Transcript
WEBVTT
00:00:00.059 --> 00:00:00.402
All right.
00:00:00.402 --> 00:00:09.368
So we are live, and today we're going to be talking about proven Amazon PPC strategies for top performers as a part of this Tactics Tuesday.
00:00:09.368 --> 00:00:13.820
Welcome everyone to the podcast and also welcome anyone who is joining us live.
00:00:13.820 --> 00:00:24.089
This is a topic that we wanted to talk about because, especially we're recording this actually on the beginning of Prime Day PPC is one of those things that is ever-changing.
00:00:24.089 --> 00:00:37.753
The beginning of Prime Day PPC is one of those things that is ever-changing, and so we really wanted to tackle this topic and provide some principles that are going to be important and useful for brand builders, no matter what stage you're at that are listening to this.
00:00:37.753 --> 00:00:55.576
So with that, matt, I'm going to turn it to actually, if you want to start, maybe a couple of things that you've learned in the last year or so as far as PPC, and maybe some common misconceptions that you think people have or mistakes that sellers make when it comes to PPC.
00:00:57.179 --> 00:01:00.966
Yeah, I would say so in my role in working in the agency space.
00:01:00.966 --> 00:01:07.522
I've probably analyzed three 400 advertising accounts over the course of the last four years.
00:01:07.522 --> 00:01:13.841
So I mean, I would say that the very first thing that most people I don't want to say get wrong.
00:01:13.841 --> 00:01:21.774
There's more than one way to skin a cat, especially in the game of Amazon PPC, but I think structure is one of the first ways that people get things wrong.
00:01:22.878 --> 00:01:27.950
I personally I don't know how you guys feel about this, but I always have single SKU campaigns first off and foremost.
00:01:27.950 --> 00:01:32.346
I also keep all of my match types separate in campaigns.
00:01:32.346 --> 00:01:37.265
Also, it just for me makes it easier to be able to isolate a match type because they perform differently.
00:01:37.265 --> 00:01:54.626
So I would say structure is probably the biggest thing in terms of things that people could do better, sellers could do better and, in my opinion, when you have the right structure, when you have the right foundation to start with, it allows you to scale a whole lot easier and it also makes the data a whole lot more clear, because you know that.
00:01:54.626 --> 00:02:00.248
You know we know that exact match converts differently than broad match and it's less.
00:02:00.248 --> 00:02:02.006
You know the CPCs are different as well.
00:02:02.006 --> 00:02:09.724
So when you get those into separate campaigns, those match types into separate campaigns.
00:02:09.724 --> 00:02:11.353
It allows you to scale and and adjust based very specifically on that match type.
00:02:11.353 --> 00:02:15.408
So I'd say that's probably the biggest thing that I see is is just the structure of the campaigns could be a whole lot better.
00:02:17.400 --> 00:02:25.259
Yeah, that's a great point is that, I mean, if you start out with a good structure, and I think it comes down to not only your match types but also even something as simple as naming conventions.
00:02:25.479 --> 00:02:42.169
So we use both video and image for sponsor brand quite a bit now, especially with brands that are more mature, and just having that breakout of, you know, putting video in the campaign name makes it so much easier to know.
00:02:42.169 --> 00:02:53.926
Like, okay, are the static images that we have or kind of the carousel image ads that we have, are those outperforming the video or is the video actually doing a lot better in telling that brand story?
00:02:53.926 --> 00:02:55.510
And it really depends on the brand.
00:02:55.510 --> 00:03:02.441
But it's important to know like which one is having a bigger impact, either video or still images.
00:03:02.441 --> 00:03:13.764
And that's just kind of one kind of micro example of why having a good campaign structure helps you manage your account, especially once you start getting, you know, numerous SKUs in there.
00:03:13.764 --> 00:03:21.943
I mean, it's one thing if you only have one, two or three SKUs, but once you start getting 10, 20, a hundred SKUs, it can get pretty complicated pretty fast if you don't have a good structure.
00:03:23.004 --> 00:03:24.508
Well, in the naming convention too.
00:03:24.508 --> 00:03:49.525
I think one of the things that's nice about that is that, whether you're viewing your campaigns within the actual sponsored ads console or whether you're downloading the reports or viewing them in some other platform, if you really have a good naming convention where, let's say, every important aspect about that campaign is detailed within the naming convention.
00:03:49.525 --> 00:03:52.375
So, in other words, is it video or image, you know.
00:03:52.375 --> 00:03:56.806
Is it sponsored products or sponsored brands or sponsored display, you know.
00:03:56.806 --> 00:03:58.169
What SKU is it?
00:03:58.169 --> 00:03:59.592
What match type is it?
00:03:59.592 --> 00:04:01.844
Is it exact broad phrase auto?
00:04:01.844 --> 00:04:04.110
You know all of those things broken out.
00:04:04.110 --> 00:04:11.520
Sometimes it can make for somewhat of a long campaign name, but at the same time, it makes it really easy when you're filtering.
00:04:11.520 --> 00:04:54.312
You know so, whether, again, whether you're in the console on Amazon or you're just viewing it, as you know, as a downloaded spreadsheet or something, if you can filter and sort and search by either you know SKU name or product name, you know so, like, for instance, we sell pool poles, we sell pool nets, we sell a brush, so we could even have, if we wanted to have it by a product type, you know we could have pole brush net, you know, or something like that, so that I can search by any one of those at any given time where I can search video, or I can search, so that I can search by any one of those at any given time where I can search video, or I can search image, or I can search auto, and then I can get all the campaigns that are related to that particular search and see how they're comparing.
00:04:54.312 --> 00:04:57.189
So the naming convention is pretty important there.
00:04:57.259 --> 00:05:14.202
But I would also say, if you're going to use naming convention, be careful that you're being very exact about using it exactly the same way every time, so you can't have poll in one campaign and then polls in another campaign, or you know like, be very careful.
00:05:14.202 --> 00:05:22.209
In fact, it would even be useful to set up a spreadsheet where you had it set up, so that you choose, like anytime you're going to set up a new campaign.
00:05:22.209 --> 00:05:27.735
You choose okay, what's the SKU, what's the type, is it sponsored brand, product, you know.
00:05:27.735 --> 00:05:28.516
Display, is it?
00:05:28.516 --> 00:05:30.242
You know, choose all the different things.
00:05:30.242 --> 00:05:36.603
And then it spits out the campaign name and you just copy the name and paste it, you know, in when you create the new campaign.
00:05:36.603 --> 00:05:47.055
That way you can't screw it up, you know, because it's really easy to mistype something or things like that and then all of a sudden now, anytime you search for that, you don't actually get the campaign that you think you're getting.
00:05:48.141 --> 00:05:55.190
Yeah, that's a good point and that actually connects really well to one of the other things that I want to make sure that we cover before we got too deep, which is, you know, one of these.
00:05:55.190 --> 00:06:08.850
I think one of the biggest mistakes that's being made out there, and kind of you know what I would call Amazon lore, is this concept, and a lot of it comes from software automation for Amazon PPC, where you know, kind of this.
00:06:08.850 --> 00:06:21.610
The conventional wisdom is, let's say, that you've got a keyword that you notice is working in an auto campaign that converts Well, you know, then the conventional wisdom is after it's got a couple of sales, we want to move that over to.
00:06:21.610 --> 00:06:29.747
You know an exact match, and the reality about it is is that you can really be running both at the same time and they have different purposes.
00:06:29.747 --> 00:06:37.826
And not only that, but you know, say you create that exact match campaign and it's got 10 keywords in it that are also performing Well.
00:06:37.826 --> 00:06:44.728
Again, that typically what the software does and kind of conventional wisdom is take that hero keyword, if you will, and put in a new campaign.
00:06:45.350 --> 00:06:49.800
The problem with that is is that there's a certain amount of kind of essentially randomness.
00:06:49.841 --> 00:06:58.569
That happens anytime you create a campaign and there's weight to campaigns that have been around for a long time, that have historical data to them.
00:06:58.629 --> 00:07:00.362
So that's why you know a couple of things.
00:07:00.382 --> 00:07:09.523
Is you know, if you're looking at, hey, I need to redo, you know a product line, or if you're looking at redoing your whole account, why you shouldn't create all your campaigns from scratch?
00:07:09.523 --> 00:07:17.930
You're better off kind of modifying and rebuilding the campaigns that have been successful in the past as opposed to just deleting them and creating new campaigns.
00:07:17.930 --> 00:08:02.473
And so with that, you know, one of the things that we've discovered that has worked really well for us is, rather than kind of migrating those hero keywords into new campaigns, keep them in the campaign that they're already working in because they've proven a certain result in there in order to be profitable, and then take the other keywords that maybe they're doing okay, maybe they're not doing well at all, and move those into the new campaign, because then you're not rolling the dice with the keywords that are working for you, that are those quote unquote heroes, and it gives you the opportunity then to test those other keywords that, hey, when they're getting a bigger portion of the impression share in the budget, do they actually perform better, and sometimes they do.
00:08:03.920 --> 00:08:05.886
Well, and also there's a certain amount of value in that.
00:08:05.925 --> 00:08:41.340
You mentioned kind of the randomness of how it can, you know, sometimes function and since the historical data is what Amazon kind of uses as part of its algorithmic, you know, determination for whether they're going to show you or not show you, what happens is when, if you've got keywords that are not performing well in one campaign and you now move them to a new campaign whatever historical data Amazon had that for some reason might have been affecting those keywords Like there's times we all know this to be true there's times you have keywords in a campaign and you know they're good, they're good keywords for your product.
00:08:41.639 --> 00:08:49.441
Like there is no reason that those keywords should not be performing well and for whatever reason, at least from an ad basis, they don't perform and you don't know why.
00:08:49.441 --> 00:09:00.505
And you can take those and move them into a new campaign and all of a sudden, now that they're in that new campaign, now they start to produce and it's because whatever historical data Amazon was using, for whatever reason, it went south.
00:09:00.505 --> 00:09:06.562
Like you threw those keywords in that campaign, whatever happened up front, it went south.
00:09:06.562 --> 00:09:10.172
And now Amazon has that data and it's like, no, we don't like those keywords.
00:09:10.172 --> 00:09:19.899
As soon as you shift them and all that historical data goes away, you'll find sometimes you place those keywords into a new campaign and all of a sudden they become a good keyword for you where they weren't before.
00:09:20.660 --> 00:09:21.640
Yeah, absolutely.
00:09:21.640 --> 00:09:25.864
And I think that that's again going back to the software piece of it.
00:09:25.864 --> 00:09:33.587
And this is why this, I think, is so important, because I feel like for years we've been promised this kind of automated solution, an easy button that you could just tap.
00:09:33.587 --> 00:09:38.571
It would automatically, you know, manage the accounts effectively or the campaigns effectively down to, you know, the CPC level.
00:09:38.571 --> 00:10:15.782
The reality of it is is that, at least from everything that I've seen right now is you know, you want to take kind of a semi-automated approach is what we find effective where, yeah, software is great at crunching numbers and telling you like, okay, here's what your, you know predicted ACOS is and you know data on, here's how many impressions you get with, you know, this cost per click versus that cost per click and those types of things.
00:10:16.344 --> 00:10:34.423
But it does take a person in the loop to kind of make those decisions from a strategy level of you know which keywords should I start, move where to create new campaigns, because a lot of the automated software is going to say, hey, this keyword is working, create a new campaign for this keyword and, like we just talked about, that's pretty much the opposite of what you want to do.
00:10:35.125 --> 00:10:48.860
And then, in addition to that, you know there's a lot of nuance with that of you know, in increasing and decreasing both your CPCs on a regular basis but then also your campaign budgets.
00:10:48.961 --> 00:10:58.210
That I think is really hard from a structural standpoint to get right if you're relying on a software to do 90% to 95% of the work for you.
00:10:58.811 --> 00:11:00.214
I know we personally use PPC.
00:11:00.214 --> 00:11:32.900
Ninja Ritu we've had her on the podcast does a fantastic job with that and one of the reasons why we like that solution so much is because it does a great job of providing that kind of the best of both worlds, if you will, because there is a component in there where you can set up the parameters to say here's what my goals are for these campaigns or the account, and then it'll show you what the automated kind of recommendations are for you know bids and budgets and that type of stuff, and you can, you know, accept all, reject, all, accept.
00:11:32.900 --> 00:11:49.951
You know you can pick which ones you want to accept and kind of filter through there and I think that that kind of gives you the best of both worlds where it's efficient from a management standpoint but also allows you to filter out those things that maybe are a little bit more nuanced or are outside of kind of the standard process.
00:11:50.793 --> 00:11:59.042
I've used three or four different softwares for my own brands and then I've seen the result of other softwares in analysis that we've done.
00:11:59.042 --> 00:12:05.865
What I can say is that bid optimization, I think, is something that software automated software can be really good at.
00:12:05.865 --> 00:12:09.730
Like you said, john, it's math I mean figuring out what to adjust your bid to is math.
00:12:09.730 --> 00:12:12.527
There's one tool I take, a metrics that I actually currently use.
00:12:12.527 --> 00:12:33.759
They crunch so much data I don't know how much ad spend they have access to over the course of a month from all of their clients, but they use that data to feed an AI that does a predictive analysis on what it thinks that the conversion rate is going to be at different times throughout the day, based on current data, yesterday's data, all the way back to a year ago data.
00:12:33.759 --> 00:12:49.115
So I think that having some sort of especially now in the days of AI where it can crunch a massive amount of data bit optimization is something that I think can be automated and I can trust software to do the bid optimization and I'm not that concerned about going in and checking that you don't.
00:12:49.115 --> 00:13:02.405
Looking at the high level, metrics is sufficient in most cases but, like you kind of alluded to, john, where I think software isn't good at and where I think there has to be a human in the loop and I've seen in my own brands where it's gone run wild is like keyword harvesting.
00:13:02.466 --> 00:13:26.068
Keyword harvesting I think is hard for software to do because I think having software make a decision based on a conversion to graduate a keyword to a manual campaign I mean I sell in the barbecue accessory space and if I get a curiosity click from a pool tool and someone buys my heat resistant glove that was searching for a pool rake I don't want to show up, I don't want to target pull rake in any of my campaigns.
00:13:26.068 --> 00:13:37.802
So actually one of the softwares that I had, I didn't have any sort of rules at all set on what kind of you know what keyword harvesting rules there were, like I want at least three conversions to happen or four conversions to happen before I graduated.
00:13:37.802 --> 00:13:43.168
But I kind of in that I got trigger shy and in that I don't let software Take a Metrics.
00:13:43.168 --> 00:13:49.956
What I use now has the option to automatically bring or converting keywords into manual campaigns.
00:13:49.956 --> 00:13:55.884
I have that completely turned off because I think that in a lot of cases humans are going to do that.
00:13:55.985 --> 00:14:12.149
Me as the human knowing my brand and knowing my consumer, can think of keywords, that kind of tangential type of keywords that I might want to show up for.
00:14:12.149 --> 00:14:13.620
I just don't think software is great at that.
00:14:13.620 --> 00:14:15.345
So I think software has its place.
00:14:15.345 --> 00:14:20.605
I think bid optimization is where you can use it effectively as long as you have the right rules in place.
00:14:20.605 --> 00:14:22.950
But I think there always has to be a human in the loop.
00:14:24.131 --> 00:14:24.953
Yeah, I think too.
00:14:24.953 --> 00:14:36.504
You know, like the, the keyword harvesting thing is definitely difficult, and this is even difficult even even if you take the automated piece out of it and you're not talking software but you're just talking agencies.
00:14:36.504 --> 00:14:43.047
You know that that are you know when, when you're not running it internally as a company and you're having an agency do it.
00:14:43.047 --> 00:15:05.750
That's not to say, don't have an agency, but make sure that you're having good conversations with them and that you're paying attention to things, because, for instance, within our category, there's a number of keywords that at first blush if you don't know the category they look like they are extremely targeted to what we sell and in reality they are extremely untargeted to what we sell.
00:15:05.750 --> 00:15:08.765
So, for instance, a good one would be a swimming pool net.
00:15:08.765 --> 00:15:11.812
Now we sell pool nets for cleaning the pool.
00:15:11.812 --> 00:15:25.383
The problem is is that what we have found over time is that swimming pool net is although occasionally people are using that phrase to search for a net that you would clean the pool with the majority of time when they're searching for that, they're actually searching for a net that you would clean the pool with.
00:15:25.383 --> 00:15:31.404
The majority of time when they're searching for that, they're actually searching for a net that you put over the pool to keep leaves out.
00:15:31.404 --> 00:15:35.379
To keep leaves and debris out of the pool, which is a totally different product, right?
00:15:36.162 --> 00:15:44.284
Well, every automated tool and every agency that we've ever tried, you know, has tried to place those keywords into our campaigns.
00:15:44.284 --> 00:15:49.283
Because, oh, we got a conversion on it, you know, and like all of a sudden, we got to pay attention because it looks very relevant.
00:15:49.283 --> 00:15:53.413
Turns out it's not, and so I would just say you know your brand best.
00:15:53.413 --> 00:16:04.830
Make sure that you're either turning off harvesting if it's automated, because I think there's going to be a lot of situations like that, or when you're dealing with an agency, you need to make sure that you're having conversations about that.
00:16:04.830 --> 00:16:08.254
You know, like those specific keywords seem like they're relevant, but they're not.
00:16:08.254 --> 00:16:17.870
This is why you know that sort of thing and just kind of keeping tabs on that because it happens, and then you end up spending a boatload of money on a keyword that you know really never should have been in a campaign at all.
00:16:19.552 --> 00:16:20.374
Yeah, a couple of things there.
00:16:20.374 --> 00:16:21.653
I mean any, I would say.
00:16:21.653 --> 00:16:29.937
You know, you know, no agency is perfect, but at the same time, any agency that's worth their salt is going to do a couple of things to make sure that that doesn't happen.
00:16:29.937 --> 00:16:39.162
Or at least this is what we do is, first of all, you know, before, if we're looking at a seriously looking at a keyword like that, we're going to actually look at what the search terms are for it.
00:16:39.162 --> 00:16:58.686
And in those search terms, if we see that the first 10, 20 products are a pool that goes over you know, or a net that goes over the pool, as opposed to you know something or you know what you're selling, then obviously you know that's what people are buying, which is why Amazon is ranking those products organically a lot higher for it.
00:16:58.686 --> 00:17:05.087
The other thing is is that you know that's why you keep tabs and you know do keyword adjustments, do keyword adjustments.
00:17:05.087 --> 00:17:16.661
And you shouldn't be spending boat loads of money on something like that, because you should be able to tell and now it depends on conversion rate within generally 10, 20 clicks, whether you're headed in the right direction or not.
00:17:16.661 --> 00:17:19.509
So that way you're not burning a whole bunch of money on keywords that aren't going to work.
00:17:20.200 --> 00:17:41.162
But in addition to that, what I would also say is there are a lot of keywords that can be like that and I think that where they fit really well into what we call a catch-all campaign, which is basically kind of these low volume keywords that we put a low bid in for that maybe you only get a sale once a week, once a month, doesn't matter.
00:17:41.202 --> 00:17:42.984
Once a week, once a month doesn't matter.
00:17:42.984 --> 00:17:51.595
But if I'm only paying, you know if I'm paying, you know, just to make the extremes, you know, say I'm paying $10 a click for that.
00:17:51.595 --> 00:18:00.804
You know that pool term, that doesn't convert very well for the product.
00:18:00.804 --> 00:18:02.207
Well, in order to be at the very top of the page, that's a problem.
00:18:02.207 --> 00:18:22.970
You know, if I'm not converting well, well, but if I'm paying 50 cents and the only people that I'm really going to show to are the ones that are diehard, you know, looking at researching, you know pool stuff and it might be a good fit for them Well, that keyword I can still get a good ACOS on at that 50 cents and drive some incremental sales by using a catch-all campaign.
00:18:22.970 --> 00:18:25.042
So that's another great strategy that people can use.
00:18:25.835 --> 00:18:26.779
Well, it's good for auto too.
00:18:27.656 --> 00:18:29.443
That similar strategy works very well.
00:18:29.443 --> 00:18:32.825
We've had a campaign that we've run for years.
00:18:32.825 --> 00:18:34.842
To be honest, we haven't really changed it much.
00:18:34.842 --> 00:18:40.142
Occasionally we adjust the bid and whatnot, but it's basically a low-bid auto.
00:18:40.142 --> 00:18:45.759
That's just a catch-all, like we just run it, and we just run it at a really low bid and let Amazon just throw whatever they want at it.
00:18:45.759 --> 00:18:53.976
And because the bid is so low, our ACOS on that is incredibly good and we actually get quite a lot of sales out of that, more than you might think.
00:18:53.976 --> 00:19:00.788
Just because we're letting Amazon kind of go wild with it to some degree, I'm just saying look, find whatever sales you can grab for us.
00:19:05.402 --> 00:19:07.512
If we can get it for that price, I'll do that all day long.
00:19:07.512 --> 00:19:18.207
Well, I think that's an important point too is that I feel like the Amazon lore out there is that the bulk of your sales should be coming from exact match and in a perfect world, everything should eventually migrate to exact match.
00:19:18.207 --> 00:19:20.057
So you're so, you know, narrowed in.
00:19:20.057 --> 00:19:21.420
The reality of it is.
00:19:21.420 --> 00:19:35.025
This is that you know, when you look every ad type and you know campaign type has its own pros and cons, and so, if we look at it, you know, talk about auto campaigns.
00:19:35.065 --> 00:19:37.394
Hey, if I can add incremental sales that are profitable, I'll use auto campaigns.
00:19:37.394 --> 00:19:39.922
For that I'll use broad, I'll use'll use broad.
00:19:39.922 --> 00:19:51.340
We do a lot of broad mod because the nice thing there is that if you look at exact match keywords, the bids are a lot higher because everybody's like oh, I got to be exact match.
00:19:51.340 --> 00:19:59.567
Well, we might go over to broad and use a broad mod where the CPC is a third of what we're paying for an exact match.
00:19:59.567 --> 00:20:14.227
So we can get some of the benefits of what we would with an exact match at a much lower price, which allows us to have a profitable ACoS that we could never get with an exact match keyword and drive sales that way.
00:20:15.095 --> 00:20:32.930
Taking that another step further, which I've seen a lot of people do, is a broad match campaign will generate a sale for a search term, and then that seller will graduate that to an exact match, but then they'll negate it in the broad match, even though that was a profitable sale.
00:20:32.930 --> 00:20:37.386
I don't understand that.
00:20:37.386 --> 00:20:39.943
I mean, yes, I understand double targeting something.
00:20:39.943 --> 00:20:45.654
I don't know that you will actually compete against yourself in that case though, but it's a profitable sale one way or the other.
00:20:45.654 --> 00:20:54.019
If it's at a good ACOS in the broad match, I'm not going to negate it just because I'm graduating it to an exact match.
00:20:54.959 --> 00:21:15.288
Based on what John just said, the likelihood is that if you do that, if you negate it and you put it into this exact match campaign, the likelihood is you're going to pay more per click in the exact match than you were in the broad and so, even though it is exact, you might end up with a higher ACOS that way than having left it.
00:21:15.288 --> 00:21:18.009
So you'd be better off leaving it and comparing.
00:21:18.009 --> 00:21:22.711
You know, over time you know how those, how those numbers, you know flesh out and how they compare.
00:21:23.832 --> 00:21:38.226
Yeah, and that's exactly what we do with a lot of campaigns is so if we see something in, you know, in a, in a broad campaign that's a keyword that's performing really well, that we're not currently running an exact match, we'll run an exact match but we're going to pay a lot more for that CPC.
00:21:38.767 --> 00:21:43.241
Now that can make a lot of sense Again, even in an exact match.
00:21:43.241 --> 00:22:10.948
I would say the benefit there, and part of the reason why the CPCs are higher, is that because it's an exact match, you have a lot more control and you're also getting keyword rank for that, because if Amazon keeps seeing that you're converting well on cleaning pool net, for example, then you're going to organically rank high for that, and so there is a compounding effect, or kind of a halo effect, if you will, that you get from exact match keywords.
00:22:10.948 --> 00:22:21.695
That is harder control if you're using broad, broad mod phrase or you know other ad types, and so there is, you know, a reason to pay even above.
00:22:21.695 --> 00:22:26.750
You know kind of what your breakeven ACOS is sometimes for exact keywords.
00:22:26.750 --> 00:22:32.125
But that doesn't mean that it always makes sense and that you should stop using the other match types.
00:22:33.214 --> 00:23:07.182
Well, also, I think it's important to remember, like if you're using it for ranking, like if you're trying to use an exact match campaign for ranking and you're using it for ranking, like if you're trying to use an exact match campaign for ranking and you're willing to spend more you know from an ACoS perspective because that's what you're trying to do well then you better be tracking to make sure that you're actually affecting your ranking, like that's actually helping, because if it's not, and you're overspending on your ACoS because that's your target but you're missing the target, well then at some point you have to decide whether you're going to turn that campaign or turn that keyword off, or turn the bid down or whatever it is you're going to do, because now you're just throwing away money if your ranking's not changing.
00:23:07.182 --> 00:23:22.176
I was also going to mention something else that I think is probably important to pay attention to, and that is we were talking earlier about the well, let's take the maybe say more extreme example of that auto campaign that we run a low bid campaign.
00:23:22.176 --> 00:23:28.278
Generally speaking, those are going to be fairly low conversion rate keywords.
00:23:28.278 --> 00:23:32.708
We're getting them at a low bid, so we get a good ACOS on those, and so they're profitable.
00:23:33.035 --> 00:23:51.006
But depending on how much of that kind of a strategy that you employ, I think that you need to pay really close attention to how does the conversion rate of your listing overall compare to the competitors in your category that you're trying to beat in terms of a ranking perspective?
00:23:51.066 --> 00:24:09.124
Because if the conversion rate of your product is fairly similar to some of those other products that really you're trying to compete against and trying to rank better than then be careful using strategies that have a low conversion rate that are going to potentially bring down the conversion rate of your product overall.
00:24:09.124 --> 00:24:12.338
You know, because you're really close.
00:24:12.338 --> 00:24:23.410
You know like if you've got a 20% conversion rate and somebody else has got a 15 and you know the sales that you're putting through are only going to bring you down marginally in that conversion rate, it probably doesn't matter.
00:24:23.410 --> 00:24:36.227
But if you're all sitting at about a 15% conversion rate and you start throwing a bunch of low conversion rate targeting at that, I have a little bit of concern over that in terms of your ranking strategy.
00:24:36.227 --> 00:24:40.665
I'd be interested to hear what your guys' thoughts are on that, but I just think it's something to be careful with.
00:24:41.134 --> 00:24:42.519
I think for me, I think it's important.
00:24:42.519 --> 00:24:51.688
So what I think it's important to pay attention to is match types as a whole and what percentage of your spend is going to each match type.
00:24:51.688 --> 00:24:59.449
I think John mentioned the lore of having all of your keywords be exact match because it's more specific.
00:24:59.449 --> 00:25:16.278
But I think having most of your spend be exact match because it's more specific and but I think having most of your spend and when I say most, maybe 60, 70% of your spend, I think should be an exact match campaigns, maybe 30%, 20 to 30% in broad and phrase, and then only like 10%, maybe even less than that in an auto, I think you should have all of those match types.
00:25:16.278 --> 00:25:27.483
But I think if you have that ratio and you're thinking about it from in terms of how much am I spending on each match type, I think that conversion rate ends up leveling out over time because most of your spend is very targeted.
00:25:27.483 --> 00:25:28.646
You have a little bit.
00:25:28.646 --> 00:25:29.296
That's research.
00:25:29.336 --> 00:25:37.903
But again, one of the things that you mentioned is making sure that if you have a campaign that you expect to increase your rank, make sure that it is increasing rank.
00:25:37.963 --> 00:25:40.068
I think that's the same with all of your campaigns.
00:25:40.068 --> 00:25:51.885
I think you should know what the purpose of every single campaign is and don't just run a campaign because someone on YouTube video told you to, but, like, actually understand what is the purpose of this specific campaign and how do I measure that?
00:25:51.885 --> 00:25:53.537
How do I measure what the results of that is?
00:25:53.537 --> 00:26:01.989
Because if, like you said, if we, if I have a ranking campaign which I love, ranking campaigns what I love about the ranking campaigns that I've been using recently are just single keyword, exact match.
00:26:01.989 --> 00:26:09.137
But, like I pay attention to first of all, what am I spending, how is it affecting it on the ranking side and how can I adjust that?
00:26:09.137 --> 00:26:10.661
Can I adjust top of search?
00:26:10.661 --> 00:26:13.327
Will that keep my conversion rate where it is?
00:26:13.327 --> 00:26:21.108
So I think knowing kind of where your spend is going and where they are going to each match type, I think kind of levels out that conversion rate over time.
00:26:22.236 --> 00:26:27.887
Well, the other thing that I would say, too, is that remember that Amazon doesn't only take into account conversion rate.
00:26:27.887 --> 00:26:33.007
I mean, there are multiple factors that are taken into account for rank.
00:26:33.007 --> 00:26:43.088
One of the other ones is sales velocity, and I think that that has a bigger impact than than we give it credit for.
00:26:43.088 --> 00:27:07.777
So, to kind of go to your example, mike, you know, if you had a product that had, you know, an 18% conversion rate but a much higher sales velocity, versus another product that had 20% conversion rate but a lower sales velocity, my guess is is that the one at 18, at a much higher sales velocity, is actually going to rank better for a lot of the keywords that are going to matter for that brand and that where customers are going to find them.
00:27:08.317 --> 00:27:08.839
Fair point.
00:27:08.839 --> 00:27:12.147
I think it's an important piece to pay attention to.
00:27:12.147 --> 00:27:15.262
Obviously, the algorithm is not focusing on any one thing.
00:27:15.262 --> 00:27:16.707
That's why it's an algorithm right.
00:27:16.707 --> 00:27:21.203
There's all sorts of parameters that are coming into play there, so sales volume is definitely important.
00:27:21.203 --> 00:27:39.268
It actually brings me back to something that we were talking about briefly before we jumped on the podcast, and that is that there's a number of other factors that come into play here, not only in terms of ranking, but in terms of CPCs on your ads and things like that, and one of them is inventory and so paying.
00:27:39.288 --> 00:27:50.345
You know, paying really close attention when you're running your PPC that you have enough inventory long-term that you can sustain.
00:27:50.345 --> 00:28:12.480
Because if you start spending on a campaign for a particular product or multiple campaigns, whatever it is, if you increase your bid or you increase your budget or whatever that is, and you're not paying, or you increase your budget or whatever that is and you're not paying attention to, well, if I do this and my sales velocity begins to increase, do I have enough inventory to actually sustain that long-term?
00:28:12.480 --> 00:28:19.404
Then all you're doing is stealing money from yourself because you might sell more right now, but you're selling it at a higher cost to yourself and then you're going to run out of inventory.
00:28:19.404 --> 00:28:32.643
And now you're screwed where you'd have been better off not raising your bid or raising your budget or whatever and just keeping the status quo so that at least you just continued to sustain and were able to maintain those sales through and not run out of inventory.
00:28:32.643 --> 00:28:38.405
The other piece of that is that ranking is very tied to inventory.
00:28:38.405 --> 00:28:55.681
There is a direct relationship between how much inventory you have for any given product and specifically we've talked about this before in terms of geo ranking and how much inventory you have in particular areas of the country, especially high population areas, that Amazon takes into account when they rank you.
00:28:56.261 --> 00:29:03.518
So your keyword ranking is not just okay, I rank in my product for this keyword.
00:29:03.577 --> 00:29:04.881
This is where I rank everywhere.
00:29:04.881 --> 00:29:19.982
I rank here for Philadelphia, I rank here for New York, I rank here for Michigan and very much it's based on what is the delivery time for that product and if there's no inventory in that area, then delivery times will be longer and your ranking will be less.
00:29:19.982 --> 00:29:31.960
Also, if Amazon sees that your inventory is low let's say you're down at 20 days of inventory or 25 or 30, then your ranking is going to be less, they will lower your rank because you have less inventory.
00:29:31.960 --> 00:29:37.859
Because ultimately what they want is they want to put products at the top of the category that they're going to be able to continue to sell.
00:29:37.901 --> 00:29:39.041
They want consistency.
00:29:39.041 --> 00:29:46.183
Like consumers like consistency, and so Amazon's not going to rank you well if you don't have enough inventory to sustain sales.
00:29:46.183 --> 00:30:01.041
So pay attention to inventory when you're managing your PPC, because you know, as that, inventory declines even if you don't go out of stock, if your inventory declines below roughly 45 days of inventory at any given point you're going to, your ranking is going to suffer.
00:30:01.041 --> 00:30:02.520
So you better be paying attention to that too.
00:30:03.876 --> 00:30:06.680
Yeah, there's a couple of things that I want to double click on there because I think that's so valuable.
00:30:06.680 --> 00:30:39.107
One is that, when it comes to Amazon, looking for that consistency I mean the days of you know where Amazon was kind of the wild wild west and they were willing to accept almost anything, and you know sellers were basically on the platform as a way of essentially creating a job, or basically over in the sense of you have to think about Amazon more now as walking into a Walmart or Target in the sense of they want to see that product in the same spot on the shelf all day, every day.
00:30:39.107 --> 00:30:59.612
Because I can tell you right now I mean I think we talked with a couple of our brand builder guests about it where you know they ran out of inventory and they couldn't fulfill inventory to I don't remember if it was Walmart or Target for like a month or two and they got completely pulled off the shelves because Walmart has a certain number that they need to hit with every single one of those slots on the shelf.
00:30:59.612 --> 00:31:19.167
Now, in theory, there's unlimited shelf space on Amazon, but in practice there's a limited number of slots on page one and Amazon they're investing in visibility in your brand just as much as you are investing in Amazon between ads and everything else that you're doing in order to build your brand on Amazon.
00:31:19.167 --> 00:31:39.114
So they look at that and they say, okay, if we don't see that consistency over time, then we're going to put somebody else in that spot that does provide that consistency, both in staying in stock but also the inventory levels that you need in order to be a consistent brand on the Amazon platform.
00:31:39.114 --> 00:31:44.310
So I think that's a really important piece for people to be aware of.
00:31:44.730 --> 00:31:53.368
The other thing that I was curious to get your guys' thoughts on this is kind of a theory that we're testing right now as we record this, for Prime Day is rather than so.
00:31:53.409 --> 00:32:00.652
Let's say that you're starting to run out of stock and you know that it's going to be slow coming in and you want to slow down your ads.
00:32:00.652 --> 00:32:27.049
Well, rather than pausing campaigns or adjusting budgets and we'll talk about budgets here in a second one thing that we're testing for Prime Day on one particular account is actually significantly reducing the bids to about 50% of what we would normally bid, and the theory behind that is that as soon as you start pausing and unpausing campaigns again, you throw that kind of unpredictability into the situation.
00:32:27.049 --> 00:32:42.856
And then the other piece is that if we significantly lower bids, what we find is that Amazon campaigns with limited bids perform actually worse than campaigns that have plenty of budget.
00:32:42.856 --> 00:32:48.893
Or, excuse me, campaigns with limited budget perform worse, is what it really boils down to.
00:32:48.893 --> 00:32:59.691
So our theory is by reducing the bids significantly, yes, we realize we're slowing down sales, but those sales that we're getting are actually much more profitable.
00:33:00.701 --> 00:33:16.500
And by keeping your budgets high, as opposed to bringing a budget down and saying I want to spend less, then you're still maintaining that differential that so far, for you, seems to be making a difference the campaigns that have budget that outstrips what Amazon says it means.
00:33:16.500 --> 00:33:33.105
So if Amazon says you need $20 for a campaign per day I don't remember if we mentioned this on the podcast or if this is pre-podcast, but we were talking about how you've been running your campaigns at like three times what Amazon recommends for that budget and that you're getting better ACOS out of that, correct.
00:33:33.144 --> 00:33:36.593
Yeah, and I think we talked about that before the podcast and that's good clarification.
00:33:36.593 --> 00:33:49.326
So yeah, we've tested anywhere from double to 10X and what we found is that sweet spot of three to five times what those campaigns typically spend in a day.
00:33:49.326 --> 00:33:51.770
If we give it so, say it's normally.
00:33:51.770 --> 00:33:55.828
Just to make the numbers easy, let's say it normally spends $100 a a day.
00:33:55.828 --> 00:34:21.864
If we give it three to $500 budget, we see it be a lot more efficient than if we gave it a $200 budget and at the same time, we don't see any gain in efficiency If we give it a thousand dollar budget or 10 X, what it means, and so that's why, with that, then, this strategy fits that very well, because, rather than reducing your budget, if you reduce the bid, then that budget still stays high and you still maintain that at least that particular benefit.
00:34:22.447 --> 00:34:36.541
Right, you get all the efficiency, but at the same time it actually becomes more efficient because, yes, you're going to slow down sales, but let's say that normally you're paying a dollar for your clicks, or well, I guess in this day and age let's call it $3.
00:34:36.541 --> 00:34:38.389
Right, you cut that.
00:34:38.389 --> 00:34:38.650
You know.
00:34:38.650 --> 00:34:46.860
So if you cut your bids in half, and let's say you now you're bidding a dollar 50, well, yeah, you're going to get a lot of less impressions, a lot less clicks, which means a lot less sales.
00:34:46.860 --> 00:34:54.108
But because those bids are so much lower, those sales are going to be a lot more profitable for you at the same time.
00:34:54.108 --> 00:35:01.396
So now you slow down your sales velocity and made your ad sales more profitable at the same time.
00:35:01.980 --> 00:35:02.621
Makes sense to me.
00:35:02.621 --> 00:35:07.012
We haven't tested that, but it makes sense as a strategy.
00:35:08.081 --> 00:35:11.911
Mike, I know that you don't run any Prime Day deals or anything along those lines.
00:35:11.911 --> 00:35:17.472
Do you adjust campaign budgets or bids specifically for Prime Day?
00:35:19.322 --> 00:35:39.248
So we actually currently have PPC Ninja managing things at the moment, and so we had a conversation with them about that for Prime Day, and I believe they are pulling back bids for Prime Day for that reason, because we're not running deals, we're not going to perform as well as those, as those brands that are running deals.
00:35:39.648 --> 00:35:51.052
Because we found that Prime Day for us and I think most of it is because we're not a we're not a spontaneous kind of a product it's not, like, you know, clothing or gadgets or things like that.
00:35:51.112 --> 00:35:51.559
You know that.
00:35:51.559 --> 00:35:53.023
You know people are excited about.
00:35:53.023 --> 00:36:09.702
You know it's a pool, and so it's one of those things where either somebody needs one or somebody doesn't for the most part, and so if they need one and I offer them a discount today, maybe I sell it to them today, but I might've sold it to them tomorrow or next week, you know, at at regular price, and they would have bought it anyways.
00:36:09.702 --> 00:36:27.338
And that's that's actually what we found last prime Prime Day we ran deals and we had a good spike in sales, but then we just had a big trough afterward that we normally don't see, and so essentially all we did was just steal those future sales and sold them now at a lower price point and at lower profit.