How High-Volume Sellers Are Managing Amazon PPC Webinar
By Andrew Maff, on April 20th, 2018 ,
On April 18th, 2018 our own Andrew Maff sat down with Franz Jordan of Sellics to talk about how high-volume Amazon sellers are managing their PPC campaigns. They discussed everything from manually optimizing to automating the process and of course, Sponsored Products, Headline and product Display ads best practices. You won't want to miss this one!
Did you miss the webinar?
No problem! You can view the whole thing right here!
Would you like the slides from this webinar?
No problem! You can view them here or head over to SlideShare to download them!
Andrew Maff: Everyone who's here, thank you so much for joining us. Obviously you are on the Amazon PPC webinar with Sellics and Seller's Choice. We're gonna try to go over everything we can touch on about automated campaigns as well as doing them manually yourself, and then obviously your automated and manual campaigns, which we'll touch on.
If you have any questions, feel free to please comment them. I will try to get them as much as possible. Franz, just so you know, I'll be reading them and I'll let you know what kind of questions come up.
There is a slight delay on the questions, so if you guys ask questions and we don't get to them right away, give us a little bit of time, we're definitely gonna try to get to them as soon as we can. We can go ahead and get started here.
Why don't we introduce ourselves?
My name is Andrew Maffetone. I am the director of marketing and operations here at Seller's Choice. I've been in eCommerce for a little over a decade now. I've been in Amazon almost the same amount of time. I've helped a lot of large sellers, a lot of small sellers. I've helped with a ton of different ones. I believe I'm pretty qualified to be on this webinar.
What about you, Franz? You wanna introduce yourself?
Franz Jordan: Sure. So my name is Franz. I'm founder and CEO of Sellics. Sellics is a software name, is a software to help you manage and grow your Amazon business. The idea is to combine everything, all the features into one software instead of using many different softwares.
One of our main features, obviously, is PPC and PPC automation, and I've been focusing a lot about PPC in the past ... Over the past two, two and a half years. I'm speaking at some events like the Prosper Show, most recently, and in general I just love to talk everything about Amazon. I feel like I talk like 100 hours a week about Amazon.
I'm very passionate about everything that's related to Amazon.
Andrew Maff: Yeah. I saw your talk at Prosper. It was great. It's nice to have a nice high level talk on PPC. As I mentioned for everyone, we'll have a Q and A at the end. I will also probably be pulling some questions throughout the webinar so we can make sure that we stay within that category the best we can.
Of course, to make sure you guys stick around, we each have our own little giveaway that we'll be doing. Ton of extra stuff for you guys to make sure this webinar was fully worth it for you. We will try to touch on every question we possibly can at the end without taking up too much of everyone's time.
So we can get started here. If you're on this webinar, there's almost a good chance that you're using Amazon PPC right now, whether you're in Seller Central or an AMS, we get a lot of complaints about ACoS standard control, or if you are a reseller, or even if you're a private label seller and you just have thousands of products and managing those campaigns can sometimes be a pain.
Competitor issues and then of course, you're probably just really good looking. That's definitely pretty accurate right there. Let's see what we got. Let's get started here.
All right. We want to keep this webinar as extremely high level as we possibly can. We really don't want to spend a lot of time going over the basics. This is basically our only slide. Everyone, hopefully you know what ACoS is, the basics on how to optimize your campaigns. If you don't know, these are your three basic ads through Amazon.
You have your sponsored products that show up right below the headline ad, which is the next one over in your search. Then you have your product display ads that show up underneath the buy box.
We're gonna try and touch on all these, and we're gonna touch on how to optimize them manually as well as how to optimize them by automating it.
We will start with sponsored products, which Franz, I believe you would agree, is probably the most utilized, correct?
Franz Jordan: Yes. I'm having some connection issues. I'll change wifi in a second, but yeah, I think sponsor products obviously the most used ads. When we see the distribution in budget, I'd say that ... Almost all sellers, I'd say 98% of Sellics use sponsor products. It's a lot lower for headline search ads. I'd say it's something like 15% of sellers use headline search ads.
Then for product display, it's hard to say. Product display, obviously, only for vendors. So, we do have a few vendors on our vendor edition using it. Typically, we see a very strong focus on sponsored products and now gradually an increase in headline search ads as well.
Andrew Maff: Yeah. Just like any other marketer, I love to test as much as I can and we always suggest to start with sponsor product ads, mainly because like when we'll go through here on how to prep things and how you can use automated ads and all that stuff, which we'll get into, I like to basically use sponsored products to start to test which specific keywords are going to work best on headline ads, and then start my headline ads with that information.
That way, I'm not starting sponsor product ads and headline ads all at the same time and just getting an amalgamation of random data that I have to try and decipher for which one's working best.
Franz Jordan: I think the big problem with headline search is just in reporting, you really have very limited reporting capabilities. Essentially, what you get is this brand revenue, which is a very broad number, you have no idea which products really got sold. I think in terms of performance measurements, sponsored products are a lot better than headline searches.
Andrew Maff: Yeah. Structure. I know this is gonna be an incredibly opinionated webinar. I think we're gonna have people who agree with some things, and some people who disagree. Franz and I here are going to try touch on what we have seen work.
This is how I always have our team structuring our clients here at Seller's Choice. I always have a manual campaign with a bunch of manual ad groups in it, then a complimentary auto campaign with those same exact ad groups replicated throughout the auto campaign.
Every manual campaign has a complimentary auto campaign. One thing that always throws everyone off and no one seems to add is I always like to have one campaign that is an all products ... every product thrown into one ad group, and it's everything in one. I always get so much crap for this one.
I consider it like a safety net. To give you all an example, the way we'll usually approach things is where the manual ... Let's say a manual keyword, I'm gonna bid around a dollar, we'll say. My automated ad group that is complimentary to that manual campaign, or manual ad group, I'm going to probably put that bid around 50 cents to start with, so that's my starting bid.
I'm then gonna take this all products campaign, I'm gonna throw every product in there, and I'm gonna put that bid at like 20 cents. Something ridiculous where one sale will just make my cost look great. Most of the time, this is a very well converting ad group for us.
The reason that this kind of comes out and works out so well is because there's ... When you start to bid at certain levels, you're starting to show for certain keywords, especially in automated level. I want to show for some of that really random stuff that maybe you wouldn't expect, but I don't want to pay a lot for it because most of the time they're not gonna convert as well.
If you consider having that one all products campaign as just that safety net, you'll actually see a nice increase. Plus, our little back up plan is for those sellers who have a huge product line. You have thousands of products and sometimes it's a little difficult to make sure certain products are being placed. It's really easy to just select all, make sure every product is being advertised, so at least you have that running while you dig in and start to add it into other ad groups.
Franz, you've probably seen more campaigns than I have, and I would love to know what your approach usually is.
Franz Jordan: Yeah, it's pretty much the same approach. Essentially the idea that we follow and generally what we see working is trying to combine the advantages of auto and manual campaigns, and the advantages of the different niche types.
A question I get asked a lot is what's the best match type? Or, which campaign type is the best? The answer is, there is no best match type. Each match type has its pros and its cons. The pro of an exact match type is that you don't have any search terms that you don't want to have in this campaign. It's very narrow.
The disadvantage, obviously, is you might be missing out on some search terms. On the broad side, the advantage is that you have all those search terms, the disadvantage that obviously there will also be some search terms that are not relevant to your product.
The structure that we suggest in general at Sellics, is obviously one auto campaign, then one manual campaign and within the manual campaign, we suggest two ad groups. One for broad keywords and one for exact keywords.
This is something ... I wasn't sure if I got this correctly. You said you have a manual ad group as well? So which match type do you use?
Andrew Maff: I actually combine the two match types. Usually the reason it ... We have a very large spreadsheet approach, which basically Sellics can more or less automate in certain aspects. When we do the spreadsheet approach, we really utilize filters and being able to structure the spreadsheet.
Having them in different ad groups doesn't really seem too necessary in the fact that you can kind of see if you're one keyword that's broad match and your one keyword that's exact match have been right next to each other so you can see which one is working. Because, we always leave that broad match on, but we just start to lower that bid and start to raise that exact match.
We like to have them in one ad group. I have seen a lot where they get split up. Personally, I'm actually not a fan of phrase, so we don't even really use much of that, if ever. But, we really see them as just being totally fine within the same ad group and being able to track them next to each other.
Franz Jordan: Yeah, okay. That makes sense and if you look at them in Excel. About phrase, we have a similar take on this. We did some analysis on the different niche tribes and we found that the difference between broad and phrase is actually very insignificant.
We try to go with broad so that you really have the very wide net essentially instead of phrase, which is a little bit stuck in between the advantage of exact and the advantages of broad.
The reason ... Or let's say, the problem that I see if you combine them, or actually, I mean that's a question, how you handle it. If you combine both in one ad group and you look at them, so when you look at the results, do you look on the keyword level or on the search term level?
Because if you look on the keyword level then obviously you will have the problem that I see is ... So, you have the same keyword on exact as a keyword. Then within the broad keyword you have it as a search term amongst other search terms as well.
For the exact same search term, it might be difficult to really get the full picture, do you see what I mean?
Andrew Maff: Yeah. We typically look at both, basically. We're looking at a search term level as well as the key term level ... the keyword level. Basically, by being able to compare the two side by side kind of helps in a structural aspect. But, I can see what you mean in having them separated into different ad groups can help in terms of when you're basically just looking at a campaign to see which one's working.
On a search term, I also see what you mean how you need to see which keyword is working, but that's kind of where the exact match and certain bidding strategies kind of tweak that.
Franz Jordan: Yeah, yeah, yeah. I mean, what's interesting about sponsored products is that you have a ton of different options how you can structure them.
Andrew Maff: Exactly, and again, like I'd said, and I'm sure we're gonna say this five more times, it's an incredibly opinionated what works and what doesn't, there nothing out there that is from Amazon that says this is what we do. It's really just ongoing testing.
Franz Jordan: Maybe as a final question. How do you handle your very large selections with match but tried products. I'm thinking of Upper L a t-shirt in 20 different sizes, five different colors, how do you handle this?
Andrew Maff: We add them all into the same ad group, however, as certain products start to stand out and certain products start to stand out 'cause they're failing, then we separate them out into their own ad group. If I have a ...
I have my Steelers cup here. Let's say I have Steelers t-shirts and I have 15 different colors and a bunch of different sizes, I'm gonna put them all in one ad group, 'cause they're all relatively the same thing and people are gonna want to search for their size. If I see that one size is working, I may separate it out and start to do keywords like medium, does really well, or something like that.
Franz Jordan: Okay. Okay. So you gradually try to narrow. That makes sense.
Andrew Maff: Yep. Let's check it out. Oo. Sorry, we didn't even touch on this one. Budget. Our suggestion on budget, and I got this question today, I've gotten this question probably ten times this week, and I've probably had this questions a million times in the past, what do I spend? What should my ... if I'm starting an account what should my monthly budget be? How can I get my budget to X.
There is so many questions on this. The way we always suggest, this one, is because depending on how you structure your campaigns and depending on how many ad groups you have in there, we always suggest to start your budget very high, but keep your bids very, very low and only slowly, slowly incrementally raise those bids so that your ads have the opportunity to run all day, but you're not spending that much.
If you put a campaign for $500 a day, that may be astronomically high for you, but if your bid's at 20 cents, you're only gonna spend like $10 a day depending on how many ad groups are in there.
Again, Franz, I'd love to hear your rebuttal on that. But, our belief is, basically allowing the ads to show all day or you're just not gonna get enough accurate data on what's working and what's not.
Franz Jordan: Yeah, I would definitely agree for new products where you're testing out a lot and you're not sure how much you can actually spend and how big the market really is.
In general, our take on the budget is that the budget ... Basically you should never reach the budget. You should never hit the budget cap, for essentially a simple reason. If you have a profitable campaign, why should you stop running profiting campaigns at three int eh afternoon, if you could generate additional profitable sales a lot longer.
If you have profiting campaigns, you don't want to stop those campaigns. If you do have unprofitable campaigns, then again, you don't want to spend your entire budget on unprofitable campaigns. You either want to turn them off or make them profitable, or you want to change something without spending the entire budget.
We really see the budget as a kind of an air bag. Typical example, instead of typing 50 cents, you bid $50, it's a fat finger kind of mistake. You have compared to the clicking on your ads, and try to go back essentially, so those are the situations where we see the budget kind of being a safety net, an airbag for, let's say, unusual situations.
In the normal course of action, we believe that if you have profitable campaigns, you never want to hit the budget if you have unprofitable campaigns, also you want to change it as fast as possible without spending the entire budget.
Andrew Maff: Correct. Yeah. That I would agree with. I think the 500 was a little bit of an exaggeration, but I do agree that keeping it within what you're usually doing, yes, obviously makes sense, but, okay, we're on the same page.
Franz Jordan: Yeah, yeah. No, I agree.
Andrew Maff: So, bidding. Another fun question. I know we get this question all the time. Lowering your bid, does not necessarily mean it's gonna lower your ACoS, and raising your bid doesn't necessarily mean you're gonna bring in more sales. Depending on that keyword ... I think that whole mantra of marketing, the word is just it depends. It depends.
It depends on the keyword. It depends on what your bid was originally at. It depends on what your ACoS is at. Raising a bid is only gonna start to show you, if it's a broad match, it's gonna start to show you more and more for other variations of that keyword, but that doesn't mean you're gonna start getting more sales. It just means you're gonna start showing for more random keywords that you didn't select in the first place.
Then lowering your bid, yes you're bidding less on certain keywords, but now if you're hitting that perfect spot, but you lower that bid, you're now basically getting taken away from those keywords that you were doing okay with in the first place.
Franz Jordan: Yeah. I mean, it's always a bit controversial, these kinds of things, but actually it makes a lot of sense. I think the key metric, essentially, that decides on how much you can bid is a conversion rate.
This is really the central metric that determines how much you can spend. If you have a 20% conversion rate, that means on average, you need five clicks to generate a sale. That's what 5% conversion rate means. If you know that you can spend $10 for each sale, then you know that you need five clicks for a sale, then essentially that means you can spend two dollars on a click.
Now, let's say you have a conversion rate of 10%, that means you need 10 clicks and you can only spend one dollar per clicks, if again you have $10 to spend.
The conversion rate is really the key driver and the bid, just lowering the bid, exactly like I said, is really not ... might not necessarily drive down your ACoS. That's for sure.
Andrew Maff: Yeah. Very good. Maintenance. Ongoing optimization. We do a few different things here, we track each individual ASIN to see how it's performing. Basically, for the conversion, first of all, if you know what that certain ASIN is converting for then you know exactly what you should be bidding.
Like you mentioned, if that ASIN has about a five percent conversion rate, so you should know exactly what your max keyword bid should be. However, if it's low or if it's high it's gonna decipher if you can lower or higher your bids on average from what you're at currently.
We optimize every week. I know this is something we kind of touched on if we were gonna talk about this or not. The way we do it, let's say on Wednesday, so today, my team is optimizing for last week. The reason being is that 48-72 hour delay that Amazon has negating their click attribution with the whole ad to cart thing, but with that delay, we always like to optimize weekly.
The main reason being A, there's that delay, so you want to have a few days pass, then B, you really don't know I have some sellers who do really, really well on Mondays, and don't do well on Fridays. I don't want to optimize for Monday and then mess up my Friday and vice versa.
If I have a full week, that at least tells me on average what my week looks like. Now, it doesn't really take into account if there's a holiday, so it was a slow day, or it was a busy day, depending on the holiday. But, getting at least that week of data really seems to at least keep things stagnant.
Franz Jordan: No, for sure. I mean, the attribution on Amazon's side is seven days by default. We looked into the data, we found that for most campaigns it's actually, you probably get 85 to 90% of sales within the first three days, so I would totally agree that three days is probably the right amount to wait because it's not as long as seven days, but still long enough to get the majority, to capture the majority of the sales.
Waiting for this three days I think is really paramount because other than this, you always think your performance is terrible. The ACoS always looks terrible from the day. It only improves over the next couple of days.
I think that if you're running your ... Or if you're managing your campaigns manually, I think you're not having this routine of getting in every Wednesday and looking at the prior week or so, having a fixed schedule definitely makes a lot of sense.
I think from an automation perspective, obviously what we strive is to essentially manage them on a daily basis because obviously a software is a lot easier to ... As a software it's a lot easier to handle this than if you do it manually. I think if I were to do it manually I wouldn't do it on a daily basis, it would be too much. I would probably have the same schedule that you just mentioned.
At once or twice a week waiting for three days prior ... for the prior week.
Andrew Maff: Yeah. We're gonna get into headline ads now. But, let's go through. We've got a ton of questions, so we're gonna try and hit as many as we can real quick. Let's see what we got.
Franz Jordan: Andrew, let me just get another internet connection because I'm having some sound issues here. I'll be back in a second, okay?
Andrew Maff: Yeah sure. All right Franz is gonna work on tweaking his wifi, so I'm gonna try and knock out some of these questions you guys have. Let's see.
Barbara asks, "Once you have a strong manual ad, do you discontinue the automatic ad?" No, I don't. Typically, what I'll do is I'll get that bid significantly lower. That automatic campaign I'll bring it down, but I never want to get rid of it because you never know if it's going to start teaching you something, teaching you something new search terms, new keywords that you should be adding, anything like that.
It can't hurt to have it running at an extremely low bid. I would say just incrementally lower it and then give it another try. Let's see what we've got.
"Why won't you create a phrase match manual campaign. I have three broad phrase exact." So, I don't like all three. I don't like phrase mainly because if you're doing long tail keywords well enough and you're not having these small keywords like shirt or cup, then phrase really isn't gonna help you too much.
I like to have them all in one campaign. Franz mentioned he likes to split them up. It's a little bit of a different approach, but to me, phrase, which actually comes into next, Jane Smith's question about "Why don't you like phrase." To me, phrase ... I don't see enough difference and I don't see a lot of results from it.
When we do phrase, I always have better broad match terms and then obviously our exact match terms do better than that. When I have phrase, it really tends to limit the amount that it can show for, and it's not getting up the algorithm any higher than I need it to, so I'd rather have the broad match and have a better chance at showing for more keywords.
Coming back on Franz?
Franz Jordan: Yes. I'm back.
Andrew Maff: Good?
Franz Jordan: Perfect now, a lot better.
Andrew Maff: There you go. Cool. Let's see. Question to Franz. "When I used your strategy broad and exact ad group in one campaign, I usually have trouble that the budget is mostly used for only one ad group and I can't get enough results from the other campaign. How do you handle with the situation when you want to divide the amounts separate campaign?"
Franz Jordan: Yeah. Essentially what you do want to have is ... Basically the recommendation that we give is to start with the same bid for the broad ad group and for the exact ad group. Then what you see is that ... Or, the ideal scenario would be that you get all the traffic for the exact keywords that you're bidding on and your exact ad group and you only get the traffic for the other search terms that you're not bidding on exactly in your broad ad group.
Typically, what you see is a rather higher concentration in the exact ad group whereas the broad ad group. This is not necessarily a bad sign. I wouldn't say that your goal should be to have an equal spend on both ad groups. I think that it's rather typical to see that the ad does spend higher in the exact ad group because those are those big fathead keywords that you have for your product, whereas the broad ones or rather the search terms that you're not bidding on exactly and this leads to typically low ad spend.
I don't think it's necessarily a problem and I don't think it should be a goal to have a 50/50 split on the budget between those two ad groups.
Andrew Maff: Okay. Let's ... We'll come back to the rest of the questions. We'll keep going here a little bit 'cause I know we're already moving along. Headline ads.
You have to have at least three products. You guys know, hopefully, how the headline ads work. We'll touch on this. I'm actually a big fan, now, of setting up these Amazon stores and pushing my headline ads to the store as opposed to letting Amazon create the landing page.
If you're in Seller Central, and you've created this landing page before, you know what I'm talking about where ... Where'd Franz go? Franz. Hopefully he'll be back.
Setting up the landing page you select your handful of products and then Amazon creates like a search term page where it's ... Or that regular search page where it's just the three products or however many products you picked.
If you create this storefront, this custom storefront where essentially you have categories, so here, we selected this iPhone 4/4s cases. I can run an ad directly to this URL and if you'll see down here at the bottom, you can actually add the custom URL here. You can actually run the ads to a more branded and a lot better looking area.
If you have access to AMS, this is a great way to do that. Unfortunately, from what we've seen, at least as of today, or at least as of when this webinar started, there is still no option to have that custom URL in Seller Central, so it's AMS as of right now. Hopefully that'll get freed up soon. We see that as a great branding opportunity as opposed to just that stagnant regular landing page that Amazon creates.
Franz Jordan: Yeah, totally. I mean, I think for online search is quite interested. When you look at the budget, how much money you spend on AMS, do you have different budgets that you plan for sponsor products and headline searches, and product display? Do you have a fixed budget in mind that you said headline searches and branding and sponsor product's performance marketing or is it just your overall AMS budget that you consider?
Andrew Maff: It kind of depends. In most cases, we're doing sponsored products on Seller Central, while doing headline ads and AMS so it allows us to pretty easily see our budget. Otherwise, we still just kind of take it as it is, where we do a combination of the metrics, but we also break them out to see how each campaign's doing, how the headlines are doing versus how the sponsored products are doing.
That way we can kind of see a little bit of that. I know there is that whole branding ... The way that they count the conversions with headline ads is a little ... is obviously different from the sponsored products.
But, if you are really trying to build a brand on Amazon, or off Amazon for that case, we really do see it as a 'you got that sale, so it is really is beneficial.' It's just with headline ads, you want to take them with a little bit of a grain of salt before you can get so excited about your ACoS.
Franz Jordan: Your fantastic ACoS.
Andrew Maff: Yeah, exactly. This is one I wanted to touch on this just because I've had a lot of sellers and a lot of headline ads that if they were your headline ads, I'm sorry, but I clicked on them.
I see these landing pages where ... In this case, we picked three relevant items. But, I see these sellers who are maybe resellers or they're private label sellers, but they have a bunch of random products that don't really go together.
They just make one headline ad where they just throw a bunch of random products in there. You really need to have that copy that you put right at the top there, right here ... Where's my mouse? Right there. It really needs to say exactly what you're about to be shown, really because if you just say here is our brand's products and no one knows what your brand is yet, you're really pushing them to a landing page that just doesn't have the products that they may looking for.
So, you're gonna be looking at a really high ACoS of just no sales from people just going in and looking at a random mix of the products that you put in there.
Franz Jordan: Yeah. I think that this is pretty straight forward, right? Nothing to add here.
Andrew Maff: Yeah. You would think that one's obvious. This one, with optimizing and with adding keywords and all that kind of stuff, it's really the same. We have the same approach we do for sponsor products. Thankfully, I think this was ...
I always feel like it was like two or three months ago, but I feel like it was a lot longer than that now where AMS finally released some more reporting that we got a while back. Now you have it in Seller Central. Where you can actually pull the keyword report and see how headline ads are doing and be able to kind of clean them up a little bit.
To us, it's the same approach. I don't know, Franz, do you have a different approach on how you optimize your headline ads?
Franz Jordan: Again, I think it's hard to track the performance in the headline ads, I think. Generally, we see that it takes less effort to optimize because you just have less possibilities to really track what's going on. I think that's really ... I understand that Amazon considers headline searches as branding ad type. I don't agree with this. I don't see why headline searches, which is at the top should be branding, and right below where your sponsored products, this is performance marketing.
I just don't follow this logic. This is the logic that they have. It's just very hard to analyze the performance and it's essentially impossible to really track the performance. It just makes it more complicated, also, to really have a consistent optimization strategy.
I'd say, yes, basically the same idea as for sponsored products, but with less overall because it's less manageable.
Andrew Maff: Yeah. Basically what we do, is we let the sponsored product ads we're running teach us which keywords we should be adding into the headline ads. You get a little extra data there, but I agree that the branding approach doesn't really help too much, which I just touched on.
One of the things we like to do as well, when you're pulling that search en report, go into your listing and make sure that those search terms are prevalent. Make sure they're in the title or they're in the images, or in the bullet points.
We always see these sellers that are optimizing their PPC campaigns, but they have all this data and they're not doing anything with it. They're not going to their listings and making sure that these keywords are prevalent. You might be getting a lot of clicks on something that you think you should be converting for and you're not, and the reason might be because that keyword isn't prevalent enough in your listing.
My suggestion would always be to take that data that you're getting from PPC and go edit it in your listing.
Franz Jordan: Absolutely. This is probably the prime strategy. I know that some strategies out there that say that's the only goal of sponsor products, to just find out those search terms and then optimize the listing base on this. I wouldn't agree with this. I think it's a performance marketing channel and it needs to perform. It's not the keyword research tool.
You can add this as an additional outcome, but this to me is not the prime reason for using sponsor products.
Andrew Maff: Yeah. Let's go onto product display ads. I'm not a fan. I'll say that right off the bat. They're really difficult to deal with. Optimizing them you really don't have too much to work with. For those of you who haven't, who don't have access to AMS, I'll try not to touch on these too long. But, essentially, you can select your own ASIN and then you can select competitor ASINs or you can select an interest.
Nine times out of 10, we're gonna select competitors ASINs. If I'm selling my Steelers cup, I'm going to run this ad on other Steelers cups out there and try to steal the sale. That's really the only best approach we've seen. Obviously, you can do complimentary products. I have my Steelers cup, maybe I want to run it on Steelers koozies or something like that.
Really those are the best approaches we have. We do these when we know we can beat out a competitor otherwise, we try not to touch them too much, just because we've seen success with them, but sometimes it's few and far between. You may be really successful with one ASIN and really not successful with another and then they just kind of break even and why'd you even bother?
Plus, I'm kind of guessing that they're gonna get rid of them. I'm not really sure. I don't know if you think that or not. I know they're putting more work into AMS, but it's still not available on Seller Central, so my theory is they might get rid of these one day, but we'll see what happens there.
All right, so we'll touch on how we're doing this manually. Then we're gonna let Franz show us how to do it on automated. Obviously, if you haven't been able to tell yet, we're big fans of the manual approach. What we'll do, is we actually pull the search and report ... the reports each week and we'll look at how each key word's doing week over week. Mainly because what you can do is ...
Especially if you're doing this manually, let's say one week you have a keyword where it's got one or two clicks, you've spent two or three dollars, and you're like, "All right, I'm gonna raise this up a little bit, let's see if I can get this keyword moving." Next week you're at four or five clicks. You're like, "All right, it's still okay. Let's raise it anoth."
Then the next one, next week you're at 15-20 bucks and you spent too much. You're like, "All right, now I want to lower it." You keep fighting with this keyword week over week, you never look to see how this keyword's done over the lifetime or how it's done over the past month to see if maybe you've spent 100 or $200 fighting with this keyword that you thought was going to make you money. That's why we always look at these week over week.
We always add the keywords when you're pulling your search term report. We typically, we do a monthly audit to kind of make sure that we're pausing correct keywords, we're adding to the search term report. This really allows us to track seasonality for things and really kind of look at ...
Sometimes the reason I think the manual approach works is, because, you have the seasonality of products. Sometimes you want to sell through a product and not others. Sometimes you want to be less profitable on certain ASINs than you do because you have a new product and you want your BSR to go up. There's a bunch of different reasons, and obviously they'll be kind of opinions here, but this is essentially how we manipulate these spreadsheets to see how these campaigns overall are doing.
This is just a screenshot of pretty much how we start to manipulate all the reports that we're getting week over week. So, Franz, I will let you take the other side of this and give us the automated stuff.
Franz Jordan: Obviously, it's a lot better to automate everything, right?
No, no, no. I think what you said is very true. I think that is married for both things. Obviously, I mean, the advantage of automatization is that if you look at sponsored products, really it means essentially when you break it down it's a constant repetition of the same steps. Essentially, you're adjusting bids, or actually have profiting key words, so you want to increase bids, see if you can get more out of it.
You have unprofitable keywords where you want to decrease the bids. You have keywords with no impressions, you want to increase the bid to see if you can get more impressions. Essentially, on a bid side, there are two things you can do. You can either increase it or decrease it.
Then, for the keywords, essentially is also the same thing. You either pause the key word or you set it negative if it's not working and you add new keywords to try new keywords. I understand that for many sellers it looks like a very complex operation and I agree that in the Amazon space, it is probably the most complex aspect of selling on Amazon.
If you break it down, essentially, it is not as crazy or as difficult. It's not rocket science as it might seem. Essentially, it's just a repetition of the same steps. If you agree that it is a repetition of the same steps, then if you go to the next slide please, it really ... You can essentially break it down to if, then, if this then that kind of relationship.
It's like an algorithm, so you have a certain condition that needs to be met within a certain time frame and if this condition's met within the time frame, then you have a certain action that you wanna take. If you can go to the next slide.
Essentially we have one example for an auto campaign. You want to set the search terms to negative that don't generate any sales. I think that this is ... I'm pretty sure you're doing something similar on the manually on the weekly basis, probably, right?
Andrew Maff: Yeah.
Franz Jordan: Essentially the condition would be, if you have zero sales, the timeframe would be after 25 clicks, and we can talk about what is an appropriate timeframe. If you have zero sales after 25 clicks, then you should just set your search term to negative. This would then be the action.
Another example would be, on the manual campaign, you want to review the bid for nonprofitable keywords, so if you have an ACoS that is above 25%, so let's assume you're aiming for 25% ACoS, you see that you have an ACoS above 25% after 25 clicks, then essentially the action is that you want to reduce your bid to see if you can get profitable there, right?
Andrew Maff: Yeah.
Franz Jordan: That's some very, I think, straight forward stuff. Now, the main question is, how do you set the right limits? That's the question. What's the right ACoS, what's the right bid, what's the right time to wait, by how much should you adjust the bids? Let's just look into those different parts.
The first question obviously is, on the condition side, what the ACoS, what ACoS limit should I use in my PPC campaigns? This is not just a question if you automate your PPC campaigns, this is just as valid if you do it manually.
There are different ACoS limits that you should have in mind, which we'll see on the next slide, Andrew, if you could just go there.
Andrew Maff: Yep.
Franz Jordan: Essentially, if we break it down on the product side, let's just assume we have a product, we have FBA fees, we have Amazon fees, we have unit costs, and we have shipping costs, or overhead costs. At the end you will have a profit margin for this product. In this case it's just an example, 20%.
You have a 20% profit margin before I sponsor products. Before any AMS spending. This is important. This is before any marketing spend. If you have a 20% profit margin, you know that you break even ACoS will be 20% because your break even ACoS is essentially your profit margin before PPC spend.
That means if you have a 20% margin and you spent 20% of your sales price on marketing, obviously, you're net profit would be zero. I think this is obvious. The question that you need to decide as a seller is, how much net margin are you aiming for after PPC, after deducting all the ad spend.
So, in this case, we just assume we want to keep a five percent net margin after PPC spend, that leaves you with a target ACoS of 15%. If you have 20% in margin, you want to keep five percent of this 20% after ad spend, that means you can invest 15% of your sales price into advertising and this essentially is then your target ACoS.
Sometimes people just ask me, "What's a good target ACoS?" It really depends on your margin that you have before ad spend and the margin that you want to keep after ad spend. There's no valid answer to this, no global answer to this.
Andrew Maff: Yeah.
Franz Jordan: That's for the condition. For the time frame, and i think that this is the most interesting question, to be very honest. The question's how long should you wait before you adjust any keywords or any bids. The first guess would be that time frame is measured in time so something like days or weeks or something.
We believe that the best way to measure the time frame essentially, is in clicks. If you could go to the next slide. The idea behind being, that you want to wait for the number of clicks or you should wait a certain number of clicks, which is the expected number of clicks that it takes to generate a sale.
We talked about this before. If you have a conversion rate of five percent, that means that every 20 clicks, on average again, should lead to one sale.
That means, if on average you have ... you need 20 clicks to generate one sale, why should you already pause a keyword after five clicks or so? That would be ... I would say this would be unfair to the product or to the keyword because the keyword and the product, they need 20 clicks on average.
If you pause a keyword after five clicks already, then this keyword didn't have the chance to really prove its worth. We in general, we suggest that you wait 100 divided by the conversion rate. This is the number of clicks then. If you have a five percent conversion rate, it's 100 divided by five which is 20 and then you would wait 20 clicks because this is the number that you would need on average to generate a sale.
If, then, the keyword doesn't perform and does not generate a sale within 20 clicks, then you can start lowering your bid afterwards because obviously your conversion rate is not as good as you expected in the beginning. Right?
Andrew Maff: Right.
Franz Jordan: Then, finally for the action, by how much should I adjust the bids? The rule of thumb that we give you is just like 30%. The reason for this is fairly simple. If you choose a lower number. Let's say you choose to adjust it by 10%, it just takes a long time until you get into significant area.
If you start at 30 cents, for instance, and you increase it by 10%, that means you go from 30 to 33. Then you go from 33 to 37, and then from 37 to 40, 41 or so. It just takes many interactions just takes a lot of clicks get to different level of ... it's a different bidding level, which is why we suggest to take 30%.
On the other side, if you take a very high number, so let's assume that would double your bids all the time, so you go from 30 cents to 60 cents, then the chances are very high that you actually jump over the right bid that you're aiming for. You're always going back and forth between some very extremes, and you don't ... you're not in the middle.30% adjustment, it seems about right. That's the rule of thumb that we give.
Some best practice, I just collected a few rules, and this is what we suggest to our users and Sellics. If you could go on the next slide. Then we have four different rules. Essentially, what you can do, in Sellics obviously, is you can replicate those rules. Then the software will automatically manage your campaigns following those rules that you've set yourself.
The recommendation that we give in general is to just combine those four different rules because those rules would really ... It's definitely not as sophisticated as when Andrew and his team is managing it, but for sellers who just want to reduce the amount of time that they manage their campaigns and say, want to get a good output, it's probably not the best output that you will get if you spend 100s of dollars a week to manage your campaigns.
It will, for many sellers, lead to really good results already. The combination is obviously, if you're ACoS is low, after I'd say, then you'd want to increase your bid to see if you can actually get more traffic on this keyword. If the ACoS is high, let's say above 25% after 20 clicks, then you want to decrease your bid.
If you're above your target ACoS, you want to decrease your bid. Then, finally, if you have zero sales after 50 clicks, then you want to pause the keyword and if you have few impressions, so if you have less than 100 impressions after seven days, again, you want to increase your bid because you want to test if you can actually get more traffic on this keyword, right?
Andrew Maff: Right.
Franz Jordan: Those four rules, if they work together, they will keep your ACoS ... they will keep your bids around a certain target ACoS. In this case, the target ACoS is 25%. Obviously, you should calculate your own target ACoS. This is why it says adjust for your own product, and same, obviously, counts for the number of clicks.
Also, the number of clicks that you should wait, depends on the conversion rate of your product. Those 20 and 50 clicks are just some best case scenarios that work for many sellers, but in an ideal world, you would calculate your own click thresholds depending on your product.
Then you set up those things in Sellics, and then Sellics will take care of this every day. We will check your campaigns and adjust the campaigns based on the rules that I just showed.
Andrew Maff: Beautiful. All right, so, that just about wraps up everything. Now, we're gonna start clawing through some questions here. Please start to send us your questions, I'm gonna get through as many as I can. Obviously, even if you don't have a question, I would love to know if you're ready to take on your Amazon PPC, if you learned enough, if you're gonna take any of the approaches that we spoke about today, or if you think it's all BS and you're not gonna do any of it. We'd love to hear that too.
Franz Jordan: That would be unfortunate.
Andrew Maff: That would be unfortunate, but hey. Sometimes it happens.
Franz Jordan: One of our BS.
Andrew Maff: All right. I've got a bunch of stuff, we're gonna be giving away. Listing optimization is obviously a very big thing, it obviously has a lot to do with your PPC. Obviously your keywords and everything like that helps with your indexing, helps with your automated keywords, your broad match, all that fun stuff.
We have our own listing optimization Ebook, it is not gated. There's nothing you gotta do, all you gotta do is go to that link and it will download. It's a bit.ly, so bit.ly/listingebook, and it's all yours, have it. If you don't download it, I'm gonna send it out in an email anyway, so whatever you want to do.
We're also gonna be giving away a handful of 30 minute PPC consultations. I'll go through your PPC as much as I can and we'll look through everything and see what we can do to help you out with where your cost is at. I'm not selling you anything, I promise. I'm just gonna be there to help.
Then we're gonna give away a PPC audit as well. This is something we do in house that's usually a pretty decent sized project for us here where essentially we go through and we're gonna pull all the data that we can over the past 60 to 90 days for depending on which report we're pulling and we're gonna show you exactly how your keywords have been doing, which keyword you should pause, which ones you should add, how you should structure your campaign.
We're gonna send you an entire report on what you're currently doing versus what we believe you should be doing, so that's SellersChoice.agency/giveaway. It's gated so the password is amazonppc.
I did look at that page a little earlier and I do apologize, it's not available on mobile right now, which I know, kill me for being a marketer and having something that's not mobile friendly. But we just created it, so please bear with me.
Franz Jordan: You're very generous here.
Andrew Maff: I try. I like to give stuff away. What can I say? We have a ton of questions. You want me to start? Let's start going through these.
Let's see the first on we got. "To collect keywords," this is from Sam, "Do you collect keywords from auto and add them to manual, and add them as a negative in auto. Amazon emailed me saying I shouldn't have two campaigns for the same ASIN. They suggested I should do either auto or manual, but not both. I'm confused."
Franz, you wanna take that one or do you want me to?
Franz Jordan: I'm surprised that Amazon's sending out this email. 'Cause I've heard many people at Amazon that's saying the exact opposite. With the first thing you said is what we would recommend. Run both. So, when you take it out of the auto campaign and you add it into your manual campaign, what we suggest is typically to put it in as negative phrase in your auto campaign so that you don't have the redundancy.
Essentially, you take it out of the auto campaign that's in the manual campaign. It's not showing up in your auto campaign anymore. Makes it a lot easier on the analytical side. If you want to track performance, it makes it a lot easier if you don't have the same keyword in two different ad groups.
Andrew Maff: Yeah. We actually have had a handful of sellers who have gone through ... It's got a name and I can't remember it. It's basically like Amazon running your campaign for you, and from what I've learned, just like Google, if you've ever had AdWords run for you from Google, they don't know what they're talking about, which shocks me.
But, every time we've had a seller do this, the seller obviously wants to give it a try and it makes total sense, but we've made the mistake many times where Amazon actually suggests a lot of crazy keywords, and a lot of really high bidding, and a lot of weird structure, and every time we just let them do it for a month or two, it went down.
If Amazon is telling you that you shouldn't have the two campaigns running at the same mason and that it's a manual versus an auto, they're wrong. Another reason would be, the auto campaigns, you'll actually show up in that similar sponsored products area in certain listings, whereas on a manual campaign, you don't show up in those listings for that. If you turn that off, you're not gonna show up there.
Let's see, I have another one coming through. "Is it okay to run and unprofitable campaign for a new product with the means to start ranking organically for a certain search term."
You want to start there?
Franz Jordan: Yeah, I mean, there's been a lot of discussion about this. This is ... I know that Chad from Scobana and I, we talked about this already, or also, the question is, how big is the impact really on the organic rankings?
I personally, I believe there is an impact on organic rankings, so yes, it does make sense. The problem with that strategy is that if you are in any given category, you will always have this one seller that's currently starting out and they will just outbid the entire market and this is driving up the bids ... You're shooting them through the roof.
The problem is that eventually, sellers will just continue to see bid more, and more, and more, and more just following this logic of improving the organic rankings. Eventually will just be too expensive to run any profitable campaigns for anyone.
Yes, I think that profitability should not be the key priority for beginners, because it is important to show Amazon your progress performing, so Amazon can gather some data, but I think you should still have a certain threshold that you're aiming for. And not go spend whatever it takes just to rank on the first page.
Andrew Maff: Yeah. Got another one. "Sellers are trying to manage launch freezes, organic ranking profitability et cetera, what is the best way to stay on the same page as a PPC manager who just sees ACoS?"
I feel like you should see more than ACoS. You should be looking at a lot more than your ACoS if you're a PPC manager.
Franz Jordan: Yeah, so I'm not exactly sure if I understood the question correctly, so maybe you can take it.
Andrew Maff: Okay, so let's try this. Sellers are trying to manage launch phrases organic ranking, profitability, et cetera. What is the best way to stay on the same page as a PPC manager who just sees ACoS. If you're a PPC manager, you really should know the other things that it gets involved with. Your listings and your organic sales, there is a huge correlation between organic and PPC.
Sometimes, someone might click and come back later and purchase. There's a lot that you still need to take into account. I think the real answer here is that you really shouldn't just be looking at just ACoS, you should be looking at a lot of other things and looking into what's actually working and that could be correlated to PPC.
Let's see what else we got. "For using automated PPC with Sellics, can we still make manual tweaks to the campaigns?"
Franz Jordan: Yeah, absolutely. Essentially, the way it works in Sellics, you can create a certain set of rules, like I just showed before and then you can activate this set of rules for a specific ad group. Those rules will be running, however, you always have the opportunity to interfere manually and to change bids at keywords, pause keywords, whatever you want to do. Whenever you do this, we will start counting it from zero again.
Let's say that you say, I don't know increase bids after 20 clicks. If you increase the bids manually already after 10 clicks, then we set the counter back to zero and start counting up to those 20 clicks again from the moment that you make a manual change.
Andrew Maff: Okay. All right, so let's knock out one more, here. We're gonna take Sign of That Solutions, 'cause he's getting a little cranky. He wants to know "Long tail keywords, broad or exact?"
Franz Jordan: From a theoretical perspective, exact, from a practical perspective broad. Theoretically, I think that in an ideal world you would go, you would only use exact, but the truth is that a long tail here, was you don't have enough traffic. If you don't have enough traffic, there's not enough clicks, you will essentially never be able to really make a decision whether it's really a good or a bad keyword.
In order to aggregate a certain number of clicks that puts you in a position where you can really make a decision whether this is a good or a bad keyword, I would go with broad for long tail keywords.
Andrew Maff: There we go. All right. So we're well over an hour here, so we're gonna go ahead and wrap this up. I am available social, it's right there. You can email me. I will try to answer as many other questions as we can, if Franz, you want to give out your information, we're obviously not gonna give it out for you.
Franz Jordan: It's email@example.com. Feel free to send me an email. Fj@sellics.com or you know, you can find me on Linkedin as well. Feel free to reach out to me any time.
Andrew Maff: Yeah. Big thank you for every one for joining us. Obviously we appreciate it. I'll be sending out the slides as well as the replay will hopefully be in the next couple days, or if not, later today.
If you've got any other questions, feel free to reach out. Thank you so much for joining us. Good luck out there. PPC's kind of a pain, but you'll figure it out and you'll get it if you're struggling with it.
Franz Jordan: Stay strong.
Andrew Maff: Stay strong. You got this, you got this. Franz, man, I really appreciate you joining us. This was awesome. This was a good one. I felt like we got a pretty good amount of information out. If we didn't answer your question, I apologize, feel free to message us. But, hope you all have a great day.
Franz, appreciate it.
Franz Jordan: Thank you. Sorry again for being late. This was very ... Being German, it's really ... It's a bad day for me. Being late to a meeting is really unacceptable as a German, so I apologize and I hope people got value out of this, despite my delay.
Andrew Maff: No worries buddy. Appreciate it. Thanks everybody. Have a good one. Good luck.
Franz Jordan: See you guys. Bye, bye.