been doing has been using AI to create product summaries. And what they're doing is they're leveraging review content. They're looking at the reviews for the product and using AI to create a product summary from that review content, which is super interesting. I think it might be a little bit concerning to some marketers because let's say you integrate some of the negative comments or negative reviews into the summary. But leaving that aside, it is interesting to see Amazon develop that way. So we asked Joe, what are some useful ways Amazon could go about integrating AI into its product going forward? Obviously, aside from the product summary we're talking about here. Here's Joe.

Joe O'Connor:

Yeah, that's a great question. I'll answer this from two perspectives. First, from looking through the lens of a marketer, one area I think would greatly benefit from the use of AI is operating within Amazon's marketing cloud product. That's the data clean room they have where a brand could go in and pull a bunch of different data points about their advertising campaign and sales and actions consumers are taking through the campaign across Amazon's various touchpoint, amazon.com, IMDB, Prime Video, et cetera, and also upload their own first party data. But right now, to pull that data out of the marketing cloud infrastructure, you need to know how to write SQL, you need to know how to code, which is not a skill that's very common among marketing managers of brands or account managers at most advertising agency. So it would be amazing for those kind of functions who would get the most use out of the insights that the tool provides, had some sort of AI assistant that they can speak to and explain this is the business challenge I'm trying to solve.

I want to gain insights into and have that AI generate the query, deliver the results, and maybe note a couple of interesting trends within the data. That would be really powerful from a marketing perspective. Now, on the consumer side of things, as an Amazon shopper, I think what would be really, really interesting and just make a lot of sense is Amazon's the biggest retailer in the world in terms of the assortment of products they carry. So I can't think of a better place to have a really efficient shopping assistant who becomes more familiar with the types of products you like, what your size of clothes you wear when you're shopping for apparel. All different brands, they have sizes that fit a little differently.

Who can help predict when you want to buy headphones based on other headphones you've bought in the past, what other electronics you've bought, what features are important to you, and can then populate those suggestions within your feed. I think that as a consumer would be really powerful and help shoppers more efficiently and effectively browse and shop around Amazon's website. So I think that as a consumer would be a really powerful use of AI.

Mordy Oberstein:

Yeah, it's a good point. One is about the insights that you get as a marketer itself. I think using AI to sort of tap into that data source that Amazon has in order to leverage that was super interesting. On the consumer side for sure, one of the issues that I have with Amazon when I shop is really trying to figure out how to get what I want. When I know I have something specific in mind, sometimes I find I have a hard time getting at exactly what I want to get at, and some kind of shopping assistant built upon AI would probably be very, very helpful.

Crystal Carter:

Yeah, absolutely. I think that is something that we are very likely to see I think in the future for a few different sites. And I would be very, very surprised if Amazon wasn't leading the charge, considering that they have so many products and so many different types of products within their catalog that it would be very easy for them to say, are you planning a barbecue? And you go, yeah, I am planning a barbecue. They're like, great, we got plates, we got cups. Here's a barbecue, here's some tiki torches, here's some fairy lights, here's some citronella candles. You're probably going to need some sunscreen, you might need... Do you know what I mean? To give you all of those things. And you can say, my budget for the whole barbecue, for instance, is $300 bucks or something. That's all I'm spending on this. I'm not spending anymore.

I want it to be nice, but work with me here or something. And then there's 15 people and all of that sort of stuff. And if they're able to pull through all of the different data sets, that would be something that would make it amazing shopping experience. And maybe it's a pipe dream, but that kind of thing is-

Mordy Oberstein:

That's great.

Crystal Carter:

That would be amazing.

Mordy Oberstein:

Wonderful. I would love that actually.

Crystal Carter:

Right? So for instance, the last time I had this kind of wild goose chase thing on Amazon was I was looking for... Because I went to the Beyonce concert, I went to the Renaissance tour. I don't know if I told you about that.

Mordy Oberstein:

You mentioned this before.

Crystal Carter:

Possibly. So I wore an outfit that was based on one of her looks from one of her music videos, and I had to use the Amazon visual search to find something. And then I was just trawling around for ages trying to find something very similar. Whereas if you had a shopping assistant, they'd say, "I want to look like this." And they would just go through all of the things and find me something.

Crystal Carter:

We would just go through all of the things and find me something, find me all of the things that were relevant. The gloves, the hat, the sunglasses, et cetera, et cetera. So that kind of tool I think also would work, and would work really well for marketers as well, because then you would be able to see gaps.

Mordy Oberstein:

Yeah. And what you're offering. Exactly.

Crystal Carter:

Right. So for instance, the barbecue example. Let's say you sell grills, but you don't sell aprons, you don't sell the brush thing that people use to get the gack off the little... You know what I'm talking about?

Mordy Oberstein:

Yeah, the metal part. Yeah, I know exactly what you're talking about.

Crystal Carter:

The metal brush. And then if you touch it scrapes your hand. Anyway, those things. So maybe see you don't sell those things, but you could. You could brand them and sell them and buy a drop or whatever. So you can see where you have gaps in the thing so that you could potentially pull through the whole thing. Or somebody could say, "My preferred brand is this. I really, really love all of the stuff." I don't know any grill makers. Do you mean grill brands?

Mordy Oberstein:

Weber, Coleman.

Crystal Carter:

Weber! I was thinking of Weber. Okay, so let's say I'm a Weber person. That's my grill of choice.

Mordy Oberstein:

Obviously.

Crystal Carter:

I only trust them to burn my hamburgers. And let's say that, no, I'm sure they don't burn their hamburgers, no shade to Weber grills. But you might say, I want that kind of stuff. Or I need new tech. I want everything Apple, because I'm on Apple. I don't want hear anything else that's not Apple. I just want Apple chargers and this and that and that and that, and that. That would be super useful.

Mordy Oberstein:

Yeah, and it's interesting also if you were able to use what the search engines do, which I always thought was interesting because the search engines kind of picked up on a gap that Amazon has, which is Amazon doesn't offer the user, let's say price averages or price trends, all that ancillary information. And what Google, and I think to an extent Bing has done is they've kind of filled that gap and said, "Hey, come to Google shopping, let's say, because we will give you price trends and price information." I think that if Amazon can leverage that by using AI, I think that'd be wonderful for the consumer, but also for marketers, because now you can go and see what the average price for the product that you're offering is, and you can undercut that, let's say. Oversimplify it, but something like that.

Elizabeth Marsten:

Right or add another value. It might be like maybe this price is slightly higher, but maybe we'll throw in some burger buns or whatever.

Mordy Oberstein:

We'll throw the brush. We'll throw the brush in with the grill.

Crystal Carter:

We'll throw the brush in with the grill, and you're like, "Oh, okay." Or we'll give you next day delivery, or we'll do worldwide shipping, or we'll do whatever. So people can balance out what they need that's most appropriate. Because sometimes cost ain't the only thing. Sometimes there's other stuff that comes into play when you're deciding to buy something.

Mordy Oberstein:

Well, I'm a pretty one dimensional shopper.

Crystal Carter:

Pretty straight down the line. It's like what does it cost.

Mordy Oberstein:

Makes things much easier.

Crystal Carter:

For the least price.

Mordy Oberstein:

Now, we wanted to ask Joe a more general question. There was an AI race going on, whether it's Google, whether it's Bing, whether it's Open AI. So we asked Joe, "Hey Joe, how do you think the great AI race will impact Amazon?" Let's cut back to Joe.

Joe O'Connor:

So two things I'm going to keep my eye on as AI continues to proliferate in relation to Amazon is I'm curious to see what happens with Alexa and how that evolves. I think in my own personal use of my own Echo Dot I have at home and how I anecdotally hear friends and family engage with Alexa, it's pretty simple and limited. Alexa, what time is it? Tell me the weather, turn on the light, set an alarm. There's not a ton of varied use cases I think that have become popular with Alexa. I think if Amazon was to integrate more generative AI within Alexa, how does that change how we engage with that product? And obviously with the popularity of it's in a lot of households that can be a really powerful tool.

I think the other area that's always interesting with Amazon is how they bring this technology into industries they operate in that don't have a legacy foundation in tech. So for example, the kind of more obvious answer, just based on the industries that are already starting to invest a lot in tech, you expect there'll be impacts to their retail website within AWS, their advertising, tech products. Even Prime video, there's a lot of talk about how AI can impact both images and video and media. But what about within the in-store shopping experience with Whole Foods and Amazon brick and mortar stores? How does AI impact that world within delivery and logistics, within home security, through Ring?

Amazon operates in so many industries that don't necessarily have a foundation in technology, whereas Amazon can really bring that into those industries. So I'm curious to see it as generative AI becomes more advanced and there's more and more applications for the use of it, how does Amazon bring those applications into those other categories? And what revolutions might those spark within some of those other categories?

Crystal Carter:

So I think this brings up a lot of interesting ideas around the additional spaces that Amazon operates in. Obviously we think a lot about how Amazon is working in the e-commerce space, but that is a single arm of their activity. So I think that he mentioned some of the delivery logistics things. He mentioned shopping, food shopping for instance in Alexa, and combining those abilities. For instance, if Amazon knows that you had a delivery from Kroger or wherever, and they know what food you got delivered and they know what appliances you have because you bought them from Amazon.

Mordy Oberstein:

No, too much.

Crystal Carter:

No, well, there's that. But there's also the case that they know all the things, and they may very well, you could set a set of skill for instance. I mean, again, this is big sky thinking or whatever. And Amazon, if you do this, please give me, I can send you my Venmo details, but if they know you have an air fryer and they know you bought a bunch of stuff, and Alexa could plan your meals for you and say, "In your air fryer, do this, this, and this with this food, do that, that and that from a generative space."

And again, that's adding more value to the tools that you're already using, which I think is where AI really in all of these guises really works well where there's a tool that you're already using, there's a tool that you're already familiar with that has a number of data sets and data points that they're already referencing. And they're able to combine those two to offer you suggestions or to otherwise enhance the experience that you already have.

So in the situation where, for instance, if you have a partner that's done the shopping, you might not know what's in the refrigerator because you weren't there in the shop. If you were in the shop and you're like, "Oh yeah, I picked up some grapes, then you might know, oh, okay, I could do a..." Does ambrosia salad have grapes in it? I never eat ambrosia salad. This is a terrible example.

Mordy Oberstein:

I don't even know what that is.

Crystal Carter:

Grapes is a terrible example.

Mordy Oberstein:

You lost me at salad.

Crystal Carter:

Okay, cheddar cheese, right? So let's say I didn't know that somebody picked up cheddar cheese, but somebody else bought some cheddar cheese. You say, "What do I cook for dinner, Alexa?" And they say, "Oh, make some macaroni and cheese." You have cheddar cheese, you have pasta, you have this, you have that. And you go, "Oh, yeah, I could."

Mordy Oberstein:

That's great. I would also like to say, don't buy that cheese because your other person in your account already bought the cheese.

Crystal Carter:

Wait, no, they don't need to buy it. But you can say, "Alexa, what can I make with it?" And they know what's in your refrigerator because you ordered it through them. And then do the thing. Or you can say, "What's a low fat meal that I can make?" They say, "Well, you could make this in your air fryer that I know that you have because you bought one from Amazon six months ago or something."

Mordy Oberstein:

You know what it'd be great for? I forgot to order peanut butter and I had to go buy peanut butter at the local corner store, but it's like double the price at the corner store. And I wish I didn't forget to buy peanut butter when I ordered my groceries online. What I would've like is for Amazon, whoever grocery store I choose to buy from to say, "Hey, wait a second. Usually you order peanut butter every week. Did you mean to leave it out?" And I would've said, "No, I didn't mean to leave it out. Thank you for saving me $2. I would've had to go buy it at the corner store, which is double the price."

Crystal Carter:

Right, precisely, precisely. So these are things that are useful. So again, with all of these AI things, what you want is something that adds to it, that adds to something that they already do well. Trying to branch out into something completely different can sometimes be a little bit of a minefield, but building on what you already have is going to make the real difference here.

Mordy Oberstein:

That's brilliant. Plugging the holes. Use the AI to plug the holes.

Crystal Carter:

Indeed.

Mordy Oberstein:

Of my grocery shopping.

Crystal Carter:

Why do you have holes in your grocery shopping?

Mordy Oberstein:

Because I forget to order the peanut butter.

Crystal Carter:

It's very difficult to carry your groceries home when there's holes in the bag.

Mordy Oberstein:

You know why? I will tell you why. I forgot. Because I like the natural peanut butter, not like this chemical eugh that my kids like. And I saw we had a bottle of peanut butter. I'm like, "Oh, okay." It looked like the natural one, but it wasn't. It got confused. So I didn't order the good peanut butter. I thought we already had the good peanut butter. I went go eat a sandwich of peanut butter, and then I realized that I didn't have the real peanut butter. If Amazon would've told me, "Schmendrick, you didn't order the peanut butter," I would've been saved.

Crystal Carter:

Exactly. These are the things. So there's the other ones where you can think of potentially having your full inventory. You can say, "Amazon, do we have this kind of peanut butter?" Like, "Alexa, do we have this kind of peanut butter?" If they know what things you're ordering.

Mordy Oberstein:

That would be great. I would do that all the time. It'd be like if it was able, and it should be able to do this relatively easily, if it understands that you order, let's say a package of cheese every other week, and you order one last week and now you're ordering one this week, it would say, "Wait, do you really need that cheese?"

Crystal Carter:

Right, or for instance, a toothpaste, "Do we have toothpaste?" "You ordered toothpaste last Tuesday, so you probably do have toothpaste, because typically speaking, you ordered toothpaste every six."

Mordy Oberstein:

I do this all the time. I'm too lazy to get up and go check in the fridge while I'm doing the order online, whether or not I actually have cheese. So now I have tons of cheese, but no peanut butter.

Crystal Carter:

Right, or somebody else does your order.

Mordy Oberstein:

Teach me AI save me.

Crystal Carter:

Or you have people who buy things in bulk, which again is something that people do on Amazon loads. They'll buy things in bulk and then they will hide some of the stuff in a cupboard, or it'll be in the garage, or it'll be in the attic or something. And so then somebody else goes, "Oh, we didn't have any toilet paper." And it's like, "We have 700 rolls of toilet paper in my special toilet paper cupboard," and people might not know that.

Mordy Oberstein:

But also being able to track what you buy, how often you buy, and watching the trends graph of that. We're spending that much money on cheese every month? Yo, slow down here. That would also be really, really helpful.

Crystal Carter:

How can I reduce the cost of my weekly shop?

Mordy Oberstein:

I feel like if Amazon listens to this podcast, I would like a finder's fee for these ideas.

Crystal Carter:

This is what I'm saying, Jeff, call us up. We can hook you up.

Mordy Oberstein:

I'm on Twitter, you can find me. I'm on Threads, you can find me. I'm on LinkedIn, you can find me there. Now, if you're looking for more SERP's Up or SERP's Up+, where can you find that? Because this is the end of our episode.

Crystal Carter:

Now is the end. The end is near.

Mordy Oberstein:

If you forget, you can have an automatic notification of AI tell you when the next episode is coming out. Or you can just subscribe over at the Wix SEO Learning Hub.

Crystal Carter:

Yeah, we had a bot write it for us. It's totally AI. The subscription. It's an AI subscription.

Mordy Oberstein:

No.

Crystal Carter:

No, it's not an AI. It's just regular I. It's just regular I.

Mordy Oberstein:

HI, human intelligence.

Crystal Carter:

Human intelligence with a little automation.

Mordy Oberstein:

Yeah, automation's always a good thing. But legitimately, if you're looking for more SERP's Up, well, thank you for joining us first off for the first ever edition of the SERP's Up+ podcast. If you're going to miss us, don't worry, we're back next week with a new episode of Regular SERP's Up where we're diving into the wide world of SEO. And we're back next month with another SERP's Up+, as we dive into the power of being authentic in your marketing with Spark Toro's owned Ran Fishkin. Look for wherever you can consume your podcast or on the Wix SEO Learning Hub over at wix.com/seo/learn. Looking to learn more about SEO and marketing, check out all the great content and webinars and articles we have over the Wix SEO Learning Hub at you guessed it at wix.com/seo/learn. Don't forget to give us a review on iTunes or a rating on Spotify. Until next time, peace, love, and marketing.