Artwork

コンテンツは Amazon FBA Seller Round Table and Amazon FBA Seller Round Table - Selling On Amazon によって提供されます。エピソード、グラフィック、ポッドキャストの説明を含むすべてのポッドキャスト コンテンツは、Amazon FBA Seller Round Table and Amazon FBA Seller Round Table - Selling On Amazon またはそのポッドキャスト プラットフォーム パートナーによって直接アップロードされ、提供されます。誰かがあなたの著作権で保護された作品をあなたの許可なく使用していると思われる場合は、ここで概説されているプロセスに従うことができますhttps://ja.player.fm/legal
Player FM -ポッドキャストアプリ
Player FMアプリでオフラインにしPlayer FMう!

Get to Know More About Handling Tickets and Amazon PPC - Amazon Marketing Tips with Danny McMillan - Part 2

41:11
 
シェア
 

Manage episode 305561841 series 2492296
コンテンツは Amazon FBA Seller Round Table and Amazon FBA Seller Round Table - Selling On Amazon によって提供されます。エピソード、グラフィック、ポッドキャストの説明を含むすべてのポッドキャスト コンテンツは、Amazon FBA Seller Round Table and Amazon FBA Seller Round Table - Selling On Amazon またはそのポッドキャスト プラットフォーム パートナーによって直接アップロードされ、提供されます。誰かがあなたの著作権で保護された作品をあなたの許可なく使用していると思われる場合は、ここで概説されているプロセスに従うことができますhttps://ja.player.fm/legal
Things we discussed in this session:

A. Part 1

Things we mention in this session of Seller Round Table:

Join us every Tuesday at 1:00 PM PST for Live Q&A and Bonus Content at https://sellerroundtable.com

Try the greatest Amazon seller tools on the planet free for 30 days at https://sellerseo.com/
Transcription in this episode:
[00:00:01] spk_0: Welcome [00:00:02] spk_1: to the seller roundtable e commerce coaching and business strategies with and er not and [00:00:07] spk_0: amy Wiis, [00:00:08] spk_1: it makes a lot of sense. The other thing that I've noticed too is um I will open uh different tickets and you know, when you, when you go through the initial ticket process, you have to pick a category and subcategory. What I'll do is I will open two simultaneous most every time. Now that I've learned this works really well as you go to simultaneous. I know like you said, Danny amazon might hate this, but if you actually are trying to get something done, you go and pick, you know, the first one you do is the one that is in the category that makes the most sense. Then the next one you go is to something that's maybe part of the problem, but not the main category and you open a ticket there as well. Um I've found that, you know, because you have now kind of two separate groups working on it, One might hit a wall and the other one might find a solution or vice versa. So that's another kind of little trick that I've, I've found out over the years with all the hundreds of tickets I've had to open. [00:00:58] spk_0: Yeah, So here's another thing which quite interest. There is no financial incentives for Associates. All the targets are based on efficiency and quality of where they have met the correct standard. Right? So on the flip side of that, if they don't meet the targets, they will get closely monitored, Right? So it actually becomes more painful from, because the QA teams are going to be more and more on their back to to meet certain standards. And then also opens up that opportunity for them to earn other money elsewhere. Which is what happened um when a bunch of the support reps were caught in this where they were doing certain sellers favors. And um, so it's like kind of bad that they don't get any kind of monetary compensation because their pay probably isn't great and you know, it kind of sets it up. Yes. So the reward of being efficient and good quality is the opportunity for lots of promotions. It's always encouraged. So that is your upside. So while she don't get a financial there, you'll get out of tier one, say a T one of seller central and then you quickly get moved up. If you're good in the excel, what you do. But yeah, you're always going to get people on the take. Now this is the thing and I don't want to tie in with the people that are on the tape because there's good and bad people on this planet as we know, right? Um, but when there is an issue with a ticket but let's just say yeah and icing is being yanked with that. As an example as a ticket. They can log into your seller central gap. So any of these reps can access any of our accounts but purely based on the condition relating to a ticket. Now, what you would say, well what happens when reps are working from home? Because a lot of them do. They distributed they have a token system which is all generated on fingerprint I. D. So with the home systems they're very secure as well. So I want to make sure that people understand that that's clear. But I found that kind of interesting as well. And the 17-18 minutes. Um He's annotate this year. So that's something how a ticket is broken down. If this is any [00:03:14] spk_1: help for [00:03:15] spk_0: The audience 17-18 minutes. Right? So there's four parts to a case ticket outcome, one pending amazon action P. A. In case amazon needs to perform an additional research and the seller needs to wait for the outcome. Number two perch in pending merchant action PMA More information is needed from the seller to be of support. The current information provides limited or incomplete For us to proceed as support. # three resolved cases done and resolved or the conclusion is that amazon point of view, we cannot be of support whatsoever reason. And we need to inform that the seller that we cannot resolve this number four reply and keep working. This is used with more time is needed to perform research and want to notify the seller in the meantime that we're working on it and we'll come back to them as soon as we have an answer and reply. So you've got a key thing there. The other thing is that annotate each of the tickets so that everyone is singing from the same hymn sheet. They use all the same S. O. Ps. And everything else. Where is the [00:04:22] spk_1: ticket information? [00:04:24] spk_0: I'm going [00:04:24] spk_1: to tell you about [00:04:25] spk_0: annotations are added to open tickets. So this is what goes on your ticket as you do it. Number one summary of the seller issue to research done, which work followed free assessing order numbers info plus S. O. P. U. R. L. S. Followed for troubleshooting, which means they've got that catalog of uh you are the S. O. Ps. And they put that in the case stick at the U. R. L. So whoever picks it up can go to it. Plus there's lots of tools use like this guy that I'm speaking to has been there three months. He's now using 60 seller tools to support at his skill level. And he's only been there three months and there's a lot more tools involved in that um tools and tt trouble ticket number when escalated ticket has been created by an associate number four proposed action, taken summary or reply question to set up. So I think every ticket that they do, they have to annotate that ticket in four parts. I think someone said about a while ago because one of the things I want to go down the road of is performance team and stuff, you might not be able to get much from there. But I remember speaking to people before, we do a lot of the suspension um reinstatements and stuff, they were always talking about annotating. So if you've got an account and you kept sending the wrong information it gets annotated. That's going to mark hard against you because you've not paid attention to the point. They may not be able to recover that suspension if you push too many times. So just be aware is that everything we do, they're being monitored. And also every ticket raised as the four parts to its summary and the problem solving issues. [00:06:04] spk_1: That's crazy. You could really go down that rabbit hole for a long, long time. There's so much there. Um The other thing I had a quick point I wanted to make though for everyone listening or watching is is I used to you know, everybody hates filling out surveys or even those little things, but when you get us make sure that you leave positive reinforcement for the the agents that give you good support and negative for the ones that don't because amazon uses that feedback loop and all that will do will support sellers better. Right? If you get people who suck and and they weave those people out, then that's that means that next time your ticket is gonna be handled by somebody who has been reviewed positively or have got has gotten positive feedback. So I encourage you guys to do that. I used to never do it, but the last couple of years I've been doing it just because I'm like, I need to provide, you know, feedback loop. So that's, you know, so amazon can actually take action on on the ones who perform well and the ones who don't. So, um, that's my asking suggestion to you guys. [00:06:58] spk_0: This is one thing I want to quickly tell you about the frustrated button. I haven't checked yet, but he was telling me because I didn't get a chance to take and I don't know if it's fully correct by a double check with him for next week's show because I'm going to do another one. But he says when you have problems and you've raised the tickets, if you hit the frustrated button right diet that the frustrated tickets are higher, higher rent. But some people like to listen, I haven't seen the frustrated, but I haven't looked yet, but I'll get that confirmation and if that's true, if there is, once you've raised the ticket and you've got, you know, like normal ego, was this any good yes or no, if there is a frustrated button there, if you are frustrated, hit that because apparently that they will rank frustrated tickets higher, which is saying I didn't know about. And the captive team thing, I remember a couple of years ago, people going, yeah, well called captive team, I need to follow out and no one knew the captive team was right now, we don't know if it's true or not, but the captive team might just be made up. It doesn't exist. It just might be a posh name for Tier two or Tier three and there's no such thing as a captive name, but it was a really good word to use a buzzword if you like, if you're selling uh masterminds, you know, do you know what I mean? I think you know, and things can come up into the amazons there that just don't make any sense to maybe not true, but it could be true, but he's been there only three months, so I'm trying to get him to dig and see if there was ever a captive team going back three or five years. [00:08:26] spk_1: It's interesting, I think there was, or like you said Danny, that might have just been kind of what people called it. But I know that there was, uh, I went to college in North Dakota without going too far and, and a lot of people don't know this, but amazon has a big foothold in North Dakota. And so I used to have phone numbers because I had friends that work there like, way in the past, but in the, in the past you would have these special phone numbers that if you knew, you could actually [00:08:47] spk_0: call and [00:08:48] spk_1: get somebody on the phone. Like I used to be able to call the catalog team directly. Like a lot of those direct connection numbers have, [00:08:54] spk_0: yeah, yeah. [00:08:57] spk_1: Have gone away. But even that though they've gone away because amazon is now, it, the word gets out and then, and then it's useless because it just gets inundated. Right? Same thing with Jeff at amazon dot com. Good luck using that these days because you know it's like the word was out. That that's how you could get a quick escalation. [00:09:12] spk_0: This is the other thing I'm like with the shows I'm going to be I have to be sensible if he turns around and gives me a bit of information and that you know tens of 1000 people jump on and it's just gonna flip. It doesn't help anyone. What I'm trying to do with the series is really educate people on so that they can take away from the shows and go do you know what? I'm gonna write my ticket like this now I'm gonna be extra nice to this person on here. Oh to resolve that if I go to this I want to give them all the ideas of like how can we create an S. O. P. L. Creating the master ticket. What is the most optimized way of developing a ticket to get the best result possible rather than saying, you know, here's the captive team if it exists there telephone number and then blurting it out on seller sessions. But do you know what I mean? It was just disappoint. It doesn't [00:10:02] spk_1: Absolutely. All right guys we're going to switch to something else. But if you want to learn more, make sure you subscribe to seller sessions. It is one of the better podcasts I've listened to many. I think Danny does a really great job. So go check that out. Great info. There are lots of great people there. Um It's on all the major platforms. Go check that out. Uh Let's transition though dan. Because one of my favorite things which is some of your favorite things is you know uh you know algorithms and PBC. So um in terms of you know, the the the A. Nine algorithm, I would love to hear, kind of your insight and maybe some of the ways that you think that it's uh progressing and maturing and changing. Um I would love to to dive into that. [00:10:46] spk_0: Okay, so effectively what I did is a few months ago I dug out the, what I found, it started with the video and which is a Doria Sonika from the A. 19. And then I started to find research papers like amazon dot science. What had worked out is A nine works completely different to what most people talk about. And once I got hold of the information, I sat down for a few weeks and going, how can I turn this into layman's term? Because for you Andy we can geek out and I'm not a developer. You are, but you're Yeah, that's great. But for general listeners, they've got to try and understand it. So for how do we take this, which would normally send a glass eye to sleep and make it more plausible? So we broke it into into different into different sections. But the key thing that I find with a nine is that it operates differently to a lot of the way that people talk about it. And so when we look at list and we think of title bullet points, blah blah blah, right, But we don't think of where did it get there? We know that the conversion make makes an impact. We know that add to cart is massive but add to cart only works as far as page one unique keywords with selves to bind past the top of page one. So we know that add a car is a massive massive signal, but when you break it out, like for instance, no one was talking about negative and positive labels. So when you think of negative and positive labels, if you're looking at your listing, you could be saying to yourself, well we have a conversion which is on the continuum. If someone lands on your on your product page and they stay there and they don't bounce off, there's a micro conversion, like a credit going on there because the engagement, you scroll your engaging, right? You add to cart, it's another signal you're engaging. So you've got to think of conversions on the, on the continuum. The biggest thing I took away from all the A. Nine stuff is amazon is obviously about relevance, but they want to know just as much about what you didn't want to, what you did, right and what that is that it falls under these training sets fall under negative and positive labels. Now, positive labels is clicks add two car engagement and things like that different. And it's all derived around behavioral aspects as part of their training sets, which make up as part as the algorithm. So when you're writing your listing and you're saying, I want to write the future benefits and everything in the bullet points, you would think about it from a copyrighting point of view and the human person reading point of view. But if you took it into Afridi mode, you'll then start to think about it from a negative label and a positive labour point of view. And then you start getting to things like the hunger score, which is very complex for people to really understand. But Hunger Score, you could uh consider that is foam. Oh, so what the hunger score does. It searches down by category, but it's looking for the stuff that you are not interested in, don't touch. And I think what that really is about is that they're trying to unearth if there's anything they've actually missed. So they're collecting all this information. So it all comes down to relevance. You've got on one side, you've got the positive label. The other side, you got the negative label. Let's look at another reason why the positive and negative labels would work if you'll send in traffic, PPC traffic to listen. We now know that if you send traffic there and the conversion rate is poor, that's going to have a negative impact organically, right? Goes back negative labels, positive labels. What is under the negative labels? Well, it's unseen. It's why is that information not being utilized, their etcetera? So when you are looking at optimizing your listings, think about it. Will this fall under a negative label? Will it fall under a positive label? On top of what you're already thinking about when you're optimizing a listed. So it goes back from being a static vector if you like that. Here's our listing is very static and we're looking at one dimensional, let's look at it from the freedom dimensional point of view. We know that most of it is based around um Machine learning goes back as far as 2016. There was using 100 machine learning models. We know the importance of being paid one. I'll give you some data here, 70% of Amazon customers never click past the first page of search results. Then you've got 35% of amazon. Shoppers click on the first product featured on the search page. Uh and the first three items displayed in search results account for 64% of the clicks. And this is why P Pcs come to more for amazon's the paper place. That is a great model for them to earn money because everyone who is ranking organically up here, they're being pushed down there in like the active position on the page, not the Organic or the ad position. Organics is sitting down in position overall of number four and number five. So a lot of the PPC clicks, the top researches could become more expensive. So this way it's so important to dial in. The other thing is if we take some stuff off the table, there's no such thing as 18. It doesn't exist. It's not in any of the science literature, none of the engineers that I've spoken to have talked about it. And just to give you one quote from amazon here, small improvements in relevance can positively impact millions of shoppers. If that's the case, why would you take a nine, Throw it in the dustbin and dial out a 10? You wouldn't and it's not structured that way. Right. What other people don't realize as well? It's based down catalogues, indexes, categories each marketplace. So when we see a push or we observe something and we see listings have a a sharp fall if you like you. Alright son, algorithm update. That isn't happening in the UK when it's on the U. S. That's not happening in Germany, these training sets and these, these periods that they are measuring this, this data and they collect that information or the behavioral day asia. Then they're going to do the push and when they do the update, it's on that marketplace. So it's not global order. These teams are working differently on the different marketplace. So indexes and everything, a different relevancy is going to be slightly different marketplaces, but overall there yes, are using the full structure of a nine and this is why add to cart works so well and it will carry on working so well for a long period of time. Could it's such a machine, you can't turn, you can't steer the ship around and start going, oh well we're not going to use add to cart no more is a really strong signal. Do you see what I mean? So this is something that has developed over time and I think it's really important that people I think about. I think my key takeaways from kind of your sum up of the in an algorithm is that number one we need to think about more than just. Okay. I have the right keywords. You know, like I have the right keywords and my listing looks good. We need to be thinking about our products from how from what page ones we want to be on. I'm always talking about them. What's your page one? Because of course top of search is so important, but the right search is so important, right? We want to be the most relevant so that they add to cart right so that we can get that going on. So I'm hearing that with the ever increasing cost to get to top of search, it's more important to really, really dial in like what are those search terms that that customer that is so relevant for your products are going after. And you know, I see people are just even more successful now. You know, we recently had some really great folks coming and talking our mastermind group about this ranking strategy that they're using and they're really focused in on some key words that are that are the most important and very few keywords to begin with before they're moving on to others. So I think when we're thinking about our keyword strategy and our advertising strategy now, more than ever, we really really need to know what our page ones are and make sure that we're paying attention to how customers interact with our listing. Not only are conversion rates, but where are we converting? How are we converting? When are we giving that awesome add to cart, you know, what about you Danny? What do you recommend for people in really reducing their costs for their PPC and choosing the right keywords so that they can get the most out of the algorithms favor. I think the newer people are definitely burning their money because they're getting these massive long keyword lists from really great for party tools. But if they're not in the brand analytics report, most of it is junk. It's garbage. Right? So I always say to people take you a long list of 100 or so Wash that list against your brand analytics report. You might find it's only five in there and so you're chasing and pushing all of this data around on hundreds of keywords which are mostly useless because they've got less than 10 searches a day or less on the us market. When you say brand analytics report, are you talking about putting your ace in into that search terms, report and seeing where you have the most lecture? Yeah. Or your broad keywords, it doesn't matter accusations or not. What you're looking for is do my keywords appear in the brand analytics report? Let's just say you sell beard oil. So you put beard oil and then you look at all the different permutations of beard, organic beard old blood uh men's beard oil and then you start to see him and you see the frequency rank, you know that's 1 28,504 and you're looking for that. But what I'm trying to get across what can be very scary for people. We know that the U. S. Market has great long term and it's good to have long term. There's no doubt about it. And it makes up sells Brandon Young does some really good stuff with that in the way that he operates and I totally agree with it. But what you got to be careful of if you've not at Brandon's level and know what you're doing right, you might have misgivings thinking great, this tool has just given me it's got 100 keywords, I've got plenty of opportunities. You put it through the brand analytics report, there's four of them are in there Then your opportunity is gone from, you know the other 96 has disappeared, you've got four and you're gonna have also ran keywords and then you die in or one or two and you might not be able to afford the PPC on it Or they may be so far back that even if you run PBC on them, you're not going to get enough cells velocity to get anywhere with those keywords. Great. You might be on page one because you know it's got 20 searches a month but that's what it's useful as a chocolate ashtray. You need get their balance right because if the keywords too big and you run PPC on it, you don't get impression share. This is why people get 20 bucks a click and then they spend all this money and they don't realize that I will keep this keyword because we have like a calculation on joined me to quickly go for a quick calculation to give people when they're looking at keywords because I think it starts to bring home. Absolutely. [00:22:11] spk_1: And the other thing quickly I want to mention just I think what what Danny is saying and correct me if I'm wrong here, Danny is rather than go after the 1000 keywords, you know that that these tools are spitting out every iteration and what they literally do is get the key word than at an A to it and do a search and then So I mean you might get one or two sales a month from those keywords. But the other thing you have to realize is the more keywords you have, the harder it is to manage, right? It's the 80 20 rule. If you [00:22:35] spk_0: have, you know, a [00:22:36] spk_1: 1000 keywords than having to go through and try to optimize all those keywords is going to be really difficult rather than really focusing in on that that precise target and then just continuously optimizing those keywords, those small sets of keywords and even on your, you know, products, [00:22:52] spk_0: your [00:22:53] spk_1: product targeting etcetera, [00:22:55] spk_0: would you mind if I share my screen? It might be a bit easier. Just love that bit of context. All right, So let me just pull this up. Can you see that? Okay, so I've just I've just mentioned here never go beyond the brand analytics report. You might you know, there's an exception to the rule and you're going to get some people go, yeah, I had a few cells but generally if it's 10 searches a day, You're looking 300 searches a month time you get into position 12 or three. A percentage of that is the click through rate from the click through rate. How many sales are gonna get off that when that shared against some people on that page? Not a lot. Right? So you got to think about it that way. This is what I'm going to give you an example of here. So ranking with amazon PBC, right? So if your PPC is competitive conversion rate and generate enough orders, it would definitely boost your organic rankings here at each bar represents one week's worth of data over their period of eight weeks. So in the left column, we see the average organic rank, right? And then where the smallest bars show rank one is as good as you're gonna get because it's gone from he's going to go, it's gone from up in the gods to position number one over a period of time. So we've got organic ranked number one on the first column, number two on the second and number one on the third column. Each of those represents eight weeks as you see them coming down into the number one position. So you can see these move quickly towards rank one over the first month. Here, the remaining columns, we see impressions orders and spends. Uh These are mostly trending upwards, possibly plateau in after the first month or so. So you can see impressions as the organic rent comes down. The impressions increase as you can see that. Not exactly, but you know, is heading that direction and then you got the oldest column and the spend column. So you can see the top one is where we spent most of the money But it got 219 orders from it. Now in order to stuff like this to work. When you got to be dialed in on your keywords, right? And to your listing should be very well optimized. This conversion rate plays a massive factor in the success as a bad conversion rate will become a hindrance to organic rank, pulling you under water every time you send non converting traffic, like I mentioned earlier in the show, conversion works on the continuum. So the algorithm is the multi faceted view of a conversion rather than what we see as a cell. Right? When a customer clicks on your product, you effectively getting a microbe conversion credit on that query. If the customer spends time when your product page and does not bounce off, you're getting another micro conversion, the more time you spend on the page in principle to hide the engagement and when the customer is a product of car again, you're getting another positive signals. So I skipped this abandonment stuff at the moment. But there's a breakdown, right? So in order to set this up, you guys know this very, very basic setup but you can do this across multiple campaigns. one key word, 1 18 1 campaign isolated. Like this means that we can measure the top of search on a granular level, right? Because for sponsored products, top research placement statistics is on the campaign level. Right? So that's why would set them up that way. So going back to the algorithm here, the ranking algorithm compares many different factors. Two of them all the history and the conversion rate for a specific product listing and that said keyword. Right? So they're all attuned in from those two components down to their individualized keywords left. This is then compared to all of your competitors on the same keyword, If you get a higher conversion rate through PPC cells, then you're gaining points on both of these metrics. And what we've seen in house. Your relevance will also get a boost from this. But then we face the problems we've ranking high velocity products which I touched on before south uh slides. This comes back down to the ranking Algo. There are only so many add positions or placement worthy rivals, which is obviously on page one. So for a high velocity keyword with established assassin have already converting. It could be extraordinary difficult to break through and get impressions even with a phenomenal conversion rate. This is why you hear people bidding $20 a click without getting any impressions. So depending on your stem manner and if you keep going over time, amazon's algo may determine to favor your campaign, but there's no telling how long this will be. It could be that one or more your competitors go out of stock and that's where you get to sneak in. Right again. This is where it could give you a nice booster. So what is the answer? Is there a special calculation or the answer is yes. Known common sense. So these are arbitrary numbers here just to make a point of the demonstration. Right? So what we have here is K one is keyword one. K two is key word to This represents the ad placements. P one, p 2, P three. Okay, so you can see that keyword one super high velocity. We've made up the numbers here to make them stand out. Right? So if you was going for placement one, you'd need to sell 1000 units today to get to placement one. You're not gonna do that right? It's unrealistic. Replacement number 2, 500 replacement number 3, 250. Again, pretty much unrealistic. Unless you've got deep pockets Then you look at keyword # two. So you're looking for the sweet spot, right? You're looking for their midtown keywords. Then you go, Okay, where is the where is the low hanging fruit? Well, position number four, I only need five cells. So if we use the data point to you, how many orders a day can the top of search placement generate if expressed at a 20% conversion rate? So to get to position number four, you might aim for at least five orders a day. Right? So it's more plausible to reach. If your conversion rate is 20%, that might mean five orders a day, divided by conversion rate is equal to 25 clicks. So if the cost per click is $5 for that position it will cost you $5 divide by conversion rate which ends up being $25 per acquisition on average. Or 100 and $25 per day. But that's only when it converts right, which comes back to the point of gaming impression share and the cost of running expensive top research campaigns and usually a market these red hot. Generally we have a higher cost per click. So how do we launch PPC and then taper off knowing where our numbers land for break even because this is kind of difficult. So we kind of come up with an option here but they all present themselves their own pros and cons. Right? So if we look at either way, you're going to need to pick your poison, adjusting the bid out of the gate. So if you make bid adjustment, this is almost certainly affect your position in the ad placement algorithm. So we could do start putting back on the daily budget versus decreasing the bid and maybe losing on the position there, at least for the budget caps. Your campaign would run in most cases, but if you lose advertising for the rest of the day after your budget is burned. So what I mean by that, if you adjust the bidding and you're in position number two, you may come down to number four because you've adjusted that cost per click bid. But if you adjust the campaign cap on a daily basis, you'll get those positions but you're likely to run out of budget at some point in the day. So that's why you play with a budget on a daily basis rather than touch the click because you want to maintain the position. But you could have a hybrid version. So twice a day you mainly adjust the top of search boost in your campaign So you can go and boost that to 900, then 12 hours later dropped to zero. And then you work out sweet spot. You play around with those two factors and find that sweet spot. And then over time you start to get your ranking costs under control. All you don't because it's always a risk, right? So you're looking to rank middle key words that say take 46 to 8 week period. By the time you get to eight weeks, you're looking to do the pullback process and Danny. I'm always surprised at how many people don't know about placements and how to adjust those percentages. So there's many ways to skin a cat as we know, but that's one way that we're trying to present something that's quite complex And there's way too many variables to say this is the absolute way around it. But by working off an assumption of a calculation at 20%. Uh, and using ad placement is your reference. It may help people that going after really big keywords realized I can't go after those keywords because a lot of people won't do that calculation. They won't sit there and think, oh yeah, in order to be successful with this and start using that keyword to ring. I mean I need to get 50 p. p. c. sales today. That's really hard. Do you know what I mean? Like when you don't think about it, especially if you're starting with zero campaign history, you've got no conversion history. You've got newly launched listed. It's no wonder you don't get the impression share. Uh, so that's why you have to be realistic and start with longer tell keywords and work your way up unless you use the beard or analogy beard all be the number one key word. But what about if it's organic beard or so you start in your campaigns with organic beard or and try and rank that and then as you start getting traction on that, you drop off the organic and then you target beard all because you've got relevance there from the organic beer ball and you've got conversion history. So what you do is you use the longer tail until you narrow down to the head tail is just as a basic example. [00:32:34] spk_1: Yeah, Danny, I love all that Danny and yeah, and what a lot of people don't realize that this is some of the stuff I've been trying to train people on for a really long time now is, you know, just because it's the top, you know, most search keyword does not make it the best keyword. Right? So I'm gonna give a quick example here. So I'm going to let people know if you you're not, you know, new at this, you know, I'm going to give a little bit of background, bear with me here in case you your experience here. But so long tail keyword generally has a greater conversion, right? But less sales. So higher conversion, less sales mid tail keyword, you get kind of a middle conversion, right? You get a decent amount of, of sales, um, but it's still not the top keyword and then a primary key word has actually a lot of times we'll have the lowest conversion, right? Because it's not specific enough. So some examples here longer tail keyword, blue water bottle with straw, Right? And then a mid tell keyword would be a blue water bottle for example. And then uh and then a primary key word would be the base keyword of water bottle. Now when you think about this in a, in context, okay, if you're going on to amazon and searching just you yourself and you go in and you say I want a blue water bottle with a straw. If a blue water bottle with a straw pops up in, you know the top three [00:33:43] spk_0: results most [00:33:44] spk_1: likely you your possibility of getting a conversion on that tail is really high. Now if you if they search for a blue water bottle and you're one of thousands that blue water bottle, but they're, you know, they're not searching for one with a specific straw or top or anything like that. You may get the conversion but it's a lot less likely than that. More specific search last. But not least if somebody just types in water bottle, they are on their an explorer mode, right? They're not necessarily ready to buy that water bottle, they're going to click through and go, oh here's a blue water bottle with a straw and then they'll go back to the search bar and say I want a blue water bottle with a straw and then convert on that keyword so that I just wanted to really make that comparison because a lot of people I think always try to go for the top tier keywords and that's where they end up just burning through a ton of money and not getting a lot of [00:34:29] spk_0: results. Exactly. And when they burn that money they haven't got the opportunity to rise again. You get very little room for manoeuvre is when you make a mistake on a launch product, in your new, new to amazon, right? Because let's just say you've got 15 to 20 K, you burn through that very easy. Free grand on PBC can be easily done. And, and, and you don't get you don't get another shot if you've got you haven't got the money to sustain those kind of losses. [00:34:58] spk_1: Yeah, absolutely. And not only that, but, you know, a lot of people don't realize a lot of times the first week or two of PBC is is literally research, right? It's literally you're spending money to figure out what works and what doesn't work. And a lot of people think that they're gonna just be successful out of the gate. So, uh, you know, just some basic thing guys, things guys to think about when you, when you're, when you're kind of testing if you're being successful or not in PBC. I see it over and over again, people dropping in, you know, after a couple of weeks saying, you know, my boss is over 100% and you know, I'm getting killed here. I'm like, you know, that's in the research phase, that's not not uncommon. That can happen, especially if you're in a a super competitive niche. And [00:35:36] spk_0: I would love to just add the biggest mistake that I'm seeing people make is using advanced PPC strategies taught by people who know what they're doing on brand new products with huge keyword list on products that have not index yet for relevancy. And so what ends up happening is they spend that whole budget and they have an index properly. And as you mentioned, Danny, you lose your chance and it ends up costing you more to relaunch and try to regain your traction because now you've told the algorithm, oh I am all of these things and half of them you're going to get bad data for. So it's like oh man, I'm seeing that people are applying these very advanced strategies and I'll take a look at their campaigns and they haven't ranked first or become relevant to the algorithm first for those super relevant medium to long tail keywords that are going to help them get that initial boost that they need. The other one is which is difficult to measure it. It can be achieved but you can't reach statistical significance with it. If you've got only long towel, let's tell you gotta stack a long tail keywords. You run your business based on that. Like in order to reach statistical significance, you need to reach one a threshold and you have a time period set that you measure over. You can't measure conversions over a nine month period and put that in as a statistical element. That is going to be a modifier for your business. It just doesn't work because there's too many different variables there and it's too many seasons that are taking place in between. It makes it very hard to manage accounts that way. If you only focus on pure long towel without any middle or head tail, whatever name you want to give the two upper tiers keywords, you know the description? [00:37:28] spk_1: Absolutely. All right, Danny. Well, we've just proven that we have to have you back on so we can dig into more of this stuff because I know that we could all geek out on this stuff for hours and hours and hours. I know I could especially the BBC and algorithm stuff, that's, you know, that's what I love to turn dials on that stuff but in the meantime, tell people where they can get a hold of your content, get ahold of you and kind of what, what's next for you? [00:37:51] spk_0: Cool. Yeah. Just seller sessions dot com visit there, add to the podcast. If you need to get older, me email Danny, it's seller sessions dot com or if it's dated brill related to PBC Danny at data brill dot com right now. And just I said of course I won't go into too much detail but booking out a loaded gigs at the moment for next year. Just working on that. We've got sell apple which will come up in um september so sell a poll number three Brandon by William will be ready to go for january next year. A lot of stuff I'm working on now is actually for next year. So I want to say too much about those things trying to get seller sessions live sorted could be sold out last May. It's there. I've just got to deliver the event. But the problem we've got is the uh the issue with Covid. So until we put the US on the green or amber list, the speakers can't travel without quarantine because it's 14 days and I can't expect them to do that. And most of the lineup are from the U. S. So that's one of the frustrations at the moment because what's been beautiful about the U. S. Like Amy, you went over and did prosper right, great. Got together with everyone. It was good and you know, I know that a few people got Covid, but that's to be expected. This is people get together, the world's got bloody open up at some point and we'll just be as careful as possible. Um, but yeah, it's that we haven't had that chance to do that in England yet. Um, so it would be nice that day comes around where everyone can fly in and we can put in a big event, big happy family, learned some content go out and party and human contact I think is so important. Engagement of human contact. [00:39:41] spk_1: Absolutely Danny, thank you so much. Really, really appreciate you being on, like I said, we got to have you on again. Um, if you guys haven't done so yet, check out seller sessions. Like I said, awesome podcast. Uh Danny has some great guests, They go deep into uh, into amazon. Um, so definitely check that out guys, and as usual, if you haven't done so yet, please rate review, subscribe to the podcast. We really appreciate all the feedback you guys give us. Uh we try to improve the best way we can, we try to bring on people that you want, uh you know, go into topics that you want to hear. So if you want to reach out, please do so to either a me or I um and let us know if you guys have any ideas on who you want on the show. Um as I mentioned earlier, you know, I'm in Idaho now and this is the, I didn't even know this before I moved here, but this is where click funnels originated and I'm starting to get to know some of the guys around here, so we're gonna have some click click funnel people um on the podcast here, some higher echelon people, which will be pretty, pretty cool. So we'll be announcing that soon guys. And as usual, thank you so much for joining us if you want to do so Live Still around table dot com forward slash live Tuesdays at two PM, Pacific time. Thank you so much, guys. And we'll see you next time on the seller roundtable. Thanks for tuning in to part One of this episode, join us every Tuesday at one PM pacific standard time for live Q and a and bonus content after the recording at cellar round table dot com, sponsored by the ultimate software tool for amazon sales and growth seller S c o dot com and amazing at home dot com.
  continue reading

395 つのエピソード

Artwork
iconシェア
 
Manage episode 305561841 series 2492296
コンテンツは Amazon FBA Seller Round Table and Amazon FBA Seller Round Table - Selling On Amazon によって提供されます。エピソード、グラフィック、ポッドキャストの説明を含むすべてのポッドキャスト コンテンツは、Amazon FBA Seller Round Table and Amazon FBA Seller Round Table - Selling On Amazon またはそのポッドキャスト プラットフォーム パートナーによって直接アップロードされ、提供されます。誰かがあなたの著作権で保護された作品をあなたの許可なく使用していると思われる場合は、ここで概説されているプロセスに従うことができますhttps://ja.player.fm/legal
Things we discussed in this session:

A. Part 1

Things we mention in this session of Seller Round Table:

Join us every Tuesday at 1:00 PM PST for Live Q&A and Bonus Content at https://sellerroundtable.com

Try the greatest Amazon seller tools on the planet free for 30 days at https://sellerseo.com/
Transcription in this episode:
[00:00:01] spk_0: Welcome [00:00:02] spk_1: to the seller roundtable e commerce coaching and business strategies with and er not and [00:00:07] spk_0: amy Wiis, [00:00:08] spk_1: it makes a lot of sense. The other thing that I've noticed too is um I will open uh different tickets and you know, when you, when you go through the initial ticket process, you have to pick a category and subcategory. What I'll do is I will open two simultaneous most every time. Now that I've learned this works really well as you go to simultaneous. I know like you said, Danny amazon might hate this, but if you actually are trying to get something done, you go and pick, you know, the first one you do is the one that is in the category that makes the most sense. Then the next one you go is to something that's maybe part of the problem, but not the main category and you open a ticket there as well. Um I've found that, you know, because you have now kind of two separate groups working on it, One might hit a wall and the other one might find a solution or vice versa. So that's another kind of little trick that I've, I've found out over the years with all the hundreds of tickets I've had to open. [00:00:58] spk_0: Yeah, So here's another thing which quite interest. There is no financial incentives for Associates. All the targets are based on efficiency and quality of where they have met the correct standard. Right? So on the flip side of that, if they don't meet the targets, they will get closely monitored, Right? So it actually becomes more painful from, because the QA teams are going to be more and more on their back to to meet certain standards. And then also opens up that opportunity for them to earn other money elsewhere. Which is what happened um when a bunch of the support reps were caught in this where they were doing certain sellers favors. And um, so it's like kind of bad that they don't get any kind of monetary compensation because their pay probably isn't great and you know, it kind of sets it up. Yes. So the reward of being efficient and good quality is the opportunity for lots of promotions. It's always encouraged. So that is your upside. So while she don't get a financial there, you'll get out of tier one, say a T one of seller central and then you quickly get moved up. If you're good in the excel, what you do. But yeah, you're always going to get people on the take. Now this is the thing and I don't want to tie in with the people that are on the tape because there's good and bad people on this planet as we know, right? Um, but when there is an issue with a ticket but let's just say yeah and icing is being yanked with that. As an example as a ticket. They can log into your seller central gap. So any of these reps can access any of our accounts but purely based on the condition relating to a ticket. Now, what you would say, well what happens when reps are working from home? Because a lot of them do. They distributed they have a token system which is all generated on fingerprint I. D. So with the home systems they're very secure as well. So I want to make sure that people understand that that's clear. But I found that kind of interesting as well. And the 17-18 minutes. Um He's annotate this year. So that's something how a ticket is broken down. If this is any [00:03:14] spk_1: help for [00:03:15] spk_0: The audience 17-18 minutes. Right? So there's four parts to a case ticket outcome, one pending amazon action P. A. In case amazon needs to perform an additional research and the seller needs to wait for the outcome. Number two perch in pending merchant action PMA More information is needed from the seller to be of support. The current information provides limited or incomplete For us to proceed as support. # three resolved cases done and resolved or the conclusion is that amazon point of view, we cannot be of support whatsoever reason. And we need to inform that the seller that we cannot resolve this number four reply and keep working. This is used with more time is needed to perform research and want to notify the seller in the meantime that we're working on it and we'll come back to them as soon as we have an answer and reply. So you've got a key thing there. The other thing is that annotate each of the tickets so that everyone is singing from the same hymn sheet. They use all the same S. O. Ps. And everything else. Where is the [00:04:22] spk_1: ticket information? [00:04:24] spk_0: I'm going [00:04:24] spk_1: to tell you about [00:04:25] spk_0: annotations are added to open tickets. So this is what goes on your ticket as you do it. Number one summary of the seller issue to research done, which work followed free assessing order numbers info plus S. O. P. U. R. L. S. Followed for troubleshooting, which means they've got that catalog of uh you are the S. O. Ps. And they put that in the case stick at the U. R. L. So whoever picks it up can go to it. Plus there's lots of tools use like this guy that I'm speaking to has been there three months. He's now using 60 seller tools to support at his skill level. And he's only been there three months and there's a lot more tools involved in that um tools and tt trouble ticket number when escalated ticket has been created by an associate number four proposed action, taken summary or reply question to set up. So I think every ticket that they do, they have to annotate that ticket in four parts. I think someone said about a while ago because one of the things I want to go down the road of is performance team and stuff, you might not be able to get much from there. But I remember speaking to people before, we do a lot of the suspension um reinstatements and stuff, they were always talking about annotating. So if you've got an account and you kept sending the wrong information it gets annotated. That's going to mark hard against you because you've not paid attention to the point. They may not be able to recover that suspension if you push too many times. So just be aware is that everything we do, they're being monitored. And also every ticket raised as the four parts to its summary and the problem solving issues. [00:06:04] spk_1: That's crazy. You could really go down that rabbit hole for a long, long time. There's so much there. Um The other thing I had a quick point I wanted to make though for everyone listening or watching is is I used to you know, everybody hates filling out surveys or even those little things, but when you get us make sure that you leave positive reinforcement for the the agents that give you good support and negative for the ones that don't because amazon uses that feedback loop and all that will do will support sellers better. Right? If you get people who suck and and they weave those people out, then that's that means that next time your ticket is gonna be handled by somebody who has been reviewed positively or have got has gotten positive feedback. So I encourage you guys to do that. I used to never do it, but the last couple of years I've been doing it just because I'm like, I need to provide, you know, feedback loop. So that's, you know, so amazon can actually take action on on the ones who perform well and the ones who don't. So, um, that's my asking suggestion to you guys. [00:06:58] spk_0: This is one thing I want to quickly tell you about the frustrated button. I haven't checked yet, but he was telling me because I didn't get a chance to take and I don't know if it's fully correct by a double check with him for next week's show because I'm going to do another one. But he says when you have problems and you've raised the tickets, if you hit the frustrated button right diet that the frustrated tickets are higher, higher rent. But some people like to listen, I haven't seen the frustrated, but I haven't looked yet, but I'll get that confirmation and if that's true, if there is, once you've raised the ticket and you've got, you know, like normal ego, was this any good yes or no, if there is a frustrated button there, if you are frustrated, hit that because apparently that they will rank frustrated tickets higher, which is saying I didn't know about. And the captive team thing, I remember a couple of years ago, people going, yeah, well called captive team, I need to follow out and no one knew the captive team was right now, we don't know if it's true or not, but the captive team might just be made up. It doesn't exist. It just might be a posh name for Tier two or Tier three and there's no such thing as a captive name, but it was a really good word to use a buzzword if you like, if you're selling uh masterminds, you know, do you know what I mean? I think you know, and things can come up into the amazons there that just don't make any sense to maybe not true, but it could be true, but he's been there only three months, so I'm trying to get him to dig and see if there was ever a captive team going back three or five years. [00:08:26] spk_1: It's interesting, I think there was, or like you said Danny, that might have just been kind of what people called it. But I know that there was, uh, I went to college in North Dakota without going too far and, and a lot of people don't know this, but amazon has a big foothold in North Dakota. And so I used to have phone numbers because I had friends that work there like, way in the past, but in the, in the past you would have these special phone numbers that if you knew, you could actually [00:08:47] spk_0: call and [00:08:48] spk_1: get somebody on the phone. Like I used to be able to call the catalog team directly. Like a lot of those direct connection numbers have, [00:08:54] spk_0: yeah, yeah. [00:08:57] spk_1: Have gone away. But even that though they've gone away because amazon is now, it, the word gets out and then, and then it's useless because it just gets inundated. Right? Same thing with Jeff at amazon dot com. Good luck using that these days because you know it's like the word was out. That that's how you could get a quick escalation. [00:09:12] spk_0: This is the other thing I'm like with the shows I'm going to be I have to be sensible if he turns around and gives me a bit of information and that you know tens of 1000 people jump on and it's just gonna flip. It doesn't help anyone. What I'm trying to do with the series is really educate people on so that they can take away from the shows and go do you know what? I'm gonna write my ticket like this now I'm gonna be extra nice to this person on here. Oh to resolve that if I go to this I want to give them all the ideas of like how can we create an S. O. P. L. Creating the master ticket. What is the most optimized way of developing a ticket to get the best result possible rather than saying, you know, here's the captive team if it exists there telephone number and then blurting it out on seller sessions. But do you know what I mean? It was just disappoint. It doesn't [00:10:02] spk_1: Absolutely. All right guys we're going to switch to something else. But if you want to learn more, make sure you subscribe to seller sessions. It is one of the better podcasts I've listened to many. I think Danny does a really great job. So go check that out. Great info. There are lots of great people there. Um It's on all the major platforms. Go check that out. Uh Let's transition though dan. Because one of my favorite things which is some of your favorite things is you know uh you know algorithms and PBC. So um in terms of you know, the the the A. Nine algorithm, I would love to hear, kind of your insight and maybe some of the ways that you think that it's uh progressing and maturing and changing. Um I would love to to dive into that. [00:10:46] spk_0: Okay, so effectively what I did is a few months ago I dug out the, what I found, it started with the video and which is a Doria Sonika from the A. 19. And then I started to find research papers like amazon dot science. What had worked out is A nine works completely different to what most people talk about. And once I got hold of the information, I sat down for a few weeks and going, how can I turn this into layman's term? Because for you Andy we can geek out and I'm not a developer. You are, but you're Yeah, that's great. But for general listeners, they've got to try and understand it. So for how do we take this, which would normally send a glass eye to sleep and make it more plausible? So we broke it into into different into different sections. But the key thing that I find with a nine is that it operates differently to a lot of the way that people talk about it. And so when we look at list and we think of title bullet points, blah blah blah, right, But we don't think of where did it get there? We know that the conversion make makes an impact. We know that add to cart is massive but add to cart only works as far as page one unique keywords with selves to bind past the top of page one. So we know that add a car is a massive massive signal, but when you break it out, like for instance, no one was talking about negative and positive labels. So when you think of negative and positive labels, if you're looking at your listing, you could be saying to yourself, well we have a conversion which is on the continuum. If someone lands on your on your product page and they stay there and they don't bounce off, there's a micro conversion, like a credit going on there because the engagement, you scroll your engaging, right? You add to cart, it's another signal you're engaging. So you've got to think of conversions on the, on the continuum. The biggest thing I took away from all the A. Nine stuff is amazon is obviously about relevance, but they want to know just as much about what you didn't want to, what you did, right and what that is that it falls under these training sets fall under negative and positive labels. Now, positive labels is clicks add two car engagement and things like that different. And it's all derived around behavioral aspects as part of their training sets, which make up as part as the algorithm. So when you're writing your listing and you're saying, I want to write the future benefits and everything in the bullet points, you would think about it from a copyrighting point of view and the human person reading point of view. But if you took it into Afridi mode, you'll then start to think about it from a negative label and a positive labour point of view. And then you start getting to things like the hunger score, which is very complex for people to really understand. But Hunger Score, you could uh consider that is foam. Oh, so what the hunger score does. It searches down by category, but it's looking for the stuff that you are not interested in, don't touch. And I think what that really is about is that they're trying to unearth if there's anything they've actually missed. So they're collecting all this information. So it all comes down to relevance. You've got on one side, you've got the positive label. The other side, you got the negative label. Let's look at another reason why the positive and negative labels would work if you'll send in traffic, PPC traffic to listen. We now know that if you send traffic there and the conversion rate is poor, that's going to have a negative impact organically, right? Goes back negative labels, positive labels. What is under the negative labels? Well, it's unseen. It's why is that information not being utilized, their etcetera? So when you are looking at optimizing your listings, think about it. Will this fall under a negative label? Will it fall under a positive label? On top of what you're already thinking about when you're optimizing a listed. So it goes back from being a static vector if you like that. Here's our listing is very static and we're looking at one dimensional, let's look at it from the freedom dimensional point of view. We know that most of it is based around um Machine learning goes back as far as 2016. There was using 100 machine learning models. We know the importance of being paid one. I'll give you some data here, 70% of Amazon customers never click past the first page of search results. Then you've got 35% of amazon. Shoppers click on the first product featured on the search page. Uh and the first three items displayed in search results account for 64% of the clicks. And this is why P Pcs come to more for amazon's the paper place. That is a great model for them to earn money because everyone who is ranking organically up here, they're being pushed down there in like the active position on the page, not the Organic or the ad position. Organics is sitting down in position overall of number four and number five. So a lot of the PPC clicks, the top researches could become more expensive. So this way it's so important to dial in. The other thing is if we take some stuff off the table, there's no such thing as 18. It doesn't exist. It's not in any of the science literature, none of the engineers that I've spoken to have talked about it. And just to give you one quote from amazon here, small improvements in relevance can positively impact millions of shoppers. If that's the case, why would you take a nine, Throw it in the dustbin and dial out a 10? You wouldn't and it's not structured that way. Right. What other people don't realize as well? It's based down catalogues, indexes, categories each marketplace. So when we see a push or we observe something and we see listings have a a sharp fall if you like you. Alright son, algorithm update. That isn't happening in the UK when it's on the U. S. That's not happening in Germany, these training sets and these, these periods that they are measuring this, this data and they collect that information or the behavioral day asia. Then they're going to do the push and when they do the update, it's on that marketplace. So it's not global order. These teams are working differently on the different marketplace. So indexes and everything, a different relevancy is going to be slightly different marketplaces, but overall there yes, are using the full structure of a nine and this is why add to cart works so well and it will carry on working so well for a long period of time. Could it's such a machine, you can't turn, you can't steer the ship around and start going, oh well we're not going to use add to cart no more is a really strong signal. Do you see what I mean? So this is something that has developed over time and I think it's really important that people I think about. I think my key takeaways from kind of your sum up of the in an algorithm is that number one we need to think about more than just. Okay. I have the right keywords. You know, like I have the right keywords and my listing looks good. We need to be thinking about our products from how from what page ones we want to be on. I'm always talking about them. What's your page one? Because of course top of search is so important, but the right search is so important, right? We want to be the most relevant so that they add to cart right so that we can get that going on. So I'm hearing that with the ever increasing cost to get to top of search, it's more important to really, really dial in like what are those search terms that that customer that is so relevant for your products are going after. And you know, I see people are just even more successful now. You know, we recently had some really great folks coming and talking our mastermind group about this ranking strategy that they're using and they're really focused in on some key words that are that are the most important and very few keywords to begin with before they're moving on to others. So I think when we're thinking about our keyword strategy and our advertising strategy now, more than ever, we really really need to know what our page ones are and make sure that we're paying attention to how customers interact with our listing. Not only are conversion rates, but where are we converting? How are we converting? When are we giving that awesome add to cart, you know, what about you Danny? What do you recommend for people in really reducing their costs for their PPC and choosing the right keywords so that they can get the most out of the algorithms favor. I think the newer people are definitely burning their money because they're getting these massive long keyword lists from really great for party tools. But if they're not in the brand analytics report, most of it is junk. It's garbage. Right? So I always say to people take you a long list of 100 or so Wash that list against your brand analytics report. You might find it's only five in there and so you're chasing and pushing all of this data around on hundreds of keywords which are mostly useless because they've got less than 10 searches a day or less on the us market. When you say brand analytics report, are you talking about putting your ace in into that search terms, report and seeing where you have the most lecture? Yeah. Or your broad keywords, it doesn't matter accusations or not. What you're looking for is do my keywords appear in the brand analytics report? Let's just say you sell beard oil. So you put beard oil and then you look at all the different permutations of beard, organic beard old blood uh men's beard oil and then you start to see him and you see the frequency rank, you know that's 1 28,504 and you're looking for that. But what I'm trying to get across what can be very scary for people. We know that the U. S. Market has great long term and it's good to have long term. There's no doubt about it. And it makes up sells Brandon Young does some really good stuff with that in the way that he operates and I totally agree with it. But what you got to be careful of if you've not at Brandon's level and know what you're doing right, you might have misgivings thinking great, this tool has just given me it's got 100 keywords, I've got plenty of opportunities. You put it through the brand analytics report, there's four of them are in there Then your opportunity is gone from, you know the other 96 has disappeared, you've got four and you're gonna have also ran keywords and then you die in or one or two and you might not be able to afford the PPC on it Or they may be so far back that even if you run PBC on them, you're not going to get enough cells velocity to get anywhere with those keywords. Great. You might be on page one because you know it's got 20 searches a month but that's what it's useful as a chocolate ashtray. You need get their balance right because if the keywords too big and you run PPC on it, you don't get impression share. This is why people get 20 bucks a click and then they spend all this money and they don't realize that I will keep this keyword because we have like a calculation on joined me to quickly go for a quick calculation to give people when they're looking at keywords because I think it starts to bring home. Absolutely. [00:22:11] spk_1: And the other thing quickly I want to mention just I think what what Danny is saying and correct me if I'm wrong here, Danny is rather than go after the 1000 keywords, you know that that these tools are spitting out every iteration and what they literally do is get the key word than at an A to it and do a search and then So I mean you might get one or two sales a month from those keywords. But the other thing you have to realize is the more keywords you have, the harder it is to manage, right? It's the 80 20 rule. If you [00:22:35] spk_0: have, you know, a [00:22:36] spk_1: 1000 keywords than having to go through and try to optimize all those keywords is going to be really difficult rather than really focusing in on that that precise target and then just continuously optimizing those keywords, those small sets of keywords and even on your, you know, products, [00:22:52] spk_0: your [00:22:53] spk_1: product targeting etcetera, [00:22:55] spk_0: would you mind if I share my screen? It might be a bit easier. Just love that bit of context. All right, So let me just pull this up. Can you see that? Okay, so I've just I've just mentioned here never go beyond the brand analytics report. You might you know, there's an exception to the rule and you're going to get some people go, yeah, I had a few cells but generally if it's 10 searches a day, You're looking 300 searches a month time you get into position 12 or three. A percentage of that is the click through rate from the click through rate. How many sales are gonna get off that when that shared against some people on that page? Not a lot. Right? So you got to think about it that way. This is what I'm going to give you an example of here. So ranking with amazon PBC, right? So if your PPC is competitive conversion rate and generate enough orders, it would definitely boost your organic rankings here at each bar represents one week's worth of data over their period of eight weeks. So in the left column, we see the average organic rank, right? And then where the smallest bars show rank one is as good as you're gonna get because it's gone from he's going to go, it's gone from up in the gods to position number one over a period of time. So we've got organic ranked number one on the first column, number two on the second and number one on the third column. Each of those represents eight weeks as you see them coming down into the number one position. So you can see these move quickly towards rank one over the first month. Here, the remaining columns, we see impressions orders and spends. Uh These are mostly trending upwards, possibly plateau in after the first month or so. So you can see impressions as the organic rent comes down. The impressions increase as you can see that. Not exactly, but you know, is heading that direction and then you got the oldest column and the spend column. So you can see the top one is where we spent most of the money But it got 219 orders from it. Now in order to stuff like this to work. When you got to be dialed in on your keywords, right? And to your listing should be very well optimized. This conversion rate plays a massive factor in the success as a bad conversion rate will become a hindrance to organic rank, pulling you under water every time you send non converting traffic, like I mentioned earlier in the show, conversion works on the continuum. So the algorithm is the multi faceted view of a conversion rather than what we see as a cell. Right? When a customer clicks on your product, you effectively getting a microbe conversion credit on that query. If the customer spends time when your product page and does not bounce off, you're getting another micro conversion, the more time you spend on the page in principle to hide the engagement and when the customer is a product of car again, you're getting another positive signals. So I skipped this abandonment stuff at the moment. But there's a breakdown, right? So in order to set this up, you guys know this very, very basic setup but you can do this across multiple campaigns. one key word, 1 18 1 campaign isolated. Like this means that we can measure the top of search on a granular level, right? Because for sponsored products, top research placement statistics is on the campaign level. Right? So that's why would set them up that way. So going back to the algorithm here, the ranking algorithm compares many different factors. Two of them all the history and the conversion rate for a specific product listing and that said keyword. Right? So they're all attuned in from those two components down to their individualized keywords left. This is then compared to all of your competitors on the same keyword, If you get a higher conversion rate through PPC cells, then you're gaining points on both of these metrics. And what we've seen in house. Your relevance will also get a boost from this. But then we face the problems we've ranking high velocity products which I touched on before south uh slides. This comes back down to the ranking Algo. There are only so many add positions or placement worthy rivals, which is obviously on page one. So for a high velocity keyword with established assassin have already converting. It could be extraordinary difficult to break through and get impressions even with a phenomenal conversion rate. This is why you hear people bidding $20 a click without getting any impressions. So depending on your stem manner and if you keep going over time, amazon's algo may determine to favor your campaign, but there's no telling how long this will be. It could be that one or more your competitors go out of stock and that's where you get to sneak in. Right again. This is where it could give you a nice booster. So what is the answer? Is there a special calculation or the answer is yes. Known common sense. So these are arbitrary numbers here just to make a point of the demonstration. Right? So what we have here is K one is keyword one. K two is key word to This represents the ad placements. P one, p 2, P three. Okay, so you can see that keyword one super high velocity. We've made up the numbers here to make them stand out. Right? So if you was going for placement one, you'd need to sell 1000 units today to get to placement one. You're not gonna do that right? It's unrealistic. Replacement number 2, 500 replacement number 3, 250. Again, pretty much unrealistic. Unless you've got deep pockets Then you look at keyword # two. So you're looking for the sweet spot, right? You're looking for their midtown keywords. Then you go, Okay, where is the where is the low hanging fruit? Well, position number four, I only need five cells. So if we use the data point to you, how many orders a day can the top of search placement generate if expressed at a 20% conversion rate? So to get to position number four, you might aim for at least five orders a day. Right? So it's more plausible to reach. If your conversion rate is 20%, that might mean five orders a day, divided by conversion rate is equal to 25 clicks. So if the cost per click is $5 for that position it will cost you $5 divide by conversion rate which ends up being $25 per acquisition on average. Or 100 and $25 per day. But that's only when it converts right, which comes back to the point of gaming impression share and the cost of running expensive top research campaigns and usually a market these red hot. Generally we have a higher cost per click. So how do we launch PPC and then taper off knowing where our numbers land for break even because this is kind of difficult. So we kind of come up with an option here but they all present themselves their own pros and cons. Right? So if we look at either way, you're going to need to pick your poison, adjusting the bid out of the gate. So if you make bid adjustment, this is almost certainly affect your position in the ad placement algorithm. So we could do start putting back on the daily budget versus decreasing the bid and maybe losing on the position there, at least for the budget caps. Your campaign would run in most cases, but if you lose advertising for the rest of the day after your budget is burned. So what I mean by that, if you adjust the bidding and you're in position number two, you may come down to number four because you've adjusted that cost per click bid. But if you adjust the campaign cap on a daily basis, you'll get those positions but you're likely to run out of budget at some point in the day. So that's why you play with a budget on a daily basis rather than touch the click because you want to maintain the position. But you could have a hybrid version. So twice a day you mainly adjust the top of search boost in your campaign So you can go and boost that to 900, then 12 hours later dropped to zero. And then you work out sweet spot. You play around with those two factors and find that sweet spot. And then over time you start to get your ranking costs under control. All you don't because it's always a risk, right? So you're looking to rank middle key words that say take 46 to 8 week period. By the time you get to eight weeks, you're looking to do the pullback process and Danny. I'm always surprised at how many people don't know about placements and how to adjust those percentages. So there's many ways to skin a cat as we know, but that's one way that we're trying to present something that's quite complex And there's way too many variables to say this is the absolute way around it. But by working off an assumption of a calculation at 20%. Uh, and using ad placement is your reference. It may help people that going after really big keywords realized I can't go after those keywords because a lot of people won't do that calculation. They won't sit there and think, oh yeah, in order to be successful with this and start using that keyword to ring. I mean I need to get 50 p. p. c. sales today. That's really hard. Do you know what I mean? Like when you don't think about it, especially if you're starting with zero campaign history, you've got no conversion history. You've got newly launched listed. It's no wonder you don't get the impression share. Uh, so that's why you have to be realistic and start with longer tell keywords and work your way up unless you use the beard or analogy beard all be the number one key word. But what about if it's organic beard or so you start in your campaigns with organic beard or and try and rank that and then as you start getting traction on that, you drop off the organic and then you target beard all because you've got relevance there from the organic beer ball and you've got conversion history. So what you do is you use the longer tail until you narrow down to the head tail is just as a basic example. [00:32:34] spk_1: Yeah, Danny, I love all that Danny and yeah, and what a lot of people don't realize that this is some of the stuff I've been trying to train people on for a really long time now is, you know, just because it's the top, you know, most search keyword does not make it the best keyword. Right? So I'm gonna give a quick example here. So I'm going to let people know if you you're not, you know, new at this, you know, I'm going to give a little bit of background, bear with me here in case you your experience here. But so long tail keyword generally has a greater conversion, right? But less sales. So higher conversion, less sales mid tail keyword, you get kind of a middle conversion, right? You get a decent amount of, of sales, um, but it's still not the top keyword and then a primary key word has actually a lot of times we'll have the lowest conversion, right? Because it's not specific enough. So some examples here longer tail keyword, blue water bottle with straw, Right? And then a mid tell keyword would be a blue water bottle for example. And then uh and then a primary key word would be the base keyword of water bottle. Now when you think about this in a, in context, okay, if you're going on to amazon and searching just you yourself and you go in and you say I want a blue water bottle with a straw. If a blue water bottle with a straw pops up in, you know the top three [00:33:43] spk_0: results most [00:33:44] spk_1: likely you your possibility of getting a conversion on that tail is really high. Now if you if they search for a blue water bottle and you're one of thousands that blue water bottle, but they're, you know, they're not searching for one with a specific straw or top or anything like that. You may get the conversion but it's a lot less likely than that. More specific search last. But not least if somebody just types in water bottle, they are on their an explorer mode, right? They're not necessarily ready to buy that water bottle, they're going to click through and go, oh here's a blue water bottle with a straw and then they'll go back to the search bar and say I want a blue water bottle with a straw and then convert on that keyword so that I just wanted to really make that comparison because a lot of people I think always try to go for the top tier keywords and that's where they end up just burning through a ton of money and not getting a lot of [00:34:29] spk_0: results. Exactly. And when they burn that money they haven't got the opportunity to rise again. You get very little room for manoeuvre is when you make a mistake on a launch product, in your new, new to amazon, right? Because let's just say you've got 15 to 20 K, you burn through that very easy. Free grand on PBC can be easily done. And, and, and you don't get you don't get another shot if you've got you haven't got the money to sustain those kind of losses. [00:34:58] spk_1: Yeah, absolutely. And not only that, but, you know, a lot of people don't realize a lot of times the first week or two of PBC is is literally research, right? It's literally you're spending money to figure out what works and what doesn't work. And a lot of people think that they're gonna just be successful out of the gate. So, uh, you know, just some basic thing guys, things guys to think about when you, when you're, when you're kind of testing if you're being successful or not in PBC. I see it over and over again, people dropping in, you know, after a couple of weeks saying, you know, my boss is over 100% and you know, I'm getting killed here. I'm like, you know, that's in the research phase, that's not not uncommon. That can happen, especially if you're in a a super competitive niche. And [00:35:36] spk_0: I would love to just add the biggest mistake that I'm seeing people make is using advanced PPC strategies taught by people who know what they're doing on brand new products with huge keyword list on products that have not index yet for relevancy. And so what ends up happening is they spend that whole budget and they have an index properly. And as you mentioned, Danny, you lose your chance and it ends up costing you more to relaunch and try to regain your traction because now you've told the algorithm, oh I am all of these things and half of them you're going to get bad data for. So it's like oh man, I'm seeing that people are applying these very advanced strategies and I'll take a look at their campaigns and they haven't ranked first or become relevant to the algorithm first for those super relevant medium to long tail keywords that are going to help them get that initial boost that they need. The other one is which is difficult to measure it. It can be achieved but you can't reach statistical significance with it. If you've got only long towel, let's tell you gotta stack a long tail keywords. You run your business based on that. Like in order to reach statistical significance, you need to reach one a threshold and you have a time period set that you measure over. You can't measure conversions over a nine month period and put that in as a statistical element. That is going to be a modifier for your business. It just doesn't work because there's too many different variables there and it's too many seasons that are taking place in between. It makes it very hard to manage accounts that way. If you only focus on pure long towel without any middle or head tail, whatever name you want to give the two upper tiers keywords, you know the description? [00:37:28] spk_1: Absolutely. All right, Danny. Well, we've just proven that we have to have you back on so we can dig into more of this stuff because I know that we could all geek out on this stuff for hours and hours and hours. I know I could especially the BBC and algorithm stuff, that's, you know, that's what I love to turn dials on that stuff but in the meantime, tell people where they can get a hold of your content, get ahold of you and kind of what, what's next for you? [00:37:51] spk_0: Cool. Yeah. Just seller sessions dot com visit there, add to the podcast. If you need to get older, me email Danny, it's seller sessions dot com or if it's dated brill related to PBC Danny at data brill dot com right now. And just I said of course I won't go into too much detail but booking out a loaded gigs at the moment for next year. Just working on that. We've got sell apple which will come up in um september so sell a poll number three Brandon by William will be ready to go for january next year. A lot of stuff I'm working on now is actually for next year. So I want to say too much about those things trying to get seller sessions live sorted could be sold out last May. It's there. I've just got to deliver the event. But the problem we've got is the uh the issue with Covid. So until we put the US on the green or amber list, the speakers can't travel without quarantine because it's 14 days and I can't expect them to do that. And most of the lineup are from the U. S. So that's one of the frustrations at the moment because what's been beautiful about the U. S. Like Amy, you went over and did prosper right, great. Got together with everyone. It was good and you know, I know that a few people got Covid, but that's to be expected. This is people get together, the world's got bloody open up at some point and we'll just be as careful as possible. Um, but yeah, it's that we haven't had that chance to do that in England yet. Um, so it would be nice that day comes around where everyone can fly in and we can put in a big event, big happy family, learned some content go out and party and human contact I think is so important. Engagement of human contact. [00:39:41] spk_1: Absolutely Danny, thank you so much. Really, really appreciate you being on, like I said, we got to have you on again. Um, if you guys haven't done so yet, check out seller sessions. Like I said, awesome podcast. Uh Danny has some great guests, They go deep into uh, into amazon. Um, so definitely check that out guys, and as usual, if you haven't done so yet, please rate review, subscribe to the podcast. We really appreciate all the feedback you guys give us. Uh we try to improve the best way we can, we try to bring on people that you want, uh you know, go into topics that you want to hear. So if you want to reach out, please do so to either a me or I um and let us know if you guys have any ideas on who you want on the show. Um as I mentioned earlier, you know, I'm in Idaho now and this is the, I didn't even know this before I moved here, but this is where click funnels originated and I'm starting to get to know some of the guys around here, so we're gonna have some click click funnel people um on the podcast here, some higher echelon people, which will be pretty, pretty cool. So we'll be announcing that soon guys. And as usual, thank you so much for joining us if you want to do so Live Still around table dot com forward slash live Tuesdays at two PM, Pacific time. Thank you so much, guys. And we'll see you next time on the seller roundtable. Thanks for tuning in to part One of this episode, join us every Tuesday at one PM pacific standard time for live Q and a and bonus content after the recording at cellar round table dot com, sponsored by the ultimate software tool for amazon sales and growth seller S c o dot com and amazing at home dot com.
  continue reading

395 つのエピソード

すべてのエピソード

×
 
Loading …

プレーヤーFMへようこそ!

Player FMは今からすぐに楽しめるために高品質のポッドキャストをウェブでスキャンしています。 これは最高のポッドキャストアプリで、Android、iPhone、そしてWebで動作します。 全ての端末で購読を同期するためにサインアップしてください。

 

クイックリファレンスガイド