#438 – Geek Out on Amazon PPC: Data-Driven Advertising Hacks with Mike Frekey
Amazon advertising expert Mike Frekey takes center stage in this episode, sharing his wealth of experience from Amazon PPC agency industry leaders like IGG and Perch. His journey through the world of Amazon advertising for over 100 brands underscores the importance of data over creativity in crafting successful strategies. Mike details the collaborative efforts with data scientists at Perch to create robust advertising campaigns using sponsored products, brands, and display ads, all while emphasizing the crucial role of numbers in spotting growth opportunities and fine-tuning advertising efforts.
Listeners will gain invaluable insights into the meticulous process of auditing Amazon accounts to spotlight top-selling products and optimize ad spend across various types and placements. Mike breaks down the significance of understanding conversion rates and differentiating between brand and non-brand targets to maximize incremental sales potential. By utilizing brand analytics and search query performance data, he highlights how to effectively assess market positioning and address product issues that may impact performance more than the advertising strategies themselves.
Adding a forward-looking dimension, Mike explores the evolving landscape of Amazon’s PPC offerings and the burgeoning role of AI in advertising. He shares his enthusiasm for Amazon Marketing Cloud audiences and budding avenues like sponsored TV, while candidly pointing out challenges such as audience targeting limitations and conversion tracking hurdles. Wrapping up, Mike imparts strategic wisdom on launching products on Amazon, from risk management to harnessing social proof, illustrating how to use Amazon’s first-party data for long-term success. This episode promises a comprehensive look at mastering the art of Amazon advertising, all seasoned with Mike’s unique insights and expertise.
In episode 438 of the AM/PM Podcast, Kevin and Mike discuss:
- 00:00 – Data Nerd’s Amazon Advertising Strategies
- 02:39 – Amazon Marketing Strategies and Data Analysis
- 06:11 – Agency Age and Automation vs. Humans
- 09:03 – Analyzing Amazon Advertising Strategies
- 12:26 – Gender-Specific Product Line Analysis
- 16:33 – Amazon PPC Strategies and Attribution Challenges
- 18:40 – Amazon Bid Boosting Challenges
- 25:06 – Campaign Monitoring and Analysis Priorities
- 27:33 – Impact of AI on Amazon Advertising
- 29:15 – Future of AI in Amazon Advertising
- 31:41 – AI Transformation in Workforce
- 35:38 – Amazon Algorithm
- 39:31 – Amazon Advertising Launch Strategy Tips
- 41:59 – Product Reviews for Amazon Success
- 47:15 – Networking and Event Promotion on LinkedIn
- 48:30 – Kevin King’s Words of Wisdom
Transcript
Kevin King:
Welcome to episode 438 of the AM/PM podcast. This week, my guest is Mike Frekey. Mike works for IGG, one of the top agencies in the Amazon sponsored products and advertising space, and he previously worked for Perch, one of the big aggregators, and has worked for a few other agencies in the past too. So we’re going to geek out on everything Amazon advertising. In the meantime, enjoy this episode with Mike.
Kevin King:
Mike Frekey, welcome to the AM/PM podcast. Great to have you on.
Mike:
Thanks Kevin, awesome to be here.
Kevin King:
Yeah, I don’t think we’ve actually met in person, but I’ve heard some good things about you, I think through Isaac and through several other people so glad to have a chance to sit down and chat.
Mike:
Likewise. I know this has been a little bit in the works, so glad we’re finally able to make this happen.
Kevin King:
I know. Now you used to be with Perch or something right?
Mike:
Correct.
Kevin King:
One of the big aggregators?
Mike:
Yep. I spent three years at Perch. I was the head of advertising and marketing operations at Perch.
Kevin King:
So what does that mean when you’re at a big aggregator? How many companies did Perch buy up end up buying up?
Mike:
Oh, that’s a really good question. I know back when I was at Perch I think the numbers I was allowed to use was anything tied to our most recent press release, which I want to say was a little over 100 different brands we had at any one time I did. The number one way to describe my role at Perch is I spent a lot of money. Anything that had to do with advertising me and my team. We tried to put together good strategies, tried to build out good tools and make sure that we could just stay on top of our portfolio as best as we could. Eventually, the role evolved. We worked a lot with pricing, we worked a lot with promotions and we worked very closely with the data science team to do all of our forecasting and trying to build out a full-fledged marketing department that could have a call it full view on not just what are we advertising, what are we selling, but how is what we’re selling staying in combination with what do we have to have in stock? What are we trying to move through? What are the more strategic goals of the company?
Kevin King:
So you’re doing individual marketing for individual SKUs and like helping set up the product, the A plus pages, or were you doing contact, dealing with the influencers or all that stuff?
Mike:
Yeah, so I’d say the vast majority of what we were working on would be a combination of pricing strategy, promotion strategy and the biggest one being the actual advertising on Amazon. So all of our sponsored products, sponsored brands, sponsored display. We did some DSP testing. So those are the main areas where we would play. We did have a really good merchandising team. They were the people who were focusing more on on page listing content and then we would just take all of those listings and try to make sure that we were building out a full fledged advertising mostly sponsored product strategy for them.
Kevin King:
So you’re more in the data than in the sexy stuff.
Mike:
I would definitely say that I’m much more of a data nerd than I am a Don Draper marketer, that’s for sure.
Kevin King:
Yeah, the data. That’s the unsexy side of marketing. That’s probably one of the most important parts that a lot of people don’t pay as much attention to as they should. Yeah.
Mike:
If you ask me, that’s the fun side of marketing. Data doesn’t lie. It can sometimes mislead you, but for the most part, data doesn’t lie, so I do like working with it.
Kevin King:
So when you’re working with all these big brands, do you find that you have favorites? You’re like, when you get to work, when you get 100 different SKUs or 100 different brands and there are some that are like man, all right, this one’s a cool one, this one I can get excited about, and some of the others, like this is leggings for old ladies or something, and this just doesn’t really get me. I mean, is there something like when you’re dealing with that many, how do you keep objective on everything, stay the same on everything?
Mike:
Yeah, well, one I will definitely say. Yes, we have our favorite products that we get to work with, either at Perch or even back in the agency world now. Anytime you get to work with products that are really fighting and competing to be top of their market, I would say it doesn’t even really matter what your market is, just the fact that you are fighting for something. That makes it a lot more fun to me. Obviously, when you’re in an aggregator world, you’re running a portfolio level business. You want to take a good view on what are your top products, what are the items that maybe they’re not top of market right now, but you have a strategic vision for where they can end up with the right strategies and you want to push on those. And then you’re going to have probably the 80 to 90% of everything else, SKU by SKU, country by country, that you’re more so focused on. Let’s pick the low hanging fruit and let’s make sure that we have a good way to opportunity spot. So if a product all of a sudden goes from being lower level to we might have something here, do we have a good way to identify that this product has a good opportunity so we can go in and spend a little bit more time.
Kevin King:
So, at its peak, how much ad spend were you managing a month.
Mike:
That’s a good question. Want to say peak, we’re probably talking. I don’t even know how much I’m allowed to say in here, but I will say in the multi-million a month range.
Kevin King:
And how’s that compared to now at the agency? Is the agency a little bit smaller about the same? A little bit bigger?
Mike:
It’s an awesome question. So at the agency. So I’m a managing partner, I have a wing of the agency to myself. On the whole, for the agency, I’d say we’re probably bigger than what we were doing at Perch. In terms of just what my one wing of the agency has, I would say it is, I’d say, probably smaller than what we were doing at Perch. We’re at least getting we’re consistently growing and I’d say we’re getting to a comparable size to what was going on from purchase side, and that’s been nice. We’ve been able to hire more and so we’ve been sort of scaling up not just the ad spend portion of it, but my favorite part is building out a team as well.
Kevin King:
So how old is the agency? It’s been around for a while. Is it relatively new?
Mike:
IG’s around five or so years old. I joined a year ago.
Kevin King:
So about COVID time roughly?
Mike:
Yeah.
Kevin King:
Maybe right before COVID started. So in the agency world, how much is automation and how much is actually humans? I mean, there’s a lot of people saying now that there’s a lot of agencies. You know one of the little. It’s probably a myth to some degree, but one of the myths is like yeah, they’re using $100 software to charge you $3,000 or something. And then the other is like, no, this guy, I got this rep, that’s really really good and he’s just crushing it for me and I’m really happy. And then others are like man, this guy, no one’s asked him a hole in the ground that I’m working with. He’s so green I’m going to have to teach him everything. So what is it about agencies? That is some of the big misconception that maybe some people have when it comes to PPC agencies.
Mike:
So this is a really, really fun question to me. One, I think every agency both does things a lot the same, but then we’ll always say, well, we’re different than those other agencies. And I’ll say that IGPPC is different than those other agencies. I’ll say the same exact line as everybody else. I think where I like to draw the line is you should use technology wherever there is good opportunity for scale and where decisions don’t need to be made. I think there’s a huge value in actual expert level people making decisions, where you should have decisions made, and here’s an example that I’ll go through. My team right now we’re using a series of bidding tools that I personally have created myself. I would not position any of this as automation. I would position this a lot as technology-assisted execution. Every single type of change that we have recommended through these tools then gets pushed out as a one pager for a brand manager, a person like me or a person on my team to go and manually review.
Mike:
Do these changes make sense? Are there any areas where we want to be changing some of the variables at play within the tool itself? And are there any areas where maybe this does make sense? But we’ve now introduced a new thought for how we want to be making our system and tools better over time and so nothing is set it and forget it. I don’t believe in this sort of idea that you can let a system run fully in the background and again to your point, just sort of you have a hundred dollar piece of software that you’re charging $3,000 for. In my mind, if you’re not aware of what changes are being made and why those changes are being made, you’re not really managing an account.
Kevin King:
And one of the things a lot of agencies do as lead generation is they do free audits. I don’t know if you guys do the free audits or not, but what does that mean? So someone listening is like I’ve heard of this, but what does that mean? What is a free audit? What is a free PPC audit? What do you do?
Mike:
What this actually means. We’re going to get access to your Amazon account. We’re going to pull a bunch of data. We’re going to try to figure out what are your top selling products. How are you advertising those products right now? And we look at a few different areas within your account, the first of which is just standard account structure. What are you spending on between sponsored products, sponsored brand, sponsored display? Where are you serving on the page for Amazon? How much is it top of search, rest of search or product pages? We’ll do some brand versus non-brand target analysis. We got to figure out what are the meaningful words within your brand name. We’re going to try to find out how much of your sales are likely incremental, because it’s more discovery-based, versus likely not incremental, because it’s tied more to people who are already searching your brand name and, of course, if they’re searching your brand name, they’re probably going to find you.
Mike:
We’re also going to take a look at where are the outliers in your account. What specific customer search terms are you losing money on? Both at a sometimes we’re going to do this campaign by campaign, sometimes we’re going to do this over the course of the entire account, just whatever makes sense to the brand that we’re looking at. And then the last and my favorite and I’ll also put an asterisk next to this, something we can only do for sellers because vendors don’t have access to this level of data, but I love to use brand analytics, search query performance data to better identify what are you doing within the context of your own market. How are the top terms that you really rank on that matter, we’ll say what’s your conversion rate versus the rest of the market with you removed from it? What is your total impression share? What’s your impressions per search? So, based on how many times people are actually searching these top five to 10 terms in Amazon, what percentage of the time are you actually showing up anywhere on page one? And then we’ll do an exercise to try to figure out. Do you have more opportunity to grow on these? Are you already spending towards your ceiling? But what does growth look like? Often comes back to having an understanding of where you live within your market. I’m a very big purchase share, market share kind of a person, so that that shines through in the office.
Kevin King:
So you spend some time on these, or is this a? Or is this like a dedicated person that does these, or do you personally do them?
Mike:
Yeah.
Kevin King:
Or is this a system, an automated system that kind of whips out a lot of it and then you guys sign off on it? That’s a lot of work for someone that may be having five or six different agencies do this to try to decide which one to go with.
Mike:
So there is a lot of work that goes into it For all of the aspects of it that are just tied to. We need to pull data, we need to make sure it’s formatted in the right way. We have a specific person on our team who pulls all of the audits for us. But then once it goes over to me, that’s where I’m doing a little bit of extra cleanup, where I personally take I don’t know, call it an hour where I’m going through your search term reports. I’m trying to figure out what are the terms that matter the most to you and break those out so I can do my own level of analysis. Sometimes, when I do this, I just grab two lists of terms that I think matter. Sometimes, when I do this, I’ll go as far out as 15 different lists of words that matter, and when we get to the audit, I’m kind of we’ll call it sprinting through the early high level sections of it, because the fun part to me is really understanding. Hey, I’ve been deep diving within your market and there are so many different stories I want to tell within your data. I don’t just want to look at all of your top thousand search terms through brand analytics. One that doesn’t really tell you all too much, because there’s going to be a lot of trash in there. There’s going to be some nuggets of truth, but when I am an example I’ll work on right now.
Mike:
We have products that are broken out. There are search terms where people are searching specifically for men versus for women. The for women’s line terrible. Very low conversion rates, very low market share. We’re spending a whole lot of money on it for not a whole lot of value, for whatever reason. The for men’s line skyrocketing. Great conversion rates, still not actually spending all too much on it. And that’s not entirely tied to search volume, that’s, in part, just tied to we’re not maximizing in that space. And I think when all you look at is, hey, how are we doing on a parent page? Okay, our TACOS is high, that’s not good, we need to pull back, then you’re going to miss the smaller aspects that are actually converting really, really well for you. And so we want to take again, not all the time in the world, we can’t take all of the time in the world, but if we’re doing this audit, we want to hopefully find a couple of those areas that matter to you.
Kevin King:
So when you’re doing these audits, do you ever sometimes see like hey, there’s not much more we can do for these people? They’re already kind of capped out and doing all right. I’m sure we can manage it, but there’s not much more we can do.
Mike:
I do. I wouldn’t say that’s super often, but I mean, this has happened to me within the last couple of months where I’ve advised people hey, your number one issue right now is your conversion rates. We can’t really do anything with ads, at least from what I’m seeing. We talk over. What are the terms that matter to you? Are you maybe just targeting the wrong terms? And if you’re targeting the wrong terms, that’s dragging down your conversion rates. But that’s not the case and these are the terms we need to succeed in. But your market is in a 15% conversion rate space. You’re at a 6% conversion rate and your pricing is reasonable. I’ve recommended to people let’s not work together, at least not right now. I think you need to be investing whatever resources you want to invest in ads. Put that into your product page, put that into better video content, better picture content, maybe. Sometimes it’s pricing based and sometimes I would say like, hey, like, go work with your manufacturer, see if there are ways that you can get a better COGS profile that will allow you to price lower. It’s not the most common thing, but sometimes we’ve had that realization hey, we can see some of these copycat people selling in the same exact niche as us and they’re just able to sell at a $3, $4 lower price point. And if we do that, we know we’re just not making any money.
Mik:
Chances are, you probably just have a bad manufacturing agreement right now, and maybe you need to change manufacturers, maybe you just need to work with your existing manufacturer and just try to negotiate for better, more favorable terms. But in either way, an advertising strategy really isn’t going to be what you need in those situations. What you need is to just better manage your assets, to put those funds towards the changes that are going to be a lot more meaningful for your business. So, yeah, we’ll either turn people down or otherwise just push them. I think we have some partners who we work with that if I find out that somebody just desperately needs a better PDP strategy, I will in fact recommend them to a different partner of mine, say like hey, work with them for a couple months and then maybe it makes sense for us to work together again. Maybe it doesn’t, I don’t know. I, generally speaking, I always want to find a way for me to work with somebody, but if I don’t think I can provide value, I’m never going to try to push somebody to working with me, because then that’s not going to be a good use of my time, that’s not going to be a good use of their resources and it’s not going to last as a long-term partnership.
Kevin King:
I think I first started using Amazon’s PPC or sponsored ads it’s just sponsored ads back then around 2015. And they actually had a second program I forget the name of it now. There’s two advertising programs. They merged them together around 2017, 2018. But I remember back then it was very basic and everybody’s like man, Amazon’s so far behind Google, Google’s so much more sophisticated. But I think over the last couple of years, Amazon’s had a huge push into all these new tools and all this new stuff. And just earlier this year at CES, they announced the retail, the PPC retail thing. I forget the exact name of it and they just keep adding and adding and adding different layers onto it and I think it’s become pretty complicated in some ways for the average person. That can’t get their head around it, and that’s why a lot of people are actually now going to agencies like yeah, you guys figure this stuff out, because there’s so many different things here. What’s got you most excited over the last year or two of what Amazon’s released on the PPC side of things?
Mike:
I’ll first say, as someone who got their start in Google Ads and then Amazon even now, Amazon is still light years behind what Google is doing.
Kevin King:
Yeah, okay.
Mike:
It’s gotten better, don’t get me wrong. It has definitely gotten better, but there are still things. For example, one of the things I’m bullish on right now in a capacity where day one out the box this is not good. This is not really a helpful thing. I think there’s a lot of hope around it and there’s a hope that it’ll get to be more like Google, and that’s the AMC audiences for bid boost within its sponsored product campaigns. So for people who are watching and they’re not super familiar. So one AMC or Amazon Marketing Cloud, is Amazon’s data wing that allows you to basically do DSP level audiences, so you can create these full audiences that are a bit more advanced than just people who have been on my product page before, people who have purchased from my brand before. You can do those audiences, but you can also try to get a lot more advanced with it too, but you can create these custom audiences.
Kevin King:
And what’s an example for those listening of one of those enhanced audiences?
Mike:
Yeah. So a great example might be. I want somebody who has been on my product page within the last seven days but has not made a purchase within the last 30 days. So you’re eliminating people who and this is going to still be a fairly basic one but you’re combining a few different factors together to identify someone who might still be in market. You can actually add in a third item where it’s not just you have not purchased from my brand, but maybe you have not purchased within a specific niche or node over that same trailing 30 days. That way, you can try to find somebody who’s still actively in their research phase and has not yet made a purchase. Otherwise, that could be a good audience for you to otherwise avoid people who’ve already made that purchase. You know that you’re probably not going to win that sale, and so why are you trying to spend more money on somebody who’s likely already satisfied and is maybe just checking up on an order or for whatever reason, going back to their product page? So that’s a type of audience that you might want to take advantage of.
Mike:
The issues I am currently having with it and who knows, everyday, Amazon claims to be working on ways that they can make things better. Right now within the context of bid boosting, that means that you can take one audience, and only one audience, and apply it to a sponsored product campaign. Within that sponsored product campaign, you can give it a bid modifier, similar to what you’re probably doing right now with a top of search or rest of search modifier. That says, when anybody is searching on any of these terms within my campaign, treat it normally. But if you are on top of search, then give it a 50% bid modifier, so that you’re getting it more aggressive by an additional 50%. We can do those same sort of tactics now via bid boost through an AMC audience. The downside to it and the reason why I’m not as bullish as I want to be well, one you’re limited to only one audience, so you can’t just take a plethora of audiences, apply them like you could on Google and sort of say, hey, Google, tell me which one of these is working best. I don’t even want to give it a modifier, I just want to throw everything out there and see which audiences convert the best for me. I want Amazon to work the same way. It doesn’t. Maybe we get there, but for the time being it doesn’t.
Mike:
The second piece of it that I really don’t like is, as of right now, we cannot see whether or not this audience is converting for you specifically within the campaign. I believe we can see how much we’re spending on that audience, but you don’t have any actual granular visibility into. Is that audience within that campaign converting at a greater rate than any other person who is searching, who is not part of that audience? So that’s a great example of something that I want to believe in. I think conceptually I am a big believer in it, but we’re just at least one level down from where we need to be to make it really useful to a marketer.
Kevin King:
Are you guys doing anything with a new like video advertising or the new uh on prime TV stuff or the Alexa placements and audio placements and all that kind of stuff that Amazon’s rolling out?
Mike:
Yeah, so we try to stay at least aware of all of the things Amazon’s rolling out. We’re running some beta tests with some of our clients right now around sponsored TV. I haven’t seen amazing performance from it. Now I will note one of the things that I’m actively working with is Amazon’s released some path to purchase or conversion path reporting where you can actually see when somebody engages with a streaming TV ad and then a sponsored product ad or vice versa. Are there ads that are call it over-reporting? It’s part of a deeper funnel, but we’re not able to see that deeper funnel through a last-click attribution model or vice versa. It’s also part of that funnel, but it’s the first click, and so we just don’t have any visibility into the purchase.
Mike:
When I’ve done this analysis, streaming TV ads do have a probably the greatest or the second biggest impact of under attributed sales. Compared to all of the other ad types. I’d say it’s probably the sponsor display is the second. Now I’d also say take that with a grain of salt because I believe that Amazon is still tracking that through purely impression based advertisements for those two. So just because somebody loaded a sponsored display ad or just because a streaming TV ad popped up on their page that maybe they weren’t even in the room they were taking their bathroom break when it showed up, who knows, Is that actually leading to the sale? Tough to say. Arguably yes, arguably no, depending on the context of the individual. But I have seen that the streaming TV ads are probably under-attributed in sales.
Mike:
Me personally, I have not seen nearly enough positive impact from streaming TV ads to be overly bullish on it. I think I’m still in a testing phase and where clients themselves have a really big appetite for it. Let’s set this up, let’s make sure that we’re staying on top of it, let’s make sure that we’re measuring things properly. If a client does not necessarily have the appetite for it, it’s going to take a very special circumstance to make me go out of my way to say, hey, this thing that you don’t believe in, that I also don’t particularly believe in. Let’s try spending your money that way. That’s probably not going to happen.
Kevin King:
Attribution can be difficult. I think sometimes people look at just one component. They just they just look at their ACOS on their, on their sponsored product ads and they don’t. Maybe they’re running all different things, but it’s hard to actually know that marketing cloud’s trying to and some of this attribution the funnel flow that you’re talking about is actually trying to attribute that, but you don’t know. Is it because they saw this Google ads? Is it because they saw this influencer thing? Is it because they saw this? And they end up here, here and here and just one of them gets that final click and often gets the attribution, when in fact it’s whole system working together and you can make some misinformed decisions sometimes based on that. What’s your recommendation on that? Should people set not just be looking at their TACOS or their ACOS, but looking at the whole granular, whole picture of their advertising and say that, look, we’re willing to spend 20 percent, uh, of every sale on ads and maybe 15%of that is on sponsored ads and the other five percent we’re going to put into these other things that may or may not be as trackable or may not get the attribution or the love that they deserve?
Mike:
Yeah, I do think that, especially as you’re trying to have a more sophisticated advertising strategy, using ACOS of ROAS alone as a metric is just impossible. You can try, but again, there’s a lot of things that it’s really nice to say that you’re doing, but if you’re not actually keeping a better visible track of how your total account is doing, it’s not going to work out all too well for you. For me, I’m always, every single day, for all of my accounts, I’m going in and taking a look at our Trailing 7, Trailing 30, and yesterday performance for both ACOS and TACOS. So at the very least we can have a good sense on. Sometimes Amazon is going to have attribution delays. Sometimes we’re just going to have either slow data or just a bad sales day. Is the changes we’re making? Are they only visibly impacting one of these two metrics? We have high ACOS, but TACOS is fairly flat. Okay, I’m not going to be overly concerned. Sometimes sales end up shifting more towards the organic side. Sometimes we might see a spike in ACOS, but it’s tied more to a. Sometimes it’s going to be a spike in conversion or a spike in cost perplex and everything else is static and we’re looking at it as like oh well, we did just try to boost some of our bids or we’re going after a new space.
Mike:
It turns out it’s a bit more expensive than we were overall thinking it would be. I want to have an understanding on what’s going on for my specific campaigns, but also the context it holds within the entire rest of the business. Now I think getting further granular than that, you kind of have to be a certain size business for it to matter. Sometimes and again we don’t discriminate I’d say we’re working with any business of a reasonable enough size to pay our fees. But outside of that, sometimes we’re working with specific product lines. We’re only spending $20 a day, $50 a day, $100 a day, and for this client, that budget might be really, really meaningful to them, and if that’s the case, it’s awesome. We’re going to take that as seriously as we possibly can. We’re going to be tracking your ACOS and your TACOS on those product lines specifically. But when you’re only spending $50 a day, it doesn’t really make sense to get super granular into every single search term that only has two clicks a week, so to say.
Kevin King:
And you guys are just doing Amazon, or do you do any Google or anything else, or is it just Amazon at your agency?
Mike:
Good question. So we do Amazon. We also do a bit of Walmart. I’d say Walmart. We do not position ourselves as experts. I’ll let anybody know right now we consider ourselves Amazon experts, Walmart I don’t know if there is an expert out there. If there is, introduce them to me. I’d love to pick their brain a bit. Otherwise I’d say that’s more of a. We’ll do this as a favor. Make sure that you have at least an ads-minded person in control of things. For Google, we actually do recommend people go out to some of our partner agencies to manage that. I think one of the traps that agencies and I say this as somebody who used to work at a different agency and then also maybe spread ourselves a little too thin out at Perch there is a trap that you can run into where you try to find an expertise in everything and some people do it and some people have really massive teams in order to pull that off.
Mike:
Our side on IGPPC, we know what we know. We want to be as good as the things that we’re good at. As physically possible. We want to reinvest in those things we’re good at being sponsored ads, sponsored brands, sponsored display. Anything else, we want to find good partner agencies that we can introduce people to, to say, if we try this, maybe we’re going to be fine, maybe we’re going to be good, but I don’t know if we’re going to be great and we think our partners deserve great. So let’s try to introduce you to another agency who’s going to be able to help you out on the things they do great that we don’t necessarily do great, and so Google would be one of those.
Kevin King:
How’s AI affecting everything you guys are doing? Because, as Amazon’s kind of riding this line right now between the old keyword way of ranking and finding keywords and opportunities to rank on keywords versus search intent so where people are actually not just typing in a keyword but they’re typing in intent and you’ve got to make sure the back end and the listing is all optimized for that and ads the way you’re running ads is probably going to start to shift a little bit too as AI becomes more and more dominant. What are you seeing in that space, or what are you expecting to happen there?
Mike:
That’s a fun one. So obviously right, AI is a big buzzword. What does it even actually mean in the confines of Amazon advertising, Amazon listings, et cetera? I think there’s a huge placement within Amazon listing optimization to better understand what are the keywords that your listing is missing for, either some sort of an SEO impact or even just like a user trying to read this page to make sure, like, hey, if I’m searching XYZ, I want to see that word prominently in the title so I can know like, oh, this is compatible. This is, in fact, the right product. So, using AI to make sure that you just have the right terms directly in your listings and that it is fairly good legibly, I think that is a really big opportunity. There are a handful of different tools out there that are trying to make that an even better user experience and I’ll be honest, my heart goes out to some of my even friends who are content writers that are seeing that part of their jobs displaced. So that’s unfortunate, but I do see at scale, where you have thousands of products and you’re just trying to make sure that you have good enough content across the board. Yeah, AI is going to be really, really useful to hit that minimum level of content before you need more, like an editor level person reviewing and making some additional changes.
Mike:
As far as ad specifically goes, I’m a little skeptical that Amazon has the tools and the levers to pull right now where AI is something that you can actually integrate with. I know there are other teams out there, agencies out there, that their entire backbone of their system is quote unquote AI. I have a little bit of my own doubt around that, in that, when you’re advertising, the advertising levers of control you have are still just. Are you showing up on top of search versus rest of search? Again, we just talked about specific audiences, but that’s still only one audience per campaign where you can’t even track purchases off of them, and so I don’t know if there’s enough data where all AI systems or machine learning systems are only as good as their data inputs and also only as good as the decisions they can make to control an output, and I don’t know if Amazon advertising has enough controllable outputs and enough meaningful inputs to make for an AI solution right now. Again, are things going to get better? Is Amazon going to give us more audience level controls where we can turn Amazon ads into more of a pure like media buying experience? Probably, at the same time, we’re talking about certain things that Google made table stakes back in 2016. It’s 2024 and we still don’t have those same capabilities on the Amazon side.
Mike:
So there’s a line that I want to draw on where I think AI can be used as a part of your strategy to make the decisions on your listings versus on the ad side of it. I’m still skeptical. I’m also still hopeful, because I think there are certain things I’m itching for Amazon to make as a part of their larger API-based database. So, for example, one of my number one metrics that I use in everything is ad rank. Where are you serving on average on any individual customer search terms, so not just at a keyword level, but on all of the different search term permutations? Are you serving in slot one, two, your top of the page on both mobile and desktop. Three, four, sometimes five, now your top of search, but not ever on mobile. Outside of that, you’re just nowhere on top of search. That is still not a metric that you can get through an API, and if you can’t have it through an API, I don’t see how you could use a really good AI software that doesn’t even know how to take into account. Are you serving on the top of search consistently or not?
Kevin King:
AI for analyzing is one aspect, and some people confuse AI with just analytics. I mean, there’s a difference between the two and some people call their analytics or their algorithms AI when it’s really not AI. But I think AI is going to eliminate you just said earlier it’s eliminating some of your friends’ jobs and copywriting stuff. I think it’s going to eliminate a lot of people’s jobs in your agency. I don’t know how big your agency is now, but let’s just say you have 20 people. I think you’re going to be able to cut that to three in the next one to three years and actually do more volume and do more, make more. Those three people, or the company, will make more money overall than when you had 20 people. And because I think agents are coming and AI agents work 24-7, 365 days a year and they’re better than humans, and so you’re going to have a series of agents that it’s starting to happen now. Really, a lot of people in the Amazon space aren’t aware of what’s going on in the outside world. I pay pretty close attention to AI and there’s agents that can be chained on specific tasks.
Kevin King:
Like you said earlier, the guy when you’re doing an audit, you have a special team member that downloads those audits and kind of prepares them, and an agent can do that instead of him and it can do it massively scaled. Then another agent goes in and does what you’re doing now, where it organizes and looks at the keywords that they’re missing, and another agent it’s a sole job and it’s trained heavily just on this it actually then sees the opportunities. And then another one goes and looks at, pulls the Helium 10 data and says, okay, this is the five keywords we need to optimize, or nope, we can’t do this guy. Tell him sorry, we can’t work with him, and on and on down the line. We have these individualized agents that all talk to each other and all can play off of each other and they’re not just going on if-then-else statements, they’re going on on complete logic and actually can work at a much faster speed and more efficiently than humans. You’re just going to have one person babysitting this.
Kevin King:
So when something goes wrong and I think that’s where we’re going to get when it comes to agencies or running PPC and the same thing goes on the flip side of product sourcing and it’s going to affect everything. It’s not just PPC, but it’s going to affect everything and I think a lot of sellers are either not aware or not paying attention to where the puck is going, and I think you, right now it’s you know you got to keep doing the status quo because that’s what’s paying your bills, but you need to also be preparing at the same time for what’s coming, and I think it’s going to be a big slap in the face to a lot of people. And I’m fully expecting a lot of $5 million sellers are doing $5 million right now and within a few years time they’re going to be all of a sudden overnight doing 500,000 because they weren’t, they weren’t paying attention and starting to prepare for what’s happening. Now you just on listing optimization side of things, you look at images. Images used to not matter at all. I mean, how many images you had was part of the, the A9 algorithm, but what was in those images really didn’t matter. And now it’s becoming more and more critical and I think it’s going to even become more critical and I think images may even be more and customer reviews and questions may even be more important than what’s in your listing and the keywords that you have in your listing. And what’s in those images is going to matter, and that’s something that I think people need to start preparing for now, so that they don’t get blindsided when that switch is made.
Mike:
No, that makes sense to me, I think. One indexing more on either the human elements of product reviews or product questions directly, as opposed to the ofcourse we’re trying to make everything look as good as possible on a listing. I’m not even going to say the AI component of it, I know. Amazon’s making constant changes to their algorithm to try to figure out how we can serve not just the best product possible but the best product for the right individual person searching. And what are the things that they look at? And so, product questions, product reviews, your point images I think all of those are going to go together to trying to serve the right product. Some of that’s also just going to be like have you been on a product page before? Have you purchased from this brand before? I think there’s one side of the upcoming changes or things that have already changed on Amazon’s algorithm side that people are probably not investing in enough right now, which is just first time customer acquisition simply on the basis of I’ve done all of the tests where I have products that are ranking, you know, number 10 or so on a search term that we really, really care about, and the second you redo that search but it’s from an Amazon account that has purchased that product before.
Mike:
I’ve done it for both been on the product page before, but definitely, if you’ve ever purchased a product page before, you are now number one. You jump ahead of everybody and the person who was like Amazon’s choice badge ahead of you lose their badging because Amazon wants you to re-engage with something that they know you’ve already purchased before and maybe that changes again. That probably will change again, but at least right now and I’d say for the foreseeable future, there is such an advantage in pushing your products out there. Like I will take a loss right now personally, like if I’m a seller, I will take a loss to acquire first time customers on more or less any product that has at least a 20% repeat purchase rate, because I know I’m going to make that up over time. Sometimes it’s through subscribe and save, but even not outside of subscribe and save, my expectation is that if I have a good product, I have to believe in my product. People are going to come back and Amazon is trying to tweak their system to make it more likely for people to come back and repurchase something.
Kevin King:
There’s a couple of points there. One is, I think you know, like back on the images, I think you’re going to start seeing product placement type of stuff become important. I think you know, just like in movies, you know someone pays to be in the next James Bond film, they pay to have their car featured in the film, or they pay for the cigarettes or the whiskey or that he’s drinking or whatever is paid. Product placement I think you’re gonna start seeing some of that on Amazon and there’s going to be a new agency side that develops that actually does that for companies, where if I’m selling a beach blanket, I’m going to want to have a picture of my beach blanket, but I’m going to partner up with someone who’s selling a complimentary product that I don’t sell. Maybe it’s a beach umbrella, for example, and I don’t sell that and I want to have their beach umbrella in my photo with my beach blanket and the two of us do that and what’s then going to happen is the Amazon’s AI is going to see those as a correlation and people start and actually start showing them together and cross-promoting, and I think you’re going to get to that point of granularity on some of this.
Kevin King:
But back to your point. The other point on spending money. I see so many people just not willing to. When they, especially on a product launch, uh, even if it’s not a recurring product, even if it’s a one-off product, they’re just not willing to spend the money. Or they do, they work with Levanta or or one of these other brands, these companies, you know archer, affiliates or some of those guys that actually stack influence, that do do the influencer marketing stuff, and they’re not willing to give away. They say I’ll give you a, give you a 10%, 20% commission to do a launch. No, you should be given 100%, give them the whole damn thing to get this launched, because it starts this whole flywheel. And the same thing happens with people when they’re launching on Amazon. No, this is my product, I can’t do more than 20% or 25%. It’s stupid, it’s stupidity. And people, they’re not willing to pay for that placement. Maybe they don’t have the budget and they shouldn’t be doing this, but this is a pay-to-play game.
Kevin King:
And what do you say to those people? Does it get frustrating for you when you have a client saying no, I just you’re like, if you would just do this, these wheels would get going and just make this investment. Yeah, it’s going to suck and you’re going to be upside down for the first month or two, but just wait till month three sees it and you see what happens. I think it was, uh, Aaron Cordes that has the number one garlic press on Amazon right now, and I think, uh, there was some case study. I think Chad Rubin or somebody did it. It was like 13 months in a row he lost money willingly lost money, or I may have that number off. It was a long period of time. Um, he lost money and then now it’s flipped and now he’s printing money. What do you say to those people that have that mindset that they’re not willing to invest and play the long game?
Mike:
So I will first say I am an incredibly risk-averse person. I understand the mindset of I only want to do things that are easy. I only want to make sure that if I can guarantee myself $2, overtaking the risk of losing my money, or $5, sure, let me try to take the guaranteed $2 in most instances with my own money. That’s my sort of investing. I’d much rather be safe with things if I can, unless I have high conviction. I think when you are acting as a seller and you have a handful of products, you need to be looking at things either through the lens of a portfolio. I’m going to have some losers. That is okay If I can put out enough losers to eventually find the handful of winners that cover my losses. That’s 99% of cases. I’d say probably the game that you’re playing, unless you’re like a true product designer, product visionary person, where you’re doing fewer items but you just have the idea and you’re able to patent it and you’re able to really get yourself out there. In that case and it’s not a commodity. Then most people we’re talking to probably don’t fall under that like, hey, let me take my idea of the shark tank and get some investors to draw up this capital right? I think for the people who are more, hey, this is generally a good product, but yeah, there’s risk because there’s going to be. There’s no real moat that we have, because anybody can also go to a manufacturer in China and get a similar product, and maybe even you have a patent, but you only have a patent on a certain aspect of it, and the idea that anybody can go out and just try to get a competing item is scary. At a certain point, you have to decide, like are you an entrepreneur, right? Like, if you’re not willing to lose money in business, business is probably not for you because you’re guaranteed to lose money at different points. No, you still want to be smart. Right, you want to have good process.
Mike:
I think one of the main things I always preach and I’d say like pretty much every aspect of my life, not just in business, is just have a good, sound process where you can look back on your decision making and say, yes, I would do this decision again, even if the outcomes aren’t necessarily there, because the outcomes aren’t always going to be there, and that’s okay as long as one don’t put all of your eggs in one basket. I will always say, like you generally don’t want to put yourself in a position where one thing going wrong is going to sink you. Yeah, don’t avoid that situation. But outside of that, no, if you have a good sound process, you know that in order to launch a product on Amazon you good sound process you know that in order to launch a product on Amazon, you need to be getting to having reviews as quickly as possible and not just having reviews but maintaining a four to four and a half star visual on your product page.
Mike:
So you should be taking all of your actions associated with like, if you’re not doing Vine, go do Vine. If you’re not launching with heavy discounting at the very beginning, so that that first week or two, as you’re pushing out as many sales as possible, you’re really trying to maximize your chances of getting reviews faster. Because once you start to have reviews, well, all of a sudden, there’s a bit more social proof and there’s more inherent trust on the listing, where we can see stars versus no stars, regardless of the number associated or the count, I should say, associated with it. So now, all of a sudden, instead of doing a 40% discount, you can go up to a 35% discount, to a 30% discount and start to get less aggressive on. I need to push out trust through price to now. I need to push out trust through maintaining a healthy review landscape and making sure that I’m showing up in the placements that matter. But that flows together to your point. It is a flywheel experience.
Kevin King:
So if I’m launching a brand new product, what would you recommend my PPC just general PPC strategy be? Would it be lower my price to break even, do $10 a day, $50 a day, $100 a day, depending on my budget and just blast out, go after three keywords or 20 keywords? How would I ramp that up? What would be my go to market for a brand new product that I don’t have a customer list on, that I don’t have an audience that I can use to promote it? What would, what would, what would you recommend?
Mike:
So, one. I’m saying figure out who your competitors that you’re trying to steal share are. Figure out what their top terms are. You can do the top search terms report. You can do that for product opportunity Explorer, but you’re going to want to find out the handful of search terms that matter. And yes, there’s always going to be a long tail of searches on Amazon and you don’t want to ignore that either. But at the same time, the way that you will be long-term successful isn’t going to be oh wow, I’m getting two sales per term across 5,000 different terms. No, you’re going to have probably a 10, 10 to 20% of your overall search terms. It might literally be five to 10 actual customer search terms that people are typing in that make up 50 plus percent of your sales. And when I know those five to 10 terms that matter, it’s one do I have visibility into what are my conversion rates versus the rest of the market? Again, this is where I’m going to plug in brand analytics search query performance. It is Amazon’s own first party data source. I’m not using any third parties to estimate this. I’m actually saying Amazon. What is the 24 hour conversion rate that the rest of the market is seeing? And I’m calculating my own conversion rates. I’m evaluating that more or less every single week and I know my conversion rates are going to be bad. But that’s why I also want to combine.
Mike:
If I’m willing to spend money to bring people to my product page, I need to be willing to match that with some level of pricing through standard pricing or promotions that will drive somebody to actually convert in line with the rest of the market. So if the rest of the market is converting 20% and I’m only converting at 5%, I have a gigantic miss on my hands. And so if I’m not willing to lower my prices and slap on a 40% coupon, if you can just do a hard lowering of the price in order to make sure that you’re actually converting at a rate to get people engaging with you. But I’m using that brand analytics data to figure out where am I having my misses versus? Is there any area where I actually have success? So I want to take those terms, I want to advertise on them. I want to take those terms. I want to advertise on them. I want to be tracking them through brand analytics. I want to make sure that they’re prominently featured in my listing content.
Mike:
So again, we’re checking all of the boxes that there’s no reason for someone to not make this purchase, and then we’re going to iterate and try to find eventual efficiencies from there. But I don’t even care about my efficiency. For the first three, four weeks plus, unless asterisk, I always want to have a budget associated with like at what point can I afford to spend this money? Versus not, because I’ve seen some instances where someone’s going to spend $5,000 to literally move 10 total units. There’s a whole lot better ways than spending $500 per product to move a product that you sell for $35, right. Like, go to influencers, give it away. Like you need to make sure that if your advertising is going to lose you money, that you’re at least spending dollars in the number one best way you can. Like one product launch that I’m actively working on is in laundry detergent. Laundry detergent is expensive because everybody knows like you use a laundry detergent, you’re not going to switch, or it’s going to take a lot for you to switch, and so everybody’s willing to lose money on that. And so now you’re combining, you’re compounding the idea of a you’re losing money on the launch anyway with the idea that pretty much everybody in this market is losing money to acquire new customers, and so you’re you’re going to be running at a double loss, right? So again, I’m probably rambling at this point, but those are the things that go through my mind.
Kevin King:
Well, Mike, this has been great. I really appreciate the insights and everything on this. If people want to reach out to you or to the agency, what’s the best way for them to do that?
Mike:
Yeah. So I would say the best person at the agency to reach out to will be Isaac. You can either. I think we have a form right on our website, igppc.com. That’s probably the number one way to reach out to make sure you actually go into our right sort of marketing funnels. But also I’m very active on LinkedIn. You can message me there. I believe it’s LinkedIn.com slash in slash. Mike Frekey. I will be very surprised if there’s more of me than me on there. And then, when in doubt, if you really need to get a hold of me, try sending an email to [email protected]. But a lot of things go to spam if you’ve never messaged me before. So try the first two methods first.
Kevin King:
Well, I appreciate you coming on and sharing man. This has been cool.
Mike:
Absolutely, Kevin. Thanks so much for having me.
Kevin King:
If you want to geek out more on Amazon advertising, be sure to check out the Billion Dollar Seller Summit Iceland, BDSS 11, happening in Iceland next month, April 4th to the 9th. Bradley Sutton from Helium 10 will be there, so will Carrie Miller, so will some of the rest of the team, as well as the top people in the Amazon space. So check out BillionDollarSellerSummit.com.
Kevin King:
That’s another great episode of the AMPM podcast in the books with Mr. Mike. If you like the AMPM podcast, be sure to subscribe to this channel, whether it’s on Apple podcast or Spotify or you’re watching on YouTube. And don’t feel afraid to share this episode with somebody else or any episode for that matter. And if you want to see previous episodes, there should be some in this channel. So just hit the name of the channel, the name of the podcast, and it should show you all the past episodes. There’s some great ones in there. We’ll be back again next week with another awesome episode. In the meantime, I have some words of wisdom for you. You know, what you want to do in business is sell people what they need, but make them think it’s what they want. Sell to people what they need but make them think it’s what they want. See you again next week.
Enjoy this episode? Want to be able to ask questions to Kevin King live in a small group with other 7 and 8-figure Amazon sellers? Join the Helium 10 Elite Mastermind and get monthly workshops, training, and networking calls with Kevin at h10.me/elite
Make sure to subscribe to the podcast on iTunes, Spotify, or wherever you listen to our podcast!
Want to absolutely start crushing it on eCommerce and make more money? Follow these steps for helpful resources to get started:
- Get the Ultimate Resource Guide from Kevin King for tools and services that he uses every day to dominate on Amazon!
- New to Selling on Amazon? Freedom Ticket offers the best tips, tricks, and strategies for beginners just starting out! Sign up for Freedom Ticket.
- Trying to Find a New Product? Get the most powerful Amazon product research tool in Black Box, available only at Helium 10! Start researching with Black Box.
- Want to Verify Your Product Idea? Use Xray in our Chrome extension to check how lucrative your next product idea is with over a dozen metrics of data! Download the Helium 10 Chrome Extension.
- The Ultimate Software Tool Suite for Amazon Sellers! Get more Helium 10 tools that can help you optimize your listings and increase sales for a low price! Sign up today!
- Does Amazon Owe YOU Money? Find Out for FREE! If you have been selling for over a year on Amazon, you may be owed money for lost or damaged inventory and not even know it. Get a FREE refund report to see how much you’re owed!
- Check out our other Amazon FBA podcasts including the Serious Sellers Podcast, as well as our Spanish and German versions!
- You can also listen to the AM/PM Podcast on YouTube here!