DoubleVerify Holdings, Inc. (DV) Earnings Call Transcript & Summary

All right. Good morning, everyone. It's a rainy day here in New York City, but we're super excited nonetheless, to be here. Many of you know me, I'm Michael Graham. I'm Internet analyst here at Canaccord. I am super fired up today because we have the dream team from DoubleVerify. We've got Mark Zagorski, the CEO; and Nicola Allais, the CFO. Gentlemen, thank you so much for joining us. It's awesome to have you. You just had really great numbers in your Q3 results, so we're going to dig into that.

Mark, maybe I'll just start with a kind of a big broad question for you. You've been at this for a long time in the digital advertising industry. Maybe just give us an overview of sort of like what's the state of digital advertising today, maybe touch on some of those really key growth pockets around social and CTV. And then kind of what is the opportunity within that for DoubleVerify.

Yes. Thanks, Michael. It's great being here. And thanks for not saying like a really, really long time. You just said really long time up in the space. So look, I think one of the things that's so exciting about digital advertising is once you think you've got it figured out, something totally new emerges, right? Everyone thought that the were going to run the table, and then all of a sudden, you have something like TikTok pop up in social and garner 1 billion monthly users in just a few years. Everyone thought they've had television figured out, it's been the same for years. And you have CTV now emerging as this powerful force that's now ad-supported and you've got applications from Hulu to Pluto and others that are now changing the way advertisers look at television. So I think the interesting part is that the landscape is always changing. And the great news for companies like DV is that, that always means there's great opportunities, right, for companies like us. And mainly due to the fact that advertisers and most of all of these changes in the digital space are advertising driven, whether it's social or CTV are continually challenged with the complexity of the space, right? And they look for partners that help them better drive outcomes across those spaces. And I think for us, we've seen that as a great opportunity. And whether it's launching new viewability and impression tracking and fraud protection across platforms like TikTok or brand suitability and brand safety solutions across connected television, there are continued opportunities for us as a company to help those advertisers drive better outcomes. And we've seen that in our growth numbers. Our social business grew 83% last quarter, right? And that's our ability to help advertisers better navigate and better deliver outcomes using our solution on social platforms. CTV continues to be a growth engine, and we grew volume on CTV, 41% in our direct business last quarter. So all of these new areas become new opportunities for advertisers and add a new opportunity for us as a company that works with them to help better deliver their ad in a safer and more secure way.

One of the things that -- that's a great overview. One of the things that has been front and center, especially this most recent quarter, but for a while now, is all the privacy and targeting headwinds that the ad industry is going through. You reiterated again on your Q3 call just recently that DoubleVerify is largely unaffected by this. Can you just kind of talk through why that's the case? And one of the things that we heard from some of the other platforms, even if they didn't have like a lot of direct exposure, was that advertising spend was a little bit softer for them in general because some advertisers were pulling back, and you definitely did not say that. So maybe just talk about that whole privacy topic?

Yes. Look, the one thing to know about DV is we always look at the what, the where and the how around digital advertising, right? We don't look at the who. And a lot of the focus on privacy and cookies and IDFAs and all these kind of trackers of individuals, really focuses on the who, right, which has largely kept us out of being affected by those changes. We help provide advertisers with a safer way to deliver their impressions by understanding context, the security of the transaction, the geography of the transaction, where things show up on a page, all of those are super important to an advertiser, because, again, it's not just about who they reach, it's about where they reach them and how they reach them. So from a mechanics perspective, all of those changes really haven't affected our solution at all. We don't use individual identifiers, we're not based on cookies, all of those. But from a positive perspective, it's made metrics and proxies for performance like context, like engagement, like attention, that much more important to an advertiser. So we've actually seen all of these changes as really being a big opportunity for the company, right? The ability for us to know where an ad shows up on a page, how that user engaged with that ad, how much attention there was, those all become huge factors in helping drive outcomes. I use this word outcomes a lot because at the end of the day, that's all advertisers care about. I want to change someone's heart and mind if I have a brand or I want to sell them a bottle of shampoo, right? That's why I advertise. And the former proxies were all about, hey, well, if I reach men between the ages of 18 and 35, I'll be able to sell more beer, right? Well, at the end of the day, if you can no longer reach that specific demographic or it's impossible for you to track against those because of identifiers and privacy rules, you need to look at other proxies for driving performance. And I think that's what provides a great opportunity for us. It's challenged a lot of the industry, which has been built on reach and frequency, and age and gender. But I think it's provided a huge opportunity for new metrics, like the ones that DV builds, to help better guide advertiser spend.

So that's a good segue into like this next question, which is I was looking at our model a number that astonishes me is you're going to measure by our model about over 4 trillion ad impressions this year in 2021, which is kind of cool. Can you just like give us an overview of the platform? Like how does your platform achieve the scale? And maybe just some of the highlights around how all the data and everything, how it all works?

Yes, it's a great question. Just taking a step back, when you look at -- I know that seems like an inconceivable number of things that we look at, like when you start getting in the trillions, there's only two things in the trillions, it's like Elon Musk pocketbook and Jeff Bezos aspirations for his next moon shot, how much it will cost him. So like it's a big number, but the fundamental notion of our business is that we have to look at things at scale because our advertisers are buying across every platform in any media, across every geography on the planet. So they do things in a large scale. So we have to cover in a large scale. Like if we have blind spots on our verification front, it really is a detriment to the advertisers, and it keeps them from having comfort and spending in those venues. So our whole premise is based on massive scale. Like it only works. You can't be a small measurement company. You can't be a small points solution that only looks at one genre. We only look at social, we only can do brand safety across video, we only are in this one area because advertisers don't spend that way. So like the whole nature of our business has to be large. And then -- so therefore, the infrastructure that we built is pretty large, too. And not to get too much in the nuts and bolts, but it's a combination of colocation facilities in which we actually own and lease servers, right? So we have hardware in the cloud. And we use the cloud and balance colocations with cloud to manage volumes, right? Because we see transaction volumes obviously change with advertising volumes. And that mix of those two things has allowed us the flexibility to scale globally. So when we move into a market, we don't have to build a huge server farm to support it, we can leverage the cloud to leverage the volumes in those markets as we determine whether or not we want to put a regional hub in there for servers. So from an infrastructure perspective, we have to think about massive scale at all times and how we balance volume loads, right? The same way power companies balance volume loads. But that is the applications that we build on top of that are really important. Like you can have a great infrastructure, but if your apps that have to do that analysis can't handle those volumes, you're in real trouble. And I think that is where our real differentiators start playing out, is our applications that sit on that, we've built them ourselves, right? They're not franking applications that we've had to [ sow ] together. We've built them knowing the fact that they have to be able to handle massive amounts of volume. They have to be owned by us. They can't be partnered. You can't partner your way to like a high-volume, consistent metric across everything. So I think that's been a key part of our philosophy since day 1. The company has been around 10 years, and we've always known that the infrastructure needs to handle lots of volume, but the apps on top need to be able to process that and they need to do that consistently across every different type of media that we're on, across every platform because that's the final part of this, is you can have lots of volume, but if there's not consistency in the metrics, when we work with someone like Unilever or Mondelez, they want a consistent metric across CTV, they want it across social, they want it across mobile app and in web and all these places we are. They want that consistent metric. It means the apps that are pulling in that data has to be able to normalize it and make it applicable to whatever platform they're on. And I think that's -- if there's a secret sauce to all of this, its scale is important and it's part of the business, but it's how -- what you do with that scale and how you analyze it, the apps on top that really make the difference.

Awesome. Okay. And Nicola, I want to bring you into the discussion here. Before we do that, I just want to mention to the audience, if you would like to ask some questions, please go ahead and submit those via the web interface, and we'll get to some of those a little bit later. I already see some coming in, so keep those coming. Nicola, let's shift into -- I want to kind of move the discussion here into talking about the business and the different products. I think it's good to set the stage and talk about DoubleVerify's 2 reporting segments, you've got Advertiser Direct and Programmatic. So maybe just talk us through those? And any relevant differences in growth rates and profitability between those 2 segments?

Sure, sure. So first of, I will call them channels rather than segments, because our desire is to be able to measure wherever our clients are spending their ad dollars. And so those are 2 separate channels where we use -- where advertisers use our data to basically, evaluate and measure the quality of their ad spend. The Direct channel is where our advertisers will use our data to measure the quality of an ad that's already run. So it's a post measurement quality of the ad spend. The Programmatic channel is where our advertisers will use our data to evaluate the quality of an ad impression even before the ad impression is bought. And if you think about those 2 channels, basically, it covers the entire decision tree. The advertiser will use our data to evaluate before they actually bid on an impression and then we can measure it on the back end once it actually run. Another way to think about the filters of those 2 channels is one is the -- one is used through partners. So the Programmatic channel, our data is used through our programmatic partners. We integrate with all the major partners in the U.S. and also outside of the U.S. And for that distribution, we obviously have a revenue share with the programmatic partners. On the Direct side, in addition to the open web, we also do work with the walled gardens. So the walled garden part of our business will fall into our Direct channel, that's obviously growing very fast. That inventory cannot be bought programmatically. So as we think about how differentiated we are and how we're able to verify everywhere for our customers, we're able to work on the walled garden part of the ecosystem in a way that the programmatic channels cannot. Mark will go, I'm sure, into the acquisition of OpenSlate, which allows us to even do more with the open -- the walled gardens, but that's kind of generally how it works. In terms of profitability, let's start with the revenue. On the revenue side, we charge a fee per transaction that we measure, and we try to be very consistent across the channels. So that's the revenue side. And if you think about us already measuring trillions of impressions, right, the cost for us to measure that extra impression is de minimis, right? We're a very profitable business just because once you have the infrastructure, the extra impression doesn't cost us that much. So once you have the MTF, the fee that we charge for a transaction, the difference between Direct and Programmatic in terms of profitability, besides the distribution fee that we pay our programmatic partners, is basically the same because everything is based on the same level of data that we use. Programmatic has been growing very fast in the past few years, and it kind of caught up to where post-measurement was, over the last 4 years, basically programmatic caught up in terms of our revenue mix to where Direct was, which is where the business originally started. And that's a little bit of a preamble what Mark will talk about when we talk about OpenSlate, which is that pre and post mix of being as much revenue coming from before an advertiser buys a bid to after that 50-50 mix is something that we think we can achieve on the walled gardens with this acquisition of OpenSlate.

That's interesting and helpful. And just a quick kind of one layer deeper there, because I'm looking at my model, and we've got you growing the Advertiser Direct 25% this year and Programmatic growing at 45% this year. Does that also mean that like the walled gardens are growing more slowly than the open web? is that the right way to interpret that?

No, I think the way I would think about it is that the wall garden is still a small part of Direct. So wall gardens, as Mark mentioned, the number of -- the volume growth on social is actually very, very large, much larger than the 20%. So the reason why you see that 20% on the direct side is because social is only right now, 30% of that channel.

Perfect. Okay. Thank you for that clarification. That's awesome. So shifting into the product set here. Your core product historically was DV Authentic Ad. It measures whether an ad has been displayed legitimately in a brand safe environment, I got that right from your website. And it's a post-bid solution. And I know we're going to talk about prebid in a minute, but maybe just to set the stage, can you talk about how DV Authentic Ad works and why do your advertisers value this product?

Yes. It's -- so if you think of the evolution of our business, we started off first looking at issues of viewability and fraud, right? Whether or not an ad is actually being viewed and it's being viewed by a human being, not a bot, that evolved into, hey, is it viewable, Is it actually a human being, not a bot, so there's no fraud in both. And then is it being displayed in a brand-safe environment, started off with brand safety, like, okay, is this just -- is this not porn, it's not liquor, it's safe or -- and then it became a brand-suitable environment, right, which became much more fluid. It was, hey, this is -- VICE is a great content, it's not fraud, and it's not pornography, but it's not appropriate for Disney ads or Disney theatrical, right? So that's brand suitability. So brand suitability started gaining steam. And then geography, we tacked on and say, look, this is -- if you can't sell your product in a certain place, you shouldn't be delivering your ad there or if it's illegal to sell them -- sell a product there, you shouldn't deliver your ad there. All of those things together as -- we used to sell them as separate entities, right? So like, hey, some people want viewability. But what became very clear is that when you put them together, you created an authentic experience. This is what advertisers wanted. They wanted the ability to say like, hey, if it doesn't -- if it's not fraud, that's just like the base level of acceptability, but I want all these other things to happen as well. So it was a natural evolution of all of these different metrics to put them into one and say, is this is an authentic experience, and can we relate that authentic experience for the advertising into performance or into an outcome. So that was the evolution of the Authentic Ad. And now it's become really the cornerstone of everything we sell. And when we look at how we do that on the postbid sign and post measurement, the data that we get from postbid on blocking against authenticity or measuring authenticity, feeds right back into our prebid solutions. So we're able to take that information of whether or not something is authentic and feed it back into authentic brand suitability, right? So that same data feeds back into our prebid capability of blocking those impressions before they're even bid on. So the Authentic Ad is really kind of the cornerstone of all the data that we put together. It's a single metric that could be consistent across all platforms. It's MRC accredited. So like we've actually created a new metric that the MRC came out and accredited and saying, yes, this is actually a standardized metric. And now we're able to use that across every different venue. So we can say this is an authentic ad on CTV, this is an authentic ad delivered across social, this is authentic ad delivered across mobile or display or video. And I think that's super important. So we've created a metric that's consistent, that actually relates directly to performance and that advertisers can now relate. And we -- I mean, I don't want to sound too bold here, but we've created a category, like we've actually created a category around authenticity that was specific to DV, and that category now has become accredited, right? We've actually built the authenticity score, an accredited data point that's used across multiple platforms.

Is authentic ad a binary measurement? Or is it a scale?

It's binary. It's either nonauthentic -- it's authentic or nonauthentic. It either meets the criteria that the advertiser -- that we've laid out or it doesn't. And I think that's a big deal because at the end of the day, it will not -- if it's not authentic, it's going to get blocked, right? It will not be delivered on certain platforms and advertisers have bought into that.

Okay. So yes, awesome. So you mentioned authentic brand safety, which is where I want to go next. And as you mentioned, it's a prebid solution. It's for programmatic campaigns. It's been a monstrous success for you guys. It grew 60% last quarter and you said it makes up more than half of your Programmatic revenue now. So maybe just kind of segue into Authentic Brand Safety, how does that build on Authentic Ad? I mean, obviously, it's prebid, but how is it built?

Yes. So I mean this is -- the success of ABS has been really extraordinary. And I mean, the fundamentals of it come down to the fact that our postbid solutions certainly are powerful, right, because they give advertiser information to better optimize our spend. So -- and it will keep ads away from bad content. But that's after they've already spent the money, right? I'm either going to get a make good or I'm going to get a block at the end. The power of our programmatic solutions in ABS in particular, the fact that I want to block -- I don't even want to spend the money there. I want you to tell me to not even spend there, like block the impression, the bid before it even happens. And that's the way it works. So through -- just taking a step back through platforms like Trade Desk and through Amazon, and through Google and others, an advertiser goes in, they basically -- they set their bidding parameters and if they're working with us, before they even bid on an impression, so a bunch of impressions come up and say we can bid on them, they make a call to us and they say, "Hey, DV, does this meet our brand suitability criteria, right? Does it actually meet?" And ABS is interesting because ABS is specific to each brand, right? Whereas the authentic ad is a metric, right? It's either good or bad. ABS is authentic brand suitability, is, hey, like we said, an ad that's delivered on VICE may be totally acceptable for Lionsgate for a horror movie that's coming out this weekend. But it's totally not suitable for the new Disney Pixar in a movie that's coming out. So that suitability, ABS is actually relative to each advertiser. So they can block that impression. They can block bidding on that impression before they even bid on it, not waste their money there and focus their dollars someplace else. So I think that the power of that product has really been in the fact that it is specific to each advertiser. They have the ability to modulate what is suitable and not suitable for them, but also it keeps them from making a mistake before they even make it. And we always like to call it like precog, if you remember, like there's a -- in Minority Report, you can [indiscernible] the person after they do it, but we're going to stop the crime before it even happens.

Yes. Does your platform differentiate between VICE and Lionsgate because the advertisers specifically prohibit certain websites and destinations? Or because your AI figures it out?

So it's -- yes, it's a little bit of both. There's the ability for them to actually block specific websites if they choose to. But more often than not, it's our AI that's actually analyzing the content and the context of the page. Because even -- so it gets even more granular than that, and I used kind of a broad example, but something like The New York Times is a totally brand-safe environment, right? But certain content on a specific page may be totally not brand suitable for an airline who advertises -- Delta advertises across New York Times, yet an article about unruly behavior on flights is not a brand-suitable environment for that advertiser, right? So we have the ability to actually read down to the page level right, or at the specific video level and tell and advertise and actually signal back like, hey, you put these specific criteria around suitability, this doesn't meet it. So don't bid on that impression. So they can certainly proactively put in, hey, we don't want these certain sites. But more often than not, because you're looking at -- you're talking about trillions, you're looking at trillions of potential content engagements that we have to look at to actually deliver back a signal and no advertisers are going to be able to pick it out. So much on one to the other day was a really good example there, think of like 30 years ago, an advertiser was, okay, I'd buy a magazine page. And I'd know, hey, here's the 12 magazines I need to be in, and I know this one is a little bit too risque for me, so I don't want to -- GQ is fine, but Esquire is not. Now it's like, okay, there's a -- it's not just an infinite number of websites and content engagements, but there's an infinite number of individual pieces of content that you have to look at to not be around. And I think that -- that's -- I mean, that's the core reason why we exist, the complexity of digital advertising, the sensitivity of advertisers around that context, has never been greater. And there's no way you can manually or mechanically do this without having like AI and heavy processing and a huge infrastructure to support it.

So how -- I'm fascinated by you've got this great authentic ad and then you kind of leg into authentic brand safety, how does your product development team even come up with that idea? What -- how does that process work?

Yes. I mean it's really advertiser first, right? They drive our development. We look to them because one of the things that I think the company has done really smartly is we haven't said, okay, let's just come up with something, right? And try to push it into our clients and see if they'll take it. Every cue that we take is from our clients saying, hey, we're -- we need a more -- for example, we moved into what we call brand suitability tiers. So on top of -- we started brand safety, like this is bad, this is good, then brand suitability, which is like, this works, this doesn't. And then they wanted another level of sensitivity, they're like, "Hey, some of this content may be kind of good but we need to have even more sensitivity within these tiers." That came from them. They told us, "Hey, can you build this out?" So we listen a lot to what our clients are looking for. The building out of authentic attention, which is our newest kind of solution, was really driven by that, too, which was a lot of greater concern from advertisers about, hey, what happens if cookies go away. What happens if I can't track a user, like how am I going to figure out what ads are going to work and what ads are not going to work and how they're going to drive an outcome. And the first part of that was like, well, we can always use context, which is super powerful, right? It's context and people have been doing it since people bought Monday Night Football because they wanted to reach men, right? That was the context. They didn't know. They didn't know exactly how then they're going to reach that, they have it down to the household, but they knew that they would reach men because according to ratings, that was predominantly men that watched it. So context made sense. But attention was like the next level of advertisers going, well, we know when people are actually engaged in our ad and we know that when it gets more screen time, that they're more likely to behave a certain way. So can we come up with an attention metric that makes sense. And that's the evolution of that product. So our team is like, okay, let's think about what we actually know. What we know how large an ad is on the page, we know how long an ad can play for if it's a video ad. We know what the other content is around it. We know that there's other competitive ads around it. We know how many ads are on a page. We know what the engagement is, whether someone clicks, rolls over it or not, all of these things become an attention metric. So we go back to our clients and say, hey, do these make sense, and we're like, yes, we know from our past history that all of these things start driving certain outcomes. So the development process really is engagement with our advertisers, listening to them, pushing it back into our system. And then the beauty of it is that massive infrastructure that we started talking about with before and all of that data that we capture, all of these products are spun out of that same data set. We already have all those cues, okay? So we use them to determine whether this is safe, let's determine whether or not it's like actually engaging. So all that's the same data flow, we just use it and we build a new product off it, okay? Let's build contextual targeting, let's build attention. And that's what's really cool about this, about our business, what gets me so excited is, like we've got an infinite amount of raw material coming in that we can -- that doesn't -- unlike a factory is like we can't build one car with it, then it's gone. It's like, no, same data point, all right, let's use it for something else. How many times can we use this data point to build something that's insightful and helpful for an advertiser and then obviously, charge them for it, which is good.

Well, yes. So that was going to be my next question. Maybe Nicola, it's a good way to bring you in here. Just talk about -- ABS was an upsell, I think, talk about how the pricing works. How much more expensive is this? Or just...

Yes. Yes. So ABS is a premium product, right? I mean Mark just went through the benefit that the advertiser gets from it. It's a premium product, which is 2 to 3x more than your basic brand safety for existing clients. Our strategy has been to upsell our existing clients first, right? From brand safety to authentic brand suitability. And that's gone extremely well. We're ending the year with the product being distributed on all the major programmatic partners. And so it will have been on all platforms for at least 12 months by the end of this year. It's a premium priced product. We've upsold our customers, which is why we are now at 50% of our revenue. I think we are -- we've always done this part, but it's now become sort of the cornerstone of every new RFP, right? We're going to go authentic brand suitability first, rather than selling the suites of the other products. The power of the tool is so big and so good that it allows us to really get the hook with the advertiser. So we see the benefit of authentic brand safety continuing in the programmatic channel, because it's going to be ABS first with our new enterprise client ones.

So on that, is it -- is the kind of conceptual math correct that if it's half of your programmatic revenue and it's twice as expensive or you said 2 to 3x, but then we can think of it as like 25% of your programmatic impressions are currently hitting the ABS buttons?

Yes, that will be -- that's generally what the math would say. It is going to be completely integrated at this point, right, we are really going with that one. That would be right in terms of the metric.

Yes, because you talked about that on the Q3 earnings and said that some of your big enterprise discussions now are like leading with ABS. Does that sort of cannibalize your pricing because it's no longer an upsell? It's the only sell? Like how does that work?

No. I think we view it differently. We view it as you're starting -- historically, this company has sold every product until you got to the suite of products, right? We offer a suite of products, which is why we're differentiated from single-point solutions. If we start with authentic brand suitability as the going-in proposition for enterprise clients, okay? What that does is that it hooks the postbid measurement as well, right? The power of ABS is that it connects prebid to postbid measurements, right? The same levers that you're going to use on one side of your decision is going to be the same on the second side. So actually leading with authentic brand safety, I think just raises the entire proposition for the customer because they'll come in and they'll want both sides, right? It's no longer an upsell, you start there and you're going to get postbid as well.

They're still going to buy both sides, and it's just...

Yes. Absolutely.

Got it. Got it. Okay, cool. So let's move into OpenSlate. I know that you're really excited about this acquisition, and it takes this prebid concept and applies it to CTV and social video. Why was OpenSlate the right platform for that? What's the opportunity there?

Yes. I mean so Nicola just came off talking about ABS and our success in Programmatic. I mean, look the Programmatic business, which we call it programmatic, but really, it's prebid, right? It's our prebid business. That went from 0 to $150 million, give or take, in 3 years, 3.5 years, right? And it's now half of our business. And the reason why, the fundamental reason why is we've mentioned before, which is people would rather stop something from happening than try to -- before it even happens, than learning about it, optimizing it and blocking it at the end of the day. So this idea of like keeping something from happening, aligning my spend with where it should go beforehand, is super powerful. And it's obviously been very -- helped our business massively over the last several years. Where OpenSlate fits in, and that's -- so we've done that in the programmatic growth. Where OpenSlate fits in, as you noted, is in the social and CTV space, they basically have the ability to do the same thing. They have the ability to go -- they work with advertisers directly and some of the bigger ones that we named. So guys like Volkswagen and Nestle and P&G and others, and they work with them to actually, on YouTube and on TikTok and on Facebook, help them align, thereby so, hey, what should I or should not buy, put the criteria in and only specifically target my impressions across those specific -- by buy across those specific impressions. And I think that we've seen -- and then the magic in this is that we're doing the post buy measurement on all those platforms, right? So we have the ability to actually do what we're doing in postbid, which is half of our business on those platforms. So the power here is connecting those 2 things, the pre and the post. So the advertiser can actually optimize, right? And that's what we've been able to do in Programmatic, right? So pre and post, you target specific things and then you learn what happens afterwards, and you're able to optimize that cycle so that what you find out in measurement, you push back into your prebid filtering. Now we'll have the ability to do that on social and social video across YouTube and Facebook by taking those parameters and pushing it back to that cycle over time. And I think what gets us so excited is we've seen the power of that cycle with our current clients. We know how the 2 solutions can work together. And the business opportunity is, I think, is really great. I mean they've had a really good business. But as I kind of started this discussion, it's hard to be a point solution in this business. It's hard to do one thing for one platform. So it's a powerful solution that fits in really nice to our broader platform that we can expand from. And one of the things I noted on the earnings call this week is the client overlap between the 2 businesses, they have 60 customers. We work with 1,000 brands, right? So our ability now to walk this solution into those 900 other customers that we work with and say, "Great, you love us in programmatic, now use us in prebid on social and CTV," I think it is just a no-brainer. It's no brainer for our clients, and I'll tell you the feedback this week after we made the announcement was exceptional. Everyone from agency heads to some of our biggest brand partners going, this is super smart, we get this, we love it, and we can't wait to talk to you. And then people want to talk to us now. We obviously -- we still have to close the deal. So we're pushing them off a bit, but to me, that's the best sign is when your customers like it, and they're not stretching their heads going, "We don't get this." They're proactively reaching out, it's a good sign.

That's awesome. I can't even -- like we've heard from advertisers a lot over the years that social video, because it's user-generated content, and it's -- some of it and inconsistent and it's video, right? It seems like a really hard thing to kind of analyze that ahead of time. I can't even imagine how that works.

Yes. It's no small feat. And if you think about it, the way that OpenSlate has tackled it by not only looking at the metadata that they get around the videos, but really analyzing the content creators too because a lot of social videos is who creates it, right? Because there's -- in the traditional world, you've got a publisher, so you know that publisher apps represents certain qualities. Disney is a certain quality, Bloomberg is this certain quality. We know what comes out of those machines. These are individuals. So really, your publisher becomes one person. So they actually look at the content creator in addition to all the other metadata that they have about the videos, like, for example, on YouTube, to determine whether or not like what category does that content creator go into, right? Where do they end up? So I think it is -- look, it is no small feat. And we were, again, talking about volumes here, you're looking at billions and billions of videos all the time. And it's -- again, this is where machines really have to play a role in this and scalability is key.

So you said, I believe that it was $15 million to $18 million in revenue, and those numbers might not be exactly right that you expect to get from OpenSlate next year. But it seems to me that the revenue potential is so much more vast than that and the connection that has to happen is getting it integrated into your product set and your sales process and everything. So just maybe talk about that process and how long it takes?

Yes. I mean, obviously, we expect the business to be generating revenue going into next year on its own. They've had some great momentum. We -- I think we're really going to see the power of the solution once we get it integrated into our platform, which we think will take around -- probably around a year to at least some level of integration. But we -- the lighter integration, i.e., kind of marketing and sales and being able to sell the solution to our clients, that's going to happen definitely more rapidly. So we see that happening in probably 2 quarters versus 4 or 5 quarters product base. And that product evolution and integration will obviously go through phases, right? There'll be lighter integrations of UIs and content, and then there'll be more heavier integrations of actually data sets and standardization of how we qualify and quantify each category of content, et cetera. So there's definitely some work to be done there, but net-net, from a basket of goods idea, it's a LEGO piece that fits into our LEGO set that we can go out and sell. They haven't been melted yet together, but it certainly clicks in nicely and our advertisers, like I said, are excited about being able to buy both solutions from the same team.

I think they call those LEGOs in London. I'm not sure.

I'll translate that, LEGOs.

So I want to ask you a philosophical question. Before Trade Desk, let's say, a lot of the DSPs were SSPs and SSPs were also DSPs and couldn't really understand who they were trying to help and Trade Desk kind of achieved a lot of success by just focusing on the demand side. And I kind of look at it legacy DoubleVerify as kind of -- okay, it's -- I'm not going in with DoubleVerify solution trying to necessarily plan my campaign, but they're going to tell me basically did I do okay afterwards. Now you have all these awesome prebid products, great technology. As you mentioned, you've got all this data and you can just build new products on top of the data. Is this -- does this start to create a channel conflict for you? Does it start to put you in competition with some of the DSPs and things like that? Or how do you think about that?

It's a great question. I mean, there's no doubt that the business is evolving, right? And we've kind of used this [ moniker ], we've gone from protection to performance, right? So this idea of like keeping people from doing bad things to then helping them do good things. And I think that evolution is not going to stop. We have too much data, there's too much demand for what we are doing and our clients wanted to happen. I think a key part of that, though, is the fact that we're doing all this independent of the media transaction, right? Our only driver, and you talked about kind of like when there is a conflict between buy and sell and are you on both sides of the equation, our only driver is to be an objective measure for our advertisers to help drive better outcomes. We don't make more money off the CPMs if they go higher or lower. We don't make money by diverting spend one place or the other. We make money by helping them -- providing them with metrics that help them do a better job. That's it, right? So there's no conflict, it's totally independent. And we do so by ensuring that our data is available everywhere where they do their buying, right? No advertiser spends all their money with Trade Desk or all their money on Facebook or all their money on YouTube. So it's essential that they have a single metric that goes across all of those platforms. So that independence and that separation from the media transaction, I think, puts us in a unique position. And that position is the fact that the advertisers look to us first to be separate from those platforms yet engaged with them. And the platforms themselves don't look at us in conflict with what they're doing, but hopefully look at us as being a partner in driving more confidence in the transactions that they're sending. And those platforms, at their heart, have to be open because their advertisers want to bring their solutions to bear, right? It's -- it'd be like having a car and saying, hey, this is a great car. You can drive it, but you can only drive it on these 5 roads, right? That's -- no one is going to buy a car that does that. So an advertiser wants to have the power to say, "Hey, look, not only am I going to spend across multiple platforms, but I want to be able to take the metrics that I believe in, that are independent across all these -- across all these different platforms and put it into places where I need it." So I think that the fact that we have the advertiser relationship, that we're independent, that they want to have that consistent metric across multiple platforms, it -- I don't say it eliminates the friction between us and those platforms, it doesn't eliminate competition. But puts us in a position where the platforms now if they're going to be really agnostic and they're going to have a good relationship with those advertisers that they have to let partners like us be part of that equation. So look, at the end of the day and also just to get the brass tacks, a lot of these solutions -- a lot of the platforms have their own embedded solutions, yet advertisers still use ours. And the reason why is because they want that independence.

Awesome. We're at the 45-minute mark, but I want to kind of go a couple more minutes if you can. Because I have a couple of questions in from the audience and on e-mail, and there are one or two more I'd like to get to you. One of the questions is about the guidance for Q4, and you mentioned CPG supply chain headwinds. Maybe you could just touch on how did you incorporate that factor into your guidance? And when do you think those headwinds start to dissipate?

Yes. So we incorporated that into our guidance, because we heard it from our customers rather than just a general macro trend that everybody is reading in the paper. And we heard it from our CPG customers where we saw in October sort of a slight reduction in what they're spending with us. It is slight, right? Our view is there was a modest reduction in the guidance, just to make sure that we're covered if it were to continue and deteriorate. That is not how we adjusted our guidance. We're not -- we did not hear from our customers that this will continue into the new year at all. It sounds like a narrow time frame of a small blip, and so we thought we should be more cautious because of it.

Just on the business model in general, I mean, you guys are a rule of 60. And like a high-quality rule of 60 too because it's like equally split between growth and profitability, which is kind of cool. Do you expect to stay this way for a while? And do you expect that mix to change much over time?

Yes. I think the -- so yes, we agree that it's obviously a quality mix. Let's start with the EBITDA part. I think on the margin side, we are going to invest where we think there is an opportunity to accelerate growth. So -- but we are very profitable as a business model. As we said, once you measure trillions of transactions, the extra cost is not that high. So -- and we did this even this year, right? We had a Q2 EBITDA margin that was below that 30 threshold. So we will do this as we see the opportunities to invest wisely in ways that accelerates our road map. On the top line, I think the way we think about the top line is the -- we grow -- most of our growth has been volume. And if you think about the digital ad spend growth, the overall digital ad spend growth is in the mid- to high teens. To us, that's where you would put sort of the floor of our growth, right? Because we're going to grow with the volume that our advertisers are spending in digital. Everything above that is acceleration based on new products, new sectors, M&A contributing. All that creates a growth pattern that's ahead of the digital ad spend growth. But this is how we think about it, right? Looking at DV, you can say that's our minimum threshold and anything above depends on these opportunities to accelerate it. I'll mention one that we haven't discussed at all, there's still plenty of social impressions that are to be verified, right? We're not yet seeing all of the volume that we could be providing brand suitability, for example, on Facebook. So there's growth that will just come. There'll be accelerators above and beyond a normal growth rate. So yes, I think there's plenty of avenues for us to continue to show strong growth.

Can you touch on just the philosophy around pricing? Because you have so many products you're building to offer and I mean, I think you're viewed as a really great bargain relative to kind of like the cost of an impression, but at the same time, advertisers would rather pay less than more. So do you get push back on your pricing? Do you -- how do you think about it?

No. We -- so we are in the mode of wanting to verify everywhere, right? And Mark has spoken about us becoming the de facto currency of measurement. And for that, the volume is really what's driving our business proposition right now. So you're right that the value proposition to the advertiser is really high. They're not paying a lot to DV for what we're providing to them in terms of avoiding that content, et cetera. The opportunity that we have on pricing is there. It's obvious, especially as more media spend goes to even higher CPM categories like CTV. Right now, our view is to try to really become -- to verify everywhere for our customers. And if keeping a consistent price that doesn't confuse or complicate the decision allows us to do that, we will continue to do that. It doesn't mean that the opportunity is not there.

That makes a lot of sense. Let's try to get two more quicker questions in, so I can let you guys go. The first one came in from the audience, just about incremental margins for each dollar spent by the same customer. I mean, I think it's 100%, but...

Very high, yes. Yes. I mean I can't say it's exactly 100%, but it's very, very high, yes.

Yes. That's pretty cool. And then just maybe the last one on international. You bought Meetrics a quarter ago to support expansion into EMEA. Just what's going on internationally? How important is that? And how is Meetrics going?

Yes. Maybe I'll just talk -- get it -- so I think we see a considerable amount of greenfield outside the U.S. If you look at right now, our revenue, and Nicola, just check me on this one, we're probably around 20% of our revenue is outside of the Americas?

That's about right, yes.

So we feel like there's definite upside there. We see APAC growth last quarter was 96%. EMEA growth was 44%. Overall, outside the U.S. growth was 60% plus. So we know there's good trajectory there. We've invested -- this year was kind of our investment year, both from the acquisition of Meetrics, but also headcount. So 55% of our headcount was outside of the Americas this year, and now 40% of our total headcount is. So I think we've set ourselves up really well for the opportunity over '22 and '23, because a lot of those headcount was sales, right, sales and client support and going out and just selling the local brands because investing in those places does two things. It gives us the ability to sell global clients, right? So the Unilevers and Mondelez of the world that are in multiple markets, and the Colgates, which we now work with in over 70 markets, and sell those enterprise clients because we've got people on the ground to support them, but also start selling to local clients from those. So when you hear a lot of the closes we had this month or in this quarter were folks like Sony Japan, right, and Afterpay in Australia and New Zealand. These are local brands that we're selling from our operations there that are supporting our global clients, but also selling locals. So we're super excited about the opportunities outside of the Americas. And that's where we've invested. That was the -- Meetrics acquisition will certainly help us in EMEA. They're super strong in Germany. And German market is a big market, I think the second biggest market in EMEA after the U.K. So there's a big opportunity for us there, and we're leaning in.

Awesome. All right. We're going to leave it there, gentlemen. Can't thank you enough for doing this. I know our audience really appreciated it. So thanks a lot. Great to see you, and we'll speak to you.

Thanks, Michael. Have a great one.

Thanks for the questions. Cheers.

Thanks.