← Marketecture: Get Smart. Fast.
Marketecture: Get Smart. Fast.
Episode 109: MMMbop with Ron Jacobson plus StackAdapt's huge funding round
00:45:32
Loading summary
Transcript149 lines
- [00:01]
Ari Papera
Exclusively on ESPN. UFC 312, Saturday. Reigning middleweight champion Dricus Du Plessis defends his title in a rematch against Sean Strickland. And Zhang Weili defends her strawweight title against undefeated Tatiana Suarez. UFC 312, Saturday at 10pm Eastern. Buy it on espnplus.com ppb.
- [00:40]
Eric Franchi
Welcome to the Marketecture podcast. I'm Ari Papera, I'm joined by Eric Franchi and our guest this week is Ron Jacobson from Rockerbox. We asked Ron on the show following last week's announcement of Google's MMM tool Meridian and just generally things being really interesting and in flux in, in the measurement area. Eric, do you do a lot of measurement? You're excited to hear from Ron?
- [01:01]
Ari Papera
Yeah, I mean, measurement is one of the big stories of the, of the year, I think. Number, number one, we have a couple of measurement companies in the, in the portfolio. So it's an area we have a, we have a thesis on. And like, Ron is the person that I think about when I think about measurement. He's got such a great purview just being the, the CEO of, of RockerBox. And his clients are actually a little bit different than, you know, the ad tech ecosystem. They tend to skew towards, you know, walled gardens and, and media. So I think he brings good fresh thinking around the topic.
- [01:31]
Eric Franchi
Yeah, he's our measurement guy. We got a guy for everything nowadays.
- [01:34]
Ron Jacobson
I didn't want to say that.
- [01:35]
Ari Papera
I don't want to say that, but he kind of is.
- [01:37]
Eric Franchi
Yeah. Also speaking of which, a lot of funding, news. So Your portfolio company, TV Scientific, raised a big B round with Roku and StackAdapt raised a fairly huge $235 million round, which was kind of eye opening.
- [01:53]
Ari Papera
We're going to spend a lot of time on that one. That shook AdTech X right up. It was highly controversial.
- [01:59]
Eric Franchi
So Marketecture also publishes the Ad Tech God podcast. You should listen because AdTech God had the CEO of StackAdapt on as a sort of emergency pod yesterday. I think he's the only podcast that had Stack Adapt's firsthand account of their funding. So check it out. Yeah, I know. He's like Ad Tech God's becoming like a little journalist. So good for him. And another reminder that he is not me or Eric. He's a different human being. He works for me. I know who he is. I signed his paychecks. He's a real human being. Works for me, not me.
- [02:32]
Ari Papera
He's also not little. Don't throw people off and thinking there's like A small person they're going to look for at the market conference.
- [02:39]
Eric Franchi
He's huge. He's huge. And last, we couldn't. We'd be remiss if we didn't give a plug for marketecture Live. So the Market Live event is coming up in March 17th. We're adding new guests every week. This week, we announced that Alex Sherman, the CEO of Bluefish, will be on with Eric Linuti from Omnicom. And what they're going to be talking about is the future of AI for brands. So if consumers aren't using search to find things, how are you going to search engine optimize? You're not. You're going to use something different. So Bluefish is one of the companies that's really innovating in that space. There are a bunch of other ones. So he will be at our conference talking about that, among many other really great conversations. So go to MarketLive.com to register now before tickets sell out. All right, let's get into it. We will be back with Ron. All right, Ron Jacobson from Rockerbox, thanks for being on.
- [03:37]
Ron Jacobson
Hey, glad to be here.
- [03:39]
Eric Franchi
So this is your first appearance on the podcast, right? We had you on the vendor interview, right? No, we've had you on both the vendor interview and the podcast. Wow.
- [03:48]
Ron Jacobson
I'm glad. I'm so memorable.
- [03:49]
Ari Papera
If you remember that. All right, if you remember that he was on before, would you have invited him back for today's episode or.
- [03:55]
Eric Franchi
It's a good question. We have to dole out the repeat visits very carefully or else it'll get stale.
- [04:03]
Ari Papera
I advocate that with so much going on in measurement right now, Ron deserves the repeat.
- [04:10]
Eric Franchi
Yeah, Ron better give us some incremental insights on this pod, or else it'll just be repeat. All right. No, we wanted to have you on in all seriousness, because measurement is a very hot topic. There's a lot going on with, in particular, media mix modeling, which I know is not exactly what you do for a living, but it's very related. So why don't you just start by giving us the state. What, in 2025? What is the state of advertising measurement and what's going on?
- [04:37]
Ron Jacobson
So there are three main things that are going on in measurement today. There is like a battle over methodologies. How are you doing your measurement? Is it minimiz modeling? Is it incrementality testing? Is it mta? Uh, so that's kind of topic number one. Topic number two is how do you actually use these different methodologies? I think this is like a really big problem. It's a lot of folks just like go and suddenly say I'm using median modeling but they have no idea how to operationalize it. So kind of what's the system day to day to leveraging measurement? And the third one, and this is just started to get some legs I think middle of last year. But it's going to heat up in 25. It's feeding back attribution results into ad buying platforms to optimize against. So those are the three main areas. It's again, which methodology, how do you actually use the data and how do you make sure your platforms are optimizing against your source of truth?
- [05:21]
Eric Franchi
So on the methodology is it, do you have to choose one? Are they either or?
- [05:25]
Ron Jacobson
I mean they aren't. I think they all bring different things to the table. I think a lot of advertisers and my purview is generally advertising facing at Rockerbox anyway we work primarily with brands so that's where we kind of get most of our information from. I think a lot of brands often are going for whatever the hottest measurement methodologies of the minute. And you're right. MMM has been very hot in the last year. Incrementality testing has also been really hot in the last year. Having said that, it's almost been like a sine wave over the course of the past five years over when they're hot, when they're cold. When they're hot, when they're cold. Incrementality testing had a really hot period like four or five years ago. Cold for two years. It's hot again now. Mmm I actually think was hotter in 23 than it was in 24. It ebbs and flows. I think a lot of advertisers just kind of run towards whatever is being talked about in the moment and then the problem is they often have no idea what they're actually trying to get out of it. I think like the classic example here is I had a client come to me talking about mediumence modeling and when we asked them why do they want it, they said to understand the path to conversion. And like that is complete opposite. MMM does. But I think that speaks a lot to just there's a lack of information on what these different methodologies do. So to your exact question, no, you don't need to, you don't need to be forced into using one. But that kind of comes into the system problem. Like if you're using multiple methodologies, how do you actually make decisions? And that that becomes really complicated and that you hear words like triangulation thrown around, which I have strong opinions about. But it's difficult to make use of these different methodologies in a practical way.
- [06:46]
Eric Franchi
Well, incrementality, let's tackle that one first. So that was really hot. I know like five, six years ago when people were talking about ghost bidding where you'd have like some, some users who were not exposed to the ads. And my understanding is that ghost bidding really fell out of favor with cookie deprecation because if you can't identify the user, you can't create a holdout group. Has that changed?
- [07:07]
Ron Jacobson
No. Ghostbitting is. You're completely right there. I'd say the hot current moment of testing is geo based tests.
- [07:13]
Eric Franchi
Right.
- [07:13]
Ron Jacobson
Testing control, figuring out statistical, statistically similar geographies and either doing a whole lot test or a scale up test. So that's kind of replaced the user based version of ghostbittding and that's very popular now and can be really, really effective. There are pros and cons against it. Cons are it takes time, actually has a revenue impact because you're potentially not serving in an area. But it's considered like the gold standard of measurement in the sense of the most accurate way to actually engage what's happening with your marketing.
- [07:41]
Eric Franchi
And when you think about incrementality testing versus say media mix modeling, aren't you also thinking about different sort of time frames of decision making? You need to make decisions on what to change in the meta UI and how to change how you're targeting whitelists and blacklists in Trade desk. That's very short term. Whereas there are other decisions you're making on an annual basis, a budgeting basis.
- [08:05]
Ron Jacobson
Yeah. And I think this is a huge part of the problem is that advertisers come and they want mmm or testing and they don't think about what am I going to do on Monday when I get to work and Wednesday I need to make changes. And neither of those methodologies can actually tackle that challenge. There's some companies out there that claim that they can do like daily mmm and things like that. I think it's all BS and honestly I think it's even just like trying to solve the wrong problem to be frank. And that's where at least the way that I view the world is. You need kind of your day to day hub of measurement. I think that has to be either like a Google Analytics type of day to day attribution. That could be app platforms themselves. They're self reported data. We can talk about the pros and cons There it could be a rocker box type thing with our sort of like MTA type products. But that's like your day to day hub of measurement. I think then you want to. What ends up coming from there is you see things that seem off. So it's just like CTV seems off or can I scale into this channel? You have these sort of questions that you want to validate and that's actually where I think MMM and testing can be really good validation frameworks to validate what you see from that day to day source of information. But again that's based off historical information. So it'll take some time for you to get that answer to validate. And then eventually once you get those responses, we see clients then calibrating it back to their daily data with some sort of multiplier.
- [09:15]
Eric Franchi
Yeah. Like one example would be if you're using a sort of a vanity style metric like a view through in an open web and that may justify spending more money, but then on a quarterly basis you run your mmms and it's like, well that's actually just not driving actual cash registers.
- [09:31]
Ron Jacobson
Yeah, that's exactly right. And I also think it's, I mean once you get to mm, it's a broader type of measurement. Right. It takes into account things that you just couldn't do on a daily basis. It takes into account competitors and brand search and inflation and like all these exogenous variables. So like it, it fundamentally answers a different problem on a different cadence. So I think that's where as a brand you need to figure out what's that day to day kind of pulse of your business. And then how do you make those quarterly biannual annual decisions that mm, incrementality testing can be really helpful in doing. But if you run your business exclusively off mmm, it's kind of hard to make a decision on any given day. You know, what the hell should I do?
- [10:03]
Eric Franchi
Yeah, exactly. It's like, you know, California is west of New York, but not clear how to get there. So why is MMM so hot right now? And we want to. Secondarily, we're going to talk about Google and Met as investments here.
- [10:15]
Ron Jacobson
I think a lot of the reason it's hot is because of their investments. I mean you have the biggest app platforms in the world, Open sourcing, Meridian with Google, Robin with Meta. So obviously the weight that they bring to the industry is huge. I had this thing called lightweight that was a market for a while. It was pseudo associated with Google. It's kind of like one of those 20% projects it definitely leveled it up and polished it with Meridian and made it a bit more sophisticated but I think by and large it's that, I mean when you have Google and Meta in market with open source MMM libraries there's obviously gonna be interest there. I think there's the question asked like why, why are they so.
- [10:47]
Eric Franchi
Yeah, why?
- [10:48]
Ron Jacobson
I mean this is my, my hypothesis at least on the Google side is MMM is ostensibly better at measuring more upper funnel channels and I think a really important upper funnel channel that Google has a lot of ownership and stake in is YouTube. So I think that if you look at long term trends, who knows where search is going with especially like gen AI and things like that. I think having a methodology market that better suits themselves through upper market is going to be helpful. And also I mean there is the whole whatever happens is cookies and with the Google opt out. So I think they need to hedge as well and just have different methodologies that they, that they feel confident in likely make themselves look good.
- [11:20]
Ari Papera
Ron, do you see clients like using these open source tools? Do you see it seems like with the Google announcement there's a partner program where a lot of like specialist agencies are going to be using these and presumably kind of creating products for clients. And what does it mean for the ecosystem of companies that just do this as a business?
- [11:38]
Ron Jacobson
It's actually a pretty interesting parallel I think in the open source versus closed source and media mix modeling to kind of like where AI is going. I actually think any company that's trying to have closed source MMM is going to be in a really tough position. Like it's just going to be very hard to keep up with Google with Meta. There's this other company, pymc that's like a pseudo open source MM company and a consultancy layer on top of that. They all have lots of contributors and they're able to advance really rapidly. So I think it's a tough position to be in if you're trying to build proprietary models these days to be frank in terms of do we see clients leveraging them? The short answer is yes. Like I've seen, I'm going to say all I'm, I'm being glib there but like almost everybody has like dipped their toe into trying one of the open source models. The problem arises with what do you do after you do the first build and that's I think where you start to realize the model is in a way the commodity, the actual business value is everything before the model's made and after the model's made is Funny, it's the same problem that MM has had for probably 10, 15, 20 years. It's how do I centralize all the data, how do I get it clean, how do I get it normalized, how do I do this in a repeatable cadence? Which is surprisingly hard, especially if you're a global company with multiple agencies across the world and different logins to Meta in the US and the UK and having to aggregate that data, it just becomes a hard challenge in that regard. The second problem is cool. I built my MM model. Now what? And is this really going to be something that gives enough confidence to our business to make fundamental changes in how we allocate our media spend? I think that there's apprehension when there's a one off model that's been made with an open source tool that's kind of like a side project in the data scientists in the company to actually make large scale business changes on top of that. And that's to your point are where I think the consultancies or Rockerbox and some of my competitors obviously have a role to play in the sense of like hey, this is the model. Here's how you actually action on it from a budget perspective, from a day.
- [13:21]
Eric Franchi
To day perspective other than just directionally. And I just have to presume that Google and Meta believe the outputs of good MMM modeling will be beneficial to them, which is a little shocking because of how much of the advertising market they currently absorb. They actually I you can infer that their belief is that they're not getting their fair share. If you are modeling. Do you think that's their point of view?
- [13:47]
Ron Jacobson
I do and I think there are two ways to, to kind of see that. One is if you look at it from like a day to day attribution perspective, MTA, you can't really get views through data from YouTube into that. So YouTube is underrepresented in third party attribution or NTA. So that's one. The second part is if you look at Meridian, actually one of the key things they launched in it was a data set around Google's. I forget exactly what they call. I think it's like their search query volume where they made a data set public, an API public where you can get really, really granular impression levels and search levels against giving queries and you can funnel that into Meridian. So I think that they actually believe that that's going to make their MMM even better for search as well. So in a way they're trying to make MMM that I think helps them out on both the search side and on the YouTube side I think less so on the search side but definitely on the YouTube side. They're underrepresented from deduplicated attribution.
- [14:35]
Eric Franchi
Right, that makes sense because MTA models would require, you know, impression level data or logs and which they don't recognize. Right. Well do people use ads data hub for that?
- [14:45]
Ron Jacobson
Some do. I mean we've explored it ourselves. It's something I'm not against doing at some point, but it's a mixed bag to be honest with you. Yeah.
- [14:53]
Eric Franchi
So in general off of mmm, but onto just general measurement, have we permanently entered the world where we're not looking at actual deterministic data and we're only going to be looking at statistical or correlations for measurement?
- [15:11]
Ron Jacobson
No, I don't think so at all. I think that everything we look at will have a statistical or correlation part on top of it. But I think there will and there, there is still deterministic aspects that are kind of the foundation, I'll say what I mean there post iOS 14.5 I actually think, you know, it was done for privacy's sake but it's actually led to a lot of platforms being more open to sharing some deterministic data with third party measurement providers. And Rockerbox has been a market with this, with partnerships with like Reddit and Pinterest and Snapchat same a few where we are now working with them and getting deterministic view through data and others are doing this as well. So I think this is a trend that kind of came out of iOS 14.5. So in a way I think we're actually getting higher fidelity data and I think a lot of advertisers are doing a much better job at capturing first party data and I'd recommend all advertisers do that. Where I'm saying that there is a probabilistic component is once you get that deterministic measurement, there's an extrapolation being on top of that extrapolation try to compensate for opt out rates, match rates and this is even done in Google and Meta's attribution. They do, I think they call it modeled conversions. So they even themselves in their deterministic attribution add a probabilistic component on top of it.
- [16:17]
Eric Franchi
Right.
- [16:18]
Ron Jacobson
So I think there definitely is still a basis of deterministic data but it's going to be layered on top of with some sort of probabilistic measurement.
- [16:24]
Eric Franchi
So in a fragmented environment I'm sort of giving you a softball for your company here. But In a fragment, if you're a CMO and you're buying on 20 different digital channels and they all don't cooperate with one another, what's the marketer to do?
- [16:38]
Ron Jacobson
Candidly, I think you need a day to day source of measurement. I think that should be some sort of attribution mta. I think Rockerbox is great obviously, but there are other companies out there and I do think over time you're going to want to validate that with minimum and incrementality testing and bring those results back to your day to day measurement. I actually think in a way, like if you have that sort of system in place and you know what you do every single day, every quarter, every half year, you could be really effective. I think when you're in a bad position is if you kind of take this term triangulation, which is this broad term that you're going to hear thrown a lot at measurement, which basically says pick whichever type of measurement you want and start there. Just use other ones whenever you kind of feel like it to kind of validate. And the problem there is like there's no system. It's just kind of like you, you, you get one number and you get another number and you're gonna just like. I've never heard anybody be able to define what triangulation actually means, even though it sounds really smart. So that's what I would not do. I would not focus on triangulation. I'd focus on just the non sexy thing. Have something you can go in every single Monday, Tuesday, Wednesday, see how your business is performing, have a method to check it and refine it based off what those checks actually report out.
- [17:39]
Eric Franchi
Yeah, what I've heard about triangulation I think is that every once in a while you do a very definitive test. For example, I once worked with a client that just turned off YouTube for two weeks entirely to just see what happened. So you do a test like that once every couple of years and that gives you a, you know, some data points you really believe in. And then you kind of use that as a rule of thumb for, for day to day work, which is less deterministic.
- [18:05]
Ron Jacobson
I think that that is close to what people will say about it. I go then, but that day to day requires something. So what is that day to day method you're using? And that's where people often will say you can just start your measurement journey with just like holdout tests, which is true, you can, you can get some information about that. But like after that test is over and a quarter's gone by, you still need that day to day data. And it's almost like with this triangulation methodology, it's almost as if you can start measurement without having someone use on a daily basis. And I think that's just wrong. And I got you.
- [18:31]
Eric Franchi
What's the role of Chrome's Privacy Sandbox? Their measurement APIs in, in this. Are you, have you tested them? Are you counting on them? Do you think they're real?
- [18:41]
Ron Jacobson
I've not tested them. I don't know anybody who really has from a measurement perspective. I know companies that purport to be, I mean they are, I'm not going to purport that are more privacy focused. Measurement companies like Anonymous, I think that was bought by Mozilla even they are more in terms of privacy enhancing technologies. You know, we set up servers that have all of the privacy compliance regulations, auto regulations, so even they are more in terms of what's the raw clean room oriented than I think leveraging one of the APIs like Privacy Sandbox.
- [19:11]
Eric Franchi
Is that because you don't believe in the browser doing it or because you just haven't prioritized spending time with it?
- [19:16]
Ron Jacobson
So it's more the latter. And for me, I'm the weird one here where Open Web Programmatic is actually surprisingly not important for my clients. The biggest weakness of the core attribution I talk about is that third party cookies are the only way by and large to measure open web view through outside of using like you know, one of the user identifiers which we haven't invested in either. The reason why we haven't invested there or in Privacy Sandbox is because it makes up like half a percentage of impressions across all of our clients. So anytime there's a choice of where do I want to focus on better measurement, is it open web viewthrough or is it better measuring CTV or better measuring viewthrough on Snap or Pinterest or TikTok? It's always the latter, never the former.
- [19:55]
Eric Franchi
It always seems to come back to the fact that there's actually two totally different advertising markets. There's the top Fortune 100 customers who are moving from TV and print to digital and have a certain point of view about it. And Open web is really important to them. And then there's, you know, everyone else who are, you know, DTC app install, you know, just all kinds of companies out there who are much more social. First a little bit of open programmatic maybe. But CTV is important.
- [20:25]
Ron Jacobson
You, I think you nailed it. Even the word clean Room. Right. Like, that comes up never in my conversations with clients. And mind you, we work with, like, some very large e comm retail clients, but it's not that Fortune 100 level clean room is not a term that they care about. So, yeah, I think you're completely right in that bifurcation.
- [20:42]
Eric Franchi
All right, so, Ron, you and I work together at AppNexus is how I got to know you. My time at AppNexus was brief, but I met some great people. So is there an AppNexus mafia startup mafia?
- [20:55]
Ron Jacobson
I'm a firm believer that there is. I mean, we can start to rattle off a couple. We have, obviously, Beeswax. We have Rockerbox. We have triple lift. Sinceran Scope.
- [21:03]
Ari Papera
Three scopes.
- [21:05]
Eric Franchi
I'm not sure that counts.
- [21:06]
Ron Jacobson
A little too close to the source. Right. There was Ezra's Avocet or Avocado. I forgot to pronounce that one.
- [21:12]
Eric Franchi
Oh, yeah, I forgot about that one.
- [21:13]
Ron Jacobson
Evil Non. I forget Orbenstein's doing a new thing. Yeah. You have Gen Chen running kinetics and stuff, so, yeah, I think there's a definite admols mafia triple lift.
- [21:26]
Eric Franchi
If you had to ask me 10 years ago, which of those startups would have the highest exit, I would not have guessed. Triple F, probably.
- [21:33]
Ron Jacobson
Timing is everything, right?
- [21:34]
Eric Franchi
Is that your bar you got to beat Eric and Ari on?
- [21:38]
Ron Jacobson
I've been trying to beat Eric and Ari my whole life and not get succeeded there, but I'm fingers crossed.
- [21:46]
Eric Franchi
All right.
- [21:47]
Ron Jacobson
I'll say this, though. I think the app Nexus mafia is split up into two different cohorts. You refer to it as Apnexis or Xander, and that, to me, is like the telltale sign of which. Which era of Apnexus you're. You're a part of.
- [21:58]
Eric Franchi
Yeah, I don't know anyone who calls it Xander.
- [22:00]
Ron Jacobson
I'm with you there.
- [22:01]
Eric Franchi
I call it Microsoft Advertising Technology. All right, on that. Let's take a quick break, and we'll come back with the news of the week.
- [22:12]
Ari Papera
All right, we are back with the news of the week. It was a big news week. We have big fundraising news, big M and A news, big tariff news, some earnings, and maybe a little bit more. But before we do, for some reason, we got a slew of corrections and updates from the listeners. So, Ari, go ahead and correct everything we've said in the past year.
- [22:31]
Eric Franchi
I know people actually listen to what I say and I'm wrong, like, at least 20% of the time, and they like to tell me that. Okay, so we're going to run through corrections. First, I had said on the podcast that Congress's ban of TikTok under PFACA is the name of the act should have been broader. And Adam Klee from Licorice corrects me and says it is broader. In fact, the President can determine that any app that distributes media has over a million users and is deemed a threat by the President and is owned by a foreign adversary can be banned. So good for Congress and good for Adam Cleve for correcting it. Another listener, I forgot who, but someone reminded me that the VPAA law, which I referred to as the Clarence Thomas law, is actually the Robert Bork law. So I got my Supreme Court justice is messed up, but otherwise I was correct on that point. And finally, this isn't a correction, but more of an enhancement. We talked about the Robinson Patman act last week, which is the act that prevents brands from favoring big retailers over small ones and its potential impact on retail media. I want to have a quick call out to James Tenser, who is a retailer guy and writes a blog, and we'll link to it in the newsletter. I had a great conversation with him and he told me that there's not actually been any enforcement or suits in this area, but people are looking at it because it's sort of a dormant act and it has been used for other things recently, most recently against Pepsi and not retail media related, but just pricing related. And so there is some concern among the retail media folk about the Robinson Patman Act. Okay, so if I put you all to sleep, I apologize. But you know, sometimes we got to just give our readers the call outs on these technical details.
- [24:12]
Ari Papera
I think given the architectural listenership, this is not at all putting people to sleep. I think we've got such deep, like subject matter experts across a lot of things, evidently that people eat this stuff up. Thank you to the listeners and the readers. Keep correcting Ari. I love it. Anyway, let's go. News of the week. So this one's big. All right, stackadapt. Prior to this week, how much did you guys think about StackAdapt?
- [24:38]
Eric Franchi
Not much.
- [24:39]
Ron Jacobson
It's actually used by my clients. But we'll talk more.
- [24:42]
Eric Franchi
I had good things. Everything I heard was good.
- [24:44]
Ari Papera
Yeah, that was, I think, the general sentiment. Hear good things. Got some good people that work there. You know, apparently like this G2 ranks them. You know, the the best UI for a DSP Stack it app this week announced a $235 million fundraise at a $2.5 billion valuation and has revenue of 500 million. This is a big business. So everybody started like talking about this both sort of like publicly and, and in the DMs Connor from Luma Shout out he found an article that points to the fact that it was indeed 500 million in net revenue and 125 million in operating profits and little investor note. A lot of this stuff was secondary both in this round and the, and the previous round. So this is a big business. This is for context, higher revenue if indeed it's apples to apples, 300 million versus 500 million than Vyant. A higher market cap than Trade Desk.
- [25:48]
Eric Franchi
When it went public. Sorry to jump in.
- [25:50]
Ari Papera
Yeah, higher than Trade Us when it, but when it went public. And way higher than Pubmatics valuation which is 776 million and it's close to magnite in terms of market cap. So this is a company that could go public if indeed they wanted to. Give me the reactions, boys.
- [26:06]
Eric Franchi
Yeah, I'll say. My initial reaction was this makes no sense. These numbers make no sense. But then I talk to people, I do a little Twitter dming and get some background. So the comp here really isn't the trade desk. The comp is basis or simplify.
- [26:22]
Ari Papera
Simplify.
- [26:23]
Eric Franchi
Yeah, yeah. So a lot of smallish medium sized regional advertiser agency clients, very high take rate. One of the things that I like to do is always evaluate DSPs based on their gross spend because that gives you kind of their relative size. So if you have 500 million in net gross spend you might say well multiply by five. So that's like 2.5 billion gross spend. But no, that's not the case. It's probably a lot less. It's probably like one to one and a half billion in gross spend which would put it in sort of the same category as like Beeswax and Yahoo in terms of size and probably basis. I don't really know what basis numbers are. So it's big, it's in that, it's like a leader in that sort of second tier of DSPs. But it's much, much, much more profitable than a lot of those other ones because of very high take rates. So overall, really interesting. I think one last point is that if you have very high take rates you don't really want to go public because people will find out.
- [27:20]
Ari Papera
A good point. Ron, what you got?
- [27:22]
Ron Jacobson
Yeah, I mean I, I, I did a bit of texting as well. People I know in stack of that. Basically I say it's just easy to run. The Trade Desk has minimums that are relatively high. Not the case with Stack Adapt. The quote was like, well, just let me run. And I think that the power of the SMB advertiser is like always underestimated. It's really what made Meta. It's really, I think, what made AdWords. And yeah, I think StackAdapt has found a. Have a.
- [27:46]
Ari Papera
Found a really good big niche for sure. I think this is also a testament to. To. And again, these are public market comps, but like the power of the demand side. You know, you look at PubMatic, it's a $300 million business with a 7, 76 market cap as of this morning. You know, this is a $500 million business with a $2.5 billion granted private valuation. So it just, I think, is a testament to, you know, if you're close to the spend, you've got a lot of optionality to build a valuable business.
- [28:16]
Eric Franchi
Yeah. Also, engineers in Canada are half off because of government subsidies, which is really. Which is. It's surprising there haven't been more Canadian ad tech companies that have gone global. There are a lot of small ones in Canada and have been, but not as many have made the leap. So that really helps with the operating profit.
- [28:34]
Ari Papera
Yeah, it's a great point. Okay, congratulations to Stack Adapt. That was awesome. All right, a couple more here. So TV Scientific raised a Series B or longtime investment of ours at the fund. I think this is interesting because if you read the article, and it was kind of buried actually in Axios, Roku participated in the round. So Roku over the past year has like, done. I don't know if it's a 180, but a big turnaround in terms of like being open, having this data cloud and now working with TV Scientific to presumably bring on a lot of performance in SMB advertising. I thought that was, you know, pretty interesting outside of the fact that, you know, a big Series B has still been somewhat rare in ad tech. So it checks a lot of boxes in terms of ecosystem momentum. What do you guys think?
- [29:22]
Eric Franchi
I think you're right. It's interesting that Roku is pursuing a diverse strategy. I would say that they launched their own performance platform and now they're investing in someone else's performance platform. And they work with DSPs and they're doing things to make the DSP channel more effective, like watermarking their data, watermarking their inventory. So I like it. It's not as doctrinaire and rigid as the strategies coming out of, let's say, Amazon for their video business.
- [29:51]
Ron Jacobson
Yeah, I mean, I think it's interesting to see what comes with OneView, how they invest there versus how much they're looking to partner with these other DSPs. I haven't heard that much about OneView lately, to be honest, but maybe they're kind of just raising the white flag.
- [30:05]
Eric Franchi
I got the impression OneView was sort of dead ish before they launched their performance program in the fall, in the spring or last winter. It felt like one of you died when they did that deal with the Trade Desk and others.
- [30:19]
Ari Papera
Yeah, Jason Pascanasi came on the pod. He was just talking about all the partnerships and being ecosystem friendly. So I think you guys are both right. Cool, we'll move on. This wasn't a funding announcement, but this kind of BW up on AdTech Twitter or AdTech X. Had you guys seen this company icon.
- [30:39]
Ron Jacobson
Prior to this week?
- [30:40]
Ari Papera
Icon me.
- [30:41]
Eric Franchi
It definitely blew up. Yeah, they have a really nice website. They claim a lot of really big brands use the product and you know.
- [30:48]
Ari Papera
Kind of fun fact like Founders Fund is Peter Thiel's fund that has backed like a bunch of like amazing businesses, doesn't invest in ad tech. They back these guys at the seed so they just like publicly launched this week. There was like a cool video from the founder. Seems to be a cool product, but just thought it was, it was kind of interesting from a number of perspectives.
- [31:07]
Eric Franchi
My initial so their headline pitch is create 30 ads a month with our AI really easily. And Joe's app asked me for my take and I gave sort of my general creative and AI take which is integration matters more than the AI. But then I actually looked at the product and it's very oriented towards like vertical video on social, which is an area that probably does not have a lot of investment. On the creative side, people just generally just use their phone and create the video. So if you can make that more efficient. It does seem like kind of a tip of the spear where there may be advertisers who don't have access to the kind of tools other people do and may be really benefited by this. So I think calling it creative is missing some of the nuance about which types of customers, what type of use cases it's automating and there may be actually some, some really beneficial trends in their direction.
- [32:03]
Ron Jacobson
The Founders Fund thing is very surprising though. Like still them investing in an ad tech, I mean creative or not generation company was not on my bingo card for this year.
- [32:13]
Ari Papera
I'm seeing it like just with, you know, some of the deals that are, that are happening. Firms that either I have not heard of or firms that previously were. We don't invest in ad tech. They seem to be circling and coming around. Just one clarification. They say three person creative teams make 30 ads per month. With Icon they can make 300. So I think that's a little bit of. Yeah, the acceleration piece. The one thing about this stuff is Meta is already offering tools like this and Meta in their last earnings report said they have 4 million clients using generative AI creative tools. So the real question is if this is geared towards social, is this going to be used by meta customers or will they just use the self serve that is free with Advantage Plus?
- [32:56]
Eric Franchi
Definitely it was very oriented towards vertical video and maybe TikTok, meta, et cetera.
- [33:03]
Ron Jacobson
The biggest thing that always hurt TikTok was creative production and I would have thought TikTok solves it first but maybe it is going to be a cross platform solution.
- [33:10]
Eric Franchi
I think there's also a frequency issue here which is if you use these systems like TikTok or Instagram reels, the creative burnout's much faster than it is in a passive environment like tv. As a consumer you can recognize an ad almost instantly when you've seen it before and skip past it. So more creative is actually a really important thing.
- [33:30]
Ari Papera
Absolutely. Okay, let's move on to M and a couple of things to tick through here. So the first thing is Outbrain Teeds Close and the new name is Teeds. They've gone with Teeds for the company name. They basically had three choices. It was called a broader company outbreak. Call the combined company Teeds or come up with a new brand. What do you guys think of the. The Tees name?
- [33:52]
Eric Franchi
It's not bad, you know, it's good they chose something like I just don't waffle. Make a choice. Go go with it but back it. So I give it an endorsement.
- [34:02]
Ron Jacobson
I'm thumbs up, I'm pro. Teeds is the name.
- [34:04]
Ari Papera
Yeah, I think, I think I'm aligned. Right, you think Teeds, you think video? The future of the business is clearly video. I think it makes a lot of sense.
- [34:12]
Eric Franchi
Do we know where the name Teeds comes from or what it means? Is it short for something? Is it some Britishism or. I don't know, what is it?
- [34:19]
Ari Papera
I don't know. But I bet there is a listener that is furiously typing a correction right now.
- [34:26]
Eric Franchi
It's not a correction. We want info. Tell us where, where's the empty and answer right now.
- [34:31]
Ari Papera
Please reach out to us. It's a great question. Another deal Coinbase acquired spindle. So 18 months or so we had Antonio Garcia Martinez. Awesome episode, by the way. And this was his company. It was Attribution for Marketing on chain. So the blurb in Coinbase says by acquiring Spindle's team in technology, we'll be able to help builders go viral on chain and fund distribution on Coinbase and across the onchain economy. What this basically means in terms of my interpretation is Coinbase has the number one app for trading crypto. They also have the number one wallet, the Coinbase Wallet, for trading crypto. They have an installed base, somewhat like Uber in terms of just having like, you know, millions and hundreds and hundreds of millions of users. Using the app, they can build basically a vertical media network for builders, quote unquote, which is effectively trying to get projects out there for people to buy and sell. So I think this is super cool and there is no better company and no better person to be leading this effort than agm. So I'm a fan. I know you guys don't care about crypto or maybe you do.
- [35:37]
Eric Franchi
Yeah, I was an investor in Spindle. I don't really know what they do, but it's.
- [35:40]
Ari Papera
Oh, congrats.
- [35:41]
Eric Franchi
Yeah, I'm going to get.
- [35:42]
Ron Jacobson
Does that make sense now?
- [35:43]
Eric Franchi
Yeah, it's retail media for crypto. I love that. That's great. Great spin on it. I'm going to have to explain to my accountants how I got capital gains and coins. It's going to be great, you know. No, I was an investor actually through Eric Seufrid's venture capital fund Heracles, which is an amazing portfolio that he's built.
- [36:02]
Ari Papera
You got coins?
- [36:03]
Eric Franchi
I think it was paid in coins or I think it might have been paid in coins.
- [36:06]
Ari Papera
Fascinating. What's up? What you got?
- [36:08]
Ron Jacobson
I'm very, I'm a big Antonio fan. I first met him when he was building out fbx and I think he's the fastest, fastest speaker I've met to date with the highest density of things. I have no idea what he's talking about, but sound very, very smart per minute. So I'm very, very pro. Antonio.
- [36:22]
Ari Papera
He's a fascinating dude. Chaos Monkeys is an incredible book if you want to learn about ad tech from the previous era. And he's, I mean, he, he's been to Ukraine, he went on Rogan. He is a wild man.
- [36:36]
Ron Jacobson
Yeah, he is like a, a modern day hunter ass in a way.
- [36:39]
Ari Papera
Absolutely. Shout out to agm, final deal. This is a little bit more of a continuation of a trend, but we're Touching on Havas acquired a retail media agency called Channel Bakers. I think Havas was the final of the hold codes that didn't make a move in acquiring a commerce or retail media agency. And this is their move. Shout out to Havas. You guys know about Channel Bakers?
- [37:02]
Ron Jacobson
Never heard of.
- [37:02]
Eric Franchi
No, I don't. I assume it's an agency that does retail media. Retail media is hard. We had Mark Menino on the, on the pod a couple months ago. Also a great episode where he sort of explained to us what the flywheel company that had been bought did. And I think that was enlightening because they do stuff besides advertising. They do things like making sure the product pages look right and making sure your reviews are lined up and you know, all these kind of things that are required to be successful in retail. So I thought that was pretty enlightening.
- [37:30]
Ari Papera
Yep, agreed. Another great episode. All right, so here's something going on we got to talk about. So Ari, can you explain to us what the de minimis exception to tariffs is in a sense?
- [37:42]
Eric Franchi
Yeah, basically this is how TEMU and their competitors have gotten so big because they ship directly from China from factories to consumers. And there is a exception to tariff rules if the goods are worth less than $600. And so that's given them a price advantage, a huge price advantage versus products that might be sold more traditionally by being shipped over to a warehouse here in the US and then and then sent to consumers. And it's generally been seen as a loophole and a bad thing for America that these manufacturers and retailers could skip on them. And so when Trump launched his first salvo in tariffs last week or on Monday, the China tariff I think was only 10% but it include a removal of the de minimis exception. I've read there's been some follow up coverage of it which is that first of all it definitely affects TEMU and those companies. It could effectively put them out of business. One of the follow on effects is that the US Post Office is not ready to implement this. It's too complicated to implement it. So they're not yet. They're letting. It hasn't actually hit the US Postal Service yet. A massive increase in the amount of paperwork you can imagine. And then secondly that TEMU is still shipping from the US where they have goods in the US Distribution. Yeah, it could be that TEMU doesn't go out of business, they just start having warehouses in the US which would be probably good for everybody if that changed their business model to be on an even keel with the other competitors.
- [39:11]
Ari Papera
Awesome explanation. Thank you. There was a good journal piece about this thing. Everybody should read. We'll point to it in the newsletter. It was almost like a thought exercise of if they actually just cut off Temu, what are the implications and then what are the broader implications of the, of the, of the tariffs? Temu is Meta's second largest client with an estimated spend of 2 billion.
- [39:35]
Eric Franchi
Yeah, it's crazy how much money they spend. Also, I guess we'll all be watching out to see their terrible ads. On the super bowl. Last year they had like three different ad slots with that horrible ad, but.
- [39:45]
Ari Papera
You remembered them fine. Seventh largest advertiser in the, in the us. So the, you know, again, the thought exercise is, does this have a real impact on the big platforms like Meta and Snap? On the surface, it appears to. But then just given the number of sheer clients, particularly Meta has you wonder if they can just like suck up all the, all the supply just with increasing, you know, their spend across millions of advertisers. So I don't know if there's like a massive impact on, on Meta.
- [40:15]
Eric Franchi
There's also another story we have not covered on the Pod, but reminds me of which is Meta changed its rules in the new year for anything health and wellness oriented. Ron's shaking heads. Maybe he has more details. But I believe what they did is they are disallowing email match on anything health and wellness oriented, which is a huge category on Instagram, et cetera. And I've seen all kinds of people freaking out about it that they're not gonna be able to spend.
- [40:38]
Ron Jacobson
Yeah, they've made it harder to. Well, exactly that. If you're in health and some financial services, bottom funnel conversions are gonna be harder to optimize towards. They don't want to accept all the metadata associated to those conversions. A lot of folks are starting to optimize more towards upper funnel, KPI's landing page visits, things like that. So it's definitely been a bit of an uproar. I think the biggest part was honestly just how it was handled and the speed at which the change was made with kind of no real heads up. Going back to the Timo point though, my gut says is Meta is fine. I've seen this time again, giant advertiser leaves. They have such a wide roster of advertisers. I feel confident they're going to be able to fill in the lack of demand.
- [41:17]
Eric Franchi
Yeah, in a sense it's good for other advertisers. So one of the effects of temu's Blast has been to raise rates. And in Meta's earnings, I think they, they claimed a 14% annual price increase on their inventory, which really hurts advertisers. So removing some pricing pressure is probably good for the marketplace.
- [41:36]
Ari Papera
Yeah, agreed. Okay, so a couple things we'll, we'll call today. So earnings are rolling in this week. Two things maybe to note. First is Critio slightly dipped in revenue like 1% but the stock popped like 20%. And this is Meghan Clarken's last earnings call as CEO. So shout out to Megan, a couple of things I picked up. So 225rmns are using critio. Their agency biz is up 50% year over year and they seem to have like really good partnerships with Shopify and Meta that I think are boosting investor confidence in Critio being a sustainable business moving forward under the CEO. New CEO.
- [42:16]
Eric Franchi
Yeah, it's great. Long term business transformation. Megan deserves a lot of credit. The new guy will get a lot of credit. I don't know particularly, I think that it's an interesting company to watch, see what moves they make moving forward.
- [42:31]
Ron Jacobson
The Shopify one was interesting for me. They're the only ones who have claimed to be part of this audiences program.
- [42:38]
Ari Papera
Third party partner.
- [42:39]
Ron Jacobson
Yeah, I actually haven't heard much of it from any of my clients or the advertisers there. So it was good press. I, I'm interested, I want to go down deeper to see what's actually happening there but, and I totally, I haven't heard anything there.
- [42:49]
Ari Papera
Yeah. In the not so good earnings news front, Omnicom made some comments around what they anticipate the cost savings to be from the acquisition of IPG and they're expecting 750 million in cost savings and the, you know, the line was client facing and revenue generating rules will generally be protected, but back office and operations are likely going to be targeted with some layoffs. So that's what happens sometimes with these big mergers.
- [43:18]
Eric Franchi
Not unexpected, right?
- [43:19]
Ari Papera
Yeah, not unexpected at all. That's how they make these things work economically at least from a corporate perspective. Anything else guys you want to call today?
- [43:27]
Eric Franchi
Let's give Steve Yap a big congratulations for finally leaving Google. So I'll call him friend of the pod. I think we made fun of him once. He's a friend of mine personally and he's been at Google for like, I don't know, 20 years or something.
- [43:39]
Ari Papera
Yeah.
- [43:39]
Eric Franchi
So he's now the CRO of Perrion.
- [43:42]
Ari Papera
Yeah. Congrats to Steve and congrats to Parrion. Sounds like you got a great leader there.
- [43:46]
Eric Franchi
All right, well, it was a great episode, Ron. Thank you so much for sharing your wisdom about measurement. Next time you come on, I'll also forget you've ever been on and I'll welcome you for the first time.
- [43:56]
Ron Jacobson
I'm looking forward to it. Thanks guys.
- [43:57]
Ari Papera
Awesome. Thank you, Ari. Thank you, Ron. We'll see you next week everybody.
- [44:01]
Ron Jacobson
Thank you for subscribing to Market.
- [44:04]
Ari Papera
New interviews are added every week at.
- [44:06]
Ron Jacobson
Marketing and your favorite podcasting app.
- [44:17]
Eric Franchi
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 weekend news at News Market tv. And if you're feeling social, we operate a vibrant Slack community that you can apply to join@adtechgod.com.