We have created a standard operating procedure for performance max. And when I say we, I mean, John really John has created a, an SOP for performance max. I'm so excited about this. I think it's a massive, big gaping hole in the industry. Uh, as of yet, there's really no scaffolding around the decisions that you're making. Um, so we wanted to share this with the world, um, with the obligatory disclaimer, and you already know what it is, I'm gonna say, but I'm gonna say it anyway, because if I don't, I still get the comments or, you know, we get our feet held to the fire. Like, well, you, you promised, um, this is, this is a set of guidelines that should be viewed as very, very loose recommendations that are meant to be broken. As often as not depending on your industry, competitive market, et cetera. So, uh, this is not gospel. This is just a, when in doubt, set of guidelines. Um, and, and, uh, I'm gonna reference the dis this disclaimer often. I, I feel, but I I've gotten it out. So, uh, oh, I guess the other thing that I should say is this is probably gonna. 30 days ago, we were telling everybody start with maximize conversions today. We're telling everybody start with maximize conversion value it's because we learned it's because we learned when we were saying start with maximize conversions it's because the campaigns that we use maximize conversions, uh, with were more successful, faster, but then we realized that long term the optimization was limited. So. You know, now it's a question of, well, do I need to prove concept quickly or do I want to optimize over time? And because we're an agency that's working towards customer retention, we wanna optimize over time, but that's a good example of, you know, there's other things that we might be doing. Um, now that we could find out, you know, days, weeks, months, years from now, our we might recommend, uh, uh, an alternative for. So, um, I, I ask for your grace. All right. That's enough. Um, here's the beginning of the standard operating P. Every performance max campaign is gonna be built with, uh, this general set of guidelines and I'll go through them in sequence. The very first one is how do I separate my asset groups? That's how that's, you know, one of the big questions that we get asked all the time for us, it's by product or service category, um, for lead generation service category, obviously, but for, for e-com, uh, you're gonna separate your asset groups by product category. This is just the most obvious it's the most basic, it's the easiest thing to do. Um, it requires the least amount of thought and most. most businesses already have their product categories defined and delineated. So it allows us to kind of like live into their presence and existing structure. Um, and that's not, you know, that's helpful not just with product categories and, and you know, their, the, the tree that they have built, um, you know, their, their supposed product site. It's also helpful because they generally, uh, separate their media and assets and, and marketing materials that way too. So, uh, separate asset groups by product categories. Now here's where things get interesting. You're gonna replicate the asset groups by each distinct and applicable audience signal. So if you have five product categories and five audience signals, you have 25 asset groups. Here's what that means. If you're selling shoes, which is one of John's favorite examples, um, one audience signal might be, um, Let's say you you're selling shoes and, and you have, uh, dressed shoes for, uh, people that are going to a wedding. Um, so you've got wedding shoes. Are there wedding shoes? There are for this example, uh, the audience signal here. If you were to do an InMarket, uh, audience for people who are shopping for wedding dresses, a separate audience signal, wouldn't be somebody who's visited the David bridals. Those would be potentially in theory, the same audience signal, depending on how granular you want to go. And then a subsequent audience signal, um, again on shoes might be people who are, uh, in the job market. You know what I mean? Like you're gonna, you're gonna choose audience signals and make sure that they're distinctly separate. So the way that you're targeting the audience signal can be bundled inside of a separate inside of the same asset group. Um, and then you're gonna separate the audience signals thematic. and then you're gonna take every audience signal that applies to all of your product categories. So, you know, maybe you have. Towed shoes, high heeled shoes, uh, big and tall shoes or whatever. And you're like, all right, do all of these apply to people? Go into weddings. Yes. Cool. I've got, um, you know, that's, uh, all five of these are replicated one time for that, uh, audience signal. And then I move on to my next audience signal. I hope that made sense. I bumbed through that, but I think I got there. Um, you're gonna have a combined asset group, which has all of your products to all of your signals. This is where we're going. Google, give me all the data you. Give me everything. Um, and then you're also gonna have a combined asset groups to where you're gonna have all of your products to each individual signal. So, and we wanna play this all product game because you wanna see, uh, you wanna see what the cross pollination looks like on a signal and product basis. Now you might run out of asset groups and if that's the case, you now have to begin making some hard decisions as to which product categories and which audience signals you think are most viable. And, and that's the way to do it. Build a hierarchy. And to be honest with you, Guess guess it, it, you know. Okay. Well, I think we're, you know, the, the, the, the wedding shoes have real high margin. I'd love to see if we can make that work awesome. But my hiking boot super low margin, and, you know, we're not very competitive there. So I might deprioritize that in a couple of different places. Um, We're always gonna turn final URL expansion on, unless you have extreme cases, uh, or, or a very strong reason not to do this. However, within final URL expansion, don't forget to exclude URLs that for whatever reason you don't think, um, should capture traffic. I'd be very careful there. I have another video on final URL expansion to the fact that we're converting off of URLs and I never expected to convert off of a B. and we can't, we can't see this in Google ads, but you can kind of piece it together with analytics. You see pages that aren't conversion pages that, that feel dangerous word to use in a data driven environment that feel like they contribute to the conversion. So testimonials case studies about us, mission, vision values, giving back those types of pages where, okay, they're not converting off of that page, but gosh, it sure feels like on my conversion path, people that land on those pages are more likely to convert later. So I would, I would caution you against. Being too aggressive with your UL exclusions, but don't forget to use 'em. Um, we're gonna use maximize conversion value and you're not gonna have a TRO as there's an exception to this. Google told us recently that if you apply no TRO as the system automatically applies a TRO as of 200%, which is really frustrating by the way. And also very stupid. Um, if I wanted a 200. TRO as, and I would've applied it 200% TRO as if your TRO as is under 200%. And then I were talking to a client about this yesterday. They have a break, even target. They're like, look, if you can go one to one, um, we're in great shape, cuz their retention is so strong. So if your TRO as is under 200% set a TRO as if it's not set, no TRO as you're going to be spending it efficiently on purpose. You're gonna spend four to six weeks potentially longer, possibly overinvesting inside of this market, but this is the only way that Google's gonna go figure out what product categories work with, what asset groups. So realize that you're not spending here to buy customers. You're spending here to learn and, um, it's funny because that, that was a pretty consistent narrative in old school advertising radio, television newspaper. It was, yeah, we're gonna do the first season is gonna figure out, you know, what, what placements workforce and the second season will, we will get a little bit more efficient. We've been spoiled by digital marketing and, and we're kind of back there now. So this is, this is, so this is a pill we've all had to swallow in the past. Not that any of us were alive back then, but you know what I mean? Um, humanity has had to contend with this before. Uh, you're also gonna include all product. Unless there are potential fulfillment issues. That's the one big caveat. If you ever think you're gonna run outta something, PAX is a scale machine. Don't. Teach Google how to sell something. You're not gonna be able to sell it kills campaigns. It's such a nightmare. And you know, it's funny cause I bring this up on sales conversations. I'm like, all right, so no fulfillment issues at all. Like, oh gosh, as much as you can sell, we can get a lot of hoorah. And then all of a sudden we start selling stuff and I'm like, Hey, we ran out of product. Skews am four and nine. And it's just like, well, we talked about this and you lied to me. So these are the best practices again. mild suggestions, by the way, they're all in one PAX campaign. I guess I haven't said that because it's sort of implied. We're not separating PAX campaigns, we're using one PAX campaign. And then there's a reason to separate PAX campaigns later. Let me move my face. Now you are going to build this and you are going to run it for six weeks, plus whatever the client's time lag is or whatever your time lag is, if you don't know your time lag. then you don't know what this number is, and you're just gonna have to figure that out on the heels of performance, max. So if you're a brand new, uh, advertiser, then I don't even wanna tell you to guess, because it's pretty difficult to determine many instances, but, um, if you run. campaigns prior to especially smart shopping, uh, because performance max and smart shopping operate in similar ecosystems in terms of their multichannel distribution, then your time lag instead of smart shopping, I think should be pretty analogous to your time lag instead of performance. Max, I don't have data to tell you that that, that they're, you know, pair pursue, but I'm just kind of assuming that that's the case. So take six weeks plus your time. Plus your performance timeline. Uh, this is especially true for lead generation, but can be true for high ticket econ. You know, we've got somebody in the medical device space and they have $30,000 medical devices. People don't buy those right outta the gate, you know, it's six weeks to, uh, allow the campaign to. to learn, reach stasis, and then you have to wait the time lag period, because the time lag is the amount of time it takes for, um, uh, the marketing endeavor to start showing what it can, what, what happened in the past to come to fruition and then your performance timeline. Here's what this means. If you're not a little scared of this, by the way, then you're not paying attention. We're gonna, this has never been the case with Google ads ever. We're gonna build it. We're gonna launch it. We're not gonna touch it for what could amount to, for some advertisers months. Yikes. I'm telling you, we have a dozen, some odd case studies now showing that this is necessary and the people that want to go play tinker, tinkerer, tinker bell, the people who wanna go play tinker bell are gonna are. You just reset, learning you kill PAX. Um, you have to let it be, I wanna sit here and just stew in that message. You have to let performance max learn. You have to let it be. Um, every time you make a change, it resets everything and, and that's only, that's not entirely true, but it's, it's worth believing that it is true. Um, I've heard it said that, you know, I, I don't believe in God, but no matter what I act as though he exists because regardless of whether or not he does exist, it's just good for me to, to act in a fashion. Uh, that keeps me right with the world as extreme as that example is, that's the way I want you to look at performance max, you make a change, it resets learning. Um, the other thing that I'll say before I move on is the amount of money you spend. is gonna be dictated by the amount of product categories. You have, the amount of audience signals you're targeting and how competitive your market is. And there's no way for me to tell you what that is. Um, if you've been running Google ads in the past, you probably have a pretty solid. And that probably will carry over from all the campaigns that performance max is gonna steal from. If you have it, you're gonna have to go figure that out. It's a hundred dollars a day at a minimum, but a hundred dollars a day, probably isn't enough to support 25 asset groups in a competitive market. So you, you're gonna start to see, you know, where you're making a splash and where you're not at all. So you build a performance, max, you follow these SOPs. You wait this amount of time, which by the way, I would love for any of you SP for, for those of you, who've actually been successful with performance, max, you arm, chair, theorists. I don't care what you have to say. The people that have been successful with performance max, I'd like to see if you disagree with any of this. This is a set of logical and educated assumptions based off of what all of our strategists cm specialists have said combined with John's leadership. And this is the output, which is pretty solid by the way, we have $54 million. In ad spend under management. And most of it now is headed in the performance max direction. So I think we have a lot of data, but I I'm still interested in being informed by other people on the outside because some of this could very, just be self fulfilling prophecy, echo, chamber, whatever this is the best we can do for now. Now you've run your PAX campaign. You've waited the adequate time. Super important. You've waited the adequate time now, what do we do if, and I'll actually start from the right and move to the left because it gets more. if you're not meeting your Roaz goals, uh, which it, which case the budget is negligible. Um, if you're not meeting your Roaz goals, then you need to figure out why. And this feels very passive aggressive, but the problem here y'all is there's no S O P for this. there can't be. If there was, I wouldn't have a job, like we wouldn't have an agency, there's a hundred billion reasons why this could potentially be the case. Um, and, and a lot of 'em aren't even in the Google ads ecosystem, like it could be on site issues, um, visibility, competition pricing, uh, go to market strategy messaging. So if you're not meeting your Roaz goals after having waited the adequate timeline, um, then. you need somebody who actually knows what the hell they're doing to review and diagnose. It might mean cuz performance max can fail. It might mean that you just need to lean into more traditional campaign types. Now, one of the questions that I got when I presented this to we'll call it a sounding board was, well, if I'm not meeting my RO as goals, I don't have a TRO as applied. So can I just apply the TRO as you sure can. That's after you're learning, um, after you're learning. So once applying T RO as if you're still not meeting a RO as goals, now we need review and diagnosis and then lean into more traditional campaigns. If you're meeting a RO as goals, but you're not meeting the budget, you come down and do this. First you're gonna clone the entire PAX campaign. This is gonna get real counterintuitive by the way, we had a bloody nerd war internally over this. Uh, you're gonna clone the entire PAX campaign in the new PAX campaign. You're gonna shut off all non-performing asset groups. So the new PAX campaign only becomes your top performers in the old PAX campaign. You're gonna shut off all of your performers. This was the thing. That scared the daylights out of everybody is, well, wait a minute, I have a campaign that's working. I'm gonna take everything out of the campaign. That's working and only turn on the things that aren't working. And then I'm gonna create this new campaign with the things that are working. But it stands to reason that because we've all seen this maybe only a billion times, uh, even when all things are equal, for whatever reason, those campaign types aren't necessarily treated as other equals. So that was the fear. We've tested it both ways. And the new campaigns with the performing assets. perform better while the old campaigns with the non-performing assets actually perform better with the old campaigns. And let me tell you why. I think this is, this is highly speculative by the way. And I'm an idiot and I don't know what I'm talking about, but this is just my best guess. I think if you take the old campaign that the original campaign, all the performing assets. are the performers cuz they're they're the lowest hanging fruit. So this old campaign went, tried against, you know, tried all of the asset groups, all the audiences, all the product categories figured out what worked and was able to make those work well. When you turn those off, the original campaign has still has a wealth of data and still made a bunch of attempts against the non-performing assets. So it's ability to make those assets perform greatly outpaces a brand new campaign with the non-performing assets. So this isn't about the performing assets. So far, what we've seen is they perform just as well in a clone campaign with everything else removed even better because they have dedicated budget. This is about your non-performing assets. If you want your non-performing assets to perform, then take away their big brothers, the, the, the assets that are performing even better, because they're the ones that are eating up all of the, all of the budget. I hope that makes sense. That's a theory. I think it's a pretty good one, but it's just a theory. So, um, new campaign. Shut off all non-performers old campaign. Shut off all performers and you have now effectively PAX sculpt. Uh, one really quick note in your fire final URL expansion for both campaign types, you're gonna turn off all of the pages that apply to the products that have been removed from each of the PEX campaigns. So the new campaign is gonna have all of the pages for that applied to any of the old, uh, uh, campaign products removed and vice versa. Um, you wanna make sure that you're your. keeping these things separate and from being incestuous. And you're gonna do that until you're RO as is you're meeting your RO as, and you're meeting your budget, which incidentally is the, the third thing that could happen from a brand new campaign. If you're meeting RO as, and you're meeting budget, drum roll, don't touch it. Stop don't touch anything. It's a crazy, don't mess with a good thing. Cause everybody's like, oh my gosh, we're making a bunch of money. What do we do? And the. not a damn thing. We're gonna sit and we're gonna watch it now. Are there. Optimizations that can be made. Yes, of course. Absolutely. A lot of 'em are actually in campaign types outside of performance. Max, you can start to see like, oh gosh, you know, this is working here and this is working there. I'm gonna build a search campaign in order to push PAX more to search. Um, there it's really fun when you get there, but in the very beginning, you're gonna optimize yourself out of performance. So don't touch anything, uh, unless you really know what the hell you're doing, which really means that you're us. Uh, More aggressive than I meant for it to be, but hopefully y'all know what I mean. Um, you can sculpt lower performing asset groups. Very, very, very, very, very carefully. A lot of that is media, uh, copy imagery offer. So there, there is some optimization to be had, but be real careful if you got a PAX campaign and all this is the things we've seen this happen, we've seen it to where it's like, oh, PAX is killing it. And then. The client comes in and starts pressing buttons, and then they tank it. Uh, we have one client in particular and this is funny. Uh, I'll go big for this one, one client in particular, we built a performing high performing PAX campaign for, and their Google rep. Their dedicated Google rep came out and said, Hey, we should. And we were like, no, we don't wanna do that. And then they convinced the client to do it killed performance. We brought it back, had to backtrack, redo everything that we did brought it back. And then their dedicated Google rep had another. And then we said, no, let's not do that. And then they did it and it killed performance. We brought it back. Third time won will be the charm. The client now understands that the Google reps. not all they're cracked up to be. Uh, you know, I, I I'm pretty mean on the Google reps. It's, it's those, it's the lower level ones. Some of the high level Google folks are freaking the smartest people I've ever met. Now, if you have, if you're meeting rows and you're meeting budget and you want to increase your budget, you have two options. Uh, it's based off of how much budget increase you, you, you plan on doing so. Are you more or less than 50% of your daily spend? Um, If you were less than 50% of your daily ad spend, then you have a maximum of one budget increase every two weeks. If you're more than 50% of your daily ad, spend a maximum of one budget increase every four weeks. We're spreading this out. Google. In writing has said, we can take any amount of addin ad spend increase. This is a lie. This is a boldface unadulterated in capitals with underlying lie. And it's shocking to me that it's, I I'll try to find a link if I can. Um, it's crazy that they'd even say that this is how you increase budget, increase in weight, increase in weight, increase in weight. And then at a certain point what's gonna happen is you're gonna find. here, you're meeting raw. You're not meeting budget. And then you scroll down and you do this, and then it gets very cyclical. Um, which I actually kind of like, it means the whole damn thing kind of makes sense. So this is what it looks like, zoomed out. Um, I'll include a link to this somehow. And. I hope that this is helpful. I imagine this is gonna be very iterative and I'd love for y'all's input, by the way. I mean this, all this, this is all so new. So don't think just because you're smaller, just cuz you have one campaign that, you know, oh, there's no way I figured out stuff that they haven't, you would be surprised. Everything that we figured out has come from. Think about this. It's all just individual. Participation and inclusion. It's, you know, a specialist figures this out and test this a strategist test that a cm recommends this. And then, and then we sew that together in this little patchwork, quilt that we're offering up. So we'd love to include you in the quilt. Um, if you have ideas or recommendations for us. And then, and then my hope is someday I'll be a little bit more robust here, but right now it's just like, it, it would be like asking the question like, oh, why didn't my Google ads campaign perform well. You know, like, I, I it's, it's impossible. Impossible question to answer. Um, that's all I got really appreciate. Uh, we just hit 10,000 subscribers by the way. So I'm pretty excited about it. Appreciate everybody's support. I hope the content that we're pushing out has been, uh, helpful and, um, I'll see you. Wait before you go, I'm constantly looking for amazing people to come join our team. So if you're passionate about Google ADSS and you're passionate about customer success, please go to so late.com/apply. And we'd love to see you as a part of the solutions 18. Also, if you like this video, give us a thumbs up. It lets the YouTube algorithm know that we actually know what we're doing. And uh, don't forget this job. We shoot a video every single day and I don't want you to miss out on any. Lastly, if you have questions, comments, concerns, confessions, or you just hate my face, my voice, go ahead and nose up in the comments. We get very little human interaction and even the heckling is something that I kind of get a kick out. So thank you so much for watching. Thank you for being subscribers. If you're a subscriber, don't forget to apply. If you're interested in working at solutions eight, otherwise I'll see you tomorrow.