Do you feel like your ad spend isn’t adding up? What if the data you’ve been relying on is actually misleading you? In this episode, we uncover the dangers of inaccurate attribution models that could be costing you thousands in missed revenue and show you a smarter way to optimize your ads.
Tier 11’s John Moran breaks down how media buyers can fall into the trap of relying on third-party platforms like Triple Whale, which often misreport up to 30% of revenue. He shares real-life case studies of clients who dramatically improved their ad performance by simply questioning their attribution model and switching to more reliable tracking methods.
If you use the wrong attribution model, you may end up running campaigns based on faulty data—putting your business at risk. Let’s teach you how to avoid these costly pitfalls and ensure your marketing strategy is powered by the most accurate insights available.
In this episode:
06:50 Case study 1: The problem with Triple Whale tagging
20:03 Case study 2: Lowering CPA from $44 to $25 in 10 days
29:56 Practical tips for media buyers to optimize their campaigns
35:52 When should you turn off Meta ads?
40:44 Using Meta’s “Creative Fatigue” feature to optimize ads
45:22 Comparing Meta ads and Google Shopping results
Resources mentioned in the episode:
Previous Ad Lab Sessions with John Moran: https://perpetualtraffic.com/?s=john+moran
Tier 11’s Data Suite: https://www.tiereleven.com/what-we-do/data-suite
Apply to Work With Tier 11: https://www.tiereleven.com/apply-now
Creative Diversification Playbook, Practitioners’ Guide From Meta: https://perpetualtraffic.com/creative-diversification-playbook
Listen to this episode on your favorite podcast channel:
Follow and listen on Apple: https://podcasts.apple.com/us/podcast/perpetual-traffic/id1022441491
Follow and listen on Spotify:
https://open.spotify.com/show/59lhtIWHw1XXsRmT5HBAuK
Subscribe and watch on YouTube:
https://www.youtube.com/@perpetual_traffic?sub_confirmation=1
We appreciate your support!
Visit our website: https://perpetualtraffic.com/
Follow us on X: https://x.com/perpetualtraf
Connect with John Moran:
Connect with Ralph Burns:
- LinkedIn – https://www.linkedin.com/in/ralphburns
- Instagram – https://www.instagram.com/ralphhburns/
- Hire Tier11 – https://www.tiereleven.com/apply-now
Connect with Lauren Petrullo:
- Instagram – https://www.instagram.com/laurenepetrullo/
- LinkedIn – https://www.linkedin.com/in/laurenpetrullo
- Consult Mongoose Media – https://mongoosemedia.us/
Thanks so much for joining us this week. Want to subscribe to Perpetual Traffic?
Have some feedback you’d like to share? Connect with us on iTunes and leave us a review!
READ THE TRANSCRIPT:
3 Mind-Boggling Ways Triple Whale Is Screwing Your Business
00:00:00:00 – 00:00:20:04
John
I wanted some more recent results to share with everyone of how the Andromeda Diversification Playbook came out and said, do it this way as it working extremely well. If you follow the rules, I could drop my cGPA from 44 to 20 5 in 10 days by simply just making more ads. That works. Here’s where a lot of media buyers will get stuck because they haven’t been trained and taught attribution.
00:00:20:04 – 00:00:27:22
John
So what I mean by this is there’s a new feature inside of meta. Okay. Watch this. You ready?
00:00:28:00 – 00:00:55:08
Ralph
Hello and welcome to the Perpetual Traffic Podcast. This is your host, Ralph Burns, founder and CEO of tier 11. Not alongside my amazing co-host Lauren Petrillo, but alongside our most popular guest on Perpetual Traffic. There is no question about it. This guy has downloads for his individual shows and for his videos are absolutely astounding because he brings it every single week and this episode is one of the best ones I think we’ve ever done.
00:00:55:08 – 00:01:15:11
Ralph
On our Friday Lives, we call it the AD Lab. Cook it up. Chaos in the Conversion kitchen. Every Friday, 230 on on YouTube. On our Facebook, you can see it all on all of our socials until 11. But if you can’t catch those, you can, you can obviously get them here, or you can get them on our YouTube channel over to 11.com/youtube.
00:01:15:11 – 00:01:40:13
Ralph
But today’s episode is all about triple. Well I get more questions about triple. Well, now everyone knows that we use Wicked Reports as our attribution. Third party software of choice because we get 99.4% accuracy on click based data. Highly accurate, way better than anything that’s out there right now. We do love the guys over at North Beam.
00:01:40:13 – 00:02:04:01
Ralph
If you’re a very high level like enterprise type client, it’s not typically the types of clients that, we serve because we’re in that that middle to upper range lot of e-commerce, service based businesses, digital products businesses, all of those triple well is probably the worst platform we’ve ever seen because they miss so much data and it’s because it gets blocked.
00:02:04:01 – 00:02:28:01
Ralph
And we go into why this is all the time. And John actually has a case study here today where he shows that nearly a quarter of the revenue, quarter of the revenue, 25%, maybe even 30% of the revenue for this particular business doesn’t even appear inside triple. Well, because of the way that they track and the way that they track is basically they get blocked by all the ad blockers.
00:02:28:01 – 00:03:02:14
Ralph
They get blocked by all the prompts, all the stuff that the tier 11 data suite ends around using edge tags and using our data warehouse, and ultimately piping that clean data through Capi imports, which John talks about here today back into the ad platforms and into our interface inside Wicked Reports, we can actually view everything, right? And today is the first time we’ve actually gone to a triple account and deciphered this and actually compared it with the source of truth inside Shopify, so that in addition to some amazing Andromeda updates, nobody’s talking about any of this Andromeda stuff.
00:03:02:14 – 00:03:30:11
Ralph
I’m still blown away. I think we’re the only ones that are talking about it because it’s so cutting edge. This content diversification strategy, and we’ll leave links in the show notes here for an article are actually the creative diversification playbook. Practitioners Guide that we got from meta. They’re actually telling us to do this. So what John is doing and what he’s teaching, all of our growth strategies and our media buyers inside tier 11 is how to do this.
00:03:30:12 – 00:03:56:09
Ralph
And even meta says this is the new way which you should run ads on meta. And when you spend, you know, a couple hundred million dollars on ads like we do, these are the things that really matter and metrics that matter. Growth at scales, which is what this podcast is all about, what this show is all about. So without further ado, we’re going to get into myself and John Morin talking about Triple Whale and the latest, greatest Andromeda update and Joy.
00:03:56:11 – 00:04:03:11
Ralph
And we are back. Hello, cooking up chaos.
00:04:03:13 – 00:04:06:03
John
Ruining media buyers lives once more.
00:04:06:04 – 00:04:17:00
Ralph
One one media buyer at a time. I used to do it this way. No. Now you need to do it this way. Yeah, to learn something new? No. Well, I guess.
00:04:17:00 – 00:04:32:04
John
I wouldn’t say all media buyers will be buyers that are stuck in their ways. This is even more stuff that we’re going to be throwing on top of things. But it’s a such as digital marketing. It’s always, you know, full time job. So that’s why I try to like even tell our clients too. It’s like, hey, like, your job isn’t to learn my job.
00:04:32:06 – 00:04:39:06
John
Your job is to get better at your job. I will be better at my job. And together you make a good product. I could sell it like, oh, it’s a match made in heaven.
00:04:39:07 – 00:04:40:06
Ralph
Yeah, yeah.
00:04:40:07 – 00:04:47:01
John
Like Ryan Dice always used to say to on my dentists, figure out digital marketing. I want them to have any.
00:04:47:03 – 00:04:47:22
Speaker 3
Oh.
00:04:48:00 – 00:05:10:19
Ralph
My God, it’s so true. We have our one on ones or every other week. One on one right before this session here. So we get to talk about what’s going on. And one of the things that I think this is the type of case where this business owner was trying to learn attribution, trying to learn marketing, as opposed to figuring out new products, servicing his customers.
00:05:10:21 – 00:05:27:20
Ralph
And then all of a sudden you come in and turn everything upside down and say, hey, if you look over here, it’s actually doing better than you think because you’re trying to learn and you’re going through a faulty attribution model. And he was using our old friends over at Triple Oil.
00:05:27:22 – 00:05:29:07
Speaker 3
Yeah, yeah.
00:05:29:07 – 00:05:33:00
Ralph
As opposed to looking inside the platform itself.
00:05:33:02 – 00:05:36:02
John
Yeah. Or like the inside of the platform with like Shopify. Right. Exactly.
00:05:36:03 – 00:05:55:04
Ralph
Like, yeah. Like Shopify, like where the money is like that’s the source of truth. Yeah. Right. You’re right. Always with them. Yeah. Your bank account will always when you’re a back end will always win. Like that’s actually where you’re making your money. That’s where people are buying your stuff. Yeah. It’s they’re not buying it through your attribution platform that doesn’t track anywhere from 20 to 40%.
00:05:55:09 – 00:06:21:20
Ralph
That’s what we’ve seen FBA for include we’ve got over on our website, if you look up to your 11 data suite, there is a graphic in there that shows the difference between data suite capturing data and what you would typically see inside Ga4. Not only is Ga4 missing 40% of the conversions because of this unknowns and directs, they’re missing another 50% of conversions because they’re not even capturing any of that.
00:06:21:22 – 00:06:47:01
Ralph
So it’s if you’re not looking at the source of truth, you’re not really looking at the right metrics in order to grow that business. And I know this is a theme that we’ve said many times here, metrics that matter, growth that scales. But show for the people at home exactly what we’re talking about here, and a real life example of how you got this dentist to start doing more dentistry, so to speak.
00:06:47:03 – 00:07:11:13
John
Pretty much all of us. Exactly right, except for different type of doctor. But yeah, it’s not. But doctor, this evolved as a client that came to me for help on Google and while we’re modifying the account for Google, most of the time the situation occurs is when a client or customer or whomever says, hey, I need to grow and I’m going to use Google to grow and treat meta and app love.
00:07:11:13 – 00:07:37:07
John
And axon, which is app loving, is now axon. I’m going to use the channels against each other. So the opposite of what was developed to do. And so when we looked at everything and I said, hey, we’ll grow Google and we’ll scale meta, but we don’t want to necessarily have a lot of attribution that interrupts what we know to be true based on what we as digital marketers expect to happen.
00:07:37:07 – 00:07:56:07
John
So what this means is that he was looking necessarily at his trip oil account for the health of his business, and I understand why, because it has all sources even of PostScript made a sale, etc. and yeah, it has a lot of information here. And that tracks all the spend tracks, all the revenue and new customers and returning customers and all the other good stuff.
00:07:56:07 – 00:08:28:12
John
The problem, though, is the way that Trip Oil tags is actually the same methodology that Ga4 and GTM will have an issue with. For those that don’t know, and this might be shocking to some, probably shocking to all in Ga4 when a user does not grant consent in Google Analytics, or which means they do not accept all cookies, that is on the little consent mode pop up, they are not counted as a user for a session based metrics, the same as consented users that basically they use model data to fill gaps.
00:08:28:12 – 00:08:48:20
John
Interestingly enough, Google Tag Manager also does the same thing. It will not count users who deny consent. So about your Google Ads, or think about anything that you’ve deployed inside of a Google Tag manager might be a triple Google tag as an example, the information or the tag will not necessarily count or track the people if they did not ask for consent.
00:08:48:20 – 00:09:10:05
John
Now for the tier 11 data suite. Obviously we use edge tagging, which doesn’t have that issue, which is why a lot of that data is a lot more more useful and accurate. But also, if you skip attribution and just say, I spent money on campaign A, and what happened to my total top line sales? And then I can measure was that a good effect overall to my business?
00:09:10:05 – 00:09:26:18
John
So just skipping attribution, and I was looking at the back end of the website when we’re making these changes and I’m like, it doesn’t look like trip oil is actually accurate at all. Because as an example, when we scaled last week from 23 to 29, we push 47% more spent into meta went from 46 grand to 67 grand.
00:09:26:18 – 00:09:31:22
John
So basically $21,000 more in spend app loving.
00:09:32:03 – 00:09:33:10
Ralph
Is that never accurate?
00:09:33:14 – 00:09:37:22
John
I’m just that number is accurate. Yes, the spend is 100% I know, I know triple well.
00:09:37:22 – 00:09:38:16
Ralph
I have no idea.
00:09:38:17 – 00:09:57:13
John
So we look at that. That was accurate. That’s I’m loving spent seven grand more and then Google stayed about to say 0.6. Now when you start to look at all of the metrics, when you see a whale tail here, that’s attributed, when you see a non whale tail, that’s not attributed, basically just means all the data in. And so we’re looking through the metrics sense like Roas is down 10%, 9%.
00:09:57:13 – 00:10:19:23
John
If you don’t contribution all purchases are only up 13. The attributed purchases up 13. Not in terms of 19. And it’s like hey, your customer value. Basically, if you’re looking at Roas attributed somehow says you made 177 non attributed says you, you went from 181 to 205. So like 13% increase in all of customer value. New and returning does not matter.
00:10:20:00 – 00:10:41:08
John
Yeah. So he’s looking at this and saying hop this is poor performance where you have 24% more spend and only 13% more revenue right now. See that 205 number? I said it doesn’t seem accurate. Why don’t you install lifetime long lifetime is like a very good tool. It will just tell you the results of all of your marketing and your health of your business.
00:10:41:12 – 00:10:42:20
John
Yeah. So when we installed.
00:10:42:20 – 00:10:44:02
Ralph
Directly from Shopify.
00:10:44:06 – 00:11:10:23
John
Exactly. Pulled from your actual customer data. So we saw a 24% increase in spend. And then lifetime during that time period saw a 25% increase in net sales. So this number is 275,000 triple. Well thought it was a 205 on the high side. So we missed about $70,000 this week or two thirds of all activity basically. About yeah, about two thirds of the increase.
00:11:10:23 – 00:11:13:12
John
So that’s why I’m like, I’m wondering where this is. About a quarter.
00:11:13:12 – 00:11:15:14
Ralph
Of the spend. That’s about a quarter of the revenue.
00:11:15:19 – 00:11:34:12
John
Yeah. It’s like this one said that it lifted up 23 grand, but it was actually fairly high. It was more like about 55,000. So when you look at this and say, okay, well let’s just see, was that healthy? We know that we pushed a bunch of money into meta. We spent 47% more, which was a 25% increase. Now Google didn’t do anything.
00:11:34:12 – 00:11:54:03
John
App loving is just gonna also rise with it because it does mostly remarketing anyway. So a 25% increase in spend, 25% increase in net sales, blended down 3% more, up a half a point. So we’re looking at this thing. All right. So we increased it by 25%. And that is up 25%. Net profit. The contribution margins up 41%.
00:11:54:03 – 00:12:13:06
John
So a clear to 30 grand. So not to mention that’s the profit that’s triple said he lost 23 grand. So don’t know how that happened. But apparently it says they lost 23 grand when they actually made 29. It’s up $9,000 additional. So we’re looking at that. And then we say okay units for orders stayed or stayed. Orders are up 27%.
00:12:13:08 – 00:12:32:11
John
Refunds are down 6%. So that’s fine. The percentage of marketing was flat. The blended cakes better, the merger is better and the marketing cost one of 25%. So like amazing 25%, marketing cost 25% more sales. But at a kakak that we like at a year that we like gave us a profit margin increase of 41. Yay! So this is very simplistic.
00:12:32:11 – 00:12:37:07
John
We push 25% there and we got 25%. Oh wait a second, wait a second.
00:12:37:09 – 00:12:42:21
Ralph
But yeah. Well then. It’s media mix modeling modeling being the last em.
00:12:43:02 – 00:12:43:20
John
Right.
00:12:43:22 – 00:12:54:16
Ralph
But this is not modeling. You’re not modeling anything. You’re actually pulling real data from the back end where people are actually buying. So that by virtue of this is really not on that.
00:12:54:18 – 00:13:04:06
John
It’s almost like me. It’s like media mix interpretation. Yeah. We’re pretty sure it came from meta. Now nothing supports that. Instead of drip oil, Roas went down 85%, says.
00:13:04:06 – 00:13:06:06
Ralph
It’s meta, is doing terrible.
00:13:06:08 – 00:13:28:15
John
All right. Yeah. So we put 47% more spending on Roas. Went down 11 oh oh. Like it’s just not true at all. And so that’s what’s interesting is like when you skip the attribution, I was play around with this earlier and I thought this was really funny. Watch this. You know what? I’m, I think I can copy and paste this into a new window, and I don’t even know how this happens, but somehow changing the attribution model changes the totals.
00:13:28:15 – 00:13:30:13
John
So I’m like, wait a minute.
00:13:30:15 – 00:13:34:09
Speaker 3
Oh, that that doesn’t make you smile. Changes the.
00:13:34:09 – 00:13:36:14
Ralph
Totals. Okay. Right. It’s not.
00:13:36:14 – 00:13:38:01
Speaker 3
Possible.
00:13:38:03 – 00:13:55:12
John
All sales change. But the attribution model when you’re talking about all sources, even organic. So I was looking through this thing and I said, wait a minute, let me, let me run a quick model here. So 23 to 29 compare previous period 23 to 29. Compare previous period ones. First click once last click. So I’m like I still have 205 and I still have 205.
00:13:55:12 – 00:14:17:09
John
Here. Like customer values is 177. Unless you go last click then the customer values 294 is 294 from a last click perspective. Which is really cool because somehow it went from 177 to 294 and we still only have 275. So it either undershoot it by 100 K or like overshoots it by 40 K, just depending upon what attribution model that you so like.
00:14:17:11 – 00:14:38:15
John
Thanks. That was fantastic. However, the total customer value is still 205. So for those out there that love to stick with an attribution model, not change it. If you’re looking at last click, you made 90 grand more than your totals, which is really interesting. Or you made basically like 25 grand less than your totals. Pick your poison. Either one of them is where early?
00:14:38:15 – 00:14:56:03
John
Fricking far off. And I also like the fact that my customer value actually went up 20% and overshot by totals. But on a first click basis, the totals were only up 11%. So you get different percentile increases and then also undershot or we overshot. So where we basically just removed everything and said let’s just use your bank account.
00:14:56:03 – 00:15:11:04
John
Remember how we’ve been teaching this for years of Ncac and Mirr and contribution margin, same stuff. It was very interesting to see how when the tag misfires and they model it incorrectly and you make very expensive decisions off of this. How wrong it really can be.
00:15:11:04 – 00:15:17:08
Ralph
And this is not an isolated case with triple L as it you probably have more experience than I do.
00:15:17:11 – 00:15:17:16
John
Yeah.
00:15:17:16 – 00:15:27:23
Ralph
I’ve been I’ve seen enough triple L accounts to know exactly like see this scenario. But you see this consistently. Yeah. You’re talking about you’re missing one quarter of their revenue.
00:15:28:01 – 00:15:32:11
John
Yeah. It’s like you can’t just fuck off 75 grand in seven day period.
00:15:32:13 – 00:15:33:09
Speaker 3
Right.
00:15:33:11 – 00:15:33:23
John
That’s a great.
00:15:33:23 – 00:15:40:11
Ralph
Yes. I mean it’s the difference between a probably a 30 K loss and a 30 K gain and contribution margin.
00:15:40:13 – 00:16:02:14
John
Right. Yeah. I mean it’s huge. And depending upon which attribution model you’re going to be overshot or undershot, but then you’re making expensive decisions off of it. So. Right. We’re probably gonna get rid of triple L pretty quickly for this client here and move into basically just live timely. But the M modeling when you’re looking at I push here this and here’s what’s funny is check this out the week before this when he says, hey, push on Google.
00:16:02:16 – 00:16:20:16
John
I’ll show you the push on Google. Watch this. If we look at the lifetime timely here and we’re running performance max with non first click contributed, which what is going to happen with Google when you spike up Google. Well if you’re using non first click non KP imports. So when this one went up 40% do we had a flat.
00:16:20:16 – 00:16:39:20
John
Basically spend the lifetime Lea where you have a flat return. So this one pushed up by eight grand. But the reduction of that two grand here was not made up by the increase in this money here. So meta pull back basically ten grand and we added nine. Okay. What was supposed to happen at the time is that I’m going to increase Google.
00:16:39:20 – 00:16:56:23
John
You increase meta and I increase Google. They didn’t increase meta. I’m like okay, so our spend is the same and our revenue is the same. CAK is the same versus the same. So it look like it looks like Google and meta are doing well. It’s just when you push one without the other, you don’t really see much result.
00:16:57:01 – 00:17:17:20
John
So when I took it from 21 up to 29 that next week, I stayed at my level of 29 or like 2930. I basically stayed at that level and then they increased meta. So we basically just increased Google. That increased meta. And now our lifetime after those two scales, now we’ve upped, leveled and we went into the same we’ve increased all of our sales by 25%.
00:17:17:22 – 00:17:33:15
John
So it’s really interesting that Google when you take Google up and you push meta down, nothing happens. But if you push meta back up, This is good. So if we were to and this is I don’t fault the guy, this is new to him. He was very nervous. I fed him an email and I said hello from 30,000ft.
00:17:33:15 – 00:17:36:18
John
You you messed up, man. He responds with, I know I came.
00:17:36:20 – 00:17:37:01
Speaker 3
I.
00:17:37:01 – 00:17:54:15
John
Was funny, we have a good relationship. He’s like, look good. I came to, I was like, okay, I need you to do this now. So now that I’ve won him over in the modeling kind of thought process, which is and Kojima that sales contribution margin compare that to your spend. Don’t look at attribution, especially from a third party platform that’s missing to one third of the data.
00:17:54:17 – 00:18:04:01
John
Now this is actually become really easy now. Now life is easy. All we had to do is just stop confusing ourself and stop paying for the confusion that it has been very, very beneficial.
00:18:04:03 – 00:18:10:12
Ralph
You don’t really even need lifetime to get that result because you could have just gone back to Shopify and just like pull the results from there.
00:18:10:17 – 00:18:12:04
Speaker 3
But yeah, I was going to say.
00:18:12:04 – 00:18:41:15
John
If I just look at, I look at here, this is the same time period 2013, I compare a previous period, 25% more sales like lifetime, says the new customer, returning customer, it went down and negative new customers orders are up, traffic is up 17% with a better conversion rate. And we see Facebook, YouTube, Instagram, like YouTube took a whole bunch of credit even though we didn’t touch YouTube because it’s a warm network you push on meta YouTube natural looks better, which is why I tell people be cautious of demand when it gets drug around by the nose from other platforms, but basically everything is up for except for Google search.
00:18:41:15 – 00:18:55:18
John
We already scale up the week before, so that’s what’s nice, is we can see through here. This is why I’m looking at his. He scaled. I look at this first. I’m like, this looks good. I happened to trip well and so like everything’s terrible. Like, well the spend I know is accurate. The revenue I know is accurate. The only thing that’s saying things are bad is Triple-A.
00:18:55:19 – 00:19:08:21
John
I’m like, let’s not look there. Let’s get lifetime late, get it installed, go retroactive, see what actually happened. And we may find out that third party attribution that is poor is not necessarily even worth our time.
00:19:08:23 – 00:19:10:23
Ralph
Why would anyone ever use triple?
00:19:10:23 – 00:19:32:19
John
Well, I don’t know if they believe it is. Hype is believed a lot. I do believe that when you’re running heavy amounts of meta, your third party attribution, if it’s going to take a third party approach and not have an actual baseline like ignore, the meta says happy to ignore the bank accounting or like life timeline or actual new customers and only track two thirds of things.
00:19:32:21 – 00:19:54:11
John
You’re tracking two thirds of a network that has 95% of its activity. Not even click based attribution at that point is like a dartboard. I could have sold ten people, or one, just depending on if I think the people that saw the ad converted anyway. So it’s like trusting view and gauge conversions. It’s like it’s directionally accurate, but you don’t want too much of that triple just underreport that.
00:19:54:13 – 00:19:58:03
Ralph
Yeah, you want the click and you want the actual click being captured.
00:19:58:05 – 00:20:05:21
John
Yeah, yeah. And or you take it like directionally, but you let your, your bank account dictate, I want to move on to something else before I get to Q&A here in ten minutes.
00:20:06:02 – 00:20:23:04
Ralph
That was great, by the way, because we’ve never really talked about, well, all that much. And it shows just all the holes that are in it. Yeah, of course we need our help. We can get our help over to 11.com/apply. Definitely check out our data suite that’s on tier 11 as well. Helps angels here. So anyway all right on to the next case study.
00:20:23:06 – 00:20:25:12
John
Mitch is here. He says I’m here. You can start also.
00:20:25:12 – 00:20:28:10
Ralph
Thank God Mitch is here.
00:20:28:12 – 00:20:28:23
Speaker 3
Hey, couple.
00:20:28:23 – 00:20:32:11
Ralph
Well, sucks. I love Mitch even more.
00:20:32:16 – 00:20:39:01
John
I optimize for view only and brand terms only on Google, says Mitch. She goes, just kidding, just kidding. I’m like, I do.
00:20:39:01 – 00:20:41:04
Speaker 3
Do. I’m like, look at all.
00:20:41:04 – 00:20:42:14
John
The million dollars that you made up.
00:20:42:16 – 00:20:46:11
Ralph
Yeah, there’s a view that means it’s coming from that platform.
00:20:46:13 – 00:20:56:07
John
First, I wanted to touch upon the creative diversification. We bet following the model that we’ve been using. And this is actually the model here used in this before. Yeah.
00:20:56:07 – 00:21:01:16
Ralph
This is an actual meta deck we’re showing like meta stuff. Yeah. This isn’t a hack John.
00:21:01:18 – 00:21:17:02
John
Is a practitioner guide like Andromeda came out in December of 2024. For those still not in the know or living under a digital marketing rock, but this is what they have producing. So here’s an example. And I won’t go through this again because we’ve already gone through this, but we’ve been following the fact that.
00:21:17:02 – 00:21:19:02
Ralph
The hammer this into people’s heads. Yeah, really.
00:21:19:02 – 00:21:19:06
John
Do.
00:21:19:06 – 00:21:22:04
Ralph
But a little watching on this channel. But it’s still people.
00:21:22:04 – 00:21:23:16
John
That are actually working at tier 11.
00:21:23:16 – 00:21:26:10
Ralph
Do some of the work. Go ahead, keep the hammer.
00:21:26:13 – 00:21:45:05
John
Hammer away five insight driven and five insights driven. Net new concepts for a month basically means like what’s your hook pain point offer? Like what is your marketing plan for this product or service? Create at least two, three, 4 or 5 would be great and that whatever starts to win when the traffic and audiences start to resonate with it, leads to sales and clicks and yeah, make better versions of the winners.
00:21:45:07 – 00:21:58:23
John
And so we said, okay, let’s practice what we preach, because you all seen me say this before, but now we’re actually going to show you the actuality. So back on September 20th, we said, hey, let’s scale up. So we added $3,500 went from 9600 to 13 grand in.
00:21:59:02 – 00:22:03:21
Ralph
Just so we’re clear, you are basically doing what meta recommends you to do.
00:22:03:23 – 00:22:23:00
John
Yeah. Okay. Crazy enough that the Andromeda model makes so much sense when you look at the literature. Absolutely. It’s basically it says, here’s all the people inside of meta. Here’s your ads. We’re going to find out what ads works to the right audiences, and we’ll deliver your ads correctly to the users. We’re like, okay, like I understand that.
00:22:23:00 – 00:22:25:17
John
Let’s and says do it this way. So we said okay.
00:22:25:20 – 00:22:28:14
Ralph
In those beautiful Nvidia chips. Right.
00:22:28:19 – 00:22:42:12
John
And the 6000 to the marketing agencies, there’s like two that are running Andromeda and it’s us and mile Radius and a part of the flight that I have personally. So they may as well be the first one to listen to Google and find out, or listen to me to find out what happens.
00:22:42:12 – 00:22:58:04
Ralph
I forget the last time we were together, but I was at a meta conference in New York and nobody I talked to everybody like every agency I talked to, like they’re not doing any of this stuff. Now, I was out with a dozen people, like including my rep, who’s brand new, like he came from TikTok. He’s a nice guy, but he’s still learning.
00:22:58:04 – 00:23:01:04
Ralph
It’s like, what do you think about the Andromeda update? His work. What?
00:23:01:06 – 00:23:02:11
John
What? It’s insane.
00:23:02:13 – 00:23:03:09
Ralph
Yeah. Like say.
00:23:03:10 – 00:23:15:13
John
No. We’re so far in in the weeds and deep into this, but I don’t know how you call yourself a professional. Not do that. It’s like you know me. Be on me. It’s like, hey, I need to write a book. I’m like, okay, hold on. I think it’s like, you know, they invented pens, right?
00:23:15:15 – 00:23:17:22
Ralph
So my way of doing it. Yeah. All right, keep going.
00:23:18:01 – 00:23:37:12
John
So we increase the ad spend 36%. We saw some really hit important diminishing return calls for result went up 61% and our results went down 15%. We said, okay, what do we do? Well, we have to iterate our creative. And so this is on September 20th. It’s 23 or 20 second when we push, we said, okay, this isn’t working.
00:23:37:12 – 00:23:40:20
Ralph
And just so clear. This is not an e-commerce client, correct?
00:23:40:22 – 00:24:03:18
John
It is. It’s a long sale cycle. They’re college courses. So you could do it lead gen. You could do lead flow. You could do e-commerce everything. So we’re basically optimizing for per click through checkout, because this account has been tracking the wrong metric for over two years and over $1 million in ad spend. So because of a long sales cycle, we basically counted a micro conversion first so that we can retrain who these actual people are.
00:24:03:18 – 00:24:21:11
John
We’re backfilling it from AOS so that it actually appends to the per click through checkout custom events through Capi. And so we’re basically importing first click copy imports. But just on a previous touchpoint simply because their tech stack, even AOS is like we can’t see the conversions, neither can no one can see the conversion. It goes through like three different websites and this loses.
00:24:21:11 – 00:24:24:20
John
Everything’s got it. We’re going to fix that. But anyway, we’re just basically counting the point where they begin.
00:24:24:20 – 00:24:27:13
Ralph
Checkout plus edge tag on this edge tag.
00:24:27:13 – 00:24:31:20
John
First click checkouts rather than edge tag. First click purchases. Exactly right. Correct.
00:24:31:21 – 00:24:32:09
Ralph
Got it.
00:24:32:11 – 00:24:49:04
John
Yep. And so we saw softening in this performance. And we basically just said well let’s follow the model that unless do we iterate our five new piece. So creative. And so that was on the 28th or the 22nd. And so after we uploaded our new creative on the 22nd and we look at the 23rd, 2420 faith and we see what happens.
00:24:49:04 – 00:24:59:10
John
So we kept our spend or that kind of spend we that was the next thing we did. We increased our spend. Again. We said, okay, let’s try to make a push. Now that we have new creative. Yeah. And that’s where we saw it’s very cool.
00:24:59:10 – 00:25:03:01
Ralph
There’s three day increments every day in three day increments. Got it. Keep going.
00:25:03:06 – 00:25:29:18
John
Yep. So now in the 23rd, 2420 fifth our CPA is down ten bucks went from 45 down to 34. And remember we were at 30 and shot up to 45. And now we’re at 45 down to 34. And then we increased our conversions by 62% with 25% or 23% outspend. We said, okay, let’s just follow this along. So the next week, fast forward to October 1st, we said, okay, let’s do some more creative iterations.
00:25:29:18 – 00:25:47:06
John
We updated all of our creative on the 30th. I don’t have the next three days since we only have the first two days after it. Okay, we’ve been looking at three day increments 23, 24, 25, three to increase 20, 2122 I only have the first second. I’ll have the third. But once I have the third, then that spend will normalize.
00:25:47:10 – 00:26:09:17
John
But it looks like we’re down 16%. We’re good today we’re back up to normal ranges. I just throw up all the imports just yet. But long story short, in the two days our CPA is now down to 25 better than it was the week and a half ago before we started the creative iteration process. And now we’re seeing a 10% increase in results and a 24% decrease in cost for results for 16% less spend.
00:26:09:19 – 00:26:31:09
John
So we’re back down now to $25 where we started at 28. So now we’re actually better since the 20th to the first. So that was 11 day period. We did three rounds of iterations and we dropped our CPA, still $3 cheaper than where we were before with an increase in ad spend and all we’re doing. And we look at the the view history here.
00:26:31:09 – 00:26:53:17
John
Let me go to the 30th so you can see this is when we are that’s all started delivery of all of our ads. We basically just updated all of our creative on the 30th. You can see here we paused the whole bunch and we added new iterations. Cool about this. As you can see, like from the 30th to the second quarter, a previous period where we said, all right, let’s start adding in new ads.
00:26:53:19 – 00:27:16:14
John
These are all starting to be reduced. Some are still working, some are having reduced. And you can see all of this negative here. -63. -69. -70. Negative. You can almost see it pausing these things for you. My spend is 0.07% difference. But these are all down 60 percentile. And then this one out all of a sudden just sucked up eight G’s immediately in the first two days and the same.
00:27:16:16 – 00:27:38:21
John
And then look at this thing. It is giving us 221 conversions at a 37, which is still $10 higher, but it has brought down the overall $16. So people are like, well, what happens if an ad sucks up the Aspen? It’s like, that’s an old that’s a on CBO. That’s you in control. You’re no longer in control anymore matters in control.
00:27:38:21 – 00:27:47:12
John
So if meta says I’m going to do it, it works and you’ll see a better result. The media buyer, for some reason, thinks that to happen. And like pause this one. It’s like, do not touch it.
00:27:47:14 – 00:27:49:05
Ralph
That would be the first reaction.
00:27:49:06 – 00:27:50:16
John
Exactly like a.
00:27:50:16 – 00:28:03:06
Ralph
Second of I’m looking at an app. I mean, it’s at the CPA or I guess we’re talking about Ncac here, obviously, but CPA, it’s not 37, it’s the total for that. I thought it was like 47 for your top point, for your top. UGC AD.
00:28:03:07 – 00:28:04:04
John
Oh 37.
00:28:04:04 – 00:28:11:05
Ralph
Yeah. Oh, it is 37 okay. Yeah, but I’ll show you why should I not turn that off? What’s the rationale for that?
00:28:11:07 – 00:28:28:18
John
Yeah. So this one is thinking about what has a high amount of ad spend when you have 8000 out of $14,000, what do we think this ad is doing to the cold traffic? Imagine we put an ad in there and it’s going to go and go out to the cold traffic and is starting the users on their journey.
00:28:28:19 – 00:28:37:18
John
It has a highest amount of ad spend and it has a high amount of engagement. But without having the frequency be crazy high. So you have a high amount of engagement.
00:28:37:20 – 00:28:56:06
Ralph
It’s now it’s engaging in the market. It’s expanding the market, it’s increasing awareness in the market. It’s getting some conversions, but it’s sucking up the ad spend for a specific reason. Because Facebook says or meta says that’s where they are in the journey. They need to see this piece of content in order to continue down the journey and ultimately make the purchase.
00:28:56:09 – 00:29:06:20
John
And we see the rest of the as here doing what? So the CPA is getting better, and we even have 20% more, 14% more. One 4357 100 how they’re using.
00:29:06:20 – 00:29:09:02
Ralph
All of these other ads, really. What’s happening?
00:29:09:08 – 00:29:27:09
John
Yeah. Now this is down checkouts for the CPA better. This is down checkouts with the CPA is better. This is down checkouts with the CPA is better. So it seems to be reshuffling the deck and the ads that they have more information on because they’ve been running longer are extremely smart because they have had longer time period.
00:29:27:09 – 00:29:44:09
John
So when we say, hey, we need to see this new ad start to reengage people. Why? Well, the week before, when we were looking at the top performing pieces of creative that had the highest amount of ad spend, it was guess what, a UGC. So what do you think we did? We went and created another, more UGC.
00:29:44:11 – 00:29:44:22
Ralph
That’s what.
00:29:44:22 – 00:29:45:17
John
Works and.
00:29:45:17 – 00:29:54:00
Ralph
That’s what’s resonating with this market. And one of the ten ad types, you just never really know which one is working. Is there a founder story in here too or. No.
00:29:54:03 – 00:30:10:05
John
Yeah, it’s we have a founder story in here as well. But the here’s what’s awesome is here’s where a lot of media buyers will get stuck because they haven’t been trained and taught attribution. And this is where if you learn attribution, everything else becomes so much easier. So don’t skip the hard part because you’re making the easy part hard.
00:30:10:06 – 00:30:16:17
John
So what I mean by this is when we are importing first click copy imports. What’s Meadow’s attribution model.
00:30:16:17 – 00:30:18:00
Ralph
Last click.
00:30:18:02 – 00:30:28:11
John
Bingo. So you’re saying, John, I need this thing to spend a lot. I need to get the first click. They need to see nothing else. Then they have to click this. And then I will pause it. Like you’re asking for the impossibility that.
00:30:28:17 – 00:30:38:05
Ralph
Everybody is going to see my ad and click the very first time. And then by John. All right. That’s no humans work things. That always happens. Never.
00:30:38:11 – 00:30:54:05
John
Yeah a time lag sale cycle is often we just forgot. So that’s what’s interesting is like if you haven’t had this resonate with the marketplace and you say no, they have to click that. Not only that, not see anything else, show up. Everything else. La la la la. That’s what I used to do, right? Yes. That’s why you’re not teaching these.
00:30:54:05 – 00:31:12:02
John
This is why we are. This is the examples that we’re going through. So it is very interesting to see how the Meta Andromeda engine is working. But if you follow the rules, I could drop my CPA from 44 to 20 5 in 10 days by simply just making more ads that work. Now, let me let me jump back into Microsoft Paint here.
00:31:12:02 – 00:31:13:20
John
They come, a graphic designer again for a moment.
00:31:13:22 – 00:31:27:18
Ralph
Well, just to your point, like you’re looking I want to get back to that one. But what you’re really looking at is the bottom line. Cost per checkout, which was $27. You don’t really care about any of the other ones. That’s your universal Ncac from all your assets.
00:31:27:20 – 00:31:38:01
John
Exactly. This is the total assets. All you put combined. Exactly. And here’s why. So if you develop an ad here, it says, great, I’m gonna take that. And I’m going to show graphic right.
00:31:38:06 – 00:31:40:02
Ralph
Shows this art skills.
00:31:40:04 – 00:31:46:04
John
Is very good. I spent six months at art school, in the parking lot, picking up my wife for two at art school.
00:31:48:22 – 00:31:51:13
Ralph
So I know there’s somewhere.
00:31:51:15 – 00:31:52:07
Speaker 3
So you develop.
00:31:52:07 – 00:32:09:10
John
An ad, it takes the ad and shows it to the people and says, hey, you know what? Out of all ten, we like half of these. So these ads here are good. And these ads over here are bad, whatever it is. So they choose the ads that are working with and then they say, okay, great. The next day they basically take those ads and they show it to them again and say, hey, they still really like those.
00:32:09:10 – 00:32:25:23
John
Now on day three, are they going to say, take those same ads and show it to the users? Again, Meta’s going to say, no, sorry, we’re cutting you off their stop. We’re not going to beat the crap out of these people. Well, what has happened so far? Well, we tested a bunch of ads and, you know, they resonated with they resonated with the fact that this has college credits that will transfer to any university.
00:32:25:23 – 00:32:31:22
John
And the next week it was college credits that transfer to any university. And the next day is college credits are transferred to the next university.
00:32:32:03 – 00:32:33:09
Speaker 3
And the next day is college credits.
00:32:33:09 – 00:32:36:22
Ralph
I know that stuff. Banner? Banner blind already? Exactly.
00:32:36:22 – 00:32:44:06
John
So what do we have to do? We have to use a new iteration. We have to use a different variation of these. What else can you tell them? You already told them that.
00:32:44:07 – 00:33:00:18
Ralph
What I told them, right? They didn’t convert on that message. However, in the back of their mind, they maybe remember it, maybe they don’t. But then meta says, let’s show you another ad with a slightly different message, because we didn’t get the result by showing you that ad that we wanted. So let’s show you another ad.
00:33:00:20 – 00:33:31:06
John
Exactly. And there’s always traffic flowing in and flowing out of the marketplace. So there’s always flowing and flowing out. And what you’re trying to do is identify your pain points, your hooks, your offers that are resonating with the marketplace. So this can become more evergreen. Once it’s evergreen, you’re like my Vermaak love company. I’ve been running the same as since May 14th, and I have 129 active ads, and all of them have under $30 CPA, and I’m always spending 250 bucks a day because we’ve kept iterating our process to what we see working well with the audience.
00:33:31:06 – 00:33:49:12
John
We said, let’s make a new version, make a new version, make a new version. And if you go all the way back here since May of 2025, you were literally going back May, June, July, August, September into October. And we look at my one and drop on a campaign and we look at my one ad set and we look at my ads that are descending my spend.
00:33:49:14 – 00:34:01:04
John
I have now spent $23,000, not a whole lot because I’m spending 20 or 250 bucks a day, but I have 130 ads. And look at the CPA. They’re just beautiful, like $25 across the board.
00:34:01:04 – 00:34:06:14
Ralph
Everything falls $125 product 125 Novi. In this case.
00:34:06:16 – 00:34:25:12
John
Yeah, a $9,000 increase and has been year over year was a $220,000 top line increase. So this all we basically did is find the 130 different accolades about our product that everyone loves. And we’re just like telling the world, or we don’t have enough inventory to increase the spend because we’re trying to keep up with demand from like plateaued.
00:34:25:14 – 00:34:43:02
John
But again, that is a creative, iterative process. I’ve been following it since for last five months. I tested the shit out of it. It works. I wanted some more recent results to share with everyone of how the Andromeda Diversification Playbook came out and said, do it this way. I said, okay, I’ll do it this way because no one else is testing this and it’s working extremely well.
00:34:43:04 – 00:34:59:16
John
When I say that media buying is tactical, media buying has been snuffed in the creative media buying has been buffed. Like that’s what we’re talking about. So when you’re like, oh, this isn’t working, think of creative first. What are your concepts hook pain points offers? Are you having good nomenclature? Are you reacting? What is positive and are you iterating that process?
00:34:59:16 – 00:35:02:08
John
If the answer’s no, you’re not using Andromeda.
00:35:02:10 – 00:35:15:18
Ralph
Yeah, snuffed and buffed. I love it now on either one of those out accounts, which we just looked at some point you turn the ads off. So I got I know you had a couple of that were off.
00:35:15:20 – 00:35:34:20
John
Right. That was not our choice. Unfortunately. We have like this one here. That person is no longer with the company. That person is no longer worth the cost. Oh I see, yeah. So we did do that one just because that one we they just didn’t like it. But all the other ones we have are still on audit except for anything that see how that was a person in the field I sell.
00:35:35:02 – 00:35:49:11
John
But that person in the field is actually the same one as this one here. That was also office. The same like person of the field is starting with it. It’s just basically they’re like, hey, we’re not allowed to use this anyone, we let them go. We can’t use our likelihood. And so we said, okay. So situationally we had it turned off.
00:35:49:11 – 00:35:50:17
John
But they had good performance.
00:35:50:19 – 00:35:59:14
Ralph
Makes sense. So in what scenario do you ever shut anything off that. Well I know I’ve asked this question before, but this is the question that keeps coming up.
00:35:59:16 – 00:36:20:23
John
I it was funny, as everyone knows now, that there is a 50 ad limit inside of Andromeda. Yes. Now, if you were following me back then, you are not included in that right now. If we look at the active ads because we’re also iterating. I have 79 active ads in one ad set. So I if you are if you started back then I have more accounts that don’t have that 50.
00:36:20:23 – 00:36:37:13
John
Then I do have the 50 restriction. Because if we’re working together back then or if we were using Andromeda when we’re started teaching it back in February, March, there was no restriction. So all those users were grandfathered in. So I can use as many as I want. Everyone else is stuck at 50 and it’s not 50 active or 50.
00:36:37:13 – 00:36:42:05
John
Pause. You can only have total, so you have to start deleting. Oh really? Really.
00:36:42:09 – 00:36:43:02
Ralph
Yeah.
00:36:43:04 – 00:37:01:17
John
So what we’ve been basically doing is after we start a five iteration process or ten iteration process, however many winners there are that we need to iterate. We look at what meta has started to ignore. So anything that basically has had tested Aspen and buried deep, like under $3 per day in the ads, is where you’re just going to start to see this basically get flatlined.
00:37:01:19 – 00:37:05:16
John
And so those are the ones that we start to choose to hack off. So if we look at my.
00:37:05:16 – 00:37:08:18
Ralph
Gut, it basically shuts them off before you do. Yeah.
00:37:08:18 – 00:37:22:21
John
Like you saw how when we add a new everything, we drop down by 60%. It’s like this is performing better than those. So as long as you keep doing that, you’re going to naturally find a divergence between the ones that continue to work and the ones I continue to suck. And that’s when you’re just going to basically hack off, add more, hack off, add more.
00:37:22:21 – 00:37:40:12
John
It becomes very systemic. So we look at this last seven days as example. We have our active ads. In this one we have 23. If you start descending by span and go down to the ones with like 30 bucks, look at the chart here and say, was there any spend that has now been ignored? Yep. 3430 started. Ignore it.
00:37:40:12 – 00:37:58:02
John
This has been flatlined into some few dollars spends. Like I said this one is on the chopping block. It’s not going to do anything for us spending $0.70 a day when we’re spending 61 grand a week, you’re not going to feel anything. So as long as meta says thank you, I don’t need that one anymore. You can delete it.
00:37:58:04 – 00:38:15:16
Ralph
But you also take that data and you say, okay, that to the creative team. Those are not the ads that I want. I want more of the ads that are the other ads. Like if it’s that if the chart is in the reverse, it just keeps going up and up and up. Obviously, like the UTC ad that we showed in the previous example.
00:38:15:18 – 00:38:25:03
Ralph
Like that’s obviously you need more. UGC, UGC is resonating with this audience or the particular message or the feature benefit that is being discussed in that UGC.
00:38:25:05 – 00:38:52:16
John
And what’s crazy is that was true up to a point. Like everything with a dog picture is UGC, you just see. UGC, UGC, non. UGC, UGC with a dog, you just see with the dog non. UGC we’ve gotten to this point of spending about $5,600,000 per month with just UGC. However, if we filter towards the ads that are I think I might be able to now use yeah, filter only selected rows if we filter all the non.
00:38:52:16 – 00:38:55:06
John
UGC in the last 30 days. You see my CPAs.
00:38:55:07 – 00:38:57:03
Ralph
233 yeah.
00:38:57:05 – 00:39:02:14
John
Oh wait I’m sorry. Let’s we have to get rid of this one. Oh that’s 318. So anyway so that wasn’t a UGC.
00:39:02:14 – 00:39:05:05
Ralph
So you’re filtering seven okay. You’re filtering. Yeah it’s got it.
00:39:05:05 – 00:39:33:13
John
Got it I should be filtering six now. So the six ads that are non UGC I get 230. Look at this. Everything else is UGC. So these six ads 230 when we remove the filter. My overall when you include UGC 371 my non UGC is $100 and $150 cheaper than my non UGC. And the reason why is it says hey, don’t just basically make the same shit three different ways, create different points of view and different variations.
00:39:33:13 – 00:39:50:01
John
So again we said, okay, we have to follow the hi fi lo fi treatment. Hi fi would be more graphics, lo fi be more. UGC we’re doing the same thing video static reels. Once we followed that, the stuff that we’ve been adding has actually been performing better than our UGC, and we’re able to scale up yet again so you can follow what works.
00:39:50:03 – 00:40:04:16
John
But this is showing us that we tested all UGC guys to a point. We’re like, well, this sucks. We try to UGC our way out of it because we thought that’s what works. And we’re like, let’s try non. UGC and all of a sudden we uplevel ourselves into the next one because those ads came in at two thirds the cost of CPA.
00:40:04:16 – 00:40:24:00
John
I was like excellent, meh. People probably need to see more than just UDC. See all these dogs, they’re all getting healthier. Buy this product. This is why thank you. Like, that was a logical sequence to see that a person’s mind had to go through problems. Solution together equals sale, not just we solve this problem with trust us, we don’t even show you the product.
00:40:24:02 – 00:40:25:07
John
People need to see the product.
00:40:25:09 – 00:40:39:17
Ralph
Right? Right. Makes sense. A couple of those ads, they were off, but you only turn them off after the spend went down to basically negligible levels where it was spending maybe 4 or $5 a day, but they had significant spend at one point in time, which is interesting.
00:40:39:20 – 00:40:40:22
John
Oh, they’re rolling.
00:40:40:22 – 00:40:44:08
Ralph
So because they’re oh, they’re cycling in and out, I gotcha.
00:40:44:10 – 00:40:54:08
John
And what we’ll be talking about soon is I’m going to do one last example, because there’s a new feature inside of meta. Okay. Watch this. You’re ready.
00:40:54:10 – 00:40:55:01
Ralph
This is I’m.
00:40:55:01 – 00:40:58:23
John
Ready. So freaking good. You were a meta media buyer, right?
00:40:59:01 – 00:41:00:12
Ralph
We are in the performance kitchen.
00:41:00:12 – 00:41:01:13
Speaker 3
Maybe you can.
00:41:01:15 – 00:41:04:05
Ralph
Learn stuff up in.
00:41:04:06 – 00:41:22:12
John
You were a meta media buyer, right? I am like, we all were. All right. Yes. So would you say that this one ad set and this 47 ads in the last 37 days, when $28,000 out of $31,000, 28 out of 31 across. How many active? One. 234. Five. I don’t know, 12 ads. Okay, let’s just do this.
00:41:22:12 – 00:41:37:14
John
Let’s look at the active delivery. So 11 ads right now, only one of 11 is working, and it’s getting all of but $3,000 on the total 31,000. This is what normally people would be like. Got to shut that one off. Right. This is what we all believe.
00:41:37:15 – 00:41:38:16
Ralph
Back in the day.
00:41:38:18 – 00:41:44:03
John
Exactly. Back in the day. That’s what the shop was. If you went back to your handbook when you were hired at like two new at a or whatever.
00:41:44:03 – 00:41:52:09
Ralph
It is and be like shut, minimize, maximize, optimize. That’s what we is that we have used different spreadsheet of like any ad that look like that minimize.
00:41:52:13 – 00:41:54:04
Speaker 3
This looks exactly.
00:41:54:05 – 00:42:05:19
John
Like you pause because like I have to get everything else to start. Yeah. Let’s look at metas do beta going to ads set go into charts. Look at there’s a creative fatigue meter.
00:42:05:21 – 00:42:06:20
Speaker 3
Whoo hoo hoo hoo.
00:42:07:00 – 00:42:17:00
John
Now this creative fatigue says I’m. Yeah. The creative fatigue meter says I’m sorry. Media buyer. You think that one ad is getting all of the Aspen is horrible? We’re only 12%.
00:42:17:00 – 00:42:21:18
Ralph
Fatigued. Oh, geez. So this.
00:42:21:18 – 00:42:33:11
John
Is basically even saying, hey, no, don’t shut that off. You’re only 12% fatigued. And some days it actually is better. 7% as of our recently last three days. Hey, why? Because Andromeda’s doing the targeting it.
00:42:33:11 – 00:42:36:22
Ralph
So that’s that the ads, people, if I’m reading this correctly, that’s at the ad set.
00:42:36:22 – 00:42:42:20
John
Low ad set level. But the asset level is only one ad as 28 out of $31,000 and.
00:42:42:20 – 00:42:48:10
Ralph
Got it at it. I see what you’re saying. Okay. How can you do that at the ad level two or no, it’s only oh yeah, it’s at low.
00:42:48:10 – 00:42:56:00
John
Unfortunately it’s just at the highest level, which makes sense because you can’t have creative fatigue at the ad. Little. It’s truly creative as a targeting. Yeah.
00:42:56:01 – 00:42:57:00
Ralph
Creative just targeting.
00:42:57:00 – 00:43:16:13
John
If that ad fatigues itself at my spend can roll to nine more ads. Easy. It just doesn’t. Why? It’s working. Remember when I said follow meta spend? Yeah. Now this is actually bringing us some of the most efficient and cack we’ve ever seen. We’ve actually shut off Google ads by 95% this. Wow. And so what I mean by we were like.
00:43:16:13 – 00:43:18:12
Ralph
All Google shopping, aren’t they?
00:43:18:12 – 00:43:34:22
John
I want performance Max counting the existing customers. Yes. Yeah. And so that’s what was very interesting is if we just look at last 30 days compared to the last. Yeah, last 30 days compared to the previous year. And then we look at all here not active averages all. And then we go into this one here for the same account.
00:43:34:22 – 00:43:52:21
John
And then we go into the campaigns here less 30 days compared to the previous year. Again previous year. And we’ll do it all here. Not just all enabled or anything. We can look at the cost and say, okay, our cost year over year went from 40 grand down to 4200, where 90% down in spend. Yeah. So I was like, this is all.
00:43:52:21 – 00:44:16:05
John
But we did copy and it’s like nothing has ever worked in here. Cut out 35 grand, hop over in the meta and we say, okay, meta, year over year we tech on our 3500. So we’re down 3500 plus 35 grand. So we’re now we’re $40,000 reduced and we’re only spending 4200 plus 30, 135. So from spending 75 down to 35 we cut off 40 G’s.
00:44:16:07 – 00:44:34:23
John
One now with the top total top line, which if we look at the last 30 days compared to previous year, if we go into analytics, go to last 30 days and then optimize turns this way. And then compared to previous year, our total top line is up 0.7% by cutting out 60% of all of our ad spend. Why?
00:44:35:01 – 00:44:55:21
John
This is when you’re looking at the actual difference between attribution contribution, everything was just sucking off the teat of the returning customers, because those are 75% of the time, focus on Ncac and profitability. You say, okay, I didn’t need the other $40,000 a month in Aspen because it was telling us, that’s useless. It was an attribution. Give us a signal reduction spend, give us a confirmation.
00:44:55:23 – 00:44:59:12
Ralph
So let me guess. Your Google ad spend is all brand name.
00:44:59:13 – 00:45:01:02
John
It’s just a p max brand. That’s all it.
00:45:01:02 – 00:45:02:04
Ralph
Is. The Max brand.
00:45:02:08 – 00:45:21:08
John
And think about this. It’s lobster. Everyone pulls lobster out of the ocean. You didn’t make it in season two. You didn’t spice it. You froze it and sent me. So all I have to do is just count on it being cheap. Well, I’m not going to take the most expensive. Never try to be number one for everyone that’s looking for the cheapest product, but I can when compared to my competitors, if I’m the only one in your feed, just shoving my amazing messaging down the.
00:45:21:11 – 00:45:32:16
Ralph
Road, who but also in the feed, they’re not necessarily in market. They’re not in that bloody red ocean of those high value keywords, which is like basically a race to the bottom. Whoever’s cheapest wins.
00:45:32:18 – 00:45:51:07
John
Yeah, and especially in Google Shopping, which is because Google Shopping didn’t work and then they switched over to performance Max. Performance max. It’s like, look at all of everything I’m doing. What happens is like if you type in like Maine lobster, everyone’s running like free one day or you get should be was like everyone is just trying to basically sell you 2 or 3 or 4 or $500 worth lobster, right?
00:45:51:07 – 00:46:11:14
John
We did a list building in meta that says, go sign up for my text messages and you’ll receive two free lobster tails. If you spend more than $100. We’re like, okay, that’s a pretty good deal. So I get two free lobster tails. I’m just trying to get lobster for cheap. But if you’re going to give me two free tails on an order of worth over $100, and you’re putting your money where your mouth is, look at our first click attributed sales.
00:46:11:18 – 00:46:19:01
John
We have $300,000 in sales with 235 new customers coming from SMS. I’m just using as list building right now.
00:46:19:03 – 00:46:20:02
Ralph
Yeah.
00:46:21:22 – 00:46:27:15
Ralph
So no, John, no John. It’s attentive. That’s getting all those customers for you.
00:46:27:17 – 00:46:28:23
John
First click text message.
00:46:28:23 – 00:46:30:12
Ralph
Click at first click.
00:46:30:14 – 00:46:50:20
John
Oh it’s first click text message. Thanks. Attribution please all when you look at that’s reassuring. If you just turned that into gospel you’re going to have a bad time. Well, I shouldn’t say this. If you’re looking at attribution as gospel, I could do that. So please don’t try to put yourself in a position that can’t be replaced by I also being wrong.
00:46:50:22 – 00:47:09:21
Ralph
Oh my God, it’s so true. Well, it’s we’re trying to teach people here on this channel and on this show is to look at the broader view and how all the channels work together. And Andromeda is amazing. Do I have that creative diversification playbook? I know I do. I need to show that around.
00:47:10:00 – 00:47:11:10
John
You actually send it to me.
00:47:11:12 – 00:47:12:21
Ralph
I sent it to you.
00:47:12:23 – 00:47:14:04
Speaker 3
Did I really? Yeah.
00:47:14:10 – 00:47:15:15
John
You’re like, hey man, we just got this.
00:47:15:15 – 00:47:17:14
Speaker 3
I was like, yes.
00:47:17:16 – 00:47:21:15
Ralph
Because it’s actually that’s the title of it is the Content Diversification Playbook.
00:47:21:17 – 00:47:27:02
John
I think it’s I think it’s a practitioner guide. Yeah. Credit diversification playbook, dash practitioner guidance. Yep.
00:47:27:02 – 00:47:39:10
Ralph
Yeah. All right I’ll have to find it. Yeah I said that to you. Haha I keep showing them like you’re oh your ten ads thing. You’re like the god of the conference a nice I never like yeah he’s making that up. I’m like I’m making it up.
00:47:39:10 – 00:47:42:00
Speaker 3
Oh I know that’s.
00:47:42:00 – 00:47:47:17
John
Why I love it is it’s so beautiful when all of our competition is like I don’t believe it. We’re like, yes, you know what? Keep thinking.
00:47:47:17 – 00:48:05:00
Ralph
That keeps I at that. I actually got it from meta, which is crazy, so I know. All right. Hope you enjoyed this week’s show. We’re going to continue to do this. Take the best episodes from our ad labs, bring those over to perpetual traffic, put them on our YouTube channel so you can actually see what’s going on. This stuff is in theory, we’re actually doing it.
00:48:05:00 – 00:48:25:22
Ralph
We’re spending our own money. John, when he first started doing this, just as a brief background, when Andromeda first started taking place the end of 2024, he has a bunch of his own private businesses. He tested it out with his own money to start before we started utilizing it, before we started deploying it inside tier 11. And that’s why this whole thing works so well.
00:48:26:01 – 00:48:48:00
Ralph
We’re not testing new strategies, you know, on client accounts. We’re testing it on our own businesses, some of which are John. And my business is actually together. So that’s how we do things here. And, that’s why we can bring you this type of content every single week with real case studies, real usage. We’re not just talking theory, we’re actually doing this shit and getting amazing results.
00:48:48:00 – 00:49:06:16
Ralph
So if you enjoy this show, make sure that you leave a comment, leave a rating or review. We’d love to have your feedback and help us get out to a wider audience to teach people how to do this stuff the right way, the right way, not the way that we used to do it back in 2015, 2016, 2017, 2018.
00:49:06:20 – 00:49:37:10
Ralph
Like how we do it today in 2025 and going into 2026, and you leave us a rating and review, we can, pass that learning along to more marketers out there and help them grow and scale their business, and using metrics that matter and growth that scales. So on behalf of my amazing co-host Lauren Petrillo, who couldn’t make it on this show, but on behalf of John Moran, who was on this show until next show, see you.
