client, here's a reason why, and this is why I'm saying you have to know the conversion path as much as you can. I mean, make it your life's mission to identify that conversion path. But that's the thing too, is we always wanna look at every single thing you can, but it will scale 15 grand clients and now a hundred grand client in a year. with just knowing that conversion path and knowing that Google's gonna miss it, knowing that eyes is fake. So once you're getting this data inside with these broad match keyword, What's your analysis here? What are your metrics? What are you looking for? What are you looking to kill? What are you looking to scale? Now what's funny is this whole account looks like absolute crap, but that's okay. last 30 days we spent 125 grand, make 288. I mean, that's not bad, but that's cause like branded is 80 k. These are just, one 50. 3 88, 1 27, 86 we're, we get $12 conversions sometimes. So when they come back and spend 20 grand, then we actually make our money so we get a four x return in real life. but that's, again, topic for a different conversation. But you're exactly right. I'm looking at what is actually happening from a, the asset group is saying, Hey, well, Google's just gonna fricking take everybody and shove 'em into the asset groups equal. And I, we've seen this actually by impression of the asset group. you can see the one asset group impressions are the exact same, no exact same, sorry, divided equally between all the other asset groups. There's really no, like, wait a minute, that one asset group has got like, you know, 80% more all the time. It doesn't work like that. We can see that when it's divided up Equally though, one or two start to perform better than the other ones. And then we can find out why. What, what's the search term? And it's about shopping in ml. And then also what is, what does Delta do with the remarketing? and is that being split between the other two? The one reason why my I have a smart shopping campaign and then a, PAX here is because these two are actually targeting different audiences, which is really crazy, but you'll see much different metrics. These are, this is a PAX with a feed and this is Pax with a feed. And this is just a feed. It's really interesting how these two are actually working together. Same products, same products. We ran a, smart shopping campaign for a while and the hold on one second. Lemme just see if this is seeming gonna be enough different. This one's running 90,000 Muse on YouTube. This one's not cost complex. About the same average cost. Way different though. You can see one doing more op on. Yep. It's beautiful. That's why I was like, what's the best? It's like, shut up. There's no the best, like things like this you can't account or predict. You just have to really understand what's going on. So that 34 cent cpc you, like you said, you got a lot of views. So it's obviously YouTube, discover display. Well you always wanna do is look at cuz average cost isn't cpc. Oh yes. And that's what's interesting is average costs and average C P C are one's counting an average engagement and one's counting the national click. Yep. Gotcha. But, but you know what dictates costs per click when you have engagement, your clickthrough rate I have a hundred views at a penny per view. And one click it's a dollar. Yes. But if I have a hundred views at a penny click, I got five clicks. Why? My CPCs are 20 cents. No, it's, it's o it's obviously in our best interest to have average cost and average CPC in our columns to say, okay, yes, this is what the CPC is, but this is what the cost is for the interaction. Yep. Yeah. Okay. That's good. And this is running feet only now it's not competing on, ad copy. Yep. I only have one ad that is, is. In both PAXs. Yep. So then once you've identified those keywords, you've gone out and built those two search campaigns. What's the structure that you do with those, those search campaigns, targeting those keywords that you're building out? This one is actually kind of, kind of crappy. it's just simple ad groups. It's simple, broad, broad match. This is going to, oh, sorry, this is right here. Brought here. this here is what's getting the majority of the actual conversions, which is really weird gift boxes. It shows up the most. Okay. And so that, that in the, in the general, this was all, all simple, broad. Okay. So building out that, that campaign, after you've identified those keywords that are converting inside the p m. And sending, setting up that campaign using broad match keywords, we can assume that will theoretically get preference from impressions and clicks over the pax. Well, what if it does is it allows it to compete. Yep. And I This is not a low spend account. What about, what about campaign objective on that search campaign? What, what are you doing with that, conversion action? I'm. No. In that. So in those search campaigns, what's your campaign objective? T ROAS max conversions. Oh, I run everything wide open. I never use TCP or T ROAS ever in any of my campaigns. Gotcha, gotcha. except for Braden Yeah. Yeah. So very cool. Yeah, that's what's interesting here. Is this, these general ones, like, remember when I saw custom cardboard boxes in Pax and that asset group? There's my, there's my keyboard here. It's gonna run that, well, not really. This campaign can spend about $25,000 in a month, though. I spend 1100 on my keyboard and it's my best performing in pax. I don't care. I'm winning one side and not really winning on the other side. It's just like you're winning twice, just harder one time. All right, cool. I, I'll take. So once you're getting this data inside with these broad match keywords, what's your analysis here? What are your metrics? What are you looking for? What are you looking to kill? What are you looking to scale? If I stay above a one, after, the initial first purchase, I four x consistently. And so my baseline is as long as the campaign or the account non-brand. Around a one on average will be, will be good. So what do you mean by that? 100% raw? You mean 100% raw? Correct. So if you look at the campaign, sorry, look at the campaign and you take out brand. Oh no, I meant to do this the last 30 days. One 60. That's gonna be a good month. Mm-hmm. that's because again, like all things so this is coming into our off. Yeah, this is coming into our off season. I'm just gonna use clicks cuz Facebook is doing enough button marketing, but that gives us a 3.27 Ros look at this, this 121 K spent, I spent 113 of that and we're at a three x return with a one. with a one x. Now, this client, here's a reason why, and this is why I'm saying, you have to know the conversion path as much as you can. I mean, make it your life's mission to identify that conversion path. You will be. I haven't touched this campaign in three months besides scaling, but what what's interesting is the difference between a first time order and a big order is $12 and 20,000. And so when you look back in here, it's like, okay, how much did I. Spent 3000, make 3000, 4,000 and make 4,000, made 3000, make 3000. And then it's like, all right, now I spent 4,000 and make 2000 and I spent 6,000 to make 34. Yay. And then you go on. This person here was probably a client I got a month and a half ago. Yep. But because we've been scaling up like this here, which is why when you know your conversion path, you're gonna know something really crazy. This is our trajectory. When I said this is like, you know, Hey, John only gets a high spend clients, well this clients start off with a 15 K per month and now we're at 123,000. Yep. And we've never gone over like a 1.5 x. Gotcha. So this is in a year too. Yep. So I always wanna make sure that if it's like, Hey, if I'm gonna say this model works, you can see the same exact structure I've been using since all year. All I did was just keep adding an aspen. Did I try anything different? Nope. What Rose looked like crap, I don't care. It's not about. So what's interesting about this is I've scaled this thing up ruler really well. Here's the best part. These people here, when I have a three Mer watch, my accrual. My accruals at 1.79, which gives me a three by cash. Why? Well Google tracks for 90 days. These people buy every three months for years. Google's gonna miss the recurring transactions cuz it can only go back three months. So when you have L T V that's longer than three months, like, hey, I have people that buy every quarter, every six months. Your Google's gonna always not take those, purchases that you've earned, it can't see them anymore. It's blind to them. Mm-hmm. So that's why the K and L T V model also works. My loss leader is a $12 sample. That's what I sell. I sell $12. But I know the conversion path. I know how long it takes. I know it can scale that. And I know that my risk can look bad because I know that every, look back here, for example, going back to last year when we started this, this is something you always wanna track me, r and it's the same things that we're we're saying all the. 400, three days of following the same exact model and just scaling it up like 10 to 20 grand every month. I have a 1 million in spend and a 2.7 million, and I'm 11% better than before scaling it up 400%. I've come launched this company essentially for them. Mm-hmm. and I meet with them every week and they, now they're just, they're on board. One nice turn looks good to me. I'm like, I've said 20 grand more. All right, I'm like, it's weird. But that's the thing too, is we always wanna look at every single thing you can, but it will scale 15 grand clients and now a hundred grand client in a year. with just knowing that conversion path and knowing that Google's gonna miss it, knowing that RO eyes is fake, looking at returning customers, not everybody's gonna be like this. Again, I understand this, but this was a company that was just right off the street and I just met him and I was like, Hey, all right, so custom made boxes. That's cool. We know this down to key exactly how we learn and where we can measure. I was just gonna say, could you, you were talking about conversion path, so once you, once you're looking at those, those metrics, you've got those conversions coming in. Could you explain to us the conversion path? Like knowing your conversion path, where are you looking at that data? Are you using analytics? Are you interpreting the data analytics? Are you do looking inside Google ads? yes. Um, all of. So top conversion path, which we already looked at last week. Yep. So this is, oh, okay. So you're looking at inside North Beam, I'm looking inside nor Beam, which is a better version of top conversion path. But you can, I, I built top conversion path in Nor Beam. Yep. For them, because I used it so much in analytics. So one thing that I noticed very often is you need to know you're new versus returning. I ask about every large. That came in when it originally came in. Well, who talked to 'em and how long ago was it? I investigated a lot. Mm-hmm. because I'm trying to find my conversion path. So when I say, Hey, it looks like the brand campaign is now at a 6,000 bro as, and there are $90,000 in conversion value. I know that that conversion value. Is coming in from a $45,000 sale on Monday that came in from brand. So I said Perfect. That came into brand at 45 grand, hopped on with Tuesday and said it was a $45,000 sale. It happened, two days ago. Where did that originally, or where did that originally come from? That that came from? Uh, I would guess inbound search. But how did they find out about you? and I investigate the hell, uh, I don't know. I think they came from, um, alright, do you have a CRM tool? Yes. Let's pull it up. And I just start, I go hunting and I find exactly where it came from. This was not even captured by NPI They talked to their, their sales team, and they remember talking to the gal and the gal and it took 11 days for 'em to convert. And she said that she was just searching around on Google and phone us. So what's nice about this is I can't attribute this now to any sort of big conversion capital. What I can tell you is that I have offline conversions that come in through the brand from people calling them veterans, not even being captured by call, track and metrics because they were searching around. So sales team is now trained to tell me exactly when every big sale comes in the history. We have it so much so that when I hop on me and you're like, Hey, John got a big sale. It looked at me 17 days ago. I'm like, perfect. Who's the salesperson? And I just dig. And this is a $15,000 per month client that's now 120 grand. But I got them to beat that cuz I treated them like $120,000 account, not a $15,000 account. So I just wanna make that distinction. This is not high spend. I scale these two. I just wanna make sure that everyone has that, that mindset that we don't discount this too hard just yet. So when you're looking at the convergent path for anything, half of your information is gonna come from the client. That's half of all of your information that you're gonna find for a convergent path. Half of it is gonna come from them. They know a lot about these customers. So, The next thing that I wanna look at though, is I study all conversion paths from anything that I can possibly find. First thing, look to the, top conversion path. Find. The model comparison tool and the top convertor path. How much is coming in from one channel? Then correcting on another channel. Find out what channels in between. Find out if there's, a brand campaign running. If there's,, I kind of need a scenario cuz it's like it gets more general. Cause all of a sudden I'm like, well, that there's six ways that are writing six ways in row. So I'll just use this one. For example, big sales. The next thing I say is, what is your most purchased, item under site? Because what I wanna. what is the, what's the most valuable thing that I can make sure is, running at a high level mailer box, they said sells way more than anything else. People really don't buy the triangle ones. This is usually your subscription services and that kind of stuff is where boxes like this are coming in from. So what then we did is we took this item, replicated it about 50 different ways in the feed to make sure that I could showcase as many different versions of this item as you can, as you can possibly see, because it's one product. So what I did is I said, okay, well we have a lot of different ways that we can actually build this thing. Yeah. It's fully customizable. I said, well, what if we had one that was this, I had a skew for this item here, and that one with a quantity of 900 skew number. Quantity of eight 50. Skew number two, we built it out in a spreadsheet. So now when you go and you type in custom mailer boxes, of course now they don't. Oh, I need a customer. I can't spell. so you got Thousand hundred. Yep. Different skews, different prices, same product. I think that's the only one that we really, that one's like doing 99% of the work. We used to have a different one, but then it got disapproved because it was a pure skew pricing. Anyway, our feet has got like 35 different versions of the same item. Gotcha. So I can also replicate myself in these here, we, we also get way out spend. By the way, we're, we're a small fish in this pond. There's a different one right. And you see how there's different variants in the imagery. So I make this thing look as nice as I can. So now I have a lot of placements. I'll share this with you, what the product list looks like. it's the same product. It's a box. They have one box under site. It's, so look at this. What's funny is our $107 one beats are 70 cent, 77 cent one. That is sweet Remember when I said test everything, like all skews? No one would've predicted that, right? No. Yeah. Okay. So now that we have a bunch of versions of what one product that I could sell, we also made those different nasty groups like we talked about. Now what we're doing is massive amount of equal marketing. What I mean equal marketing is, you're gonna notice something here. Actually, I wanna, I wanna see if anybody can guess. This part here, when I say massive amount of equal marketing focus on the ad spend and you will see that these are equal pushes and pulls and sharing. Here's what I mean. This is the beauty of Google. Brand campaign. We hack right off. We're not gonna be talking about that one here. I have a PAX that's doing YouTube, that's cold outbound. I have a search campaign and a shopping campaign that are doing also inbound and cold. So we have cold. We have cold, and then I have my remarketing here. So what I mean by cold and cold is I'm using less inbound search. And more inbound search here, but you have a 1750. So 1750. Once we know that, that here's 1750, and then we have an eight 50 plus a 900, which equals 1750. So I'm equally pushing as hard on inbound search as I am on all other channels. The reason why is the big leader that I was getting was pmax. The second step in the sequence was Sno Prox, SNO General, and then. So what's interesting is the conversion path was I had high amount of cheap clicks that, got a bunch of new users that also were coming back with non-branded demo search clicks. But I also had a high amount of good, L t V from search that actually was better than Pax, but Pax had better CAC So watch this in Nor Beam. When just using click only cause I really don't care about views. Here's what I was analyzing and here's the reason for the split. Let me go back a little further. Cause at the time when we made this decision was before Black Friday, December, Monday, all the, all the crap that just went happened when went sideways, that screws up all my data. But if I look at just the platform, whatever. So here's some things I was looking at in the last 60 days. My m e r. Bad. Good, good. Okay, so search one on m e r. Let's look at cac. Bad. Good. Good. Okay, so Cack wins. Now let's look at transactions. All right, now this one's good. High amount of learning. Bad. Bad. Okay, so this is equal. These two these two are sharing users, cuz one's broad one's phrase. This one's also gonna be kind of sharing the same ones. Mer, good, bad, good. So now I'm second place here. Transactions more here than more here, than more here than these two are gonna share users. So at all things being equal, I'm gonna. equal. I have a, better e r here, but this is also because these people were coming in by a search after also clicking on pm access and top conversion path. My CAC wasn't wildly enough different for one versus the other. My transactions weren't wildly enough different for one versus the other. My e R wasn't wildly different one versus the other. My transactions weren't wildly different, one versus the other. When you combine these of two and now what is better Search or PVAs? Yes. They get equal 1750. I don't care. One if one's, five or 10 points better because you give another three weeks and the other one's five or 10 points better. So when I told the client, I said, I'm gonna take all of your spend. I'm gonna put an equal between pmax and search. I'm gonna leave this one for remarketing. This one for inbound, this one here I have a multi-prong approach here I have, I'm gonna capture everything I have cold. Cold. Cold. Sharing the same Aspen, and then warm, warm, warm. And then this is the test we just started. Three warm, three cold equal split. Oh, we looked at this through the customer paths. you'll see it at Snow Products brand General Pax. General Pax Brand. Brand. What one I did Gotcha. And we could, I spent, I spent, I got $18,000 off of a person. And which campaign won? Could General have shown up for the same as Pax? Yes. Could Pax show up the same thing as branded terms, or not brand terms, but products? Yes. So these are all equal paths that all point back to me. So really all you're looking at, when you're looking at customer paths inside North Beam or analytics, you're looking for confirmation that your campaigns are all interacting with each other. They will. Yep. They will. But what you don't wanna do is take a look at, like, Andrew as an example. I'm not using this, I'm not saying that this is right or wrong, but what I'm saying is if you're like, Hey, if I'm using this here, if you analyze your conversion path, In nor you're gonna find out that a lot of times you're gonna find some commonality between campaign types. Gotcha, gotcha. Being more equal is going to make sure that you win because by sheer luck of the draw that you have a conversion path that looks like this one here, this would completely change the thought you had yesterday on what should get the ad spend. Gotcha. Which is why conversion path is important. We had a three directs brand, emails direct, and then they ended on pmax. That was five grand for that one guy. Look at this conversion path for 3,600 days. This conversion path here does not match any other client's conversion path, but if I could show up equally, I don't make the wrong. On a person with what they're going to choose to interact with this business. Yep. Conversion path dictates all