← Marketecture: Get Smart. Fast.
Marketecture: Get Smart. Fast.
Episode 143: Omar Tawakol on Using AI to Scale Product Placement
00:57:07
Loading summary
Transcript152 lines
- [00:01]
A
Exclusive inventory, fragmented tech, high minimums. Programmatic is built for the few, but advertisers need scale without the complexity. Adlib sits above it all, orchestrating activation, optimizing 24, 7, 365 and giving you back 40% of the time spent juggling multiple DSPs and media partners. 1 Login every platform.
- [00:29]
B
No trade offs.
- [00:30]
A
Go to getadlib.com G-E-T a d l.
- [00:35]
C
I b.com this podcast is brought to you by Audiohook, the leading independent audio DSP. Audio Hook has direct publisher integrations into all major podcast and streaming radio platforms, providing 40% more inventory than what could be accessed in omnichannel DSPs. What's more, audiobook has full transcripts on more than 90% of all podcast inventory, enabling advanced contextual targeting and brand suitability. Audiohook is so confident that in addition to CPM buys, they offer the industry's only pay for performance option where brands can scale audio and podcasting with peace of mind knowing they are only paying for outcomes. Visit audiohook.com to learn more. That's audiohook.com welcome to the market Podcast. This is Ari Paparo. Today we have a great episode. Omar, come in. Omar Tawakal, the CEO of Rembrandt, formerly the CEO of BlueKai and well known figure in the ad tech circles, will be joining us as our guest Eric will be joining as well. It's exciting conversation about product placement and how AI can help with product placement, which is a subject that's this podcast is brought to you by Playwire. What if the real competitive advantage isn't choosing between human judgment and AI but knowing when to use each? Playwire is launching a self service platform with rules based control and machine learning optimization. No more one size fits all monetization. At Marketector Live we'll reveal performance data including how one publisher saw 168% higher CPMs by choosing where to use human oversight versus automation. See the platform@playwire.com that's playwire.com Probably a lot of people haven't thought about so much and also they recently did some acquisitions which is unusual in an early stage company. We also are celebrating Advertising Week or talking about it at least. So we're recording this on Thursday of Advertising Week. A lot of stuff going on, a lot of parties, a lot of events so we get the perspective on that. Thank you to those of you who came to the Monday Ad Tech God party at the Virgin Hotel. It was lit, it was fun. I know some folks didn't get in. It might be a little sore. Sorry, there's only limited capacity. And the next thing up on the agenda from Architecture, of course, is our live event, Architecture Live, coming up on October 27th. That is selling out pretty quickly. I would suggest if you are planning on going, you should definitely buy a ticket. I know I've said that before, but one day the tickets are just going to be gone and then you're going to be texting me, asking me to get in, and I will not be able to help you. It's also worth noting that for qualified brands and agencies, it is a complimentary ticket. So if you're listening to this and you're in those categories, you should probably register and we will approve qualified members of those professions. With that said, let's just jump right in with Omar Talakal, the CEO of Rembrandt. All right. Welcome, Omar to the Market podcast. Thanks for coming.
- [03:55]
A
Thanks for having me.
- [03:56]
C
So we just saw you at Advertising Week. Eric and I saw your talk. So it's new and fresh, but we were supposed to. The people who saw you were going to give you a hard time about your really bad headshot. So do you have a professional headshot somewhere hidden out there?
- [04:12]
A
I probably do. I think that was a Facebook headshot from like 15 years ago.
- [04:18]
C
Yeah, yeah. I mean, there's a lot of headshot fraud going on where people look like young and thin in their headshot and then you see them and you're like, wow, that was a long time ago. But not for you. It just was a bad, low quality photo. You look the same, so you know, you're aging well. I guess it's the lesson here.
- [04:36]
A
Apparently.
- [04:36]
B
So what a bit of great no brainer. No brainer. Generative AI tool. Like, everybody should do this for their speakers. Like, let's build this.
- [04:46]
C
We should age them. We should. We should, you know, give us the year your age and the AI will aid you appropriately. So. So, Omar, you've been in the news a lot. We want to talk about your company, Rembrandt, which Disclosure. Eric, you're an investor in this, right?
- [05:05]
B
Yes, I am.
- [05:06]
C
All right. That's just always our disclaimer. Unusually, I'm not involved in this one, so that's a little bit of a twist. And it's an interesting space and has done some acquisitions lately which we want to hear about. Private to private. Acquisitions are always kind of an interesting subject. So why don't you just start with giving us the elevator pitch? We'd love to hear what the story Is, Yeah.
- [05:26]
A
So the background of the story is the three of us on this call grew up with programmatic in our careers going from 0 to 250 billion. And all the while, this industry has become much better at showing the right ad to the right person at the right place, with the right data, with the right price. And during the last 20 years, consumers have proven that they just want to run away from ads. You walk into a room at Advertising Week yesterday and ask the people how many people here are paying for subscriptions to Netflix and Amazon prime and stuff like that. And virtually the whole room raised their hand and that's Advertising Week. So here are the professionals who spend their day allocating budgets to literally go home and pay to say, I don't want to see that stuff. And you know, we saw this and we noticed it and so we basically said, why bet against consumers? That doesn't seem like a good idea in the long run. And in particular, we noticed this 10 year trend, that 10 years ago, 80% of time on video was spent in high ad load environments where you're talking about 17 minutes per hour of having to sit through ads or walk up and get a drink or not pay attention. But you are subjected to that and then 10 years later it's actually flipped. 80% of the time is in low to no ad load. So you're talking about, on average it's now like five to seven minutes as opposed to 17. So the consumer's already spoken. And the people who benefit from this are streamers who figured out they can make a lot of money outside of advertising. And so with those facts behind us, we built a company to say maybe there's another way. Maybe we can embed a brand opportunity in a video in a way that doesn't interrupt the consumer, but also allows you to get a branding moment for the advertiser. So that's kind of the guy.
- [07:24]
C
It makes total sense. I mean, we've been hearing about that for years. You watch the Sopranos and suspiciously they're drinking a lot of Tropicana orange juice in all the scenes. It's like, you know, it's like how. I don't know, I'm not going to put a joke in there because I'll get in trouble. But, you know, so what, what's different? What's, what's changed in the AI world?
- [07:46]
A
Yeah, a lot. Lots changes. One last thing about that, that, that occurred to me more recently even than, than we started it is that I'm happy to pay for subscriptions and a Lot of the execs I know are happy, but if you want to produce a solution for all of humanity where there are a bunch of people who just don't want to afford to pay for it, there's got to be a very, very good way. And so in order to do that, we needed to find an at scale way to place brands and content that isn't high friction. So now I'm going to get to your question. What we're doing is not new. There has been physical product placement for a long time. Physical product placement from the very, very beginning. And except you have to negotiate two years in advance of a show to get into it. And with influencer stuff it was six to seven weeks. So that existed then for the last 10 to 15 years. There was a crop of companies who came after the virtual product placement area. But there was two or three things that they did that made it a very manual high friction proposition. The first thing they did is they use VFX to actually place the product in, which meant you literally had to manipulate every frame. And you know, you depending upon on the content could be like, you know, 32 frames a second or less. So you've got a lot to manipulate. Sometimes it could take 30 to $40,000 in three weeks for people to do things like this for high, high end content, so high friction to place it. But there was also a second type of friction that was maybe even more insidious and that is matching friction. So an advertiser, you know, comes at, wow, this virtual product placement's great. I want to place my can of, you know, some drink. I won't say which brand. On Tuesday, taco scenes like, okay, great. That really cuts the inventory down to a way that, that makes the publisher say, yeah, thanks a lot. I'm not sure this is an interesting conversation. So that's one matching problem. The other matching problem that's also troubling was that you have actors and directors who have opinions about, you know, well, would my actor really drink, you know, bubbly or Pepsi or like what, what really goes in the scene. And so when you combine that kind of friction with the VFX friction, think about how much back and forth would happen for a buy. You cannot do that at scale and programmatically. So those are the more interesting problems we set out to solve to Rembrandt, which is, which is one use AI to replace vfx. That's the clearest one. And you couldn't have done that before. Literally the last couple of years were the first and even this year there were some advancements that I could talk about later that really did a massive unlock for us. So that's the first problem. But I think the more interesting business solves are really along moving the matching friction. And that's where I think a lot of startups have died. They figured that they could pay the costs of VFX by having VFX people in lower cost countries, but they didn't think through the higher friction in the selling process around matching and they didn't solve that. And so those are the problems we can talk about.
- [11:04]
C
So this is like a three sided market, right? You have the rights owners, the distribution partners and the advertisers. Is that a good way to think about it?
- [11:12]
A
We attacked it first as two sided. How did we do that? And this was a very interesting inventive step. Everybody else had gone to movie producers. We started at first with influencers because influencers were single decision makers that would come into our platform, pull a virtual Pepsi off the shelf and then once they embed it in their video, they would upload it into YouTube or TikTok and at that point the platforms aren't involved.
- [11:40]
C
Okay, so this is pre production. So or during production you're influencer, you're recording your video and in that workflow you log in the redbrand, grab a Pepsi offer effectively and upload your video and boom, the Pepsi's in the frame.
- [11:57]
A
Totally. And then we just oauth into the platforms to get impressions and charge the advertisers. That's the first couple years this is gonna be added to veto.
- [12:04]
C
I'm gonna ask the Juicero question, why can't I just hold up a Pepsi? I need AI for this.
- [12:10]
A
Yeah, yeah, great point. Because what's happening is. Yeah, yeah, and totally, totally get that. You separate out your, your. Well first of all, the easy answer for that, let me just give you the easy answers fast forward. We now have TV content and we can do it all programmatically and we can have Eric, half a Pepsi and you have a Coke and all post production. So that, that, that's the long term answer for sure. But the short term answer for us is these creators would go ahead and create their content completely without even thinking about what product's going to be put in. It wasn't even the influencer, it was the influencer. This ops person, two hours before uploading, everything's already done, came in and said, okay, which offer do I want? And they would put it in. And the interesting thing is they would be able to put it in sometimes for like 15 minutes. Right, right. So you're not going to go back in time and say to hold up the can for 15 minutes. Right. You've got, you've got much more separation of content creation. But the real answer is you can now apply this to a catalog and programmatically put two different products for two different viewers.
- [13:22]
C
Yeah, right. And, and, but it's not real time. Right. It's a currently, it's a pre production, it's a post production process.
- [13:30]
A
That's what it was for the first two years. We now have wired in actual programmatic that you can, you can in real time shift one thing to the others. Now I want to be clear about what I mean by in real time. It may be some number of minutes before a particular part of show that shows that you look and say, oh, these two advertisers won the bid. You bake in two variants so that a minute later the auction can decide which one to show. Right.
- [14:04]
C
I can imagine maybe this is beyond the tech right now, but you could have sort of a Twitch streamer who takes their camera and somehow, you know, plugged into your system so it's adding in the, the product placement before it ever gets streamed to Twitch.
- [14:17]
A
Correct. Yeah, we don't, we haven't been doing the streaming part. Yeah, but we've been doing the stream. Let me say that wrong. We do streaming. Of course, we haven't been doing the live streaming part yet.
- [14:31]
C
And what, how do the advertisers think about this? Is it amount of time on the screen that determines the currency? Or is it the. Or we're, it can't be raw impressions, right?
- [14:40]
A
Yeah, great. It is, it is impressions. You need to have minimum time on screen and minimum percentage of screen for it to make sense. And it depends upon the context. But you know, you're talking about, it'd be great to have a screen with, you know, 20% of it captured by, you know, a billboard or a product. Sometimes product is too small for 20%, but basically you want a minimum percent of pixels and you want to be, you know, on screen for five to second. Minimum seconds, minimum. Those are the kinds of things that you get. Now in our influencer product, we would sometimes be on screen for tens of minutes. But, but in professional TV shows where the camera's constantly moving, going from scene to scene, it's a different proposition. But I think you're, you, you're, you're getting into some kind of, the, the more interesting part of like why would it advertise to do this. Like, what. What is the real value prop to them?
- [15:35]
C
Yeah. What is it?
- [15:39]
B
Yeah.
- [15:39]
A
So this is the area I think virtual product placement has struggled in also the most, is that it's. It's a. It's a wide canvas. You can do anything. And sometimes people just get attracted to the sex appeal of, oh, wow, that's cool. Let me just show up in the show. But what we have found is that the job to be done that's really interesting for advertisers specifically for virtual placement is when they have new news. So, you know, Sprite now with Terry Lincoln now has an electric model, a particular, you know, promotion. New news is like a good piece. You know, 30 to 40% of a CPG's, you know, budget may have that style to it in their advertising. And this is where what we've discovered is the opportunity to put a very large billboard that naturally sits in an outdoor city scene, where you can say that, except it won't get skipped. There's nothing else to look at on the scene that has a message to it. So the message recall is through the roof. And we found that this performs even better than the 3D product placement that happens to be sitting, you know, the can on the table, which is the more sexy thing the brand wants. But the. But the actual billboard placement in an outdoor scene has very strong attention. You don't have an actor sitting in the scene taking away the attention. It's a beautiful scape. And all you see is right now with cherry. So that one satisfies the delivery of a message that isn't skipped, that's in a natural environment that looks like it belongs there.
- [17:17]
C
I feel like I could ask you, like, a million questions about this, about brand safety and all that stuff. I'm going to kind of leave that to the. Leave that to the reader as an exercise. I'm particularly interested in the pricing issue, though, because you go from basically, here's a half million dollars, you're going to use my car in your movie two years from now, which. How do you know that's the right price to a CPM model? That calculation breaks my spreadsheet. How do you convince the buyer of the correct price?
- [17:49]
A
Yeah, so first of all, we want this to all be programmatic, right? Different shows should have different price, different audiences and all of that. So we want to get out of that business and let the market set the price. That's the first thing. But how do you know that it's working? So, for example, there are brands that are Doing a very significant part, 15% of the budget in some countries of their linear TV budget, doing exactly this strategy. Because what they've done is they've connected to Kantar not at the campaign level, but at the brand tracker level to look and see how much is this technique moving the needle for them. And so I think the importance for this to scale is integration into the tools the brand advertisers using so that they can do their attribution, they can do their modeling and really understand how much they're spending on this. And that's why we've started to believe that even though you can do anything with virtual product placement, narrowing the field down to something like new news with a specific job to be done with a specific format, it's a billboard, makes the job a lot easier on the adoption side. So that you can precisely answer your questions, how much should I spend on this?
- [19:04]
C
Or is this coming from the, from the product placement budgets or from the influencer budgets?
- [19:09]
A
We are very focused on media budgets and we've, as you know, with a couple acquisitions we've done. Most of the virtual product placement we're doing now is in tv. Our work in the influencer is shifting to the kind of stuff we do on top of the space back platform. And so, yeah, let's talk about that.
- [19:33]
C
So Spaceback, longtime member of our sort of extended ad tech community, you acquired them. What was the thinking on the acquisition and what's next for it?
- [19:46]
A
Yeah, this was a deal we were really excited about because about mid last year we launched a new product called Virtual Product Ads. And what virtual product ads did is it solved a brand safety problem that we had with influencers where an influencer would buy a product for us for the first two years, but they didn't get to see the video they were in. Right. Because the influencer was, hadn't launched it yet. And they, two hours before they launch it, they come to us and they're, we're gonna sit around waiting for a brand to approve it or not. So they had to approve the influencer, but they couldn't approve the video. And some brands were really scared. You could see it. Like, I remember talking with like the head of investment at one of the, one of the agencies and they, they were like visibly scared. They're like, oh God, but what if they do something stupid? So rather than try to solve that with technology, what we did is we moved to a model of doing back catalog only, meaning the video had already been posted. You get to approve it and choose it and then we'll virtually place in it. And we launched this product and we integrated the trade desk. And the very moment that we launched it, we saw that the announcement. There was a company called Spaceback there with us. And so we started talking to these guys and we're like, wow, they're doing exactly what we do, except. Except they're not doing the virtual product placement part. They're basically decorating the video, putting the product next to the video in a frame. Like you have a CTV frame and that you have the. The. The vertical that shows, you know, Instagram with someone talking, doing something. And then to the right of it, they put a product. We're like, oh, wow, These guys have 3000 advertisers and are profitable, doing exactly what we just launched. Wow. What if we add AI to that? And not only would we add virtual placement in it, what if we started using AI to generate new scenes and stuff like that? The vision aligned really well with Casey and team. They had a good go to Market Motion that they've tested over years. And they solved one problem that we forgot to solve, which was what that is they only used videos that had already been approved. So then when you went to a media budget, that person didn't have to have an extra friction step of getting approvals. They could just run a campaign. We had the extra step because it was an influencer from our network. And we noticed that that was slowing down how long it took to launch a campaign. So we're like, wow. And they solved the actual problem that we have, man. This is peanut butter and jelly.
- [22:10]
C
And that was your second acquisition of the summer. Right. Which sort of explains why you were deferring on coming on this podcast in the month of August and September. You were a little busy. What was the other acquisition?
- [22:24]
A
Yeah, so that one was a little bit different. That was no equity involved. And so what that one was a company called Myriad, who had been in this domain from 11 years and was very, very different from us. First of all, the reason we bought them is they had a huge number of very good publisher deals with inventory sitting in their system from TV shows ready to have VPP done. And we had been vpp, Virtual Product Placement. Placement. Thank you.
- [22:55]
C
You're venting new acronyms here. Sorry, Right on the show.
- [22:58]
A
Yeah, yeah. Virtual Product Placement.
- [23:00]
B
New to us. Sorry.
- [23:04]
A
So, yeah, so they've been doing virtual product placement, but they're literally the opposite of us in the sense that. How did they start as a company? They won an Oscar for doing the wings in Now I'm forgetting the show where she's a ballerina, Black Swan, was that right. And they did some really good VFX there and it was beautiful artwork. And they said, wow, why don't we turn this in the business? Except instead of, you know, doing art in a movie, why don't we do with products? And our mission in life has always been to not do manual artwork and take this to the point where it's programmatic, done all with AI. But they had 11, 12 years to build up publisher deals and relationships, and so they accelerated our entrance. We always wanted to enter into the professional area once we knew the AI was ready. And it just became ready this year. I'll tell you why.
- [24:03]
C
So to be clear, they're doing kind of what you would call the old fashioned VFX style virtual product placement.
- [24:11]
A
Correct. And just to be fair to them, they weren't all old fashioned. They did use AI to determine, this is a good movie, this is brand safe, it has a kitchen scene, and the kitchen scene lasts 20 seconds. And please approve this kitchen scene. So, you know, they had technology too. They just didn't have the technology that did the AI that was needed to finish the job and make it programmatic. But they had all the publisher deals and there was an opportunity for us to buy that asset and just massively accelerate our time to market on having inventory and having publisher relationships outside of influencers. So that's why we did that one. And, you know, I guess I have a bias because when we were acquired at BlueKai by Oracle, one of the tools that Larry Ellison made available, and it was literally the first day I met him, the day of the deal closed, he looked at me and said, hey, we buy, we hired, we acquired Blue. Kai, I want you to build this group called the Oracle Data Cloud. Tell me who you're going to acquire. And I remember I looked at him and said, you mean to tell you now? And the corp dev guy grabbed my hand and squeezed it to tell me to shut up. And I said, how about I come back next week? That gave me a chance to go to the corp dev team and say, here's my map for the world. Do you guys agree? We went back to Larry and great things happened.
- [25:37]
B
Omar, one of the things that I learned from you early on in Rembrandt's life was the size of the, I guess you would call it the product placement market as a whole. I said 50 billion yesterday, like on a panel. I think that's. It's more like 30 something.
- [25:55]
A
Yeah, there's also, it's in the tens of billions.
- [25:59]
B
It's an enormous market.
- [26:00]
A
It's enormous market and it's global, but give you a sense of like Casino Royale. And in Casino Royale you have the Aston Martin that, that deal took a long time and it's a lot of money Aston Martin put in that. But if you think about the humans that did the deal, because I've actually met them, great people, they think very differently than we think. We want a media deal that can be done. They want a deal that takes time where they can have dinners with the folks, maybe show up at the opening of the movie, take a selfie with the actor. Like these guys think in a very different way than we think. And so we do not target them as the kind of folk we are trying to build a media business. We want this to be fast and efficient.
- [26:45]
C
In defense of that mode, that Aston Martin mode, like having, being able to have your executives, you know, take photos with the stars, has enormous positive effects. It's not just vanity, it's, it's channel sales, it's pr. There's, it's part of a campaign. Right?
- [27:03]
A
Yeah. And they should continue to do that. We have no, no issue with what they're doing. If you think about the model that we're trying to do is the separation of supply and demand. So they could be programmatic and that's why we choose stuff like transition scenes. There's no actors in the scene that can be approved once. And as long as it's a brand that, that, that you have on our approved list, it could be any of a thousand brands that show up in a billboard. You're really not going to care.
- [27:28]
C
So I want to talk about this private to private transactions. You bought two companies as a venture funded, non pe funded company. It's pretty unusual. I've tried to do these deals in my personal career and it's often really hard because the two sides have really. You're stocking, swap, you're swapping stock and both sides have different opinions about what their stock is worth. Can you give us some like gritty details just without numbers, Just like what's it like negotiating this, how do you get to the table, et cetera?
- [27:58]
A
It is hard. It is definitely hard. And you're doing the exact issue is how do you determine relative valuation. So before you get there, you really need to be sure that you have some sort of vision alignment and that you're doing something that's going to be sensible to you. So at BlueKai, we did one of these acquisitions because we needed to build out the DMP fast. And there was this team that was ex Amazon that had a advertising facing analytic data platform that massively accelerated how fast we became a dmp. And those guys were just awesome. They like, they stayed with the company, they built up so much with me. And so, so that one worked really well. Then when we did voice, you know, we didn't have a mobile app. We were all kind of a meeting agent that came into the meeting via the web. And this little three person startup had all these users that were growing fast. They had no underlying AI and we had the AI, but they have really good mobile tech. We acquired them and you know, some of those people are still with me here. Multiple companies later doing stuff. So that is the like more important part than the relative value part. But the relative value part is real. And so anytime you have a signal from investors who have recently invested, that just makes it unemotional. You just go in and say, look, that's what we were priced and this is what you were priced. So yeah, we can make some adjustments, but if either side isn't going to accept that, then maybe you go out and do a, do a little investment to, to let the outside world determine this, if you can.
- [29:44]
C
Yeah. So Eric, from your perspective, you're an angel investor, seed investor in the company, and the CEO comes to you and says, we're gonna have a significant dilution event, I guess would be the best way to put it, in order to grow, you know, how, how do you look at that?
- [30:01]
B
Yeah, I mean it's, it's, it's said differently what Omar said, which is you need to understand how these businesses fit better together. How there's either immediate or like, you know, credible value creation over, you know, a defined time period. And then ultimately you need to bet on the acquirer doing good acquisitions. Right. Because as we know across our careers, most acquisitions don't work out. So you need to feel really good about the acquiring company having the skills and the roadmap to make these things work. And as Omar said before, he's done this multiple times. So we had real confidence and we were an investor in Spaceback, by the.
- [30:45]
C
Way, so of course you were.
- [30:47]
B
Yeah. Being a believer, I know this then is, this is now an N equals one example. But we have strong belief in Rembrandt and where they're going. We knew that Spaceback was a great company. We would, we love the idea of Spaceback coming to Rembrandt and Rembrandt getting, getting Additional capabilities. So this one, it's a little bit of, you know, inside baseball, you know, perfect down the, down the fairway. I don't know what you, what you use the, the, the analogy for, but yeah, it's about having, you know, just like the credibility to understand where this.
- [31:17]
C
Might fit the one.
- [31:19]
A
The one thing I'd add to that, Ari, about making this stuff work is I saw so many acquisitions of Cisco and Oracle and there's one thing that they did sometimes that I always saw, predicts failure and something they didn't do, sometimes that always predicted success and that is come in as a conquering nation, right? Acquire the company, assume you know better, divide them up and say you need to do marketing our way, you need to do product our way, rebuild it on our stack, resell the way we sell, let my sales team sell it. You don't even need to have a company meeting anymore because these guys went one way, these guys went the other. Like, why the hell do the acquisition, right? You're such a conquering hero like, and so, and people are full of egos. So being able to find a way to allow people to keep the magic that they have, yet still explore synergy is the hardest part.
- [32:13]
C
Yeah, I heard this funny story from a former Blue Kai person that like the day the deal closed, the first thing that happened was that the Oracle lawyers show up at the office and make it very clear that dogs are, aren't welcome. There's a no dog policy in the office. Any you want to comment on Oracle's hatred of dogs?
- [32:32]
A
Yeah, I don't know, I don't remember that. But I'll tell you one thing I can say about the Oracle experience, which proves my point. They divided bluekai into two parts. One part went into the Oracle marketing cloud and one part went into the Oracle data cloud. They basically said, Omar, you go around the marketing the data cloud part and to be honest, I love the DMP part even more because it was faster growing and is what drove us probability. But as a point of fact, this was a social experiment. The part that went into Zeta cloud was untouched. We could do exactly what we used to do. And the part that went into the marketing cloud was divided up into seven pieces. So the higher growing part after acquisition slowed down and the slower growth area accelerated.
- [33:20]
C
Interesting, interesting.
- [33:21]
A
That is amazing social experiment.
- [33:23]
C
It's like a split twins experiment. So, all right, I think that's a great stopping point. So we're going to take a quick break and we're going to come back with a lot of advertising Week, trade desk, various retail media networks, usual around the bases news. We'll be right back.
- [33:45]
A
All right.
- [33:45]
B
And we are back. It is Advertising Week, so that means there are like so many partnership announcements and product announcements and deals and rumors. We'll try to get to as many as possible and we'll try to actually talk about Advertising Week a little bit because we were all there. But before we do, Omar, you had one more thing to add.
- [34:03]
A
Yeah, I was just going to add one last thing about Spaceback is when we did the deal, we knew we were going to merge these companies based on what we saw about the product and sat with that leadership team. But last week we. We actually did a company to company thing where we were with the entire company for a few days out in Charleston, and they're. They're better than the Diligent showed this is. And, and it reminds me of like the old days of Blue Kai. Our go to market team was just such an awesome team and they stuck together. Well, that. That's the. What I see in that space back team. The people element of it is. Is really cool.
- [34:42]
C
Anyway, I met Casey back when he first. First first started, and he said, you know, imagine if instead of banner ads, it was a valentine from your girlfriend. And I was like, that's a dumb idea, but nice guy. I'm sure it'll work out. So congratulations.
- [35:02]
A
Yeah, he worked out, but it worked out.
- [35:05]
C
Yeah, it worked out.
- [35:08]
A
Okay.
- [35:08]
B
All right, let's talk Advertising Week. Give the listeners who may or may not have been there in New York a vibe check. Why don't we start with you, Ari?
- [35:19]
C
Yeah, It's Advertising week, is big and messy and takes over the whole city, and I love it. I actually have come around. I used to be the kind of person, you know, regular listeners may know. I like to complain a little bit. I'm a kvetter. I'm a New Yorker. I've come around, I'm like, okay, fine, I don't love the venue. The emerald people know the venue sucks. They're not dumb, they're smart. It's just. It is what it is. But then you go out and you see seven people you barely know on the streets on 6th Avenue and they go to a party. Then, you know, I did a hackathon. I was a judge of a hackathon. You know, all kinds of interesting stuff going on and, you know, it's fun, it's exciting.
- [35:58]
B
I'm shocked. I'm shocked. The Kavetcher in chief is coming around. That's great. What about you, Omar?
- [36:08]
A
Well, this was my favorite advertising week for a very silly reason. I happened to have breakfast with the CEO of Emerald, who's an old buddy of mine. And what an amazing individual this guy is at a ve. And so after breakfast he walked me into the show and so I got to get a slightly different experience than the usual times. I was like, wow, this was cool. So. But the content was great and I think this is one of their biggest in terms of number of attendees. So good stuff. And I agree with you about, you know, venue, but I'll keep my mouth shut.
- [36:50]
C
Yeah, I talked to Herve about it as well. He's super nice guy, really interesting. He treated me and some other CEOs to a little dinner. Thanks to the Emerald team for that. But yeah, they, they're aware that the venue is not ideal and they're looking.
- [37:04]
B
At options for those who are not in New York or haven't attended. The venue is in I think a defunct department store, Manhattan Middle, in the middle of the Manhattan mall in the middle of Herald Square. So, okay, so. So it's a defunct mall. So picture being at a mall having multi levels, escalators, like storefronts on all sides, just kind of gutted out and filled with advertising and advertising technology, content and vendors and chaos. It's pretty unique.
- [37:40]
C
The mall is a big improvement because it used to be in a movie theater on the Upper west side. The sixplex on Link Lincoln Center. That was terrible. So the mall is an improvement. Also side note, many New Yorkers know this. The mall was built using the money that Mel DeMarcos and her husband stole from the CIA back in the 80s as Ronald Reagan connection to all of this.
- [38:03]
B
Final question on the advertising reform before we get into some of the news of the week. Anything interesting in terms of conversations? You had things on people's minds, you know, like general vibe check around the industry.
- [38:17]
C
I think that the trade desk drama continues to be a conversation that the people who I know like to talk about, you know, buy side, sell side, blah blah, blah. It's a little, it's a little tiresome, but that's what people want to talk about. And for me it's all antitrust, all the time. And so, you know, I've become a very boring guest to have at your party because that's all I want to talk about.
- [38:37]
B
That's all anybody wants you to talk about since they got.
- [38:41]
C
Yeah, basically.
- [38:41]
B
What about you, Omar?
- [38:45]
A
Yeah, I mean, obviously I, I just born the joined the board of the trade desk. So I I do pay attention to that and, and while I won't talk about the trade desk, I will comment that Jeff's the same guy that I, I really liked back in the day when he was so deep on this idea of an open, transparent, fair Internet and nothing's changed. She really, really cares about that. So it's refreshing to see somebody who's had such a long held vision on where they want to go. But other than that, I won't talk about their actual strategy.
- [39:25]
C
Omar, are you going to get Jeff on this podcast? Because he's been flirting with us for a really long time and he remains not on this podcast.
- [39:35]
A
I'll certainly ask.
- [39:37]
C
Thank you. Appreciate it.
- [39:39]
B
Thank you listeners. You can see how hard we've been working to try to get Jeff Green on for you. We're actually putting in the work. I would say just one addition on the conversations. It's a big difference from last year. I think last year one of the things that we were commenting on was that there's not enough conversation about AI. This year it's flipped. It's all AI and I think a lot more examples of how things are coming to market and a lot of questions around the next 12 months and what really is going to take off, how these shifts around traffic dropping from search and coming from LLMs, all that stuff seems to be quite imminent and a lot of conversations around who might be the agentic winners. So I was pleased to see that.
- [40:33]
A
Yeah, I agree with you. Lots of discussion about what it means, what the disruption around agentix going to be for marketing. And the other area that I saw, and I may be biased, is a lot of talk about the creative unit and in content. So everything we do is about placing it in content. But if you think about what's going on, LLMs, it's another variant of in content, which is what would it mean for you to have a sponsored sentence? And you know, what can you do even if you're not sponsoring, just to make sure you're influencing the sentence. So there's a lot of attention to that.
- [41:04]
C
I also saw a presentation from, I forget her name, but the person who's the head of publisher relations at Perplexity. And it was very different conversation from maybe a year ago. They were not talking about ads, they were talking about, you know, their browser and revenue sharing on paid subscriptions from the browser and stuff like that. And she, she said like you may still see sponsored ads in the search results, but I don't expect to see them for very long, more or less Maybe I'm mischaracterizing, but that it was a pretty big change in Perplexity's attitude towards ads.
- [41:35]
A
Yeah, yeah, I was there.
- [41:37]
B
Yeah, I was there. I got the same takeaway. And then, you know, finally just, you know, the. The what, What. What are ads in a. In. In completely virtual environments. I finally got access to the Sora app and I've been playing around with it, and it's TikTok, but it's all AI, and you can imagine, you know, what this becomes from a media platform and what ads might look like. It's like a whole birth of new categories or still kind of rebuilding the other categories.
- [42:05]
C
Yeah, I mean, I think that it definitely is closer to the future than Facebook's vibes, which is just AI with no human touch or very little human touch. I don't know if it's the right thing. I mean, TikTok Plus AI sounds like the right combo. AI Plus TikTok seems maybe the wrong way. I don't know.
- [42:23]
A
We'll see.
- [42:25]
B
The way they're doing it is so cool.
- [42:26]
C
Omar, this must be magic for you. Like, you're, like, just champing at the bit. Give me an API and I'll put those ads in there.
- [42:33]
A
Yeah, yeah, totally. Of course, that's more content. And we're actually even partnering with a company that, you know, will take all this video of a real world, you know, actress, influencer, and then. And then kind of not quite like it's similar to, like, Sora in approach, but then they create branded content for brands in different parts of the world based on this. So so many people are attacking this from different angles.
- [43:02]
B
Yeah. Final thing, on Sora, I like the way they're giving the opportunity to put you in videos with your friends or with creators that have opted in to allow you to insert them into your videos. So, like, I've been creating videos of me fighting Jake Paul for the past week. It's so. It's so hilarious.
- [43:24]
A
Our kids are not going to know the difference between reality and It's. It's amazing.
- [43:31]
B
Yeah, for sure.
- [43:32]
A
This is going to be really bad. I mean, if you remember, you know, starting in, like, 2008, because of the iPhone, there was a generation that. That. That grew up on that. And as those people grew up and became, you know, teens, the amount of, like, anxiety and depression skyrocketed worldwide. Right. And that's because people started, instead of having real world interactions, watching all these curated interactions and comparing their real life to the curated ones. They're like, wow, why does my life suck? So think about what's going to happen now to, you know, the kids who grew up on, on this like world. There's going to be some interesting ramifications. Agreed.
- [44:11]
B
All right, let's get into the news. Amazon DSP named transition partner for Microsoft Invest. I don't know if this was a surprise so much as the next announcement out of Amazon over the course of a month or so. I think also what was interesting is Microsoft Monetize the SSP was named a preferred SSP in Amazon supply program. So while they're sunsetting the buy side, they're not exiting advertising or media completely because they're kind of continuing to monetize on the supply side. What was your take on this one, Ari?
- [44:52]
C
I think sounds like a quid pro quo preferred transition partner that doesn't really mean very much. Right. It's maybe some download upload capabilities, get on a joint phone call. You know, customers can transition wherever they want. Right?
- [45:08]
B
True, true. But if they make it easy, they could retain some good business.
- [45:14]
C
Yeah, yeah. I mean, yeah, every dollar matters. Right. And so good for Amazon. You know, they have the hustle, they're making deals. It seems, seems like another piece of mini bad news for the trade desk in their ongoing competition with Amazon that, that Jeff Green denies is happening. So yeah, so good for Amazon.
- [45:37]
B
Let's talk about ttd. So last week was the two posts about the industry and the wrapper open ads. So open ads creates a separate branched off auction based on pre bids codes that still uses the transaction IDs. Is there anything more to cover on that one? I mean people are talking about this all week. Just want to touch base on it.
- [46:00]
C
Yeah, I talked to some people about this and it doesn't seem like it's a big market mover in any way. I think publishers are cautious. They're saying, well, why should I have another wrapper or why should I, why should I switch wrappers or why should I have another wrapper? Because currently most publishers have two wrappers. They have pre bid plus Amazon's tam. So do you want a third one? If you do have a third one, why, what's the benefit? Am I going to be forced to use open ads in order to benefit from open path, the inventory? A lot of open questions and people waiting, waiting and seeing to what the reality is.
- [46:37]
B
Yeah, makes sense. One other announcement on the TTD front that was actually really cool. So they're offering programmatic sponsored ads on the GoPuff app via Kodi, who is a specialized ad Server for commerce media and retail media. So you know, one of the rubs against sponsored ads are that they're hard to buy programmatically. They're not yet scaled. You can't use rtb. This seems to be one of the first forays into, into doing that. So I was impressed with this one.
- [47:10]
C
No comment. I don't, don't have any opinion about this. I don't know what Gopuff is, I don't know what Cody is. So I'm very the wrong person to ask.
- [47:19]
B
Oh, Gopuff is I think a delivery app and Cody is like I said, it's sort of like they're a competitor to Kevlar. So it's you know, ad server and monetization for RMNs.
- [47:32]
C
I'm just an old man trying to keep up with these youngins and their ad tech companies.
- [47:37]
B
This is one of the two. Go ahead.
- [47:45]
A
Yeah, yeah, I'm, I'm, I'm, I'm unfortunately less familiar with all the details there but appreciate you highlighting a good piece of news. So I like it guys.
- [47:57]
B
There's two fast growing categories. It's all things video and CTV and there's retail commerce media. This is a good sign in the direction of commerce.
- [48:07]
C
Congratulations.
- [48:08]
B
Thank you. I feel like I'm getting congratulated. Let's move on. Index launched data Marketplace a week after TTD had some significant changes in their marketplace for data. This one for sell side curation. So interesting on this front. Index is not collecting any deal curation fees for use of the data marketplace. So it feels like this is a win win for publishers.
- [48:32]
C
Yeah. Index from the start has been very much highlighting their deals fee structure as an advantage in their curation marketplace. I cannot comment on how real it is how the other people price. But Andrew Casale definitely feels like their fee transparency and lack of fees in various paths is a competitive advantage where they'll make money on selling more inventory. They don't want to make money on the data. So that's an interesting positioning and I don't know what the direct competitors are doing in the fee area.
- [49:04]
B
Yeah. The other announcement Index is Firefox's first programmatic partner.
- [49:10]
C
Yeah, I like this one too. So Firefox had a effort to bring in ads maybe five or six years ago. They didn't really go anywhere. They hired. What's his name. Well, Darren Herman. Thank you to run that. And, and you know it came under the same sort of problems of like we don't like ads but we want ads. They should be good ads. Not Bad ads and it sort of, I think it didn't really go anywhere. So nice to see them opening up. This is, you know, more of this kind of hedged garden approach, limited availability, you know, probably a lot of curation, old fashioned curation around the quality of the ads. You know, it seems like it makes sense.
- [49:54]
B
Yeah. Firefox is, it's not clear exactly what market share they have. It's 2 to 7% on desktop, depending on who you ask, and about 0.5 to 1% on mobile devices. So while those are small numbers, it's still probably a decent total audience size.
- [50:14]
A
Good inventory. Yeah. Yeah.
- [50:17]
B
All right, back to RMNs, three RMNs, maybe two and a half RMNs launched this week. So the first, I thought this one was interesting. PayPal launched an SMB ad network. So it is all of the SMBs that have a digital presence that work with PayPal. They can start getting ads at the point of checkout, which is actually a great place to put ads. Non intrusive, kind of all incremental, and then channel those profits into spending money on PayPal ads or to their bottom line or something like that. You know, question around, is this inventory as good as, you know, a sponsored listing that's you know, super, super endemic. But I think it's a good effort to try to, you know, build up some supply in a unique place. What do you guys think?
- [51:08]
C
Yeah, I mean, Dr. Mark is on the move. He will be at Marketecture live. I'll be asking him about this program on the stage. But yeah, like PayPal has this great advantage which is huge reach, you know, enormous number of retailers use them. But the disadvantage is fragmented reach. It's a lot of smaller ones. So this is kind of a just give me the money program. You know, you put it into your app and boom, you see some ads and you make some money. Competitive with rokt, I guess, or okt. So seems like a no brainer.
- [51:40]
A
Yeah, and it also depends on that the, the granularity of the targeting option on the checkout page because it's very granular and you can aggregate enough that could be pretty helpful.
- [51:54]
B
Yeah, yeah. From the release it said that they're giving the SMBs control over categories. So if you are a pizza place, you won't have other pizza places showing ads on your property, which I think is a good thing. But the real key to your point, Omar, is how much is PayPal using the transactional data to inform these ads? And you can do some presumably really, really interesting things with, you know, understanding Payments. So it's one to watch. Two others, Amex launched Amex ads.
- [52:28]
C
Good name, good name.
- [52:30]
B
It is a good name. I would not get confused about Amex Ads. Who owns it? And I think it's starting with Amex Travel. This is maybe the inverse or opposite of PayPal where this is Amex users, which apparently it's 30 million US and obviously with the different cards they probably can target based on certain demographics and household income. Amex Travel obviously can be tailored to high end. So this is a cool one, I thought.
- [53:01]
C
Yeah, I mean the whole finance sector is diving right in, right?
- [53:05]
B
Yeah, yeah, for sure. On the heels of Chase and others. And you got duolingo. So Duolingo launched ads. This one like Amex. It was overtly saying this is direct sold. We don't want programmatic. We're not doing programmatic. We're being highly conscious of creative and format and so on and so forth. So it's an interesting, you know, maybe mini trend of, you know, dipping your toes into ads but you know, being pretty highbrow about it. We don't want any of that programmatic stuff.
- [53:40]
C
You know, they missed the opportunity here. The opportunity is to make people translate brand messages like, you know, say in Espanol. Tropicana is delicioso. That's the opportunity here.
- [53:54]
B
What a no brainer. I mean, maybe they are. If not tall Art. He's got ideas because.
- [54:02]
A
Because everybody wants to hear my terrible French accent.
- [54:08]
B
Amazing. All right, what else we got? Applovin gets investigated by the SEC over data collection practices. So the stock dropped something like 20% after being up massively 80% this year and 700% last year. The SEC noted that they are responding to whistleblower and short seller complaints and the company is not being accused of anything. But such a stock that is, you know, such a high flyer. I think it's, you know, super subject to anything like this.
- [54:44]
C
Well, I mean they're, they are being accused of something, just not by the sec. A lot of people are accusing them.
- [54:50]
B
Of something, but investigated by the sec. Super clear about this. Yeah, it continues to be the one that everybody's watching for sure.
- [55:00]
A
Yeah, I mean I heard that a lot yesterday.
- [55:05]
C
There's some interesting lesson here about the fact that like the mar everyone mobile app market is huge. Like there's a lot of value being created there by sellers and buyers and intermediaries. But if you somehow like concentrate it into a flywheel that really works, the value just explodes. It's sort of like the hydrogen bomb approach. You have to set off. You have to concentrate it to get the value. And I can't reconcile my brain with a $200 billion market cap for this. But at the same time, right, the numbers are there. So, you know, we. We are yet to learn the lessons here. It's a work in progress, I guess, is the best I can offer.
- [55:50]
B
Yeah, for sure.
- [55:51]
A
And it's a complex ecosystem that's hard to the regular people to understand.
- [55:56]
C
Yes.
- [55:58]
A
And when a lot of money goes into a complex ecosystem, you remember mortgage backed securities, There was lots of opportunity for people to package odd things. So I actually am not accusing them of anything, honestly, I'm not. I'm just saying it's a complex ecosystem. So.
- [56:13]
C
Yeah. So we've been going pretty long. I think we gotta call it there, Eric.
- [56:20]
B
No problem. Thank you, Omar. This is great, man.
- [56:23]
A
Awesome, guys.
- [56:24]
C
Yeah, Omar, this is fantastic. I feel like I really learned a lot because unlike Eric, I have not been involved with the company. I didn't really know what you were doing, so it's super interesting for me.
- [56:33]
B
Cool.
- [56:33]
A
Thanks for having us.
- [56:34]
B
We'll see you next week, everybody.
- [56:37]
A
Thank you for subscribing to marketecture.
- [56:39]
C
New interviews are added every week at marketecture TV and your favorite podcasting app.
- [56:48]
A
Foreign.
- [56:52]
C
Thank you for listening to the marketecture podcast. New episodes come out every Friday and an insightful vendor interview is published each Monday. You can subscribe to our library of hundreds of executive interviews at marketecture tv. You can also sign up for free for our weekly newsletter with my original strategic insights on the week's news@news.marketing. and if you're feeling social, we operate a vibrant Slack community that you can apply to join@adtechgod.com.