#107 - Why Placement ACoS Doesn't Matter on Amazon PPC

#107 - Why Placement ACoS Doesn't Matter on Amazon PPC Speaker 2: People think I'm crazy because I don't consider placement ACOS when making optimizations. Speaker 1: Wait, you don't consider placement ACOS when making optimizations? You're crazy. Speaker 2: We're about to break down how ACOS based placement optimizations are a huge mistake. Today on That Amazon Ads Podcast. Alexa, play That Amazon Ads Podcast. Unknown Speaker: Which one would you like to hear? Speaker 2: The best one. Unknown Speaker: Okay, now playing that Amazon ads podcast. These gentlemen are completely changing the game. Speaker 1: After listening to that Amazon ads podcast, my ads are finally profitable. Unknown Speaker: I also heard they're pretty cute. Speaker 1: All right, Stephen, let's get into this. This is going to surprise a lot of people. I know a lot of people that look at their placement optimizations, they're looking for low ACOS and they're adjusting based on that. Why would you not consider ACOS whenever you're making your placement adjustments? Speaker 2: Yeah, so a lot of people are doing this the wrong way. The reason why we're even doing this episode is because I've gotten so many people who are using AdLabs lately thinking AdLabs is doing it the wrong way because they're so used to doing it their way and it all comes down to, from the very beginning, people are just calculating the placements incorrectly. So we have a entire episode. Go back and watch that. It's episode 44 of That Amazon Ads Podcast titled How to Optimize Placements for Sponsored Products and Sponsored Brands. So everything that you're about to hear in this episode is really just answering the questions from episode 44, but very important that you watch that first. So I think we're just gonna jump into a screen share, Andrew, and give an example. Actually, before we do that, I've got a question for you, Andrew. What is the default CPC for a placement? Speaker 1: Default CPC. I'm not really sure what you mean by that. I would just say whatever your bids are currently set to. Plus your placements. Speaker 2: Basically, yeah, it's a trick question because there is no CPC for placements. So people are always trying to control placement ACOS. You know, ACOS, we also have the episode on revenue per click and breaking everything down there. But people are always trying to control the CPCs On placements to lower the ACOS without controlling the keyword bids, which is where the CPCs come from. The placement modifiers are only increases to the keyword bids. So you really cannot separate placements from keywords. They have to go hand in hand. And so everything kind of comes back to the revenue per click from the keyword bids to the placement level adjustments. And so with that, let us actually just jump into a quick screen share and we'll show off what we have. All right, so I've got pulled up the placement adjustments example, and this is just a single campaign. Optimizing AdLabs just to give you a really good picture of where people kind of get confused, which is they're looking at the top of search, rest of search, and product pages, and they are seeing that the lowest A cost is on product pages. So they say, wow, it's only 10% A cost. Why do we not increase on product pages? We should be increasing there. This is showing a decrease. We're decreasing from it was 4%. We're pulling it down to 0% when that's the lowest ACoS. Let's just say in this case, we're saying our target ACoS is 15%. So the ACoS on top of search is already at 15%. So they're saying, you know, why are you increasing on top of search? It's 15% ACoS. You shouldn't be doing that. Product pages is low ACoS. You should be increasing here. All right. With that logic, what they would be doing is they would probably just keep this. They would probably not change these. And then because the ACoS here needs to be 50% higher, they would then probably, I'm actually just gonna do it. So what they would recommend is they would say put this to 29, put this to zero, and do a 50% increase here, right? Because they're thinking, they're trying to optimize based off of the ACoS. Here is why that is a problem. We're not going off of the ACOS. We've got different keyword bids. The base bids in this campaign are around 75 cents. That's like what the average CPC is and that's why The placement on product pages is the lowest is because the average, when there was virtually no CPC there, same with rest of search, we've got some items that are averaging $1.30 bid, some other targets that are averaging $0.75 bid. That is what's creating this lower A costs is just a lower CPC bid, which is not a factor of the adjustments going on with these placements. It's the keyword bids within this campaign that are affecting that. The solution here is to increase the ACoS on product pages to get that up to 15%. You're actually going to have a lot of low ACoS targets in this campaign that are going to need their bids increased. But it's the conversion rate that we are following. If the conversion rate on product pages is the lowest, and we actually have a better conversion rate on rest of search and top of search, It makes more sense to be increasing for top of search if the top of search conversion rates practically double, right? It's getting very high up there. I mean, I think the math was around 89 cents, at least based on the revenue per click, 89% higher. And that's why this adjustment came out to 89% previously. So that's what we're trying to get to. And I'm just going to pause there, Andrew, and see if you've got questions or if you forecast that people listening are having questions at this point. Speaker 1: I would just say that out of all the accounts that we look at on a regular basis when doing demos of AdLabs, this is one of the most commonly misconstrued areas where they aren't currently optimizing placements appropriately. And so if you log into your account and you pull a placement report, or if you're using AdLabs, you look at your placements, and you're demoing or whatever, you're probably going to find that the spend allocation across these placements does not quite match up with the conversion rate difference between those. And so there's a lot of inefficiencies that come up by putting so much budget. And you'll see a lot of times, see a lot of people spending too much on product pages where conversion rates are generally just a lot lower. And so Yeah, I think that you explained that really well and it's just a matter of how people are kind of thinking about how to optimize these placements. They're not really considering that bids and placements are a cohesive unit and their ad spend is just being forced into placements where they're just a lot less efficient. And so with doing it this way, you're able to get that ad spend where it's more efficient and direct those ad dollars more effectively. Speaker 2: Absolutely. And something else that's very important to note with these ACOS based adjustments. Let's pretend my target ACOS was not 15%. Let's pretend it was 30%. If my target ACoS was 30%, then what people would think is, oh, to get these ACoSs up to target, we need to double the spend up here to get that ACoS from 15 to 30. So they would do a 100% increase here, they would do a 100% increase here, and then they would do a 200% increase here. I don't think I need to explain why that's a bad idea, but it's a very bad idea because all you're doing is you're globally increasing all of the bids in this entire campaign by 100 to 200 percent, trying to pull the total campaign ACOS up to, you know, You know, 30%, which yeah, that will probably, you know, get you up to 30%, but you're not have like, there are different keywords that have different ACOS within that campaign. You're probably gonna have some keywords that are high ACOS, other keywords that are low ACOS, and that's ultimately how you're going to be controlling these CPCs. Inversely, let's say the target ACoS was 10%, okay, so if we gotta pull this down, well, we can set this to zero, we can set this to zero, but we can't go below zero, so there's no way now to get that ACoS down to 10% if you're just trying to control the placement settings, and this one could be 0%. If you're trying to lower ACoS on these placements to 10%, you can't. You have to lower the base bids. And point being, It does not matter what the target ACoS is. When you go back and watch our episode 44, you will see what the target ACoS is, is not a variable in calculating the placement settings. If I am targeting a, let's actually just do it. Let's back up here. Let's now target a 100% ACoS and say, you know, yeah, whatever. Prioritization doesn't matter here. 100% ACoS. What do these placements come out to? Let's filter down to just these placements again. It's the exact same. It's $0.2589. Why is that? Because the only thing that matters is the relative difference in conversion rate or more specifically the RPC. Because there's certain levels of some placements are performing better than others. So the placement with the worst performing conversion rate gets set to zero. That's where all the base bids in this campaign are revolved around. Now to control the ACoS on this placement, we are now controlling the bids within that placement with the lowest conversion rate. So if there's some low ACoS keywords that need some bid increases, we're gonna be increasing and decreasing on the keyword bids themselves. And the placements are only based on relative difference In that conversion rate, so the higher performing conversion rate placements will always be getting those bid increases no matter what. And if the ACoS is too high, you're going to control that ACoS at the keyword bid level. So I hope that makes sense. I've just got one final thing to say on this topic and then we'll pretty much wrap it up. But Andrew, anything to add so far? Speaker 1: No, let's hear what you got to say. Speaker 2: So the final thing is that sometimes people are paying attention to the delta column and saying, you know, let's just say hypothetically this, let's say previously this was set to 129%, okay? If this was set to 129% and we are targeting, you know, higher ACoS, whatever, when the calculation comes out to 89%, because that's what the conversion rate is, 89%, higher than the worst performing, if this was historically at 129%, And we're now changing it to 89%. The delta would show a decrease, right? It would be showing that we have reduced it from 129% increase down to 89% increase. So sometimes people freak out that the delta is negative, that we are reducing the adjustment when we are below the target ACOS or whatever. But you just have to keep in mind that it's still a net positive increase. So it's still an increase, even though it's not as much of an increase as before. And it's probably because before, when you were using a different date range, the relative conversion rates between these replacements were different. And so from the last optimization to this optimization, those spreads between the three different conversion rates is gonna cause some fluctuations. Sometimes they're gonna come a lot closer together. And so all three placements are gonna be a lot closer together, just small zero to 10% increases across the board. Other times you're gonna have really big spreads across these conversion rates, so much larger increases. So you just have to keep that in mind. And there's also just like the ACOS here, Sometimes your bids have been reduced. So maybe like if I look at this from, you know, I've done optimizations on this account from within the last, you know, I've done multiple optimizations this month. But if I zoom out to like a 90 day, or let's just say last 30 days. So if I pull this into like last 30 days, Yes, in this case, this kind of interesting product page is now performing a little bit better. So that conversion is actually better on product pages compared to these other placements. So here we're seeing that decrease on the top of search a little bit. It's still getting a net increase, but I did not quite get to the example I was hoping to find. But what I was going to say is that you have done changes to the bids within this timeframe. And so As that ACoS is, that ACoS itself is a historical value for whatever that window is. And the bid adjustments here are not historical values. These are just current data sets. So sometimes you're going to be decreasing on a low ACoS placement or vice versa. You're going to be increasing on a high ACoS placement because whatever the current value was, was originally calculated on a completely different data set. And the calculations that you are now doing is a completely new data set. And that's another big reason why we don't look at the current value at all, because the current value is not a historical data point. And we also have an episode on, should you increase bids on high cost keywords? I think it's like, when should you increase bids on high ACOS keywords? And it's that same concept that people who are using the bid as a variable in their calculations and saying, I'm going to increase or decrease the bid based on if I'm above or below my target ACOS. And same with these placement bid adjustments. If I increase or decrease the adjustment just based on it, that's incorrect. You don't want to increase or decrease from the current bid or adjustment for either a keyword bid or placement. Because that is not a data point. That is a current status. It's not a historical data point. So you can't use that. It's not reliable. Someone, Andrew, could have come in here and changed this setting to, okay, let's say that the ACoS here is 14%. I want to get it to 15%. Andrew could have this morning changed the current value to 900% increase. Okay, let's say 800% increase. Do I still want to increase even higher just because the ACoS is low? No, because at 800% is a bad calculation. Who knows how we got there, if someone messed up or if it was from a different timeframe. That is why you always want to calculate it fresh, always use the revenue per click from within that time period. This episode started off really strong and to the point, and then at the very end there, I did my classic move. I just get really detailed and technical and probably lost half the audience. So if that didn't make sense, please leave questions and let me know what you're thinking. Speaker 1: That's right. You heard it here first, everybody. If you want us to take a look at your account, make sure you sign up for a free trial of AdLabs. Make sure you like and subscribe for the podcast and catch future episodes. We'll see you next time. Speaker 2: Peace out.