011: Turning Small Tweaks into Massive Amazon Successes

Speaker 1: 0:00

Welcome to the Brand Fortress HQ podcast and in this Amazon Tactics Tuesday we're going to be talking about cumulative effects and we're going to be talking about how you get that flywheel moving. So this is a discussion that we've been having before we started to record and before we get kind of into the nitty gritty, I'm actually going to hand it off to you, mike first, where we're talking about singular effects versus compounding effects on Amazon and share a little bit with our listeners kind of how you think about those two different aspects of improving your listings and really moving that flywheel forward.

Speaker 2: 0:31

Yeah for sure. So I guess really what it comes down to is. I mean, it's like you know, everybody's heard you know that kind of saying. You know, focus on the 20% that gives you the 80% of the results. And I think there are definitely certain things on Amazon where you know, addressing one singular component has downstream effects on multiple other algorithmic factors and so you get kind of this flywheel you know cumulative effect. That happens, whereas there are other things that maybe they're important things to do on Amazon. But if, given the choice between them, you might want to choose that kind of flywheel effect option versus, say, something that really addresses more of a singular type thing you know, like, for instance, adding, you know, virtual bundles to your listing, like it's useful, there's some help there, but it doesn't necessarily have massive cumulative or snowball like effects on your listing and so you have to make those decisions. You know, which of those aspects of your listing or of your Amazon strategy are you going to focus on?

Speaker 1: 1:32

So, when you're weighing those out, can you give some examples of the ones that you see at the top of your list when you think about the cumulative effects?

Speaker 2: 1:42

Yeah, for sure. I mean you know the easy ones, I think, the ones that you know. It's interesting because there are elements that everybody knows you should work on and to some degree we all do but most of the time we don't place nearly enough emphasis on them and we don't come back to them frequently enough, and that would be our title and our main image. So the reason I would consider those both to be this kind of or at least that they have the potential for having this kind of flywheel effect on your listing and this cumulative, you know growth factor is. Let's take your title, for instance. If your title is written properly, you know it's got a hook, it's using the proper keywords that are going to get you the most response. You know it's got your USP in there, so that there's a differentiator between you and your competitors. Like there's certain things that should be in that title. And if you nail it properly, the interesting thing is is that okay? So the title shows up in the search results, which obviously then is going to have an effect on your click through rate, so the number of sessions that you're going to actually get to the listing. But then, once somebody gets to the listing. Already your title has prepared them for what they're going to see on the listing. Hopefully, and if you presented the hook properly and you presented your USP properly, they're actually going to spend more time on your listing looking for the details that prove out that USP, that actually give them what they want to know about whatever the hook was. So they're going to spend more time on your page, which is an actual factor for Amazon. They want to know how much time is somebody spending on your page. But then beyond that, the more time they spend in your page, the more likely they actually purchase, right. So your conversion rate goes up and your title had that effect, right it? You prep them with that title so it not only affects your your click through rate, but if you've done it properly, it also affects your conversion rate. So what's interesting there is click through rate gives you more sessions. Conversion rate takes more of those sessions and turns them into orders, which means you have orders of magnitude higher sales volume just from a CTR and CVR improvement right. That compounds on each other. But then you have downstream effects from that. So now we've got the algorithm in terms of search ranking not only takes into account click through rate also takes into account conversion rate, also takes into account sales volume and sales volume trajectory. Like those are all factors in the ranking algorithm. So you just improved CTR CVR, you improved your sales volume and your sales volume trajectory is on the upswing right. So all four of those factors, which are major aspects in the ranking algorithm, are better, which means you're going to rank better, and not just for a single keyword, but for every keyword that you're indexed for you're going to rank better. What does ranking better mean? It means more sessions, which means more sales, and all of those sessions now have a better conversion rate than they would have had before. So you get more sales out of those sessions than you would have gotten, which raises your BSR even further, which is a ranking factor which is going to, you know. So you've got this massive cumulative flywheel that goes on from one single change. So if you're not running Amazon experiments on your title all the time to see what you could do better, how you could improve the hook, how you could improve the keywords that you're using, or you know whatever, whatever it is, then you're losing out on that flywheel effect. And again, you know, title is not very sexy, right Like it's. You know there's a lot of these hacks that you see. You know like everybody wants to, you know to focus in on the hacks and some of them are great, but those foundational basics. If you're not focusing in on those things that give you the flywheel, you're missing massive opportunities.

Speaker 1: 5:38

Yeah. So there's a couple of things I want to double click on there. The first is is that I think this idea is this is one of the biggest principles to understand on Amazon, which is incremental improvement can lead to exponential sales. So if you think about it, of hey, I increased my conversion rate by 2%. That may not seem like a big deal, but over the course of 30 to 60 days it may be enough to get you ahead of your competition to where your product goes from fifth organically for some of your major keywords to second organically, and because there's such an exponential difference between those being in the top three versus being in number 10 or below, that just those incremental improvements can give you a massive boost in sales.

Speaker 2: 6:31

Well, and if you think about it too and I wanna hear kind of what man has to say too, but something to think about too when you're looking at the search listings, especially on mobile and mobile, since mobile has become such a massive part of the sales cycle, even more than desktop, for sure. Now, if you look at that and you look at how many sponsored listings show up in there, the further down you are. Like the difference between being number four and being number three is actually more like the difference between being number seven and number three, because you've got multiple sponsored listings that show up in between there. So there's more scrolling necessary. So you know, like, what you're talking about. If the person who's above you and the number you know if you're at number four and somebody else is number three, if the differential between those two listings your listing and theirs is this very minor, you know conversion rate adjustment, like maybe there's only two percent difference. You know between your conversion rate and theirs, but all of a sudden now you make that differential with your title and now you went from number four to number three, but again, that's more like being from number seven to number three. So who knows, you know it might be two, three times as many sales between you know, with that little bit of a jump.

Speaker 3: 7:48

Yeah, one of the things that stood out with what you talked about is the kind of the structure of the title and what's important to put in there. I see so many titles in the search results that are still a keyword salad of just a bunch of keywords that if you read it as a sentence don't really make a whole lot of sense. Like that. You know it's part art, part science with your title, the same thing with your main image, and you know that's. I think so little people pay little attention to the fact that you have to be able. You know, yes, amazon's algorithm is important, but also so is the consumer that ends up hitting the buy now button. So if that doesn't make sense to them and it's just a bunch of keywords that you feel are important, you're missing out on part of the art, part of that listing. So you know it's super important to also pay attention to that when you're doing your experiments is how can you also speak to the consumer at the same time, because that's gonna increase your conversion rate as well.

Speaker 1: 8:45

Yeah, and a couple of things that I think are really important. There is and Mike you pointed out, which is manage my experiments, which I think is probably one of the most powerful and underutilized tools that Amazon gives you, because it really is the only way to accurately split, test different parts of your listing. And it's important to know that it's not just your title or main image although those generally have the biggest impact on your click, your rate and conversion that we're talking about and kind of moving those metrics. But now they've opened up secondary images. They've opened up other aspects. You know I think your A plus content has been open for a while, so you know there's if your product is eligible. You really should be running some sort of experiment all the time, all the time, and I've run enough these experiments to know that you know you never know what works and what doesn't until you test it. You know things that you think are gonna do amazing end up doing worse than you know kind of an okay title that you have. And you know other things that you're like well, this seems a little goofy, but you know we'll give it a try end up converting really, really well.

Speaker 3: 9:54

That's a mantra that I say to myself all the time. It doesn't matter what I think, it matters what they think, absolutely absolutely.

Speaker 2: 10:02

You know, and actually it's very interesting, like I really feel like the best marketers are the ones that really take that to heart and recognize that Like, yes, you should be thinking through what the copy looks like and what it is that you're writing, and you know who is your target and you know what do they care about. All of those things are critical, obviously, and you want somebody who understands that to be writing your copy. But, by the same token, if you don't understand that very foundational principle that you're not selling to you, you're selling to the customer and you don't actually know what they want until they tell you and the easiest way for them to tell you is for you to run the experiment and listen to tell you Yep, test it.

Speaker 1: 10:46

Yeah, well, and I do want to talk a little bit more tactics for titles, specifically, because that's something that I just have a strong interest in because I think it's very underutilized. But before I do that, you know, one thing that also comes to my mind is that I think a lot of people don't know is when it comes to main image, you know there's a lot of questions now like hey, is my main image terms of service compliant? You know what are the boundaries? Because I mean, it is kind of there's got to be some gray area, and I think one of the things that Manage my Experiments does is helps eliminate that gray area, because they all get reviewed by Amazon before you can have that new main image tested. So if it go, if you're able to set up an experiment with that new main image, you're almost guaranteed that Amazon's gonna see it as terms of service compliant.

Speaker 3: 11:41

Yeah.

Speaker 1: 11:42

I don't know if it's a different team or a different algorithm.

Speaker 3: 11:46

I don't know if it's a different team or algorithm, but I've actually had better luck in getting images that kind of were on the gray side of things approved in Manage my Experiments than I have just uploading it to the backend of the listing. I've had better luck with those kind of gray area ones than I have with Manage my Experiments. So there's that too.

Speaker 2: 12:06

So here's a question for you, matt, that I would be interested to know what your experience is there? So, in those situations where you were running a semi gray area type image on Manage Experiments and you got it through the filter right, so you got to run the experiment, how many of those images there were, those kind of more gray images ended up being the winner of the experiment and went live on your listing and went through just fine, even though if you had just simply uploaded them directly to your listing, they wouldn't have Very high.

Speaker 3: 12:39

I mean, I don't want to say 100%, because I there's probably one or two that I'm not remembering, but if they went through. First of all, the images that I use and manage my experiments. In my opinion and I've run enough to know that the gray area that we're talking about is, like you know, a, a, a recipe. Not a recipe, but an ingredient that's let's call it like a topical ingredient that's inside of the lotion, having that like poke out of behind, from behind the packaging that's kind of the stuff that I'm talking about or having food inside of a container. I knew that from running as much of these experiments that it was going to be the winner. But once I was able to sneak it through and manage my experiments and it won I don't I can only think of one time that it didn't actually go through and was went through and was the ended up being the listing or the image that showed up on the listing. So very, very high likelihood that if it wins, it's going to show up on your listing, cool.

Speaker 1: 13:38

Yeah, so I would talk about main image, I don't. I do want to come back to titles, because I think that there's a lot there. You know, the first piece I would say again, for conversions manage my experiments is a fantastic way to see, you know, to actually test which title converts better than the others. And then I also want to come back to what you talked about, matt, which is that combination of art and science, and I think this is what trips a lot of people up is is they're like well, I'm just going to throw my best keywords into there, and a problem with that is that, like you said, at the end of the day somebody has to click on that and buy that product and you know your title is probably 80% of the copy of what they see, and so it really does have to have your, you know, unique selling proposition and you know, or what differentiates you and what's the benefit to the customer, along with those keywords. And what I would add into that is that really needs to be well, two things One, it needs to be in the first 80 characters, because I think what a lot of people forget is that you know 60 to 70% of our shoppers now are on mobile and those titles, yeah, you can put 150, 160 characters in your title, but on mobile they get truncated down to about 80 characters. So, yeah, it's great to put those keywords in there, especially some of the keywords that you know get you the visibility that you need. But just be aware that customers, you know, may not see those benefits if you put them at the very end of your title and so they may not even click on your product to know hey, this is the right product for me.

Speaker 2: 15:19

Yeah, I think I mean and that's really, you know, it's very key, right is just knowing there are sections of your title. You know, like I mean you should be dividing your title up into, to some degree, sections, and that first section is okay. I've got my mobile, visible portion, you know. That's where I need the hook. I need the USP and maybe I need the at least one best keyword that I can use to describe my item. That's a clear description of what it is. It's obvious that that keyword relates to what I'm selling. It's a high volume keyword and it's one that if somebody sees that keyword in my title, it you know, and that's what they search for it is very likely that they're interested in my product, and so it preps them, you know. So, if I've got the USP, I've got the hook and I've got the keyword. And also, I think it's relevant here to take a step back and recognize too, if your brand is not really a recognized brand that people are searching for. If you can get away with it, do not put your brand at the front of the title. In my opinion there might be some people who would disagree with me, but I say, put it in the at the end. You know, you know if they're not looking for you now you've just taken up especially. I mean like okay, if your brand is like six characters, fine, maybe put it at the beginning. But you know, like pro top products, you know, like our brand, it takes up like 15 characters at the beginning of our title. Now, for us, nowadays they're searching for our brand, like a lot of people are searching for a brand, so I use it. Now sometimes I just put pro tough, if I can get away with it, because they still know it's us. You know, but just be careful with that, because you use a lot of characters, the brand name, and if nobody's really looking for your brand, you need to decide whether you actually want your brand up front.

Speaker 3: 17:10

Yeah, categories exist now that don't require it. Every category that I sell and it requires the brand name in the front.

Speaker 2: 17:16

now, I've been able to get away with it. I still don't always enter stuff at the beginning. Now I will say it's dependent. It changes listing to listing. Sometimes it goes, sometimes it doesn't so interesting.

Speaker 1: 17:29

Yeah, so doing some testing with this here. My experience with that has been and I've never I've yet to have this happen. I've only heard it from other people. So I've done both brand up front or, you know, just straight into the benefits. And, like you said, mike, there's definitely an advantage if you're not a big brand name of and or if you have a long brand name and you know it takes up 20 characters. I mean that really leaves you with very little real estate to go through those benefits. So my understanding in and again, category to category. But as far as the Amazon rules, you're supposed to have your brand name up front and there is a small chance that if you don't have your brand name up, brand name up front, that Amazon will lock you out of making any changes to that listing for not complying to that rule. Now, I think that that's one of those. You know kind of one in 10,000, you know type issues. So it's important to be aware that what Amazon wants and what the potential consequences are. So that way you can weigh it with, kind of, where you're at in your brand and, like you said, maybe you start out when you're just getting out of the gate and you really focus on the benefits and you don't have that brand name up front and then, as the product matures, you can always make that change in order to eliminate that risk. Right, yeah.

Speaker 3: 18:55

It's such valuable real estate. Those first 1520 characters, like we've already mentioned, I mean, if you can get away with it, then I mean certainly that. I mean for me, even if I have a brand name that is starting to get recognizable, like I would rather use those first, you know, 1520 characters for as many as the hook and the USP and all of that stuff. So yeah, if it's certainly if you get away, though, but I can't in my categories, unfortunately.

Speaker 1: 19:21

Yeah, the other thing that I would encourage while I guess what are your guys thoughts around, because this is one of the areas where I wish manage my experiments was better. What are your thoughts around measuring CTR for a new title, because obviously you can see conversion rates, but manage my experiments doesn't give you any click through rate data.

Speaker 2: 19:49

Now do you get I can't remember with with manage my experience do you get, though number of sessions that went through that they're going to be equal? It's an equal number of sessions. Yeah, I don't know, I'm not sure if there's a good way. Honestly, I mean the only thing that I could think, and even then it's not perfect. But if the experiment that you're running is actually current title versus new title, then you could theoretically look at what was your CTR run up. If you went into your, say, your ads, or even your search career performance, I guess basically you could look at your run up CTR for the previous week on that product and then you could compare it to the CTR of that product during the week that the experiment ran. And whatever that differential is, you would have to attribute and it would actually be technically double that right, because only 50% of the traffic over that seven days of the experiment went to the new one. So if your CTR went up by, say we'll just make it easy numbers if it went up by 10 points, then technically that's a 20 point increase in CTR. I think that's how the math would work out, because only half of your traffic was actually running through the higher CTR. Title Does that make sense, would that be true?

Speaker 1: 21:25

Yeah, I think, because we do look at the search career report to kind of get a gauge on that, although we do take a little bit longer look than at least a few weeks prior, to kind of try to get what that baseline is, to identify what that baseline is and then look at it as it's running the experiment. But yeah, you bring up a good point in the sense of if your click through rate went up 10 points, then considering that that traffic is split between your old title and the new one that you're testing, really your results are going to be in one direction or the other double what you're seeing in your search query report. And then I think the other kind of detail that's important for folks to know is that use your search query report or use your Amazon ads, but don't compare those two numbers together because they're measured very, very differently. So you could pull for those. If you're looking at click through rates for your ads and if you've got exact match hero campaigns, then you can do that and get some data off of that to compare your click through rates. But don't combine that with your search query report because those numbers are calculated differently.

Speaker 2: 22:45

Yeah, for sure. The other thing that you'd have to pay attention to also if you're going to try to use that data in that way to verify a CTR improvement or decrease would be you have to be careful that you hadn't made any significant changes on the PPC side in terms of what keywords you were advertising on or not, because if you're looking at an overall CTR and previously you were advertising on some keyword, that was crappy CTR and you just removed it from its campaign all of a sudden. Now you've skewed those numbers already and it had nothing to do with the title. So you'd have to be careful there that you're not corrupting that data.

Speaker 1: 23:23

Yeah, I really think that the search query report is a much better way of tracking that than trying to do it through ads and search query.

Speaker 2: 23:31

correct me if I'm wrong SQP. Is that solely organic traffic that's being tracked there, or is it cumulative Everything?

Speaker 1: 23:44

I don't remember off the top of my head.

Speaker 2: 23:47

Because that would be. The only thing is that potentially now at least, it would be less corrupted than the PPC data, which is specific, and if you had changed out your keywords or anything in a significant way that could be affected. Sqp would be less, because at least half of it, hopefully, is organic. But if it's all organic, then that would eliminate that issue entirely and I'm not sure the answer would be simple.

Speaker 1: 24:10

Yeah, that's a really good question. I'll have to look. For some reason, my guess is that it is organic traffic. It's not total traffic. But I'd have to take a look at the Amazon console because I think if you dig into some of the details it does tell you whether it's organic traffic or total traffic.

Speaker 2: 24:32

Yeah, I wanted to pop back to something that we talked about a few minutes ago. You were talking about the. Sometimes those small changes can have more of an exponential effect. We kind of referred to that idea of even just a slight ranking change could have a very significant effect. I wanted to relate that to. We've had a conversation with Brian Johnson about his tool, deepm. If he was talking about ranking and how you basically have ranking shelves where you might have a cluster of, say, 10 or 20 competitors who essentially are we'll say, for lack of any better way to say it identical in terms of algorithmically how Amazon is looking at them. They kind of formed this little shelf of very similar metrics. If you're sitting below that shelf and you make a very small change, you could easily jump 20 spots because you jumped that entire shelf, because all of those products are basically all in that grouping of metrics and you leaf frog that entire shelf. Same principle here a very small change, especially like a title change that could improve CTR, cvr, sessions, orders, all of that which all of a sudden now puts you above in multiple categories of everybody. In that shelf you could jump from number 25 to number five.

Speaker 1: 26:13

Yeah, I think that's an important point too. It really comes back to we like to think of things as linear, but Amazon works in exponentials and understanding that incremental improvement can have a huge impact.

Speaker 3: 26:26

That was such a fascinating concept that makes total sense, but not one that I had ever thought of before he mentioned it. Imagine the effect of that. You just talked about the difference of seven and three being much more prominent on mobile, but imagine a 10 position change or a 15 position change because you've increased over that shelf. That's massive.

Speaker 2: 26:50

It truly is massive, especially because, like we were saying, amazon throws so many sponsored listings in there these days that if you're number 20, you're actually more like number 35 at this point because of all the sponsored listings.

Speaker 1: 27:06

For the folks that are listening. What do you think would be a good takeaway item for them to where they can at least start down this path in order to take advantage of these incremental improvements?

Speaker 3: 27:22

The first thing is test, use, manage my experiment. That's first. Off and foremost is, if you're not test, always testing, you should always have at least one experiment going, always, always, whether it be your main image or your title, something should be testing. It's fascinating to me how many sellers aren't using that tool effectively.

Speaker 2: 27:41

I would take that a step further and I'll put myself on the hot seat with this, because we have not been very good about consistently using manage my experiments. It's like here I'm teaching this and it's true. There's nothing. All of what I'm saying is absolutely true, but, just like every other business, you're always constrained by how many man hours do you have available to you, based on the size of your team and what other projects you're working on and whatnot, and so some things get left, because the title and main image oftentimes are just so obvious that it doesn't even happen. And manage my experiments, the same way. We are looking at the possibility of even bringing in somebody that that's all their job is is to do experiments on all of our listings all the time and be managing them, because if we don't have the manpower on our team to do it it is such an important piece of what we should be doing then we better bring somebody on. Even if and it doesn't have to be like manage my experience it's not that hard, like I mean literally it's just pick something to change like look at a competitor, see what they're doing that looks like maybe it's better than what you're doing. Implement that in your title or on your listing. Run the experiment and let Amazon tell you which one is the better one. You don't have to hire a rockstar to do that, like you can hire a fairly basic freelancer that has good English, and that's sufficient. So if you don't have enough hours in the day, bring somebody on the team to do it, because you need to be doing this.

Speaker 1: 29:21

Yeah, I will push back a little bit on that though, mike, just because you know what what I see is. You know, a lot of people get frustrated because they don't get results from it, because they're not doing it. In thinking about it systematically, as far as, like doing the research up front of looking at their search query report to know, like, hey, which are the most important keywords, which ones am I actually, you know, finding that formula where I'm not, I'm getting more clicks or my click rate is higher than my impression rate, you know pulling down that data of, like you said, from the customer's perspective, looking through your reviews and your competitors reviews, to know, like, hey, what do customers love about my product but you know they don't like about my competitors product, where you can identify those pain points, and doing that research. I think it's definitely trainable to do, which is completely fair.

Speaker 2: 30:11

I agree with you. I agree with you that those are all great things that you probably should be doing. What I'm saying is if the difference between you not doing it and doing it is that you don't have somebody on your team and you don't really feel like you have the money to invest in somebody who's a little bit higher level and kind of understand some of those you know principles literally Just having somebody throw up two titles that even just that person thinks maybe one's better than the other. We already said earlier in this conversation you don't ever know and I would be shocked if a Per, if you had somebody on your team that was running experiments on your title and they just once a week change the title to something else that they thought might operate better, and they had even just a rudimentary understanding of how titles operate and how copy should be right, in Three months you probably still have a better title than you had three months ago. Mm-hmm, that's getting better click-through, rating, conversion rate, even if they know almost nothing. Now if you got somebody on your team who knows what they're doing, it could be exponentially better than that. But I'm just saying, taking those two scenarios I'm not going to do it or I am going to do it.

Speaker 1: 31:26

Yeah, it's definitely better to you know, move forward rather than worrying about perfect. I'll agree with that, although I will say, just so, that way my experience has been that you know, with those tests, you generally need at least a few weeks worth of data before Amazon Experiments has enough to really tell you whether what you've done is better or not. So I mean, yeah, it's good to be doing something, but also understand that you probably, you know, one experiment a month is probably more realistic than one experiment a week. If you're using manage my experiments.

Speaker 2: 32:03

Yeah, I think I mean ultimately it depends on volume, right? I mean like, yeah, yeah, more volume got you've got going through your listing, the faster Amazon can determine and also the larger the differential in performance between the two. No, if you got a title that's just way out performing it dance, far less data for Amazon to determine, that you know. And if there's only a slight difference, then you need a lot more data. So, but if you got volume running through, you know you can do an experiment in a week and you can probably get to it to to a result. But if it's a lower volume product, expect that it's going to take longer. But either way, whether it takes a week or a month, you still should be running it, yeah.

Speaker 1: 32:37

Yeah, that's. Yeah, that's a hundred percent. The takeaway is that you know for anybody who's listening to this is you should be. I would start with your hero products, so figure out what those you know top. You know you probably know off the top of your head what your top two, three or five products are. Start there and Then work your way down and you know, work to a point where you really should every product that is eligible for it. You should be running some sort of Experiment all the time, whether that's on the title or the main image, because it can only help you when it comes to, you know, conversion rate and moving, making that incremental Improvement that can exponentially boost your sales. Yeah, yeah, for sure, fantastic. Well, thank you everybody who for listening, love to hear your feedback on and results from running manage my experiments. And until next time, we'll have next week another Amazon tactics Tuesday.