Keep the PPC Decision Log

Outcome: Record what changed, why it changed, source data, expected result, and the next verification date.

What to do in this lesson

Watch: start with the lesson video.
Learn: use the summary and key points to capture the operating principle.
Do: complete the action steps against one real product, SKU, campaign, supplier, or workflow.
Submit: write one action card with owner, evidence, next step, risk, status, and review date.

Watch

Hosted on Google Drive.

Lesson summary

This lesson teaches a systematic approach to maintaining and adjusting Amazon sponsored product campaigns, with a focus on evaluating keyword performance, making bid adjustments, and transitioning keywords between campaigns to optimize campaign structure and profitability.

The key concepts covered include analyzing conversion rates, cost per click, and placement data to identify underperforming keywords, as well as techniques for shifting impressions and clicks towards higher-performing placements like top of search.

The lesson also covers the importance of maintaining a tight, relevant keyword set in ‘discovery’ campaigns and quickly identifying and removing irrelevant keywords to prevent wasted spend.

Key points

Analyze conversion rate, cost per click, and placement data to identify underperforming keywords
Adjust bids and placement targeting to shift impressions and clicks towards higher-performing placements like top of search
Maintain a tight, relevant keyword set in ‘discovery’ campaigns by quickly identifying and removing irrelevant keywords
Use the search term report to proactively manage and control the growth of the keyword list
Focus on ‘momentum over structure’ when making changes to avoid disrupting successful campaigns

Lesson notes

In this lesson you will discover the right actions to take in different scenarios Access the placement bid calculator here Scenario 1 - Main High Volume Keyword [Video] Access the Impressions SOP here Scenario 2 - Medium Volume / Broad / Non-relevant [Video] Mixed and Multi-keyword Campaigns [Video] Discovery Campaign Maintenance [Videoo] BONUS: Pro Tip on Top Keyword Maintenance Using the 80/20 Rule

Action checklist

Review your campaign performance data and identify underperforming keywords
Adjust bids and placement targeting to optimize for higher-converting placements
Regularly review the search term report and remove irrelevant keywords

Full transcript

Open transcript

So let’s go through scenario one. So this is the top level keyword that I’m going after for my specific product. It’s probably one of my top three keywords. That’s really all the context you need. It is fairly relevant. It’s more of a root keyword, but because I’m more of an established product, I can maintain some of this. And regardless of if you have the credibility or not, if you have a good listing or a good product and good momentum, Amazon will typically favor you, give you a little bit lower cost per click and allow you to perform and show up for this specific keyword. All of that said, given the context, I’m going to take you through the maintenance breakdown of how I would adjust or leave it alone, depending on what’s happening here. So for this specific keyword, I have, as you can see here, I have about a 30% conversion rate here. Same thing goes with down here. It’s a little bit lower than 30%, but still pretty justified. What I might use this as is a signifier that I could actually go and search for this specific keyword, look at the competitors that are actually aligned with this keyword or showing up for this keyword and actually target them in product targeting ads specifically for better control. So I’m not saying that you stifled this momentum. What I would say is almost using this as a broad campaign where because this is a sponsored product ad specifically designed to go after a keyword, you want to make sure that you have keyword driven sales, meaning top search or rest of search going through this specific campaign that goes through most of all of campaigns that go through sponsored product and keyword driven campaigns. However, because it’s the top keyword, I’m using this data here as performing well on product pages as basically a justification to start maybe product targeting ads aligned with any product showing up under this specific keyword. Now the filter that you want to put on that is you still want to make sure that you’re not focused on the competitors that you probably can’t compete with because you really don’t know based on this data snapshot, whether or not you’re competing with the top competitors or whether or not Amazon’s targeting maybe the lower competitors, maybe bottom of page, middle of page. So you really want to be smart on the way that you attack this, but it does give you some justification to go gain control on product targeting, targeting specifically so that you’re not allowing Amazon to do both of those targeting techniques on this specific campaign because you really want to keep this focus on keyword targeting. With that said, I always adhere to momentum over structure. I’m not trying to stifle my growth or my momentum based on that. I’m going to go start those campaigns in addition to this rather than going and changing all of my bids for this. However, if this was skewed in any sort of way where you wanted to make changes, what you would do is you would go to the actual targeting, you would look at the cost per click. I do have this cool bit adjustment calculator here. I’m going to go ahead and duplicate it and move it over. And really what we’re going to do is we’re going to go through a little, a simple breakdown. So we’re going to go through the campaign structure. So if you want to, if you want to move these clicks toward top of search because you’re not getting the momentum that you want at a product pages, maybe this, this is a hundred and thirty four clicks with no sales or a very, very low conversion rate. If this conversion rate does not meet or exceed or at least within like five or ten percent of the overall conversion of this top of search or rest of search, you want to make sure you’re moving this click off of this campaign. So because I’m paying eighty nine cents a click, what I would do is I would go in here. I would make sure to put in the bid adjustment to something lower than eighty nine cents, meaning I might do something like eighty five cents just to make sure. Now what I’m also going to do is I’m going to look at the placements and I’m going to see what am I paying top of search? I’m paying two thirty one. What we’re trying to do here is we’re trying to move the impressions and the placements from the product pages to top of search to ensure that we sort of have the eighty twenty effect going in motion here. You really want eighty percent of your clicks to go through top of search or rest of search when you have a keyword driven sponsored product campaign started. So I’m going to take this two thirty one and I’m going to put that in as my target bid two point three one and it’s going to tell me that I needed a hundred and seventy two percent top of search. Now all I’m doing here is I’m going to go do my bid adjustment to that eighty five cents like I talked about but because I already have a hundred ninety five percent top of search I’m going to just basically go ahead and go change this to eighty five cents and call it a day. So hopefully you understand the breakdown. I basically took the ninety six cent bid that I had to start. I wanted to change the adjustment from the placement ad sorry the placement ad of eighty nine cents I wanted to go below that to eighty five and I wanted to change the adjustment so that Amazon’s not looking to place me in the same placement in the carousel as they did for the specific clicks and sales. Since I’ve lowered this bid to eighty five cents I want to make sure that my multiplier is meeting or exceeding this two thirty one bid which it already is. So because the bid adjustment is already above at one ninety five versus one seventy two I don’t need to do anything else other than to lower my bid. So hopefully that makes sense again in this situation you don’t necessarily need to do that but because I really do want my focus to be on keyword driven sales I’m going to take a little bit of emphasis off of the product pages because I really want to maintain that eighty twenty effect as it pertains to my main keywords and keyword driven results meaning the keyword search top of search or rest of search is getting the click and the sale and the impression on those specific keywords. So hopefully that makes sense. In the next couple of videos I’m going to go through more scenarios to break down a couple different iterations of this. Hey guys Brendan here and I want to take you through sort of a scenario too. This is going to be where we talk through sort of the filtering criteria of how to assess your campaign build that context around the keyword or the campaign you’re adjusting and why you’re making those adjustments and how to make those adjustments as well given that it’s a different scenario. So I’ve sorted these by performance top down by a cost so I just basically do highest to lowest on a cost it’s still a good performance indicator and then you’re basically just looking for performance indicators like overall conversion and maybe click through rate although once you get into the campaign you’ll really be able to diagnose whether the click through rates really a problem or not based on your placements. But what you are looking for here is more performance based on conversion so how many clicks and how many orders came out of this as it relates to the a cost typically when the conversions low the a cost will be high the rise will obviously be low as well. So you’re going to want to take a look at these sort of one by one basis and build context around them. So I’m going to dive into this campaign and we’re going to take a look at the actual data behind this. So I change this before just to go through the exercise so forgive me for changing it back but basically I want to assess this putting it through the three filtering criteria. So for this keyword I’m going to build context around why I’m making decisions I’m making. We’ve already eliminated that this is not a placement issue here. I’m actually going to pull in a little bit more data like we did on the other. So I can have a 14 day trend line and really build the context around this like it showed in the actual campaign performance history. So I see here that this is immediately not a placement issue of the 80-20 rule exists. I am getting most of my impressions through top of search so I’m not too worried about the placements piece. I’m going to limit that as a problem for this specific campaign and keyword. What I am going to do though is ask myself is this the right keyword to be going after? So the context on this keyword specifically is this is not a keyword that I probably would have gone after tier one or when I launched or to build up my credibility overall. This is a keyword that came in kind of tier two, tier three and I’m using those loosely just to say this is not a keyword I would have gone after at launch. So it is one of those sort of testing keywords, it’s sort of irrelevant. So it’s maybe not right keyword wrong timing or low credibility because I do have credibility and I’ve established this product so it really comes down to is this the right keyword to go after with this much aggressiveness or not and it is not. So what I want to do based on the conversion rate and the performance is I want to lower this cost per click and I want to really shift the impressions and the clicks that Amazon’s giving me back downward. So what I want to do is take you through the exercise of how you would shift the cost per click down versus shifting it from placement to top of search and things like that. So you’re still going to go back and use your bid adjustment because we’re going to use the top of search percentage to make that happen. You’re going to put in your original bid which I didn’t show you but it is 90 cents and then you’re going to put in the dollar value or the cost per click that you want to achieve and then that will then give you the bid adjustment that you need to put in. So I put in a dollar 25 as a cost per click that I would like to see. Again, there’s no perfect magic number here. It’s really about just taking action and setting the precedent to Amazon that you want those impressions and those clicks to shift. So I put a dollar 25 in here and then I can see here that my percentage is 39 percent. So I’m going to go back to my placements and put in here 39 percent which is why you guys saw 39 when I came in here to start the video, didn’t save it. And then that should reset the campaign so that you don’t pick up these clicks and you don’t pick up this lower conversion rate. So you’re really just shifting the emphasis to Amazon so that you can save on this a cost maybe show up instead of top of search, maybe show up middle of page, so on and so forth. So hopefully that gives you guys some good criteria to go off of. Again, I want to reiterate that make make decisions based on having enough data as well. It’s really important to build the context not only on the keyword and why you’re doing what you’re doing, but also have enough data to inform the decision. So this is where it’s really critical to be able to be in a mode like this because if I would have looked at maybe a seven day snapshot, it actually has worse data on the most recent seven days. So I could have taken action on 13 clicks, but it’s good to build enough context to really make an informed decision over time as well. So make sure that you are setting these date ranges in a way that allows you to pick up enough data to make an informed decision. Obviously put it through those excuse me contextual pieces of is it the right keyword or is it too irrelevant is it the right keyword, but maybe I have I don’t have the credibility to go after it or is it a placement issue. So in the next couple videos, I’m going to go through a couple different scenarios. So stay tuned. Hey guys, Brendan here and I want to go through a mixed keyword campaign scenario and sort of the maintenance structure and the decision making processes that I’m going to go through in order to properly maintain this campaign. The example here shows a polarization really of what can happen when one keyword or a couple keywords starts to pick up momentum and then what that does in your decision making process as far as all of the data being merged from a placement standpoint. So I’ll go through this stat line real quick. Obviously you can see there’s about 80% or 90% of the overall volume and the clicks and sales are going through this top keyword here. I’ll tell you why in a minute, but these other four keywords here are pretty low volume, decent stat line overall, nothing too polarizing with a cost. So I’m not I’m not truly worried about maintenance because performance is bad. However, in order to change it on a per keyword basis, I need to have isolated data so that I can make informed decisions. When I start to go over to the placement piece, you can see here we have some some polarity on the product pages and while we can assume and and figure out that most of the impressions generated on product pages are from that top keyword based on the total impressions it had. And I’ll show you that here because you can see it had about 15,000 overall impressions. So we can pretty much assume that the majority of these impressions are coming from that top keyword here. But we don’t what we can’t do though is now make decisions on how we adjust the cost per click because this placement data here is at a campaign level, not at a keyword level because when I go back to this top keyword, I’m only paying an average of $1.13 per click. My bid is actually already lower than the average cost per click on product pages. So it really leaves me without an ability to make accurate decisions. I basically be guessing on where I move the bid, you know, how I adjust the percentage and it really doesn’t give me a good data pool to start making those those changes in those maintenance decisions. So what I’m going to do in this scenario is I’m going to move out these four keywords for sure. Now, again, it put everything through momentum over structure because this is the best producing campaign. I’m going to keep this in this original campaign and I’m going to move these other four keywords out of this campaign. Now, yes, there’s momentum here, but I always use the 10% rule. If it doesn’t make up at least 10% of your overall PPC sales, then I’m okay with moving it immediately rather than worrying about the momentum that’s been created on this specific keyword. So I’m going to move these out and then I’m going to start to make decisions over the next week or so. So I’m going to give it maybe three, five, seven days if needed to start getting these clicks back. And as you can see, if I go back to just the two-week snapshot that we’ve been working off of, there’s a pretty good amount of impressions that I’ve come through here. Now this was a little bit of Christmas and stuff too, so it could be a little bit polarizing. But again, I want to be able to see this stat line for what it is and make justified decisions, having only this keyword as sort of my placement decision maker. So I’m going to basically make those changes. I’m not going to go in here and start changing anything because the stat line isn’t bad. I’m not seeing 100% a cost or anything that’s super out of line or out of whack. But I also want to call out one other piece of this. So when I went into the search terms piece, what I can see here is that there was one keyword that was driving the majority of this. Now this was the singular version of the plural version that I had in my campaign. So oftentimes what you guys will see is maybe your singular version or like that root keyword, whatever it may be, you can put that in a campaign that has a lot of volume associated with it. However, in each campaign, you can pick up the singular and the plural version. So one, I was running after the plural version of the specific keyword. It was kind of a long tail keyword. But it started to pick up the singular version in this campaign and Amazon started moving the impressions of the singular keyword into this campaign as well. So there might be some duplication in essence. If you have the singular version of the keyword running in another campaign. So you want to make sure you’re also pairing up with that to make sure that you’re not mixing the data up. That one is an outperforming the other. I would assume this is probably outperforming whatever that singular version is. I didn’t go look at that. It’s something I’ll do after this video. However, I want to call that out because this is really why so many impressions started coming here is because of the higher volume keyword, the singular version started going through the plural version, causing sort of a swing in momentum on this specific campaign. So it’s worth calling that out. Again, hopefully it makes sense what we’re going to do here as an action plan. Again, I’m not going to start guessing and making changes right now, just given that the stat line isn’t terrible. But I do want to be able to control this for ranking purposes, really understand where my volume’s coming from, where my clicks and my impressions and my placements are coming from and kind of go from there. So hopefully that makes sense, guys, again, let me know if you have any questions on that specifically, but I’ll see you in the next video. What’s up, guys? Brendan here. Let me go through a discovery campaign breakdown and sort of maintenance structure. My goal with this video is really to give you a more simplified approach to doing your maintenance on these types of campaigns. For those of you that haven’t seen any training or don’t come to the BBC work parties or things like that, you may not have heard about sort of the budget allocation around this. And these types of campaigns should really just be sort of a safety net that helps you find keywords over time. It should not be where the majority of your budget is spent. I think because of the keyword research and the techniques that we teach in Scaler-OS around keyword research, you should be targeting a lot more sort of 80, 90, 95% of your keywords ahead of time rather than sort of doing the old method of climbing the ladder, going from auto to broad to phrase, so on and so forth. So if you flip the model around and you’re really attacking your keywords from an exact standpoint early on, this should really just be a net that you create as consumer buying habits change or they type in different keywords for their products. You should just be able to catch those little things as they come. And so we’ll start from there. I want to implore you to check out a piece of your broad or phrase or auto campaigns that has to do with this search term field. So I’m going to dive into the search term field here. And what this is, it’s actually sort of a real time search term report. So rather than getting the stagnant data on a very specific date range, you can actually come in here and start to direct the date range to see how much data is flown or flown has come through this campaign over time. And it really gives you that real time feel because I want to really stop the track on having to wait for X amount of clicks and all of this stuff and all of those old maintenance structures because what this should do for you is if we’re putting it through the same criteria that we’ve talked about, is this a relevant keyword for my product? Yes or no? You can answer that by the words that are written inside the keyword. Do they describe your product or do they not? If they don’t, they need to be eliminated and negative because it’s exacted. It doesn’t matter how many clicks have gone through it. If it’s an orange and you don’t sell an orange, if it’s a four pack and you don’t sell a four pack, get rid of it, right? So we want to really take a hyper focused approach on trying to eliminate these keywords or not even not just eliminate but pull them out as quick as possible. So whether you’re negative exacting them, whether you’re testing them inside an exact campaign, you want to make those decisions even if there’s only one click available. Because as soon as that keyword shows up, if it doesn’t show a lot of volume associated with it, which it shouldn’t because you have most of those keywords already running in your campaign, if it’s showing you that it’s relevant, I would go ahead and move that into one of your lower volume campaigns, make sure that you’re hyper targeting it from an exact standpoint, and then really try to eliminate this list growing and growing and growing without any sort of control. It’s almost like weed management for your yard. Because what you’re also preventing here as well is sort of the death by a thousand cuts over a thousand days, right? So you can pick up a lot of $1.52 clicks over time if you’re not coming in here and continually moving these keywords out. And as we’ve sort of alluded to in previous videos, because of the placement filtering, because of all of the dynamics that come along with having multiple keywords running in one specific campaign, you’re also trying to eliminate that back build of data pooling and that sort of muddiness that it creates like we talked about in the last video with the mixed keyword campaign. So hopefully that makes sense. I would implore you to go through this. If you haven’t done this before, if you’re not very good at going through your broader, discovery campaigns and doing this, I would make sure to start from a heavy date range. Again, this date range piece gives you really the pivot ability so you can go back as far as 65 days and it allows you to do that here. And I would implore you to do that. Start from there. Start to go one by one by keyword. Does it match my product? Does it not? The other two filters are really not as important and I’ll show you what I mean by that in a minute. So does it fit your product or does it not? Yes or no. If you’re unsure about it, move it out anyway and test it. And you’re allowing the data to pick up over time. It’s sort of a low risk, high reward because in most cases, 99% of the time, these should be low volume keywords. If you’re doing your structure the way that we’ve taught with the keyword research and so you shouldn’t be finding a lot of outliers here. But it gives you the opportunity to control the data early on because of placements, because of all the outliers. So I want to make sure I’m honing in on that. And also getting you guys away from the search term report that really just creates a stagnant approach to how you’re making decisions on these keywords. So go through that exercise. The other piece of this too from a campaign maintenance perspective is I’m going to on a discovery campaign still put this top of search emphasis. And the reason why is because especially for sponsored product, keyword driven campaigns, I want to make sure that I’m still getting keyword searches and keyword search emphasis put on my campaigns. So I would put a small emphasis still on top of search here. Obviously, you can see the sort of 80, 20 rule still exists with this campaign. Not anything to polarizing again over time. It’s not picking up a ton. But I’m going to move these keywords out sooner rather than later so that I can control the data and then move on. So hopefully that makes sense again. Let me know if you guys have questions on that and we’ll go from there.

Resources

Source lesson: PPC Optimization Part 8 Identifying Problems
Resources: none attached yet.

Track: 04 — PPC Control Loop
Module: Optimization Rhythm