Integral Ad Science Holding Corp. (IAS) Earnings Call Transcript & Summary

Hey, welcome to day 2. Everyone is kind of still fresh. That's nice. I want to start bigger picture with you Lisa first.

So when you joined IAS a good few years ago, you had a long career in advertisement. You've been some very, very big shops before as well. Like talk a little bit about like how the industry changed from like when you joined to where we are now because it seems that there are so many different factors and so many different evolutions that we're seeing at the same time. They must have been very exciting for you, but like maybe bring us on that journey a little bit.

Sure. Thank you for having us, Raimo. And hi, everyone. I'm Lisa Utzschneider, CEO of Integral Ad Science and I've been at IAS, I can't believe it, almost 5 years, I joined early 2019. And in terms of over the last almost 5 years, from a macro perspective, some of the shifts that we made, there are a few. So given the fact that over the course of the last few years, we've lived through a global pandemic when you think of all of the events that have happened at a broader macro societal level, that has definitely shifted what consumers are doing online, how marketers are investing in digital advertising, and it's been a tailwind for IAS' business. So I'll pick out 2 macro shifts in particular. Through the pandemic and coming out, explosive adoption of social platform, short form video, in particular, consumers are spending more and more of their time on their phones consuming social media, generating content -- user-generated content. And I would say the second trend coming out of the pandemic is the explosive growth of CTV and viewing streamed content. Both of those vectors again have been a tailwind for our business, especially when it comes to video and brand safety and brand stability because marketers, they want to be where the consumers are. The consumers are on the social platforms, but they just want to make sure that their brands run adjacent to brand safe and suitable content.

Yes. And then do you -- if you think about that time period, did you also see like a change in the industry -- not in the industry, it's more like what the players and behavior of certain players in the market to make it kind of more difficult for advertisers to kind of have -- make sure that the right content is kind of displayed and the viewability is here, I'm thinking of a lot more bad actors in there that help people kind of want to make money off advertisement. Has there been a -- I mean, it's always been like some sort of problem, but it does feel the last 2 years, it kind of got a lot worse.

Yes. I agree with that. So I do think there's much greater awareness by the brands in the importance of running their ads in high-quality media, the importance of sophisticated fraud detection exactly what you're speaking to. And that's been cited as almost 1/3 of every dollar that goes to waste due to fraudulent activity. That's a big problem. And then also, again, the dynamic nature of user-generated content in live to social platforms. It poses both a challenge for the brands, but also a massive opportunity because it's an incredible environment to be able to engage with consumers.

Yes. Okay. And then from your perspective, now like you're running IAS -- you have been running IAS for the last 5 years, with all the kind of opportunities, or like -- for you that's actually opportunities because you're helping advertisers. Like how do you prioritize that because there seems to be like you could go in so many ways. As you just mentioned, CTV, short-term videos, et cetera. Like how do you prioritize where you want to play, how you invest your dollars?

So I'm a big believer in placing a few bets and doubling down and investing in those bets. And our few bets right now that are definitely paying off. You can see it in our numbers, which Tania would be happy to walk you through, is doubling down in the large social platforms, getting in the live feeds, running our multimedia classification technology, those live platforms include Meta, YouTube, TikTok and X. In addition to that, ensuring on the programmatic or optimization side we're also providing solutions, again, for marketers to ensure on a prebid level that they are able to bid on inventory that is all of the things we've talked about related to verification, that lands them in a place as higher-quality media that ultimately drives higher ROI. And what I hear all the time from brands is they want greater transparency. They want to understand where their ads are running programmatically. And if those ads are delivering the highest ROIs. We're very focused on transparency too.

Yes. And you kind of mentioned, Tania, already, Tania, like if you think about last quarter, like results were -- looked very, very strong for you guys, your share price reaction, like if I look at your share price performance this year kind of shows that people are getting excited about your space again. And can you maybe summarize a little bit of what you saw last quarter.

Third quarter, we were really, really pleased with the third quarter results coming in at overall 19% growth much better than expected, and there were many levers that drove that performance both on the optimization front and on the measurement front. On the measurement front, we are firing on all cylinders with social -- social accelerated. The growth of social media revenue for us accelerated to 41% from 33% in the second quarter. And social has now just crossed more than half of our revenue at 51% in the third quarter. And then on the optimization front, so a real acceleration in the growth rate from 10% in the second quarter to 21% in the third quarter. And that was also broad-based as well. It was increased adoption of our products. It was enhanced integrations with some of our DSPs. We saw some benefits from one of our second largest vertical, which was travel and entertainment. And we also saw the benefits of some of the investments that we made in our mid-market strategy and our go-to-market strategy and in the products, and we saw the benefit of that in the third quarter. So really pleased with third quarter results.

Yes. Okay. Perfect. And you mentioned social and short-form videos are like the big things -- maybe me as a dad as well, I see my girls kind of on that all day long, like can you talk a little bit about on that. Were you surprised by how quickly that category has been taking off like? And how do you -- where are we in terms of that in a way it almost feels like taking over the market a little bit. Like what are you seeing there?

Sure. The way we like to frame the social platforms are big building blocks, medium-sized building blocks. So the big building blocks being the large social platforms I mentioned before. TikTok, our TMQ or total media quality technology is running in the live feed of TikTok, identifying image, text, video, audio within the live feed, over 50 markets, over 90 languages with very high accuracy rate, and also very granular sophisticated technology backed by AI. YouTube, second big building block. The technology is running in YouTube in over 30 languages. X -- X, we announced an exclusive partnership with X in August. We are now in beta, running prebid brand safety and suitability in the immersive video feed not in the current feed of all the texts that have been happening, but we're in beta with X. And then the fourth big building block is Meta. We're currently in beta in Meta where we're running our TMQ technology with a handful of advertisers in Facebook feed, Instagram feed and Meta reel. So those are the big blocks and then medium, but with lots of opportunity, short-form video, so both YouTube and Meta reels have launched a product to directly compete with TikTok's short-form video, both the companies have discussed these products and the adoption rate on their last 2 earnings calls publicly. Meta reels, we launched viewability, invalid traffic in Meta reals end of second quarter, as I mentioned, we're now in beta with the brand safety in Meta reels and YouTube Shorts we're running viewability and invalid traffic today.

And from your perspective, I mean, to understand these videos is obviously more work than for you guys as well than kind of understanding the picture or a text. Like can you charge more for that? Or is it just for you like more volume?

Good question. So as I mentioned before, our total media quality or TMQ technology is highly sophisticated, highly granular backed by AI. So we are classifying real time frame by frame that's every single second for example, of a TikTok 30-second video, video image, audio and text. We can also identify brands and celebrities and we're proving out in TikTok today that because of this sophisticated technology, we are showing 3x higher-quality media that we are able to identify for the Cokes and the Nestles of the world, which means they're moving further away from the inappropriate content and heading into running their ads in higher-quality media and our thesis is higher-quality media leads to higher ROI. One of the thing I actually think that it's important to announce because we announced it publicly about 5, 6 weeks ago. We also hired a lead data scientist from Meta, his name is Kumaresh Singh, who is leading our data science team. He built -- he was on the team that built at Meta reels and built the model of Meta reels, so he is a really important hire for us as we continue to invest in all things related to data science and AI, in particular, for the social platforms.

And I mean this -- that's where I'm as the software guy get excited because I get the science behind it and be able to do that in real time because it's not like 10 minutes later, and that doesn't help anyone for you guys, must be amazing like the investment in tech must be kind of staggering for your guys.

Yes, it is. An example that I shared before that TMQ alone in TikTok when we leveraged open AI, on top of our AI-based TMQ, we went from 4 languages to over 90 languages with 24 hours, with high accuracy of those 90 languages. It's just one example of how we're leveraging open AI to accelerate our product road map, accelerate speed to market and continue to drive value for our customers. And getting back to your point about price because we are proving out that it's identifying higher-quality media, we are able to charge more for TMQ product because we are able to demonstrate higher value for the marketers.

And does it have -- if you think about the -- I mean, there's price and there's volume. Do these videos, will they kind of drive overall volume? Or is it just the replacement of like I used to kind of put an image out now, I put a video out. Is it driving the overall volume in the market? Or is it just a replacement?

It's driving volume as well. And we saw that. We saw that in the third quarter with measurement volume accelerating in the third quarter, our prices was consistent in the third quarter versus last year, but definitely seeing it both on the TMQ price increase with more innovative products that are driving this return on ad spend with advertisers that Lisa talked about, but also on the volume front.

Yes. Okay. And then, Lisa, you mentioned AI as one part of it, like from your perspective, because you're dealing a lot with content, and there's like AI-generated content, which kind of you want to identify because it's actually not real, but you guys can also use AI to kind of do a better -- like improve like how you kind of operate, like talk a little bit about investments and you mentioned earlier the hire is there, but like it seems such a kind of interesting for all subjects.

Sure. A great example is upcoming U.S. elections in 2024 is top of mind for the brands. And I personally spend a lot of time with the brands and they're asking a lot about upcoming elections, our investments in misinformation and deepfakes. And we are deeply investing in both now. Our goal for '24 is that 1/3 of our engineering team are data scientists led by Kumaresh who I mentioned before, and misinformation alone, I mean, as you know, Raimo, no one's cracked that code. No one's cracked it. Even the social platforms in terms of whether it's election season or beyond the election season information that is not accurate or is not real or is misrepresented, just continues to explode across the Internet. And we are heads down focused on rolling out a misinformation category at the beginning of next year. I won't share which social platform will roll it out first, but we want to get it out ahead of the elections so that marketers are able to say, yes, in addition to those other categories like adult, hate speech, all the others, violence, yes, I do not want my brands adjacent to any misinformation happening on the social platform. And the way that's built, it is a combination of human-based and AI technology.

Yes. That's good -- sorry, and I apologize, Tania, for the question, but as financial guys always like, so if you have this extra feature now coming up, does that mean you charge differently or you just kind of improve like how differentiated you are in the market?

It's the same pricing model, fixed CPMs with our clients. But volume is not based in the period, and our revenue was re-occurring based on the stickiness of our relationships.

Yes. Okay. Correct. And then you mentioned some of the newer products coming out that they are working on. Meta comes up a lot when I talk with investors. Can you talk a little bit -- maybe just to level set like what are you doing with Meta at the moment? And why might some people be excited about next year with the extension of the Meta product offering?

Sure. So Meta continues to be a really important strategic partner. Up until this beta I had mentioned that is currently running, we have been offering verification and invalid traffic solutions across the Meta portfolio. But what the brands have really been waiting for, is for, as I mentioned before, Meta to open up their live feeds, Facebook news feeds, Instagram feed and Meta reels so that we're able to launch our total medial quality technology that's already running in TikTok, already running in YouTube, so that we're able to classify and identify inappropriate content. All of that classification that categories are based on industry standards called GARM, so the tech is built to GARM, but marketers are really chomping at the bit for this. And the one other thing to call out, that is a little different from the Tiktok launch is that Meta we already have a robust advertiser base in place already using IAS' solutions on Meta, so the launch, it will be a ramp, but we don't need to go out like we had to do with TikTok and find advertisers to test the beta. We already have a robust piece in Meta.

And remind us like how big is Meta in comparison to for example the TikTok volumes et cetera, like just to get an idea of -- get some vague idea about scale.

Yes, we won't, we don't break out the volumes. What I'm comfortable breaking out, and this is rough estimates of just total advertising for the platforms -- for the social platforms. So Meta, and you have to correct me if I don't have this right, Meta is roughly $133 billion in advertising revenue annually. TikTok, last time they disclosed it roughly $10 billion to $12 billion in ad inventory, but again I'm not comfortable sharing volumes.

Yes, yes, yes. Fair enough, like, but it gives us an idea about the scale. So no wonder people are excited. So that suggests higher -- a good bit of higher volume. So with part of the work that you're doing to enable much higher scalability of your product to be able to deliver that, or is it just the APIs to go into the Meta system and be able to work with, like what's the -- if you do the work there now, like what's the big difficult part to solve?

Yes, good question. So the beauty of Meta, as I mentioned before, is the TMQ product is humming. It's humming in TikTok, it's humming in YouTube. And so -- and Meta and the brands, I mean they'll benefit from the fact that the tech is already running. We do get asked sometimes a question about processing costs because we are processing frame by frame every single second. It is more granular than our competitors are processing, the way we are processing the inventory but this is a curb to the cost, right, as the technology gets smarter and smarter the costs come down, which we're already spoken and again, we'll just continue to invest in innovation and continue to invest in launching new segments like I mentioned before, misinformation, to ensure that we're offering value to the brands.

Okay. I wanted to move on a little bit like the other big area that is exciting at the moment that you mentioned as well was CTV. Talk us a little bit through where we are in the evolution of that. I mean the -- if you think about it, you looked at Microsoft helping Netflix initially, but then people are not quite happy with kind of the volumes in there like what are you seeing on that evolution of CTV as the kind of potential opportunity for you guys?

Yes. So there's so much opportunity with CTV. I'd say CTV is a long game. And with CTV, the way to think about it is premium inventory and programmatic inventory. And in the premium inventory, Netflix being the perfect example, is we were 1 of 3 measurement providers selected at the launch of Netflix's ad supported tier. It takes time to get that tier up and running, both from getting the pricing right for Netflix, getting the user adoption right and then getting the advertising engagement in place. At launch, we launched viewability and invalid traffic detection in all 12 of Netflix's market. They've been a great partner and we'll just continue to partner with them. The other thing that's very encouraging is that brands share that, that was 1 of the #1 asks of Netflix is that they launch with measurment verification provider. So that's just going to take some time to ramp. And in terms of programmatic, as you know, we acquired Publica, a leading CTV platform over 2 years ago now, where we are working closely with other OTT CTV platforms like the Samsungs of the world. We have an exclusive partnership with Samsung where because of Publica and the integrations we have with Publica and Samsung, we're helping Samsung drive higher optimization and yield for their programmatic CTV inventory through header bidding. We are serving Samsung's ads, we have SSAI stitching where we're stitching creative real time in the Samsung's stream. But I also think equally if not more important it gives IAS access to massive amounts of programmatic CTV data, very rich data that we're able to marry up with the IAS data and help marketers just give them greater transparency into where their ads are running in programmatic CTV, because that's the #1 thing that they ask for, so -- yes, go ahead.

And if you look at your CTV journey, and I'm speaking now as a consumer, it does feel like we are still in the early stages because it's like if I compare to like the linear TV, like it still feels like almost early innings or...

It's very early innings. And I can't emphasize enough having worked at Amazon for 6 years and seeing Netflix's play right now, getting that user price point right and getting the user adoption right of that ad supported tier is really important. You've got to get that ramp going and then the marketers the advertising on top of it and then the ripple effect to IAS.

Yes. Okay. Last few minutes, I will shift gears a little bit like we are in the important part of the year for the advertisers like Black Friday, Cyber Monday et cetera, and you talking to a lot of the advertisers and some that you would see it, I don't want you to guide now, I am not interested in numbers, but what are you seeing out there in the field in terms of like how this is playing out?

Sure. So from a macro perspective, in fourth quarter, from the brands what we're hearing is again greater scrutiny on the efficiency in ROI. ROI is everything, that they get the best bank for their buck. Greater transparency in where they're making their investments, holding on to their budgets. They might be shifting their budgets both in terms of ensuring that their consumer base they're keeping them in the boat and keeping them engaged, placing their bets on the big platforms. And then the other thing I hear is less appetite to invest in the shiny objects that are the unknown thing and doubling down in what's known and again drive greater ROI.

Yes. Okay. And then Tania, like if you think about the better revenue drove better profitability. Like can you talk a little bit about the balance between like the growth and the investments that you're doing.

Yes, we had a very good year in '23 with EBITDA margins higher every quarter of 2023 so far. Third quarter EBITDA margin was 34%, up 400 basis points from third quarter 2022. There's 2 dynamics that's happening there. One, we are very -- we're disciplined and we're focused on productivity and efficiency of the investments that we're making. But at the same time, we're reinvesting for the growth of the business. I mean it's balanced. We're balancing top line growth with as you talked about multiple opportunities for growth, large and expansive TAMs. We want to be investing to drive that top line, but we want to do it in a way to sustain our margins and one other thing that happened this year is we were investing in R&D, and we're doing it, we're investing in innovation. We're investing in our product development. And under GAAP, we can capitalize more of that R&D investments, we saw an uplift from that in 2023. Now we're at this higher level. In 2023, we expect to continue that under GAAP, but we don't expect to see big margin shifts next year.

Yes, yes. Fair enough. And then the improvements that you saw this year, like since you joined as a CFO, were they like special, it was just like a little bit more on prioritization? Or like how do you achieve this -- those better numbers?

Yes. It's a focus of balancing growth and EBITDA margin. We're rule of 49 with our guidance this year with 15% at the midpoint and 34% EBITDA margin. So we're really focused on making the investments to drive topline, but also doing it in an efficient way to continue top line growth without big shifts in margins.

Yes. And how do you guys prioritize with this go-to-market angle and there's the product side, and on both sides, you can actually spend a fair amount. Like if you think about it like how do you get that balance right?

Okay. I could say that I mentioned this before placing a few bets in the areas that drive tremendous value for our advertiser base that are with the platforms that are global with massive amounts of runway and scale and ensuring that we're investing in data science and innovation both in the near term and for the long term.

Yes. And one of the things you kind of pointed out on the mid and -- go-to-market and mid tier segment was doing better. Like what did you do there?

Yes, great question. So mid tier, we're seeing a nice uptick with the mid tier segment. We define it as North of the top 100 accounts. We're seeing adoption, both in optimization or a programmatic offering context control. We're seeing adoption both in the U.S. and internationally in mid-market, our investments in automation and programmatic specialties in particular are paying out for mid-market, and we see a long runway with mid-market, both in programmatic, in retail media and in the social platforms.

Last question for me and then I'll let you go. Like where are -- if you think about clients and advertisements, where are we in the adoption of solutions like you guys, because in a way, at the moment, the market is down to 2 players like you and another one. If I look at you client list and then if you think of the people that could advertise, there seems to be still a lot of...

There is a lot of runway. I would say there is saturation with the top 100 Fortune 500 brands but when you get north of that top 100, there is a ton of runway, both in the US and internationally.

Yes. And then you guys can remind us, I think you were -- internationally, you were kind of already built out more...

Correct. Yes, international, our footprint is a big differentiator for IAS, representing 30% of our revenue, having deep footprint both EMEA, APAC and we continue to invest in emerging markets, and we're seeing really, really good growth both in markets like Latin America and Southeast Asia.

Yes. Good, perfect. I think our time is up. Thank you.

Thank you.

Thank you.