Equation of Excellence

Rabah (00:01.018)
All right, three, two, one. All right, welcome back folks. Probably, and I might catch some heat for this, my favorite New Yorker, just an absolute mensch of a human, building an absolute empire, 26 year old, 30 under 30, not in jail, Dan Pantella, everybody. Dan, welcome to the show.

Dan Pantelo (00:24.63)
Thanks for having me. I hope you didn't jinx the whole jail thing.

Rabah (00:27.13)
no. Is there something you're not telling me, Dan? Exactly. Yeah. You backed off that meme and it's kind of hurt a little bit, but peak shareholder value. It took over Twitter, which was kind of cool to see that you had the power to insert something in the zeitgeist.

Dan Pantelo (00:30.326)
No, no, no, we're clean, I promise. We deliver value to shareholders.

Dan Pantelo (00:48.502)
Yeah, that was fun. We had a lot of fun with the shareholder value stuff on Twitter. If you know, you know. But I feel like it got a little played out, so we had to lean off the gas a little bit. It felt like it got a little too zeitgeisty.

Rabah (00:55.578)
If you know your now.

Rabah (01:05.178)
I love that for you. It's like they shouldn't have made the old school sequels and they should have just kept it old school where it's just like sometimes, yeah, I love that for you. And I also apologize. I'm actually in the city, but I'm up in Murray Hill and you're down in your cool little office and life just punches you in the face. It's the old game plan and then the Mike Tyson punches you in the face. But that is not gonna stop us from having the most incredible show. For people that don't know you, before we get into Mara Pipe and your recent successes,

Dan Pantelo (01:10.166)
Exactly. Yeah.

Rabah (01:34.65)
How did you get into the game? Because you have a pretty cool past, so I want to make sure you have some time to explain the intricacies on how you've gotten to where you've gotten.

Dan Pantelo (01:45.046)
Yeah, thank you, man. I started in like, so in my, before Marpite, I was an agency owner and we did, we made performance creative and ran paid paid social for, for direct to consumer brands. I started out, I was just like an SEO contractor in college to just like make some, some, you know, weekend money and pay off some college bills. But as soon as I graduated college, I wiggled my way into this,

incubator for DTC brands. This was back when WeWork was cool. Some of you might remember, WeWork had an incubator, a startup incubator called Area 51. And it was like super cool and mysterious. Yeah, yeah. And there was like this like, you know, intense application process. And so we like went, I went through that and like totally kind of somehow got in as a

Rabah (02:19.29)
Yes.

Rabah (02:28.442)
It's a great name.

Dan Pantelo (02:42.454)
You know, we had, I had like an e -commerce brand on the side and kind of focused mostly on that. It was really like a dropshipping type of thing, but we somehow got in and then all of these brands that were in there were running a ton of Facebook ads and they were like, can you run Facebook ads for us? And we were sort of the right place at the right time. And so we grew, we scaled an agency pretty quickly. and it got, you know,

essentially we doubled down on making performance creative. We found that that was like the thing we were best at. And you know, over time we like were looking at these ad accounts and specifically for e -commerce brands, we started to notice this thing that was like a lot of people didn't really talk about that.

that we really couldn't influence as the creative people, which was catalog ads, right? So like, it's been growing so much over the past few years, but back in like 2017, 2018, it was still pretty nascent. And you know, back in those years, we would see maybe like 5%, 10 % of the ad account is like going towards catalog ads, right? And...

Rabah (03:35.946)
Yeah. Yep.

Dan Pantelo (03:58.07)
You know, everyone like knows what I think if you don't know what catalog ads look like, they are not like the designed image or video ads that you would see. They are literally they usually look like just product images on a white background and are usually in some sort of like carousel form. And they just look like the way, you know, a product image would look like on a product page, right? Like as you're checking out.

Rabah (04:25.946)
Ideally, sometimes when they have the wrong dimensions, they look all zoomed in and look terrible in the DPA carousel. But anyways, I broke your flow. Keep going.

Dan Pantelo (04:33.334)
That's right. Yeah. No, no, no. There's a million problems with them. We'll break up. You know, you've only scratched a surface here. We'll break into that. But we saw like, okay, you know, first few years, five, 10 % of ad spend going to catalog ads. Then we just see this like totally like this entity of the catalog campaign just grow in ad accounts quarter over quarter.

Right. And with e -commerce brands who we were around for a while, we saw like the bigger the e -commerce brand, the more they spend on catalog ads. Right. And some earlier brands have like, you know, a huge percentage of their, you know, their total ad spend going to these catalog ads that creatives have no control over. Like these are cool brands. Like, you know, we're talking like sexy websites, right? Like a really good UGC stuff. And so the creative team is like really on it for a lot of these brands.

And then these catalog ads are just like, no one has any creative control over what they look like. We really can't influence whether they do well or not. And so it just brought up all these questions like, what is this? How do we optimize it? Why does it work so well? Why is it such a different performance profile than like, you know, regular image and video ads? You know, and despite not looking good, right, it typically works well. How does that happen? And how can we...

How can we help brands make this better? Because normally, I think a lot of people in the space look at catalog ads as something you just set it up, throw money at it, and see if it does well or not, and then that's it. It is what it is. You can't really optimize it or make it better. And that's the misconception we had also. But then as we dug more into catalog ads and we saw brands have a lot of questions come up about them.

Rabah (06:09.21)
Exactly.

Dan Pantelo (06:26.966)
a lot of brain marketers find problems with how catalog ads looked, right? We sort of as an agency just dove into a lot of that problem. And, and then, and then we started building some solutions for it. you know, and we thought this would be like our secret weapon as an agency. but, lo and behold, this, you know, catalog ads being the, the fast growing category that they are.

Rabah (06:44.506)
Yeah.

Dan Pantelo (06:54.326)
you know, we just, everyone wanted to just like use our like solutions that we built and not deal with the agency. And at that point we realized like, okay, we're actually in the software business now. So we just spun it out as its own product. That was how Marpite was born. And today we run catalog ads for some of the largest brands in the world. So that's sort of the high level overview.

Rabah (07:01.21)
Yes.

Rabah (07:21.274)
Dude, that's insane. How interesting. See, but I think some of the best products are born out of kind of dog fooding. And then you just realize, there's a better business model to this. So let's go do that instead of have a services business. You pivot into the SaaS business. And I think that's so brilliant, man. It's such a cool story.

Dan Pantelo (07:39.798)
We just built it to solve our own problems initially. And then it turns out it solves a lot of other people's problems too. So yeah.

Rabah (07:44.314)
Exactly.

Rabah (07:49.946)
I love that value for value baby. You're quite a good investor. I like how you see the market, but tell me like one of the most worthwhile investments you've ever made, whether it be like investment in money, time, energy, like, is there anything that comes to mind?

Dan Pantelo (08:05.042)
Hmm. Yeah, I mean, I think probably well, number one, I'd say that in every day investing energy into like taking this risk early on in my career, when I was still young, taking a disproportionately high amount of risk to invest a lot of my time and energy into building like a SaaS company that works in the e commerce space was one of the best decisions I've ever made.

I think if we drill down into that, that's broad, right? That's like kind of an obvious one, but if we drill down further, you know, Marpipe went through a lot of, when we were initially scaling up, we went through a lot of, we went through like, I guess, product pivots, right? Trying to figure out exactly the right way to attack this market. Cause we knew we had great creative automation software.

But like we didn't really know exactly how to productize it. And that sort of like good tech, like it's really not a build it and they will come scenario. That's a big misconception. You really have to figure out the right way, like all of these little tiny details around like pricing structure. What do you price around? Like how do you build these features? Right? Like that fine tuning process before we found true product market fit took.

Rabah (09:10.906)
Yeah.

Dan Pantelo (09:29.814)
I'd say maybe three years of just like fine tuning and really going flat, like churning some customers, getting some new ones, but not really seeing too much breakthrough until we really hit that inflection point. And I think like, you know, it was really tempting to give up, you know, cause three years of effort, right? Like, and you're not really seeing that inflection point that other folks see. You know, it's really tempting to want to throw in the towel. We actually got a...

Rabah (09:31.546)
Yeah.

Dan Pantelo (09:58.614)
a pretty good acquisition offer that came in from a large public company at this time. And we actually were going to take it and just actually kind of just exit. But during two months of due diligence, we doubled our revenue. And we hit that inflection point right as we were about to sell. And we came back to them and we were like, hey, revenues double.

would you pay us, you've got to pay us double the price now. And they didn't want to pay that. So we decided that what we thought was a good deal is no longer a good deal. And we kept going and the growth then since then has been nuts. So I'm happy I made that decision.

Rabah (10:29.946)
Yeah, yeah.

Rabah (10:44.154)
That is awesome. Dude, that's so, I love the way you're thinking about it too. So jobs has a thing where a lot of people will think from the tech and go backward to the consumer versus you really need to think of the consumer and then build into the tech. Because the worst thing you can do is just supply side innovation where you just makes really cool shit that nobody cares or has a demand for. And I think that's, in my opinion, the hardest part about.

product is saying no to the great ideas. Like the bad ideas, it's easy. Like that's dumb. But the great ideas, not the 100x or 10x, but just the great ideas and just staying focused. It's so challenging. And to your point with those longer feedback loops, dude, you start to get a little, you know, a little sweat on the palms. Like, is this the right bet? Like what's happening here? It's been three years. What are we doing? How are we pivoted? my gosh, I can get this exit. And then all of a sudden the universe is like, all right, you've been knocking at the door long enough. Now you're through. And now the hard parts are.

Dan Pantelo (11:26.806)
Hmm.

Dan Pantelo (11:37.43)
Yeah.

Rabah (11:39.45)
Not that they're easy, but I find, you know me, I'm quasi hippie dippy, and I find when you start kinda walking down the right path, the hard stuff starts to become a little easier and you start to get these little, I don't wanna say layups, but nods from the universe, and you're just like, all right, I see you.

Dan Pantelo (11:55.958)
Yeah, yeah, that's absolutely. I'm curious, when you guys are thinking about product at Firmat, right? Firmat. So do you guys prioritize like roadmap based on customer feedback directly or is there other like secret spice that goes into that?

Rabah (12:02.138)
Yeah, yeah, yeah. Yeah.

Rabah (12:12.794)
It's a synthesis. So we have a head of product and we have a canny. So canny's taking in like all the requests from everybody. And so ish somewhat we're more so we basically synthesize what people say and what people do. So we're looking at like post hog, we're looking at the actual usage data and then we're looking at the asks and then we'll quasi prioritize it.

Dan Pantelo (12:32.214)
Yeah.

Rabah (12:38.458)
With that being said, we have a few pretty big brands and clients that are on kind of like a quasi different package. So sometimes we'll do some bespoke stuff for them that's basically like hard -coded quote unquote. And then if we see a lot of lift there, then we'll move some of those hard -coded things into like a productization type of roadmap that then gets released to all the Firm -Out users if that makes any sense at all. So quasi, yeah, so quasi but like.

Dan Pantelo (12:51.638)
Mm.

Dan Pantelo (13:04.342)
No, that checks out. Yeah, yeah, that makes sense. We've got a similar program here.

Rabah (13:10.85)
Right. Yeah, because I think the challenge is you always want to listen to your customers. But if you don't have super elite product managers, I'm sure we both do. But I don't want to even say it like that. But if you don't have a good mechanism to actually synthesize that information into product output,

you can sometimes start just building stuff that just doesn't make a lot of sense. And then as your product gets more and more complex, there's just the worst thing you can tell, like an engineer or developer is like, why can't you do that? Can't you just do that in five minutes? Because you don't understand the knock -on effects of that one ask type of thing. But now that's a really good question.

Dan Pantelo (13:53.782)
Yeah. Yeah. It's also about, I think, like, niching down is important for like, like, okay, there's been a lot of hate, I think in the Twitter sphere amongst like DTC operators, there's been a lot of like, polarizing opinions about SaaS lately, right? And, and I think that like, when you can be when you can offer a product that's the best at one very specific niche thing,

rather than something that is like, yeah, you wanna use that platform and that platform, we have everything, right? We have like a little bit of everything and we kind of suck at most of it versus like a one pinpoint solution, right? We found that that really is appealing to like brand owners who are looking for like a specific solution to a specific problem and don't wanna be like cross -sold, upsold on a bunch of random stuff they don't need.

Rabah (14:31.77)
Yes.

Rabah (14:46.682)
Dude, I think that's nail on the head and I think the other knock -on effects or tailwinds you get from that is in the marketing department as well, where now you can really start to talk to the people that use the product. You can really build relationships with people that would possibly use that product versus being everyone to everything is, it's...

Yeah, I don't know. I couldn't agree with you more. I think it's absolutely the path. The challenge is when you go raise money, you have to show that the pain point, like you're like the, you know, GLP one drugs of the DTC market or whatever, where it's like, hey, we can actually make a ton of money off this one thing. Cause then you get into tam constraints when you're raising money. But I think that's spot on, man. I mean, the biggest thing that I learned at triple whale was cost isn't just money. Cost is expense, focus and bandwidth. And like,

If you focus on bandwidth are almost more expensive than the money because every dollar is quasi equal, right? Like, but what makes those dollars more valuable are the people that are deploying them. If you can deploy those dollars more efficiently, even though you have a hundred thousand dollars and I only have $50 ,000, if I can deploy them to X more efficient than you do or more effective. And so that's why I think hiring a team and getting killers is just so important because everybody can, can kind of spend a little bit, you know what I mean? Especially like post a where.

or at least even seed where you're starting to cash flow and you know some money's coming, then it's like, man, let's hit some rocket boosters.

Dan Pantelo (16:11.51)
Yeah.

Yeah, yeah, totally, totally. And I think spend optimizing, like getting more, I think there's like, there's two categories of products. There's like the infrastructure level stuff, like a feed management platform. Everybody needs one, right? And it's not, you're not investing in that for performance. You're not like, you know, you're buying it because you need the infrastructure. Like you need a product that does this. Like everyone needs like a

Rabah (16:30.682)
Yes.

Rabah (16:35.994)
Yes.

Dan Pantelo (16:43.414)
delivery system, a 3PL, these are infrastructure things. You're not investing them for better performance. But then there's like the risky, like more, I guess products, maybe risk is not the right word, but like you're buying products in order to get more bang for your, to increase performance, right? And so I think like a lot of more tech and ad tech tools fall into that category where we're not held to, like we're held to a different standard.

Rabah (17:01.914)
Yes.

Rabah (17:10.714)
100%.

Dan Pantelo (17:11.19)
where it's like we absolutely like we need to prove that we can deliver a lot of optimization value rather than just like serve a specific like base level function. And adding value to DPA is actually very easy to do. So we found ourselves in a lucky category.

Rabah (17:27.45)
Well, look at that. It's almost like you're reading the show notes. Dude, that's a perfect segue. So let's jump into Marpite. Talk to people for people that show. Yeah. Give us that kind of 30 second. What are DPA ads or Daba ads still in the mix, et cetera. Kind of just give us the whole ecosystem and then go into why it works and why it's so powerful.

Dan Pantelo (17:48.886)
Yeah, yeah. So, so, when the, what Marpipe is matters way less than, than what the problem we solve, which is the problem that most brands have. Right. And it's, it's really simple, right? Like cat. So there's, there's three types of, like ad types, meaning like creative, you can upload into an ad platform. One is video to a static image. And one is, and the three is a product catalog.

Right? And everyone talks about images and video. And everyone's always making a bunch of video and image content because you can upload videos and images as your ad creative. People don't really talk about catalogs as a creative input because like the concept, the idea that you can do much with it to begin with, you know, people don't even know what you can do. Right? And so like typically if you're, so if you're a Shopify brand,

You know, you by default have a, you have like a PIM, a product information management system. This is like your feed management tool, right? Shopify has a very basic built -in one that every brand starts out with. It's like a baby PIM. There's plugins that are more sophisticated things. And then there's like more like wider, so like if you're a bigger brand, you're almost definitely using like a Feedonomics or like a product app or Go Data Feed, one of those.

so, so those, those are like the source of truth that like connects like your warehouse to your front end website, right? So like, you know, if something runs out of stock in your warehouse, it shows that on your website, you're immediately for that product shows out of stock, right? Zero units of someone can't select and buy something that isn't in stock, right? Things like that. So it's kind of like a one large glorified spreadsheet that has like, for every skew that you have, you know, it has data. So for every skew, you have a product image.

that product image populates on your product page on the website, right? You have a price, you have the product title, all of this data about every product, right? So when you upload a product feed, which is this live updating spreadsheet, into an ad platform as an input, which by the way, every major ad platform accepts this, right? So obviously meta, but TikTok, Snap, Pinterest, Reddit even.

Dan Pantelo (20:14.934)
Every major ad platform accepts catalogs as an upload. The way that it works is completely different than how video and image ads work in terms of the delivery, right? And that's what makes them work well despite looking so bad. So when you upload a catalog, ad platform reads the catalog to understand what you're selling, right? So like,

when you upload a catalog, it'll say like, okay, there's a product in here that's like a brown leather shoe that's between the $85 and $95 price range, right? And also the size 10 is in stock, right? And so if you are, if Facebook knows you are a size 10 looking for brown leather shoes in this price range, right? It will look, it will, it'll know what kind of product you're already searching for and it'll scour. It's like,

universal archive of catalogs that has access to and serve you similar products. It's a very similar experience to like on Amazon, you know, when you look at a product and right below it, there's like, you may also like, and it's very similar products and it shows you a comparison on like, here's different price ranges for similar products and different pros and cons, right? This is the advertising version of that, right? Where like,

Rabah (21:18.682)
us.

Rabah (21:22.234)
Yup.

Dan Pantelo (21:34.55)
And the cool part is that the delivery is smart, meaning like if Facebook knows you're a size 10 shoe, but they are out of size 10, it won't serve your ad, right? So the delivery is much more informed because it has a lot more data that it doesn't know. Like if you're running like a brand ad, that's like video UGC, it doesn't know any of that information, right? And so it's not using that in the targeting and delivery.

Rabah (21:44.218)
Serve that ad.

Dan Pantelo (22:03.926)
So catalog ads have better delivery, better, right? But the thing is they also like, and so one of the things that's important to notice, the more information you have about your products in your catalog, the better that delivery becomes, right? So you want to, if you have like, if you're using, if you're running catalog ads with the direct Shopify integration, no go, that's not a good idea. You want to be able to include as much,

product information as you can, which these other like feed management plugins and service providers will allow you to do. And it's worth, it has a lot of bang for its buck in terms of if you're trying to scale catalog ads, it's very worth it. Now, the other thing is like how it looks, right? So, not too long ago, Meta just released a report that said that in their e -commerce category, almost 70 %...

Rabah (22:51.93)
Yes.

Dan Pantelo (23:02.006)
of all of the ad spend in the e -commerce category is catalog ads. And just recently at the Meta Performance Summit, just a few days ago, catalog ads are a huge focus because they're doubling down on ASC. So Advantage plus shopping campaigns work really well with catalog ad campaigns. They want to keep the convert, especially if you have Meta shops activated. So all of these things, they feed each other. They're closing that attribution loop. And so...

Rabah (23:02.394)
So I've seen that too.

Rabah (23:13.754)
Yes.

Rabah (23:24.314)
Yep.

Dan Pantelo (23:30.71)
So it's becoming, it's a really popular unit and it's becoming more popular. It's growing and Facebook is investing more into the capabilities there. The thing is like, you're spending all this money on just the product image on a white background. That is the ad creative, right? It is literally, it is missing all of the elements that we all know as marketers, influence, performance, copywriting, your brand logo, basic brand elements. You know how much of brand lift you're missing out by just not even including your logo?

Rabah (23:45.978)
Right.

Dan Pantelo (23:59.478)
in those visuals, right? So just being able to even do really simple things like throw in a lifestyle image for context, right? Add branding, add some copywriting, throw in price, throw in testimonials, value props, all these basic creative things that we all know lift performance everywhere else. Obviously lift performance here too, but if we have like a few hundred SKUs in a product catalog, how are we gonna control what that look like? A redesign, the whole fit, right? Like so.

So, so Marpipe is kind of like a Canva for product catalogs. That's how some of our customers have put it, right? So instead of just plugging your raw feed directly into an ad platform, first, which, which would serve your like raw product images, you just plug it into Marpipe. And then from there, you have full creative control where you can like click one button and remove all the white background. So they're transparent.

Rabah (24:32.474)
I love that.

Dan Pantelo (24:55.19)
Then throw in a branded background. Try a design that has a testimonial in it. Try a design that has value prop one, value prop two, you know, throw in some lifestyle imagery. Just make it look like how your static image ads would look. And, you know, usually, and then just test that against your unedited ones. Really straightforward premise. Get creative control over these and watch the impact that has on performance. Right? So that's the premise at a high level.

Rabah (25:22.97)
I love that. I never thought of the kind of like almost catalog SEO in a weird way, right? Where it's like if your catalog doesn't have the info or it's not scheming in the proper way, that Facebook searcher when that impression comes available isn't gonna be able to find your catalog as easily or find it at all. Do you know or do you ever suggest just running DPA or almost all the accounts mixed?

Dan Pantelo (25:52.598)
Well, we have some ad accounts that are almost 100 % DPA. And these are some larger, like large advertisers. Yeah. You know, spending over like over a hundred K, several hundred thousand dollars on just DPA. Usually these are in categories that have, you know, I wish I could name drop, but I just, I like, these are big, these are big brands that everyone knows. And,

Rabah (25:59.546)
Really?

Rabah (26:09.018)
Yeah.

Rabah (26:15.93)
Yeah, for sure.

Dan Pantelo (26:22.134)
I will say that the category where that's most popular is like the bigger the caps, the bigger your product catalog, the better it's going to perform. So usually this would happen with.

Rabah (26:29.274)
Tracking, yeah, yeah. So super high skew stores where you're talking like, yeah, yeah, that's like a hardware store maybe or a department store or something that just has a banana set of skews.

Dan Pantelo (26:35.574)
Yeah.

Like, yeah.

Dan Pantelo (26:44.214)
So really like apparel brands, home goods, furniture brands, hardware, jewelry is actually a really big category here because every like for all the rings and watches, they have different sizes and stuff and marketplaces. So, yeah.

Rabah (26:47.546)
Yep.

Yep.

Rabah (26:57.146)
It makes tons of sense, yeah.

Marketplace is another one. Do you guys still work with small catalogs or is it just kind of can and kill a mosquito?

Dan Pantelo (27:09.526)
No, we do and this is actually, you know, I was just talking to the founder of Huron the other day, right? Matt, dude, yeah, Matt's awesome. We were just doing a demo with him and his team on Friday and you know, his question was like, hey man, are catalog ads for us because we've tried them before and they didn't work. And I'm like, how many SKUs do you have? And they're like, we have 20 SKUs.

Rabah (27:17.082)
Yeah, Matt, Matt's a G, he's the best. I'm supposed to meet up with him, but it's life.

Rabah (27:25.498)
Yeah, wonderful human.

Dan Pantelo (27:39.35)
And that's a small skew count, right? And so, you know, what I say in that scenario is, like, you are more likely to succeed with catalog ads if you have a high skew count. However, that doesn't mean that catalog ads can't work for you. It's just, it's worth testing. It's worth like trying and giving it a good spin. But, you know, like we see kind of like 50 -50.

Rabah (27:41.402)
Right.

Rabah (27:54.554)
Yeah, I see what you're saying.

Dan Pantelo (28:08.406)
chances of success with like brands that have smaller catalog. So there are some like one skew stores that do well with catalog ads and actually crush it there, but it's way, it's more rare. and like really like when we say, Hey, you should probably, you should test it. A lot of people don't test catalog ads the right way. And that's like where a lot of problems that people don't even know are problems start. Right. So we'll have people come to us and be like,

Rabah (28:25.53)
Yeah, yeah.

Dan Pantelo (28:37.814)
Hey man, like I have a lot of SKUs. I'm like a perfect content. I'm under a Harrell brand or whatever, like I'm perfect for this. but we've tried catalog aids and they just don't work for us. And I'm like, I don't know. I call bullshit, right? Let's, let's look at the ad account. Let's pop it open and see. And one of, unfortunately, a lot of like marketers are very, very well versed with setting up image and video ads, right? But not so well versed with catalog ad setup and catalog ad strategy. And, and there's a lot of, we encounter like.

Rabah (28:43.354)
Yes.

Rabah (29:01.946)
Yes. Yes.

Dan Pantelo (29:06.582)
basic setup problems all the time. Where like, for instance, like the product IDs in your pixel on your website need to match the product IDs in your product catalog that you upload. If they don't match exactly, you're losing attribution, you're losing signal and will not optimize, right? Like basic things like that, that are a little more like technical to audit that like...

You know, no marketer wants to spend their day looking at pixel and catalog setup stuff, right? So, so unfortunately, a lot of people get things wrong here, but that's what we do all day, every day. We love doing that. That's our like one thing. So, we got really good at like auditing, you know, a catalog ad campaign to figure out why it doesn't work. And usually it's just unblocking a few simple setup mistakes. And then they're like, it works for us now.

Rabah (29:36.058)
percent.

Rabah (29:56.378)
I love that. And so how do you test? Like what are you doing in terms of testing? Are you basically the control is essentially like the feed, AKA the catalog. So you're just testing feeds against each other or like take me through how you think of that test. Cause I didn't even think about that. Yeah. What are you testing? How do you test? Give, give the people some, some free game out there as the kids say.

Dan Pantelo (30:14.934)
Yeah.

Dan Pantelo (30:18.998)
So here's some pregame. So when you're running catalogs, right, one of the things that a lot of people don't know is that catalogs, when you create a source feed in an ad platform, that source feed accumulates learnings the same way that like, if you could kind of think of it like the same way like a pixel does, right? You have to season it, right? The more spend you run against it,

Rabah (30:42.394)
Right. Yep.

Dan Pantelo (30:47.766)
the more it learns and those learnings stay at the catalog level, right? So creating new catalogs is bad for performance, right? Like you don't want to create a bunch of new catalogs. Back in the day, pre -MarPipe, if you wanted to test new catalog designs, right? You had to create, like you had to, so let's say I have a hundred SKUs. I'd have to go,

Rabah (30:52.57)
Got it. Yep.

Rabah (31:13.466)
Yes.

Dan Pantelo (31:15.414)
If I was to do this the manual way, I'd have to go have my designer like maybe make five templates, right? That we would test. and then we'd have to redesign every skew image through that template. So that's 500 new images. My designer would have to create. I'd have to put those into spreadsheets and then upload those spreadsheets as new source feeds. Right. and those source feeds are not, when you launch them and you start spending against the new source feeds, they are not going to compete anywhere near the performance profile of your seasoned historic one.

Rabah (31:26.33)
Yes.

Rabah (31:34.65)
Yes.

Dan Pantelo (31:45.046)
And you might think, changing the creative on my catalog guys doesn't have an impact. No, it's because you're creating new feeds. Never do that. Right. So what you should do is in commerce and metacommerce manager, there for every feed, there's a supplemental feed section where if you copy and paste a feed link into there, it will override the visual. If it has new visuals, it'll override the visuals of the previous feed for those skews instead of.

Rabah (31:51.546)
Yes.

Rabah (32:11.418)
Got it.

Dan Pantelo (32:13.718)
running, like instead of being a new cattle, so it's the same source feed, but it'll override the visuals. So it's not showing just the unedited product images, showing the thing that you designed when that skew is shown, right? So the thing is though, you could still only test like one design at a time. Like let's say I didn't go through all that work, I can still only test like one, there's only room for one supplemental feed link at a time.

Rabah (32:33.466)
Exactly.

Dan Pantelo (32:40.022)
What we did here in collaboration with some engineers at Meta was we built out, we essentially have like a special supplemental link with Morepipe where when you drop that in, let's say you created five designs, right? With like for your, for like five different design treatments for your feed, you would be able to drop the Morepipe link in the supplemental feed section. And how it works is when you create a new ad connected to that source feed.

There's a drop down menu and you can select which out of up to 20 designs you want to run for that one at the ad level. So then you can create and test up to 20 designs simultaneously from the same source feed. So you never have to create a new feed. You keep accumulating learnings and seasoning that one source feed and it really shortcuts. There's no learning curve. There's no warm up period. It's just straight to performance.

Rabah (33:16.922)
Guess.

Rabah (33:35.162)
Wow, I'm sold. I'm buying two. I'm in this car today, Mr. Pantello. That's it. That's the path, man. That's the pitch. That's no notes. I think that's amazing. And it makes tons of sense. It makes a ton of sense. And then to your point, it's almost that Peter principle. Like the bigger you are, the more impact you're going to be able to have because you just have this monster data fly loop. I'm just feeding data into the most.

Dan Pantelo (33:38.198)
Hahaha!

Rabah (34:04.314)
performant data machines on Facebook in terms of DPA and my website. So I mean, that makes a ton of sense. I love that pixel analogy. I'm gonna steal that from you. Cause that totally landed for me. Cause if I was gonna have a cook off against you and you made me start, we have the same brand, same creative, whatever, but you made me run an ad account from a brand new pixel and you're using the ad accounts pixel that's been on. It's like, yeah, there's no way I'm gonna beat you or at best I'm like, it's gonna be very challenging. And.

Dan Pantelo (34:10.134)
Yeah.

Dan Pantelo (34:29.878)
There's no way.

Rabah (34:33.274)
To your point, that's just an apples to oranges comparison.

Dan Pantelo (34:37.398)
You know, and a lot of like right now, a lot of brand marketers don't even like, let me instead of brand marketers, let me rephrase it as marketers in general, because it's the same with performance marketers. Like there's still a lot of product education to be done around catalog. Like they're spending so much of their budget on catalog ads and still are not even aware that they're able to influence how they look. They just think, that just is what it is. There's no way for us to brand that.

Right. And so we're still doing a ton of like product education. Like, like we'll, we will take a screenshot of cat of like terrible looking catalog ads, but most of them look terrible. And we'll just send that to the CMO and be like, did you know this is what your catalog ads look like? Are you guys spending a lot on catalog ads? And it'll be like, you know, it's, it's like a kind of gotcha like, like, Hey, like, are you okay with this? Like, are you, do you back how this looks? Right. And so they'll reach out and be like, how, you know, all right, tell us how.

Rabah (35:16.506)
Yeah. Yeah.

Rabah (35:23.514)
stress.

Rabah (35:31.874)
That's brilliant.

Dan Pantelo (35:36.47)
How do you fix that? Yeah.

Rabah (35:36.602)
I'm in. Send the demo link, you son of a bitch. You got me. You got me. I guess what are you most excited about with Marpie? Everything is just humming, man. Everything is humming and it feels like you guys are just on top of the world right now. What's next? What are you stoked about?

Dan Pantelo (35:57.59)
Look, things are going great. We've never seen this amount of inbound. You know, we're growing really quickly and things are going great. But there's a lot more work for us to do on the product innovation side. And those are some of the things that we're most excited about. So one of the things is video DPA is a new category that just became a thing.

Rabah (36:13.882)
Sure.

Rabah (36:22.874)
Interesting.

Dan Pantelo (36:27.094)
meta invested in it and just rolled it out very recently. And so it's actually very interesting because if you could have the same delivery mechanism that DPA has, right? But DPA just serves the product image of the SKU, right? With Markpike, you can control what those product images look like. But if you have a video...

Rabah (36:38.042)
Yes.

Dan Pantelo (36:51.798)
for a product, like you know how sometimes product pages have a little GIF or a little video in the product page where it'll just be like image one will be the product photo, maybe there's some lifestyle photos in there and then at the end maybe there's like a little video of the product, it's like maybe a 360 or something, right? So what Meta is now allowing people to do is run video DPA where if you do have a video for every skew in your feed, you can select that video to run.

Rabah (36:55.802)
Yeah, 100%. 100%.

Rabah (37:02.298)
Exactly.

Dan Pantelo (37:20.406)
So it's the same delivery benefit of DPA, but it has motion, which has to go way higher click rate, right? And typically. So the problem is that almost no brands have a video for every skew in their feed, especially if you have large feeds, right? So what if you could use generative AI to create a video?

Rabah (37:22.266)
Yeah.

Rabah (37:39.162)
Correct. Yeah.

Dan Pantelo (37:48.31)
for every video in your SKU, thus allowing large brands with large catalogs to be able to take advantage of Video DPA. That's the next horizon. We're about to unlock that for folks. And I know that we're doing some beta testing around it right now. And folks are really excited about that. Yeah.

Rabah (38:05.146)
Dude, that is something awesome. And also, shame is plug, if your clients are using Firmout, they can change all the PDPs they want without affecting actually the main site. So they can test whatever they want. On the double test, the DPA to the Firmout love. OK, one last question, then we'll get into the last question. I know you're like three years old, but what would you like the old cheesy kind of.

Is there anything that you would change in terms of like your trajectory? Would you like, what advice would you give to young gun 20 year old Dan just ready to take on the world with your perspective now, if that makes sense.

Dan Pantelo (38:44.086)
man. I mean, honestly, like, I think that the way that we played this, like, we failed a lot. We spun our wheels a lot, like I touched on earlier. And like, I just definitely, there were some moments where, especially when clients churn, such a gut punch, right?

Rabah (39:11.578)
Yeah, it sucks. Yeah.

Dan Pantelo (39:13.75)
And just makes you really want to kind of like give up, right? And then this is going to sound so corny and generic, right? But like, this is a lonely journey. It's a lonely journey, you know, for founders of all types of companies, not just SaaS, right? And, and a lot of people, I think like a lot of people throw in the towel right before they're successful and good things take a long time to build. Like one of, one of the biggest, it's like,

You know those apps that blow up out of nowhere overnight, like AirChat or like, you know, like they blow up over, like the faster it blows up, the faster it gets taken apart and like, and has a downfall. Right. And so building slow and steady and consistently and not like feeling like a failure. If you don't like cash in on that, like overnight growth, kind of like success story that, VCs want to really push, then, and, and just like,

Rabah (39:44.89)
Yes. Yes.

Rabah (40:07.482)
Yes.

Dan Pantelo (40:09.822)
and making profitability cool again, right? These boring fundamental stuff that is way more, like totally doesn't sound exciting to a young new entrepreneur is actually the truth, right? That is the truth, that's the path. And I used to think, I used to want an empire build and think like, hey, people would be like, how big is SmartPipe? And when people ask that question, like how big is your company, they're asking you how many employees do you have?

Rabah (40:22.17)
Yes.

Dan Pantelo (40:40.086)
Right? Like that's a terrible question to ask to assess the success of a company. Right? Like I having less, having less team members is cooler than having like a large team. That's like mediocre, right? Having like a few, a few savages, right? And so these sorts of fundamentals on like profitability, revenue per employee, right? Like pace, the growth, those things go really long way and assure our clients that we'll be around for a really long time.

Rabah (40:43.514)
Terrible.

Rabah (40:52.474)
I couldn't agree with you more.

Rabah (41:07.898)
Dude, so well put. I couldn't agree with you more. I mean, it's almost bringing to mind, I'm like a quasi F1 fan and the Montreal race was just this weekend. And there was these two guys that without getting super nerdy, if people don't really care about racing too long, didn't read these two guys had these wet tires on that they could have way more grip than everybody else. But that was because the weather was there. And so they were just ripping through the field.

And then they ended up getting the exact same places by the end of the race. And I think sometimes those guys on the wet tires are kind of analogous to those fads, not trends, where you're like, it's not your race. You know what I mean? The comparison thing can just be so, it's just so not there where you're exactly right, man. I think that.

There's this weird, like, I don't know, cause I get the same thing too. Sometimes you're just like, dude, and I just have to tell myself, like, that's not our race, man. Stay focused, stay winning. And I think the other thing that people index a lot on is they think success like just happens and there it's just, dude, like, you know, overnight success, three years in the making kind of stuff. It's just like this, if you can get those aggregate wins, I think that is not only a better.

Dan Pantelo (42:16.31)
Yeah.

Rabah (42:22.202)
way to live, but like psychologically more advantage when you're on this roller coaster of highs and lows and like, you don't feel like you can control the system. That's like, Rashibe one of his biggest things is like, it doesn't matter if things go wrong, but it's like, okay, did we address it? Do we fix it? Do we know why it went wrong? And then we can make it. But it's when you're in this just chaos coaster of winning, losing, winning, losing, but you're just trying to capture a lightning in a bottle versus building a lightning factory. The latter is way more the path.

than the former, in my opinion, in my experience.

Dan Pantelo (42:52.854)
That's, I love that analogy. I'm going to steal that one. I love that. Capturing lightning in a bottle as opposed to building a lightning factory. That's a good one. That's a rabbiism.

Rabah (43:01.242)
That's a good one. Let's go. It took 43 minutes, but we got a bar. Okay, DP, let's wrap this up. I ask everybody the final question of the podcast. What is excellence a function of for you? What are the inputs to excellence?

Dan Pantelo (43:16.958)
Excellent to me. It's a function of controlling what you can control, not controlling what you can't. Maintaining a loop. I heard this analogy with someone the other day was like, you know, when you're learning like a new instrument, right? Or like, you're learning, like if you're learning the drums, right? Like people who just start for the first time learning, they're like holding the drumstick too tight.

Right? Like a real expert has a real loose grip, like same with a tennis racket and same with like a guitar. Like the experts are like loose with it. Right? Don't like try to like control every little tiny factor and like squeeze tight. Right? And you only learn that over time. Cause I think initially you tried to like maintain that like really hard grip and it's just impossible. It doesn't come out, doesn't come out with finesse. Right? And so just.

loosening the grip is a function you learn over time, but it's really about controlling what you can control and totally not focusing on the things that you can't. Yeah.

Rabah (44:26.362)
Nailed it, man. Control what you can control. That is a stoicism, very stoic ideal, and you know I'm down with that. And so I used to be, obviously, you gotta bring everything back to me. I used to be really into mountain climbing, like I dirtbagged for like six months. And you see that very prevalent in mountain climbing. When you're starting out, you really don't have a bunch of gear.

And then when you get good, you get a ton of gear. And then when you get great, you use like three things. And so it's just the exact same thing that you're talking about of like the beginner, the expert and the master. And it's like that bell curve of like, as you start getting really great, you only need like two things. So the guy went and climbed with would have like three things. I'm just fucking geared to the gills. Perfect, perfect analogy.

Dan Pantelo (44:52.342)
hahahaha

Dan Pantelo (44:59.03)
Yeah.

Dan Pantelo (45:03.542)
Exactly.

Dan Pantelo (45:07.702)
Yeah, yeah, you can compare that directly. You could compare that directly to the CEO who looks at like 37 different metrics and reports versus like the master CEO just looks at like boils the whole business down to like two, right? Yeah.

Rabah (45:20.89)
That's exactly right. That's a perfect way to bring it back. Dan Pantello, everyone. One of my favorite humans on the planet. Dan, tell people how they can follow you. How can they get more involved with Marpie? This time's yours, my friend.

Dan Pantelo (45:28.662)
Ha ha.

Dan Pantelo (45:33.782)
Yeah, man, thank you. If you've made it this far, give me a follow and hang out with me on Twitter. There's some smart stuff, there's some shit posting. We have a good time. I'm at DanPantello on Twitter and obviously marpipe .com if you want to check out what we could do for your brand. Yeah, thanks, man.

Rabah (45:54.81)
Amazing, dude, thank you so much for the time. Make sure to go book a demo, firm .commerce .com. We'll make you help the money printer go brr. We also have a great newsletter that goes out every Monday called the Geometry of Growth. You can sign up right on the site, firm .commerce .com. Dan, you're the best. You are just awesome. Dude, I love to see you winning and you're just such a muse for me. And keep on keeping on. Sorry I couldn't make it into the office next time. This New York trip got a little crazy, but.

You're always so good to me when I come out here and love this city. Love you and love all of you folks. Thanks for tuning in. Make sure to smash the old subscribe button. This will be on YouTube. If you want to see Dan's wonderful hair and office. and then we're obviously on your favorite pod catchers everywhere. So thanks for tuning in. We'll see y 'all on the flip. Bye bye.

Dan Pantelo (46:22.486)
Love you, brother.

Dan Pantelo (46:39.03)
Cheers.