Lazar and Brent are talking about auto campaigns in the episode number 6. Why are they important? How can they help you with researching new relevant keywords and increasing sales? Listen to this episode to find out!
This episode is about the auto campaigns which are easiest to set up and optimize. This type of campaign is a great start if you want to advertise new products or catch relevant keywords which you can use for manual campaigns. Dive into this new PPC adventure and get some useful tips and practical examples from these two experienced PPC agency owners. Don't forget to check out our website and follow us on Facebook!
Lazar: (00:00)
Hey guys, welcome to the Wild PPC Bunch podcast. My name is Lazar and I'm a PPC nerd. I have over 10 years of experience in online advertising. And currently I'm the owner of the growing Amazon advertising agency called Sellers Alley.
Brent: (00:14)
And I'm Brent, the owner of AMZ Pathfinder. I started this company five years ago and we've been working in online advertising since 2013
Lazar: (00:21)
Every week, we will spend around 30 minutes covering one topic and it will get nerdy, I promise. We'll prepare a topic, covering everything from PPC basics, in-depth strategy and current trends.
Brent: (00:32)
One thing's for sure you won't be bored and you'll hear insights, tactics and ideas straight from two experienced agency owners. So strap in for the ride and enjoy.
Lazar: (00:50)
Hey guys. So today we're going to talk about one interesting topic. Um, it's automatic campaigns and what's their real value. So Brent, you have been using automatic campaigns since like the start of time.
Brent: (01:08)
Since day one. Uh, they have been there since the inception of Sponsored Products, which I think was late 2014. Along with very basic, uh, manual campaigns, which were broad match only at that time if my memory serves correctly. But yeah, they were definitely like one of the, uh, the grandparents of the advertising we have on the Amazon platform these days.
Lazar: (01:25)
Yeah. I remember first time when I saw automatic campaigns, just like you said, like four or five years ago or something.. Before that, I used to work on AdWords and Google ads. And, um, like for the guys that are coming to Amazon space and they don't know the match types and campaign options and all of it, i's more or less something like, um, DSA campaigns and campaigns ,like dynamic search ads. That's something that I try to link together because, uh, automatic campaigns are more or less there to give Amazon an option, algorithm an option, to try to target your search queries by what you have in your product listing, basically.
Brent: (02:12)
Yeah. And I think, um, they all have a really low barrier to entry, right? So even if your average, Joe, Amazon seller, you don't know much about PPC, cause that's not your specialty, you know, you're focused on product and sourcing and running other parts of the business. You can get an auto campaign set up in five minutes and that can start driving sales and at least visibility and, um, you know, impressions within that hour. Right. And you can start to see that data and pull that out of there. So yeah, really low barrier to entry and to go back to the Google background, uh, aspect, I mean, uh, I also have that kind of same background. And when I first encountered auto campaigns with Amazon, I was immediately suspicious actually, because anytime that the advertising platform was to like manage the keywords for you. Um, if you have any experience with PPC advertising, you know that a lot of stuff's going to get caught in there. That's like not the best. Um, and at that time, uh, we'll get to this later, there was no ability to set negatives for ASINs, for instance. So you'd just be spending against things and voluntarily you didn't want to. So there are some drawbacks to them. Um, but actually in Amazon's ecosystem, which is a very closed loop with advertising auto campaigns do have a lot of things to bring to the table if they're properly used.
Lazar: (03:25)
Yeah, exactly. I completely agree with you. As you mentioned, there are no brainer for everybody that starts selling on Amazon. It's super easy to set up automatic campaign. There's literally not much that you can do. Well now, as Amazon is adding more and more features. Oh, and we're going to cover all the features in a couple of minutes, but like you have more and more and more options with automatic campaigns. And even though you're letting Amazon, uh, target customers, basically you, you have a lot of control, like what kind of customer you're going to target? What, what kinds of stuff you don't want to show up for and so on. So basically when you create your perfect setup of, of the account, one of the first things that you want to have, there is basically automatic campaigns. Uh, we use them as safety nets. So we use them in whole bunch of different strategies. So one of them is, uh, the one when it's just like safety net with low bids. And like, that's it, you have manual campaigns that are triggered by keywords and you can use for some other campaigns, ASIN targeting and so on. Like, uh, but for automatic campaigns you just leave everything to Amazon and you harvest what you get there.
Brent: (04:46)
Yeah, exactly. So they go from what we call the pipeline at my company. But you have this similar process, I'm sure. It's called many things, but essentially you're taking the search terms and now ASIN targets that are working and you want to take those and migrate them into campaigns where they're, uh, isolated, where they have more dedicated budget. You can control the bids and exposure better. So auto campaigns are in a way, uh, one of the main harvesting methods for that, um, at least that's, that's how we run it. And that's, that's probably the primary use of them I would say.
Lazar: (05:17)
Yeah, for those that don't know how to harvest them, basically you go to search term reports and take the data from automatic campaigns and basically that's it. There's not much there. So you take the information about search terms or ASINs and like for ages, there, there was a huge problem. We mentioned, I think in a previous episode about how to negate ASINs in campaign. And I, I remember telling you last time that one of the guys that we worked with heard somewhere that if you add negative ASIN or something like that into product targeting campaign, you end up like teaching algorithm that your automatic campaign shouldn't show up for ASIN. But finally they solved it. There is no witchcraft.
Brent: (06:09)
That was an interesting theory and a fun one to bat around it. You know, I don't have any proof either way, but yeah, I think the end, the end of the story is that late July, late July of 2020 they finally solved, putting that in air quotes, the problem that they probably could have solved earlier. And the implementation of it is quite good. I mean, you can add a whole bulk list of them in there. You can do a, uh, you know, just a bunch of like, uh, ASINs broken up by a line break. Um, and then it's quite clear that you're excluding targets specifically to product targets, ASINs. Uh, the actual keywords is another sub tab inside the little tool to do it. So I'm really happy that's finally happened and, you know, put out common, all points bullets, everyone on the team, day happened and like, okay, implement this as fast as humanly possible because this is going to be immediate cost savings, uh, help us with duplication of search term issues and like exposure issues. Um, and yeah, just basically get it going as soon as possible.
Lazar: (07:10)
Yeah. When it comes to strategies about, about automatic campaigns, I know that people are using them for literally everything or anything as safety nets, like regular ones. So you got one product, one ASIN in one ad group that's targeting methods. Or some of them are using one product per campaign, just like single ad group campaign with, with one ASIN. Some others are branching out with more, more ASINs inside of one campaign. Some of the interesting theories and some of the interesting strategies that we tested and that we saw in the accounts is having all of the products in one ad group.
Brent: (07:54)
Right? Yeah. It's such a fun one to do because it's actually like, it's kind of like naughty, you know, it's going against a lot of the best practices that we usually preach and observe. But we have one of those catch all campaigns. We call them catch all. And I really like your safety net kind of analogy that we cause we do that too, but we don't have such a nice name for it. Um, the catchall, it's just like, yeah, put everything in there, 15 cents, US let it turn. Cause it's going to come back with a really great return on ad spend typically. But, you know, you're not going to have any targeted efforts that are going to boost keyword rank, or really blow out the doors of sales. But, uh, yeah, it's a catchall, you guys do the same thing.
Lazar: (08:30)
Yeah. The other thing that is like super similar with what you're doing is, um, a nickel campaigns. I've heard so many people creating, um, auto campaigns with like one ASIN in them and like, just like the regular ones. But they, they literally place the lowest possible bid on the market and they keep it that way just to see if something is going to happen. It's like from 2 cents to 10 cents, depending on the niche, like where you, where you can place the lowest possible bid and you keep it that way. From our experience, like you get one or two sales per week, that way, and ACoS is never above 5%
Brent: (09:14)
Sure. Because you just got lucky like one or two times. And, uh, yeah, continuing on the, uh, on the topic of strategies. I mean, two things we've also tried and don't use as widely, um, is kinda more of the Brian Johnson school of thought with water falling, right? So you might have one campaign that has a low budget and really, really high bids for auto. And then after that spends or is exhausted and hits its cap for the day, um, you know, it goes over to the like normal bid level ones, and then those just run and turn continuously. Um, another strategy we've used is like a 72 hour like marathon approach. So you say: Hey, this product, what if we just went in and we put a high budget, very high bids, and we ran this auto for 72 hours and we just take all that data from that giant burst of exposure and then process that and see if we turn up anything new from that. That's an interesting strategy we've used intermittently. I wouldn't say that's something we do regularly, but that could be, that could be fun if you have the budget to support it and the clients on board with the idea. Um, so I also put that out there. It's just like a tactic.
Lazar: (10:20)
I'm not really sure. I think Danny McMillan is a fan of that one.
Brent: (10:24)
Yeah, we probably got that from Danny. Yeah. I mean, we talk frequently enough that I've definitely had a mind meld with him over some things.
Lazar: (10:31)
Yeah. That might happen. So, and what do you think about adding a negative, exact keywords from other campaigns? Like everybody should add negative keywords in automatic campaigns for something that is irrelevant, but what, what are your thoughts about basically adding negative keywords that are positive keywords basically in other campaigns?
Brent: (10:56)
Right. So you're using negatives basically as traffic police, you know, they're not there to block the bad spend. They're there just to redirect traffic. Personally, a big, big fan and proponent of it. Um, and the software we use, it helps automate some of that process, um, to, to make sure that the traffic is flowing in the right direction. I've heard some people say, and let me know what you think about this, that like that's bad to do because you're, you're training Amazon to not show your, uh, products for a particular ASIN or keyword, but I've never seen any evidence to substantiate that personally. And that's not really how, you know, uh, negative keywords are set to work. I will say, like in any account, the distribution of negatives should be much greater in the match types where it's more open and kind of like a, you know, not as guided spent. So your broad and your auto campaigns are gonna have by far the most negatives phrase...
Lazar: (11:50)
Well, what we usually do we don't negate those keywords necessarily and by default. We, we let them work at the beginning and if we want to optimize, spend more, and if we want to have more like passive approach or something, in that case, we would negate them. But one of the things that I heard from many sellers saying about not doing it and... It kinda makes sense. Automatic campaigns sometime have some ad positions that are not the same as for Sponsored Product campaigns with manual targeting, with keyword targeting. So that's an interesting idea. I'm not really sure about that, but it kind of makes sense in case Amazon..Because they are testing like whole bunch of stuff all the time. Like one of the latest things that they saw, I don't know if you saw it, when you go to product listing, I think, or on the Amazon, like on the search result page on the right side, you see what you added into cart?
Brent: (13:01)
Uh, yeah. I saw that two days ago, uh, when I had accidentally added a product, I was looking at into the cart.
Lazar: (13:07)
At one point, probably they're going to add some, some ads there. So I'm pretty sure that's just, the idea, it's like just the options. And that's the main reason why I don't feel like we should do it all the time. Like if you have the budgets, if the numbers are okay, I would still keep automatic campaigns working just because of some crazy stuff that Amazon is doing from time to time.
Brent: (13:33)
Good point. Yeah and I agree. I think the place where they're going to test new ad placements for Sponsored Products is probably going to be for the auto efforts first. Because I know that autos show up on, um, I mean, this might be outdated information at this point, but all autos will show up on checkout pages. Like if someone buys something and they go to a checkout page, I think only auto campaigns get placement there. Um, you know, feel free to correct me. Anyone who has a better idea about that, because Amazon is very close to the vest with these things when it comes to ad placement locations. But I think that auto is the first to receive those new rotations and changes. So I see the validity to that.
Lazar: (14:10)
Yeah. There's one thing that's definitely changing stuff a lot. That happened like couple of months ago, maybe a bit more, they added close and loose match type. What's your, what's your experience with closing loose match when it comes to automatic campaigns? I have mixed feelings.
Brent: (14:28)
Yeah, sure. And I'm going to blow your mind. I'm going to blow your mind Lazar because I looked up Amazon's blog post on this and yeah, it was actually August of last year. So yeah. That's how fast time moves with Amazon advertising. You're like, oh, that feature just came out and then you look at it and you're like: oh, no, actually it's been out for a year.
Lazar: (14:48)
The thing is like, Amazon is adding a whole bunch of stuff and they're moving everything and everywhere. And like, I didn't literally have idea what happened last. So yeah, I'm amazed knowing that it happened a year ago.
Brent: (15:02)
Yeah, it's been a whole year. So when this came out, I was like, really happy about it because we have close and loose, which are keyword types. Right. So in auto we get keyword and ASIN and then you have substitutes and compliments, which are ASINs or product targeting. And that wasn't clear at first, but after we analyzed the data like, Oh, okay, so these four categories. And, uh, what we'll do then for our best practice, we'll build individual ad groups per the different targeting types. Amazon actually calls them match types in the blog post, but they're not really match types if you ask me, they're targeting types. And so, um, we'll build it for individual ad groups. And the reason for that is we can then set up the ad group level negatives. Even if you have them all on the same ad group, you can still switch them on and off, and you can still moderate and, you know, mess with the bids individually. That's useful. Um, but we just do a different ad groups, not all the time, but you know, most of the time. And we typically find that the, um, I think the, uh, the compliments for ASINs is usually like one of the worst performing and loose match is usually the biggest spending for keywords and also the worst performing. So sometimes we'll end up with just close and substitutes or we'll, you know, disable all of them except for clothes. And that can have a very interesting impact on the auto campaign. Cause it's like a, you know, it's an auto campaign that's only very narrowly targeted towards one of these four options.
Lazar: (16:26)
Yeah. We have literally the same experience when it comes to numbers with which types are working the best. The only difference with us is we are doing it to the campaign level. We're not creating different ad groups. We literally create separate campaigns for it. I know that if you're super neat and you don't want to have million campaigns in the account, it's not the best way how to do it. But on the other hand, I feel more secure because, we can, we can monitor budgets better. I don't know. It doesn't necessarily mean that it's correct way how to do it, but overall I feel super comfortable that way. It's more or less the same thing. Like, should you place different match types in different ad groups or in different campaigns? It's about destructuring.
Brent: (17:16)
Yeah, and if budget is your main reason for doing it, then it's totally valid. So what you're saying, you do different campaigns for each of those targeting types? Yep. Yeah. Wow. Yeah. That's definitely gonna multiply the campaign load, but if you can handle it then, Hey, you know, why not? This is why people like agencies to handle that complexity.
Lazar: (17:31)
Yeah. And it's really good if, if you know how to use bulk files, you don't even need to use any specific software or any specific tool. Just have Excel and that's it. Just download your account spreadsheet and add new campaigns. These ones are easiest to define. Oh yeah. We never discussed about how you see, you know, that asterisks symbol in search terms.
Brent: (17:56)
Yeah. That's worth discussing and also what match type it is. So, uh, yeah. Why don't you, why don't you go for it then?
Lazar: (18:02)
Yeah. For the asterisks when you start advertising and when you start looking at search term report, that's more or less first thing that you're not aware of and you have a huge question mark, next to it. Like not knowing, what is it. Basically that's triggered from automatic campaigns because there is no keyword there that triggered it. So it's switched with a symbol.
Brent: (18:26)
Yeah. That's probably, I would say Lazar, you might agree with the number one most asked question on any PPC Facebook group for the past 4 years. There's always like: why is there an asterisk? Because that's what the keyword is being shown as. Now, the match type Amazon changed this, uh, I want to say sometime late 2018 do not have that blog post in front of me. Um, if they even made one, but it's now just dash, right? So you have, you know, your broad, your phrase, your exact, and then you have dash, so that's the match type. But the actual keyword that it shows up in the keyword report is an asterisk. Yeah. So if you're wondering what the heck that is and why you have $7,000 spend against asterisk, that does happen sometimes.
Lazar: (19:06)
There was another one that was frequently questioned. I think it was about deal of the day or something like that. I think it's, it's had in search term reports.
Lazar: (19:19)
Oh yeah. Talk about ad placements. So if your product is featured as deal of the day and you have auto campaigns running, then those ads will show up on like the deal of the day slot along with the product is my understanding. So then that, that keyword that comes from that is the deal of the day. And so that's only, you know, for that day obstensively, it's the deal of the day, but usually get a lot of spend for that keyword. And so the, the metrics are usually like not the best. I mean, you get good sales through deal of the day usually, but the placement for that ad for that brief period of time. Um, yeah, that's a great point. I forgot about that entirely.
Lazar: (20:00)
Yeah. That's a really good advice. Like slow down with your automatic campaigns when you have deal of the day, or you can start crying when you see your search term report, like a couple of days after the deal of the day.
Brent: (20:13)
Yeah. I mean, ultimately it's up to you, but like, if you have deal to day going and you have an auto campaign, know that that auto campaign is suddenly going to get a lot more spend and exposure during that period.
Lazar: (20:21)
Yeah. And you already decreased the price for the product for deal today obviously and your margins are super low during that period. Like, I know people going to deal of the day just to get traction for the future. Basically it's not about making money on that day, but to get some reviews, to get some...But you know, when you do some sale and like for deal of the day, or B black Friday or Cyber Monday or Q4 sale or whatever..Like you know, when you get reviews, they're not that good. Like they're not as good as those that you get when you sell something for the full price. It's sounds really bad for my side but like people that, that go to, to buy stuff on sale usually can place, like, not as good review comparing to the guys that are buying that specific product during the year.
Brent: (21:23)
You think it's a different kind of customer quality, perhaps? You know, and those are verified reviews, but those people might have different expectations surrounding..
Lazar: (21:30)
Yeah. Yeah, exactly. It sounds really bad. I know that I sound super bad, but that's the truth.
Brent: (21:38)
I, I don't think it's that bad. It's an observation. It's a viewpoint, which I think is probably a valid one. You know, people who, all kinds of people shop on Amazon. Obviously there's millions and millions of different individuals.
Lazar: (21:50)
Yeah. And do you know that the Germans, like when you're selling on German market, one of the things that basically is a common trade there is that you get the most negative reviews there in Germany. Just because Germans are people that want to have a full information about the product. And when you don't meet their expectations, they don't mind like writing a negative review. And it's really hard to convince them that you can improve anything there. So yeah, like...
Brent: (22:26)
They have high expectations that need to be met. Well, I wonder what the average reviews, like, you know, per Amazon category, if you went and you looked at all the toothbrushes on Amazon Germany versus America..You're like, well, in America, they're all five star, 4.7, you know, and then in Germany it's like, they're generally speaking 3.9. Maybe the overall averages for the marketplaces are lower. So therefore it's kind of like a different competitive landscape.
Lazar: (22:53)
Yeah. So..
Brent: (22:54)
We should do an episode on reviews.
Lazar: (22:58)
That was my next sentence. Like, we kind of changed the topic, so let's not do that. But what's, what's good about Amazon Germany or Amazon Japan, especially, or some marketplaces where you don't know the language, you should definitely start with automatic campaigns. Like, that's the main advice if you're not having anybody in the team that speaks that language or if you're not hiring an agency that is doing translations or if you're not hiring like us. Yeah it's one of the main KPIs for you guys that you have native speakers for you.
Brent: (23:40)
I mean, we, but we also equally rely on auto campaigns and, you know, France. Like, it doesn't matter if you have a French native, but like, yeah. They still run autos and it's useful to see what the search term ecosystem is out there. Um, because of course, just like in English, even a native speaker is not going to know all the potential millions of keyword combinations for the thousands of products in a catalog. Like that's just not, not possible. And this is where the autos come in are useful. But yeah, if you're say an English native, and you're going into a Japan.. Unless, you know, Japanese or have someone on a team that's going to be pretty tough. So, uh, you know, Google translate will not save you. Uh, in all cases here and autos are going to do some heavy lifting.
Lazar: (24:19)
Especially when you have a kanji and hiragana and katakana, they're all mixed up. It's so confusing. I studied Japanese at the university. So I kind of understand some, some of the stuff there, I'm not super fluent in Japanese, but I kind of understand like half of the things that I see in the search term report, but still. It's so tricky. And like, um, the idea like are, uh, the terms in Katakana going to be triggered if you wrote the keyword in kanji or like whole bunch of different stuff. It's super, super tricky.
Brent: (25:03)
That a lot of Japanese will mix in English with their Japanese. Yeah. So then there's like technically four different ways to go about it. Um, and it, we should have a Nick Kets on the show sometimes..He's, he's a English guy in Japan and knows a lot about that market.
Lazar: (25:18)
Yeah. I think he even visited me in Belgrade like maybe six months ago, maybe a bit more. Yeah, it's a bit more because of corona and all of that.
Brent: (25:30)
Yeah. All right. But let's get back on topic though. That's a good point Lazar. One thing I want to talk about before we close out and before we talked too much, uh, too much Nick up too much is a recent blog post from Ad Badger, a friend of the show, Michael Erickson, who is actually referencing another post by Destiny Wesean, also friend of the show. We all know each other in this business, but this is a great post because what the theory of it was basically why use auto campaigns for a new product launch. So what do you think about launching products with auto in 2020, you know, mid 2020, where we sit right now. What's your thoughts about that Lazar?
Lazar: (26:09)
Like when you don't have other campaigns or you combine automatic campaign with other campaigns as well?
Brent: (26:19)
Well I don't know, how do you do it?
Lazar: (26:19)
When it comes to us, we create push campaigns. We create other campaigns. And like at the beginning, I remember when you and me used to work together, one of the main ideas at that time, and like, that's the old strategy, we used to create automatic campaigns and then wait a bit. And then from there, we would start with manual campaigns. But at this point we definitely create automatic campaigns, push campaigns and some like basic manual campaigns and ASIN targeting campaigns.
Brent: (26:51)
So we do exactly the same thing. I think in 2020, there's, you know, to start with just an auto is really selling yourself and the product short. Because if you have some theories, you know, you have some hypothesis about keywords that you think are gonna work and when, why not test those right from the get go? And I'm assuming that this product has no reviews or maybe very few reviews, right? So it's a pretty like new product, um, you know, a handful of them or something, not a lot of sales history. And the central thesis of this blog post is basically go ahead and look at what suggested keywords Amazon has for the product. Um, and in this case they use toothpaste. Um, and so like they use Colgate that had a bunch of data, right? Thousands of reviews versus a new product that had no reviews and so what Amazon pulls out and puts in auto campaigns is based on the products content, but also its sales history. So they had a lot more to go off with Colgate. And so the auto results from there were far more relevant than the ones for the brand new product, which we're actually a little bit confused by what was in the title and what was in the content, because yet again, maybe the copy for that new product is not refined as it could be. Although it's, you know, it's, it's a good place, but it's not as refined as it will be in six months after you get more data. So Amazon hasn't really done an amazing job with indexing that new product. And so those results in auto are going to be like, you know, mediocre. And so, uh, I think their conclusion is basically like, well, uh, autos are useful, but really rely more on manual these days instead of focusing on auto. Uh, do I suggest that like a seller still uses auto for as part of a launch? Yeah, cause we do it, you do it. And so like it's a component, but is it the only thing you should do? Is it the 2018 strategy of seven auto and wait for a week or two? No, I don't think so. Not anymore.
Lazar: (28:39)
Yeah. I completely agree with you. Maybe we should have an episode with like all of the strategies, how to launch a product. Because I know that there are so many like big names in the market that they have their own strategies. We basically have some, some of our own strategies how we do it. So like, I know that Brandon Young has one, I know the Trion has another one, Ben Cumming has a third version. There's, there's a lot of different stuff that you can do there. And like maybe we should create some idea why you should do or create specific campaign or why should you push in one way or another? Or like, why should you target to top of the search instead of like something else or so on..
Brent: (29:25)
Sides and downsides? Yeah. Well, you heard it here first folks. That's one of the future episodes for sure.
Lazar: (29:29)
Yeah. We have a list of them.
Brent: (29:35)
Yeah. So I don't really have too much else to say about autos. I think we've covered pretty much everything. All the different targeting types, upsides, downsides, budgeting, building. Uh, anything else off the top of your head Lazar?
Lazar: (29:45)
No, I think we're good. I literally think that we're good. There is nothing else that you can say about automatic campaigns. Probably if you want to prove us wrong and if you want to tell us more about automatic campaigns or you want to ask us something, feel free to send an email, to ask@wildppcbunch.com or just write to Brent or me. We would be more than glad to, to talk about that specific topic or like cover that specific question during our podcast episode.
Brent: (30:17)
Yeah. We'd love that actually, to get some, some feedback from listeners. I know we've got some emails, but I think you mentioned Lazar, was maybe a spam. I got excited when you told me, but I was like, Oh no. Uh, so yeah, we, we want to get some listener feedback and if you have any complaints specifically just email Lazar.
Lazar: (30:35)
For complaints, you can go with Brent. Okay. I think we're done for now. Thank you so much for listening and let's let's talk next week. That's it.
Brent: (30:50)
All right. Sounds good.
Lazar: (30:51)
Thanks. Bye.