The latest episode of The Seller’s Edge podcast features Elizabeth Greene from Jungler, an expert in Amazon Ads. While the discussion touches a number of topics, it explores the balance between data and instincts, the nuances of catchall auto-campaigns, and the risks of AI targeting. Read the full transcript of the episode below.
Episode 17 of The Seller’s Edge, Elizabeth and Jonathan talk about:
- 01:15 – Balancing Data with Instincts
- 03:26 – Why Data is Often Misleading on Amazon
- 08:07 – Catchall Auto-Campaigns for Your Entire Catalog
- 10:50 – How to Avoid Ad Waste
- 12:05 – Catchall Auto-Campaigns for Individual Products
- 12:46 – Same Keyword, Different Ad Type
- 15:34 – Common Advertising Mistakes that Sellers Make
- 16:17 – “Ranking Doesn’t Equal Sales. Sales Equal Sales.”
- 17:29 – Focus on Growth at Product Level Rather than at Brand Level
- 18:37 – Competitor Conquesting (Test It and See!)
- 20:42 – How Much Does Organic Cannibalization Matter?
- 23:51 – Sending Positive Signals to the Algorithm
- 27:53 – Amazon’s push to get Sellers Using Prime Video Ads
- 32:06 – The Evolution of Amazon’s Ads
- 36:13 – What Should Sellers be Prepared for?
Key Takeaways:
- Develop a Holistic Amazon Ad Strategy. Consider factors like budget allocation, targeting methods, creative development, and data analysis.
- Adapt Ad Strategies to Your Brand and Product Lifecycle. Established brands might prioritize brand awareness, while private label sellers might focus heavily on conversion rates.
- Embrace Data, but Trust Your Instincts. Track key metrics and assume patterns will repeat, but experiment with different approaches, and iterate based on the results.
- Prepare for Amazon Ads to Evolve. Stay informed about new features, targeting options, and potential changes to search functionality.
- Put Money into Strong Product Listing Content. High-quality listings will not only improve organic ranking but they will also reinforce your ad campaigns.
- Be Wary of AI Targeting. As AI plays a bigger role, you may be asked to relinquish some control over targeting decisions. Keep a close eye on it.
- Test Prime Video Ads. Consider the potential benefits and costs of incorporating ad types available on Prime Video into your strategy.
Full Transcript of Episode:
JONATHAN: It is a very complex ecosystem. I mean, even when you’d go into, like, you know, you’ll have a product where it’ll be very similar to another product that you’ve worked with, but it’ll be in a different subcategory. And so that throws it off. So there’s all these little nuances that you have to consider. So you can’t always go on your gut. I’m curious, how do you balance that data piece with the creative piece? Is there a threshold where you’re like, all right, I need to use this much percentage of data, and then I can use this percentage. Much of a hunch.
ELIZABETH GREENE: So I think you always have to work off a hunch. Like a lot of people who are trying to get into Amazon advertising and maybe starting down the road, I’m like, look, best practices are there for a reason. It’s because they’ve been built off of what we’ve all observed to work the most often. And so it makes sense to start there and we’ll do that as well. And so in some ways, you kind of start with what you intuitively, either through past learnings or maybe you have your playbook or something. I assume, given everything I know about this product, about this market, about the advertising space, this should work. That’s my hypothesis. And then you enter that hypothesis and you, you see if it works, right? You get the feedback. I think where a lot of people struggle, because I talk to marines all the time, and where they really struggle is when that hypothesis does not work, they struggle to figure out why it doesn’t work. And that’s actually where the data comes into play. And then what I found, at least for me and how my brain works to be the most useful, in which I’ve now started to coach the team underneath me to view things in this manner is breaking down the actionable data pieces to understand what it is that I can control. Because really, when you talk about data analytics and looking at the data and being data driven, it’s like I’m analyzing what is and what isn’t working so I can then take action because data analytics is great, but you have to walk away with something you can do about it. So when a seller goes, oh, my sales are going down, like I’m doing an audit for a brand right now. And they’re like, look, we had all these issues and we know that there’s you know, we’ve struggled with this. Or they had hijackers and things and they’re like, you know, but my sales have gone down and we’ve troubleshooted all these issues, and yet sales still continuously are declining. And at this point, we’re kind of at a loss, you know, and we know our accounts heavily relying on ads, and so therefore, we need to go in and, you know, we’re trying to figure this out. We just, we can’t. Right? And it’s because they’re looking at a sales, the sales in itself, knowing if they’re going up or down is helpful to know if you’re tracking towards a goal. It’s not actually helpful in terms of determining what’s actually happening in the account. So what I’ve been doing is just trying to understand what are the actual actionable inputs to that equation of sales. So a classic example would be that thing I told you where somebody’s like, say, bids went down and ACoS went up, and they’re like, okay, so I’m looking at ACoS cost. One up. Great, like, what do I have? Like, why? What do I do about it? Right? That’s not, that’s not helpful, because logic would dictate, okay, ACoS goes up, I’m spending too much lower bids. But, like, I already did that and I don’t. So it’s, it’s, there’s this, like, this frustration of misunderstanding of, like, what do I do about this? And so I start breaking it, because the ACoS is a formula, right? A lot of these are division problems. And to be honest, with really simple division problems, which can allow you to, like, break things up even further. So if you go, okay, cost is a function of ad spend divided by ad sales. Okay? So it’s either ad spend went up, ad sales remain the same. Ad spend went down. Ad sales went down. So you’re like, okay, it’s a function of these two things. Great, okay, so it’s most likely ad spend went down, right, because I lowered my bid. But then probably total sales went down. Okay, so why? What is that function? I was pondering over this one day, and I’m like, well, actually, okay, you can break down these two equations. So sales at the end of the day is like price, right? That’s pretty binary. Typically speaking, people play the price, but not too crazy. So then how can I make sense of this ad spend part of the equation? And so it’s like, okay, so what are the inputs on ad spend? Well, actually, you can. It’s how cost per click, right. How much I’m paying, which is directly connected to bid in some manner. And then my, how many clicks I’m getting. Okay, so I could say clicks go down or whatever. And I was like, well, that’s still not as helpful because I feel like there’s some way that I need to tie this to conversion rates of products. Oh, well, actually you can take one divided by the conversion rate and that gives you how many clicks on average it takes to convert. Oh, so then we can view our ad spend in terms of cost per click multiplied by one divided by our conversion rate. And then we could say, did the conversion rate decrease? That is, does it take me more clicks on average to make a sale? Or is it that I’m paying more per click that’s much more actionable? If my conversion rate has decreased, then it’s either where I’m showing up on the platform, it’s what targets I’m going after. I’m going after things that are not converting. Well, that’s something I can take action on. Cost per clicks are increasing. Did I put some variables in there that cause cost per clicks to increase? Did I increase bids? That’s helped me understand. Now when someone goes, I decrease this and ACoS goes up instead of me going up. Well, it probably is because sales went down. No. Check these two numbers, map them over time, see which one created the variable, and then you know what to go in and fix. So it’s just trying to break things down to like root causes in a way that gives you much more clarity instead of like, ACoS went up. That’s not good. What in the world do I do about it? I have no idea.
JONATHAN: I love that you broke it down like that. Most people don’t. So I really appreciate you getting into the nuts and bolts because I think that that’s the most important part of it, is like getting into the data parts of and dissecting things. I love that you mentioned the hunch in creativity and then kind of using data as like a post mortem tool to figure out why something happened, which is great because I’ve never really thought about it like that. But data only gets you so far, especially in a system where things are controlled, right? Like data is great when you’re working in a program where everything is going to be as predictable as you expect it to be. But in something like Amazon, where a competitor comes in and throws a 50 cent product on like an average $25 product, it really screws things up and can mess with your strategy. So I really like it, you really have to rely on that. I’m curious, somebody you recently mentioned to me running an auto campaign on all of your skus, and I was curious what your thoughts are on that.
ELIZABETH GREENE: So that’s one of those things on, for clarity, are we talking about the catch all auto I’ve been talking about, or are we talking about just an automatic campaign for every single product?
JONATHAN: Let’s hear your take on both of them!
ELIZABETH GREENE: So I love to catch all autos. I both love and despise them. It’s one of those things that we have best practices, you know, things that we know we should be doing, and then we have the things that work. Right. And I’m a huge believer of when you have a hunch and you have something that you know should be the right way to, like, logically this makes sense, but yet when I test it, my performance is not what I want. And then I have this thing that I love to hate because it is not, not my best practice, not my best structure, but it crushes it every time. I’m like, why am I going to shoot myself in the foot from getting amazing performance and amazing results just because, quote, it doesn’t follow what I logically believe should be happening here. And so that’s actually going to play into running auto campaigns for every single asin as well. But, um, so what a catch all auto is, in some ways it feels like a cheat code. So. And it works best when you have a very large catalog. Basically what you do, easiest thing in the world, you take your entire catalog and you dump it into an auto campaign in the same ad group. You’re not creating separate ad groups, you’re throwing it all in there. You are putting dynamic up and down bidding, or you can use down only depending on the performance. And you are setting a low bid. Now, low is going to be relative to your account. In most accounts you might have a little bit of 15, $0.20, maybe thirty cents. And in some accounts, like say, supplements or where your average cost per click is like $5, maybe a low for you would be like $0.75 or something a little bit relative. And then you leave it alone and you let it create amazing ACOS orders for you. Now, if there are search terms that pop up that get a ton of clicks and no orders, you know, excessive clicks and orders add negatives to it, right? If the ACoS is not steller, for whatever reason in your account, you know, you can lower bids on these things. And now I will caution you oftentimes these are phenomenal campaigns and they’ll get like sometimes like sub 5% ACoS. Or if your average, ACoS in the account is like say 25%, you might be getting like 15 or 10% on this and your natural inclination is going to say, oh my gosh, this is amazing. I’m only spending $0.30, which is what’s going to happen if I spend, you know, $0.40 on that. Sorry, $0.60. Still going to work. Still going to work. Right. And so you’re going to increase it. Don’t do that, because what you’re going to do is you’re actually going to change the strategy. So at some point you might scale to like a dollar and then that no longer works. It’s not your low bid catch all anymore. You’ve kind of pushed it to be more aggressive and that has a high potential to come back and bite you down the road. So optimize down. If something’s not working, be careful how high you optimize in terms of bids. You might scale the budgets a little bit if the budgets keep running out, but that’s kind of the catch all now. That’s a terrible solution if you want to be strategic with how you’re going to dictate budget to individual products. If you are trying to, say, gather data and have a definitive knowledge of which product triggered what search terms you can, then, you know, kind of repurpose those as keywords in your account. So this structure under, quote, best practices is terrible. That’s what, we don’t run a lot of them, but what we found is consistently that they work really, really well in accounts. And so we’ll, we’ll put those up there. This is another great strategy if you have a bunch of products that you’re not advertising because you can, it doesn’t make sense to put a lot of ad dollars behind them. You can actually create one of these campaigns specifically for sort of your problem child products you’re not really sure what to do with, and then sort of see if anything works well and then move on to like intentionally setting more ad strategy behind them. That’s a really good tactic. And then as far as going through and then launching autos for each product, that is something that we do. And there’s a lot of people who will treat automatic or broad match campaigns or sometimes phrase matched campaigns to some extent as sort of, they’re only good for discovery campaigns, like, that’s all they’re good for. And there’s this logic out there that once I’ve discovered all the possible keywords in my space, therefore I will move to the ultimate form of exact match and I will not have to have these discovery campaigns anymore and that will be found again if we’re optimizing for things that work. That’s a flawed logic. The reason why is because what we found is infuriating as it can be. Oftentimes you can get much lower costs per clicks through these broader ad types. Same keyword, same product, same biddenness. For some reason, it’s just going to work better in the broader targeting. Now, what we are aware of and do want to be careful with, because this can be a double edged sword, is if you optimize again, oh my gosh, this is working. Increase bid. Oh my gosh, this is working. Increase bids. The thing with broader targeting types, autos kind of being like the largest form of broader because it can trigger a heck ton of search terms, is if you keep increasing your bids, you’re telling Amazon, I’m willing to pay more, I’m willing to pay more. I’m going to pay more for any potential search that you might possibly want to trigger for me. And at some point that can come back to bite you where Amos goes, oh, thank you very much. You’re willing to pay dollar two. Let me show you all of these places. And then that one search that was working very well for you, that you were paying dollar two for when it was working, all of a sudden, that traffic and that ad spend is going to a whole bunch of other searches that you don’t want. You’re like, I want you to spend on that. But Amazon’s like, you put me in control and I’m going to put your budget all over these places, right? And so you can end up in that scenario where sort of like stuck between this rock and the hard place. Like I can’t turn it off because this one thing is really, really working. But then also I’m stuck with all of this ad waste and like, it’s very difficult to climb out of that hole. You can, but it’s painful. I know we’ve done it. And so for that case, you do want to be cautious and cognizant of that long term potential if you continuously focus most of your time, because again, you’re getting more cost efficient cost per click sometimes. So natural inclination will be to optimize towards that, which would be logical except for sort of this long term potential cost you’re going to pay at the end of it. And actually, where I find this comes back to bite you the hardest is post Q4, because during Q4, conversion rates are up, traffic is up, everything’s hunky dory, everything’s amazing. Your broads, your autos are crushing it. So you’re optimizing your optimization, you’re like, let me get more of this. This is amazing. And then as soon as those ship times hit past Christmas and your conversion rates plummet, you have all of these search terms that are super expensive, triggering through all of these variety of places. And it is a recipe for an immense amount of ad waste and a very, very painful Q1. It’s definitely something to keep in mind if you’re going to kind of continuously run autos.
JONATHAN: Yeah, I was about to ask how seasonality plays into that and you answered that very hard. Yeah. When sellers come to you, is there something more common than not that mistakes that they’re making that you’ve seen just again and again?
ELIZABETH GREENE: Yeah, I think it’s running really far in one direction or the other. So, for example, what I’ve seen is that anyone who is only focused on exact matches, solely exact matches, nothing else. All ranking strategy ads are only for ranking. Oftentimes they do hold very good ranking. They struggle with driving the amount of traffic that’s needed to hit that sales growth that they want. So they’re like, I’m ranking really well and I cannot understand why I can’t get sales to grow the way that I want to. Like I have the, I have the placement with gifts. And so actually my tagline for that, as much as that sounds super cliche, is like ranking doesn’t equal sales, sales equal sales. Ranking helps with sales, don’t get me wrong. But it is not the end all, be all to driving sales. And then on the flip side, there’s other accounts where all they’re running is broader strategies. And most of these accounts, while they do have some form of good traffic flows, they struggle to rank. Because again, what I talked about the broader and like Amazon kind of like pushing out all the platforms, great, not so great if you want to do like strategic rank strategies. And that’s where the exact match does come into play. It’s more of a spend control function than anything. And so what we found is that you really need these two pieces of the puzzle if you want to run sort of a more long term growth strategy. So I’d say that was, that would be one of the main things. The other to kind of go hand in hand, one of them would be not evaluating each individual product performance. I’m trying to keep the account as a whole. You still want to track account level metrics because you know, products are going to roll up to that. You want to make sure you’re not, you know, sort of losing money on the averages. But if you’re not evaluating how each individual product is doing and then adjusting your ads based on that product performance, oftentimes you have really good top line KPI’s like brands are like, well, you know, I’m doing pretty good. I’ve been, I’m actually going to evaluate one. Now that they’re struggling for growth. Our total ACoS I think is like around 12%, pretty healthy. And they’re not totally dissatisfied. They might want to get a point or two down. They’re trying to figure out why, why can’t we grow? And I’m looking at it and I just, you know, went through and I said, well, one of your products is driving 20% of your sales growth. You’re only spending 3% of your ad budget on it and actually totally cost on that individual product, 6%. There’s a place to go, right? And so if you start looking at things on a chunk down level, instead of going, how do I grow this account? It’s like, how do I grow this product? Again, when looking at actionable strategies, it’s way more actionable than how do I grow my overall account. So I think those two things really, and then a close third which kind of plans to strategy number two is if you look at the data and say I want to grow this product and then there is not an ad structure that allows you to go in and easily specifically push budget toward an individual product, then that becomes a problem and you have to go through a whole restructure phase before you can actually take moves.
JONATHAN: When do you think it’s smart to focus on something like competitor Conquest or something like that?
ELIZABETH GREENE: I think it makes sense when your budget allows it and that’s where you feel like you’re going to get your best growth factor from. We’re always looking at it from the lens of ad spend efficiencies and what we’re trying to do to track towards goals. If that’s going to fit well into that picture, 100%. And sometimes that’s out of the gate. Typically speaking, if you’re doing more conquest, I would define conquesting as going after somebody else’s branded search or potentially doing product targeting. Product targeting tends to convert better than going after somebody else’s branded terms. So oftentimes we’ll start there just because we know there’s typically, again, you follow your hunches, going to be a better ROI perspective. But oftentimes what we’ll do is if we want to test, say, competitors branded terms. The thing, I mean, you started off this saying, like, you know, being really scrappy and small. That’s actually one of the benefits of Amazon ads is because it’s a pay per click model and you can actually set really low budgets on campaigns. Maybe allocate $10 a day just as tasks and just say like, hey, can I convert at all for this? What does the cost per click look like? What does my conversion rate look like? Going back to sort of using those things to forecast ACoS that’ll give you a really good idea. Like, does this even make sense? Oh, wait, no. I have like a hundred percent ACoS on my burden. Maybe it doesn’t make sense for me to allocate ad budgets unless I absolutely want to crush them. Those are fun. Like, we’ve, we’ve had, some of our brands are like, hey, you know, they’re trying to go after us. We’re like, hey, let’s hit them with everything we got, right? We got brand new campaigns on them. We’ve got product targeting, we got sponsored display ads, like everything in the kitchen, seeing the entire catalog, like, let’s own their listing, right? Those are fun if you can make sure you get the ROI from it. So it’s really just looking at the stage of the product. What does it need for that growth factor? Do we think that campaign is going to get us there? And what’s the performance look like? And then kind of amalgamation of all of those things, you typically work out to the right answer.
JONATHAN: Alright, good advice. I like that. And then as far as PPC absorption or, you know, like cannibalization of organic rank, like, how, how much are you looking at that and how much do you think it matters?
ELIZABETH GREENE: It does matter. It doesn’t always matter as much as people think. A lot of these things are ‘test it and see’. I did an entire podcast on this. Let’s see if I can condense into brief moments on what we talked about. So cannibalization is a thing. Cannibalization is not a thing if you do not have organic ranking. And so then the question then becomes, how do I track organic ranking at scale? If you have one product it is pretty easy, right? If you have 20 products it gets a little bit more difficult. If you have thousands or tens of thousands of products, it seems seemingly impossible. And so the way that we would do this is we would track something called ad sale percentage on each product. So that is simply your ad sales divided by your total sales equation gives you what percentage of my total sales am I driving from advertising. Right. And also, you want to make sure that you calculate this on a per listing or parent ASIn level. There’s some discrepancies you can end up with if you’re doing it on a SKu or child asin level. Also, it makes it much more scalable to do it at parent level anyways. And so with this, you’re saying, do I have a high ad sale percentage? Now, full disclosure, ad. So the percentage these days is getting closer to 50%. A couple years ago, we would have said 60/40, meaning only 40% or 30% coming from advertising and you wouldn’t want the majority. Now we’re creeping up there. Okay. And so if you are over a, I would say 50. If you’re nearing 60 to 70, that means you’re heavily relying on ads to drive sales. Now, there’s two reasons why this might be happening. One of them is cannibalization, and the second one is you don’t have any ranking. So this allows you to just have a number at scale and then say, okay, these are the products I need to check. So way easier than you probably have like four or five products to check, instead of like trying to dissect an entire catalog. So then you go, okay, so I have crap, I have 70% at 0% a share. Okay, so what are my main terms and where are my rankings? So where am I ranking and where am I advertising? And if you have absolutely no ranking, then the reason why your product is reliant on ads to drive sales is because you don’t have any visibility outside of advertising, no visibility on the product. So in this case, your goals are probably to do more ranking strategies, to try and push a border network ranking and then to gain that visibility. So then you can correct for that. The second reason is, if you go and you do all of your analysis and you say, well, okay, so I have a really good ranking. I’m ranking very high on all my key terms, and I’m also heavily advertising this product, so you might probably have a cannibalization situation. So if that is the case, what is the solution for that? The solution is to test pulling back on ads. Let’s say test, don’t, don’t turn everything off. Don’t pause it. Don’t go cold turkey, because there is this infuriating scenario in which ads help you drive ranking and the ad sales and positive shopper interactions with advertising will sort of, I say, like add to the cumulative pot of all of the positive interactions that happen when a shopper searches that search page and runs across your product again, either through advertising, organic, all these things together will send positive signals to the algorithm and help you push up in ranking. If I take away those positive interactions from advertising, is all of the positive interactions that now are only happening through that organic position enough to outpace the competition and help me maintain where I’m ranking on that search page? The answer is I don’t know. And there is a point of diminishing returns in which I continuously pay for advertising and I increase what I’m paying for advertising and my organic position maybe doesn’t move as quickly as I would like it to, and maybe I pull back, but then my ranking diminishes, but then my profits are better. It is a fun dynamic to try and figure out, right? And how do you also figure it out at scale cost, a massive catalog of products. So you can either drive yourself mad and try and figure it out, which at some point the amount that you can dissect the data to the level in which you need to is impossible. And so here’s my simple way of doing that. What you’re going to do is you’re going to identify where in your ad structure you are aggressively advertising to help maintain that ranking. And you can do two things. One is going to drive you crazy. I don’t recommend it unless it is a very specific key term that you know, as soon as you drop off that your toast is to like, where am I organically ranking? Where is my advertising showing up? Go and learn my bid. Wait a little bit. Go check it. Where is it? You know, like, so I can show up lower on the page. Right. Another way that will naturally do it. Again, you’re not going to like being strategic, but you can just simply lower the ACoS target of that keyword. So if right now I’m running it at a 40% ACoS because it’s a ranking keyword, can I run it at a 35% ACoS? That demographic, no matter what bid algorithm you’re using, most of them will say ACoS too high, lower bid. So this will naturally lower your bids. I don’t recommend saying, oh, 40% ACoS, let’s go down to 15. That’s going to put you off the page. That’s going to do that. Wild swings so it’s like, okay, so maybe it’s a 40. What about 37? Okay, that’s natural. And then you watch your sales for the product and you watch your organic rankings for the product. And as long as those two things remain fine, I’m totally good. So a little bit more. And so that’s more of a, again, but it’s a test in c. And if you test that and all of a sudden sales plummet, ranking plummets, you’re like, well, even though I know that my ad is next to my organic spot, when I lower my ad, I lose all momentum and I lose all sales and all visibility. So therefore, even though, again, my brain would logically say there is cannibalization, I’ve run that test and I need those ad sales to sustain me. And so, no, there’s not. And again, that’s across all keywords, all placements, all searches, all products. It’s insane.
JONATHAN: Yeah, it’s interesting, the lowering your, the money that you put into advertising and, like, where you’re going to rank for that because I found good results with people who, you know, their sponsored placement could be halfway down the page and it does better than it does at the top of the page based on whatever the category is, what the price points are, all that stuff. And, like, how customers are interacting with that page or what they’re looking for. Yeah. Which brings me to a point that it’s funny because you talk about things that infuriate you. And I know that we were talking about consumer data and just like, how it can be misleading. I’m curious. I’m someone now that they’ve gone to putting the ads on Prime Video and, you know, the shoppable carousels. I don’t know how reliable Prime Video data is just because anytime I see, I mean, in any household, there’s always like five different people, ranging from the age of five to like 70, using the Prime Video account. So it’s like, I don’t know if you’re targeting the right person when they click on a video. I’m just curious what your thoughts are on how reliable consumer behavior data is and what specific sources you’re looking at.
ELIZABETH GREENE: Amazon has been pushing everyone to use Prime Video ads, and I do find it interesting that they rolled it out inside of the ad console. So they’re trying to make it very much available to, you know, sort of the average seller, which long term I think is very interesting. I mean, running tv ads, barrier to entry, good lord. I mean, it was expensive. I’m just the creative alone to develop it, then to produce it, and then to be able to get it on a video platform somewhere. People talk about like back in the day running video or commercial ads in the wee hours of the morning because that was all they could afford. I mean, and now we have it inside the ad console for very little cost. And yeah, again, you probably want to be a better creative, right? And the creatives you’re going to use for those ads are going to be different. A lot of our sellers are very founder led brands. They’re still much more like return conscious and more conversion rate conscious. So we have tested these slightly, but a lot of our sellers are a little bit wary. I mean, to your point, honestly. Rightfully so, in some cases. Like what’s the actual return of this? I’m not really sure. You know, it is a VCP model, which makes sense for the ad type that it is. But there is sort of an inherent distress in the Amazon ad community of VCPM models. There’s a lot of these things like VCPMS paper impressions that are very common in other ad platforms, but very seem very unfamiliar to just sort of the very conversion rate focus that Amazon advertising historically has been. And so anything where someone says build brand awareness, you know, you’ll find that most private label sellers specifically, like there’s obviously large brands who run very expensive ad campaigns on other platforms and they’re very aware of kind of what that concept is and the return potential on, you know, those sort of awareness plays. But inside the private label community, there’s very much a build brand awareness equals wasted ad spend, sort of. And so I think you really just have to go into it kind of understanding what it is. But I find it very intriguing with the video ads that have come out in sponsored brand ads and now the ability to run banner ads with video and then sponsor, display video and Amazon rolling out more ways because they do recognize that there is an inherent difficulty in developing good creatives when it comes to video for a smaller brand. Even getting them done, quote, economically, could still cost you upwards of two grand for sort of a, a basic product, sort of commercialized lifestyle video, which depending on what the product’s bringing in, may or may not make sense at scale. And so they’ve sort of developed like video creators. They now have their beta program. They’re being connected with canva, which I think is super interesting. So they’re trying to make it more accessible to sort of run these commercialized ads to sort of the average seller, which I think is amazing. And it’s a really cool way to democratize, which has traditionally been a very difficult and expensive space to get into. So I’m saying the future looks bright right now. I think there’s a lot of advertisers who that doesn’t really fit into our more conversion rate return conscious playbook right now that still a lot of us are running and so we’re not really running these at scale across most of our brands, I do think, to your point, right, like the data associated with when anything new comes out, it’s super interesting, it’s fun to try, but it’s not always super polished. The data is always a little bit hard to understand how it’s calculated and really to figure out the return of it. And so, yeah, it’s something we’re keeping an eye on. But I wouldn’t say it is like a big focus of ours.
JONATHAN: I think time is going to tell on that, just because knowing where Amazon was and where it’s at and how it’s evolved, like, it’s just these things play out over time. What do you see next? Like, are they going to start having pop ups on product pages? Like, I just don’t know what else they can be doing.
ELIZABETH GREENE: Oh, goodness, I hope not. So, I mean, the evolution of Amazon ads, you know, it went from very basic to they rolled out a lot of new levers for control. So things like splitting out targeting types and autos placement percentages, different bidding strategy, they’re rolling out in native rules which to be honest, we don’t use. And I find that very clunky to use. But honestly, all they have to do is give me a consolidated view that shows me where all my rules are applied and a way to set a sort of template for a rule and bulk apply it. And I would be a happy girl the way it’s rolled out. Like it’s, it’s way too hard to put up, it’s way too hard to take down and it’s way too hard to find out what campaigns you have rules set on. So I’ll leave it. Thank you. But so there’s this very like, okay, we’re going to give you more options for control. Then they started rolling out with sponsored display ads and sponsored brand ads. So they brought those into the seller central platform, sorry, campaign manager from vendor Central. And then they started giving us more creatives and then we had videos and then we had lifestyle images and now lifestyle images are required for sponsor burnout. So then there was this push for more creativity and more brand telling. So more data points associated with brand stores. And really let’s build your brand on Amazon because that was a difficulty for a lot of brands. Historically it was like Amazon owns everything. So like let me give you more options to create brands. And then we had this whole evolution of more data points coming up, which I found very interesting. So we had things like search query performance reports that came out, blew everyone’s minds. We had brain analytics launched in the ad console that was given to us. So a lot of product opportunity explorers. So a lot of things like here’s how you’re benchmarking against competition, which is something that sellers historically had always wondered about and had no real definitive solutions to look at. No, we have that. It’s amazing. And sort of what we’re finding out now is this push which I find interesting and also somewhat infuriating as somebody who feels like they know what the heck they’re doing with Amazon advertising is this new push to give up more control of the targeting to Amazon. So they rolled out, not even rolled out, they updated how broad match works in sponsored product apps. They’re expanding how the targeting works in sponsored brand ads. I jokingly say that match types almost don’t matter in sponsored brand ads anymore. Like they kind of do. But honestly, the way they function now, like they don’t. Which again, to me as somebody who feels like I want my ad placed here, stop trying to tell me what to do and like to put it there. I feel like I know what I’m doing. But this comes also at a time where Amazon is exploring more into how to leverage AI to enhance their search capabilities. And so I can see these two. And when Amazon rolls things out, especially when it comes to like a new targeting or like a new way that the targeting works historically, like it’s been, it’s been slightly clunky, it’s a little bit annoying, especially when Amazon tries to take more control over things. There’s these new target types and sponsored brand ads. It’s related to your branded page. Like, oh, we’ll show you those performed terribly. Like the return is like what is this? Right? Like we don’t, we don’t even use them. After some limited tests, maybe eventually we will, maybe they’ll get better, right? So we’re just keeping, we’re keeping an eye to the ground. And I think that’s where I was talking with somebody who was on the Google Ads platform. So it’s very much following sort of the trajectory, like Amazon advertising slightly behind, you know, where Google pay per clicks has been around for a very long time. And so you can kind of take a lot of clues and insights from what that platform has done. And they experienced a similar shift where Google Ads was, you know, sort of like broadening out, taking more control over where things are being targeted. A lot of the Google Ads managers like, dude, we know what we’re doing. Stop. And I think we, as Amazon advertisers feel kind of the same way. And I was talking with a lot of colleagues when sort of they started this push. And all of us are like, cynically, even though we kind of feel a little bit hypocritical because it’s like biting the hand that feeds us, but we’re like, Amazon just wants more ads. They just want more money. Their ad platform is crushing it in terms of profitability. They’re just getting greedy. And that was kind of my sentiment, although I talked to some other people who had familiarity with sort of like the Google Ads world a little bit more so than I do. And their thinking was, yes, I’m sure that is a factor. Like, I’m not naive to know that that wasn’t discussed as a selling point at some form, but with their increasing sophistication and their understanding of consumer behavior on their platform, they’re trying to optimize for sort of, I want to say, like the lowest common denominator, but like the most, the most base and entry positions. So as they sort of democratize these targeting types and as they make it way more accessible to sort of the beginner seller, there is a potential that maybe those decisions in the beginning are maybe not done as well. The targeting isn’t done as well. So the idea would be, and if you talk to anyone or read any of the documentation, like, isn’t this great? We can help you out. And we’re like, thank you. No, thank you. I’ve seen what you suggest for me, but for someone getting into it, I have no idea where I should even start. This is a good place to start. Yes, please, I could use your help. And so it will be interesting to see how they integrate the world of AI with advertising and how that will show up. Some people say search is going away and it will no longer be what it is now. To some point, I think they’re right. But then again, I feel as if the Amazon ad platform will always survive in some capacity because it’s such a cash cow. Like, if it wasn’t, if they were losing money, yeah, they might say, like, this is not worth it for us. Let’s just simply optimize for the consumer above all else. I have pretty confident in saying I think advertising will be a significant factor in how that will be shaped in the future. I just, I honestly have no idea what it looks like, which is terrifying and exciting.
JONATHAN: Agreed. It’s funny, I attended the Amazon dev summit like last month and their team definitely wants to create more transparency around the documentation and make it a lot more efficient for ads to be run. And it’s funny because they brought up the fact that they are working vehemently to make sure that broad match and phrase match mean exactly what they say. So hopefully, I mean, that’s what they say. Hopefully that will be on the horizon soon. But yeah, I mean, I don’t know. It’s interesting because we were talking about the prime video and it’s like maybe there’s just a space for different tiers of sellers. Like you’re an upper echelon, you’re gonna be adding like running ads on prime video and if you’re a scrappy starting seller. Sponsored placements?
ELIZABETH GREENE: Yeah, yeah, I think so. And it is interesting because sponsored placements are, you have product pages but primarily focused on search. And if search pages change, I don’t know if you’ve seen the demo with rufus and how that sort of interaction is. It’s almost kind of like a separate thing from searchers at this point, which shoppers can elect to interact with or then go away from, but they pretty much type in queries and then, you know, it surfaces products, which as a consumer, I think that’s amazing. I would love that. Right? Because it’s annoying if you’re like, I want this specific coffee maker with these exact criteria and you’re like, I have to like go through every single one of your detail pages and I’m a Photoshopper. So like, if you don’t show it to me in the infographic, forget it. Like, I’m not digging through your dog, you know, like your bullet points to try and find the dimensions. Like give it to me or just, no, I’m on to the next one. Right. So for me to go in and like, I know exactly what I want here, please give me this. I think that’s amazing. But like, do we have sponsor placements in that? Like, how does that work if you’re completely bypassing search pages? And again, right now, at least the way they demoed it was, you know, you, it would pop up and say, do you want to interact with this? And then if no, then you could go back to, you know, to the traditional search grid and browsing. So it wasn’t like one is taking over, it’s one’s enhancing like another option, basically. And so how that plays into it all again, it’s a whole new world that I don’t think any of us, while we’re all excited about the potential of AI, I don’t think anyone really has any idea where it’s going to even end up.
JONATHAN: I’m still waiting for someone to be in a house and reorder a product on Alexa and Alexa will be like, I actually found a product at 4.8 stars and it’s a dollar cheaper. Would you like that one? I’m surprised they haven’t gotten there yet. This has been great. You’re a wealth of information, a fountain of information. So I appreciate that I’m going to have to have you back just to pick your brain more because I feel like I only got to half the stuff that I actually wanted to ask you.
ELIZABETH GREENE: Thanks. I appreciate it. Thanks for having me on.
