About what Perion does. You had a big earnings announcement this morning where you sort of blew away all your numbers, raised guidance, raised margins. Why do you not talk about sort of what Perion does and what it is over-delivering on and what you see as the outlook for 2025 now?
Yeah, absolutely. Thanks for having me. You know, with Perion, we've made quite a pivot in the last year. We've moved into becoming a fully full-stack platform for CMOs. The way we think about it is, you know, where CROs have Salesforce, COOs have Monday.com, CTOs have Jira, now CMOs can finally have their own platform, right? That goes, we're not trying to replace others. We're not going to try to replace our DV or The Trade Desk. What we're saying is we're going to connect all of them. We just announced an integration partnership with The Trade Desk. We're connecting all those technologies, but the CMO would have one unified platform that he can run all his campaigns, all the strategy from planning to execution, and then get reports across all of them.
Now, within that, you know, in the past year, we've actually enhanced our leadership. One of them is here, Stephen Yap, that used to do all the go-to-market and sales for DV360 at Google for 17 years. And now, you know, moved to our company today to do the exact same thing for our platform. We have top-tier new management. Within that, we just announced we bought an AI-first company that does custom algorithms for closed gardens, right? For YouTube, Facebook, Instagram. Through that, we can actually increase our total addressable market dramatically. We were only on out-of-home CTV and open web, which we all refer to as open web. Now we're moving into closed gardens. That is a very good shift for us.
Okay. One of the things you said to me at CES January, five months ago, was like, the open internet is dead. In answer to my, what's the most provocative thing you're saying? That was very provocative coming from an open internet player. Definitely moved to the top of my ranking of most provocative things said at CES. My question is, I looked at the open web numbers you guys reported this morning, and I think they were down 26%-28%, but they were down 60-something percent last quarter. The open internet was dead for six months, and now you're going to be growing it again? What changed?
You have an amazing memory. Yeah, I think, you know, open web is absolutely, you know, not the future. There's a lot of money to be made on open web, and that's great. When you look at agencies, 80% actually comes from social, right? Now when you look at brands and agencies, you see that more and more money is actually shifting from brand awareness into performance, right? Those are the two shifts that we're focusing on. How do we get better solutions for closed garden? How do we focus on performance, right? How do we make sure that the advertiser is getting the right ROI while reducing waste? They're going to get exactly what they need so they can come back and, you know, build more and more and more of their campaigns on our platform.
Okay. Okay. So one of the things you just said where you're pivoting to is the CMO-first target market, where you're going to become the platform that they can do everything on. One of the things that Trade Desk was just saying is that he feels he has four constituencies. One's the Chief Financial Officer, then the CMO, then I think somebody, and then the guy that executes programmatics. Maybe it's the person that knows about programmatic. Anyway, there's four. My question is, CMO feels pretty high in the ranking. Does that mean your client size is smaller, which is why you're dealing with the CMO and not somebody at a lower level?
That's a good question. So think about it, you know, Salesforce, everybody's using Salesforce on a sales team, right? I mean, Steve has 100 people on your team. Everybody's using that. But Steve needs to wake up in the morning, open Salesforce, look at the pipeline, look at the pacing, look at the forecast, right? That's the way we're thinking about Perion. The CMO logs in and sees the orchestration of everything that goes on, right? The programmatic people are also going to use that, right? The planners are going to use it. Data analysts are going to use that. There are a lot of different layers, exactly like in Salesforce, right? I think what we figured out in the past year, CMOs are not able to talk with CFOs about the results. There's a fundamental issue because look at what we're reporting.
We're reporting on viewability, CTR. I want to see one CMO that goes to the CFO and say, you won't believe, with the $100 million that you gave me for advertising, I got amazing CTR and viewability. That's exactly the point where it's getting replaced, right? Who cares? Like, we're selling shoes. How many shoes did I sell? Right? Anything else doesn't matter. I think that's the fundamental issue we have in online advertising, digital advertising. We're reporting the wrong things. That doesn't help the CMO prove that he's investing right, right? Maybe he is, but there's just no way to help him say that, right? That's where Perion comes in. We're actually translating the metrics that we know now into actual insights, into uplifts, right? What are the outcomes that you're going to get if you're going to invest in those campaigns?
What would an example be? Ultimately, does the CFO in this worldview only hear sales? Ultimately, the CMO's job, Chief Marketing Officer's job, is to translate impressions and, you know, brand lift and incrementality, things that maybe he's solving for. He has to somehow translate those into sales revenue because that's the only thing the Chief Financial Officer understands. If so, how does Perion help him do that?
Right. So for us, it's a natural progression for what we've started a few years back. Our retail media is actually, how do we connect their data to see an uplift in sales, right? So we started very simple a few years back, just taking some data and saying, you know what, we're going to do that campaign. You're going to tell us, did you sell more, whatever it was, Hershey's or whatever, in that area? Now it's getting more and more sophisticated. We're actually integrating their promotions and we're integrating their data in a lot of the cases to see, do you actually see an uplift? Not an uplift in people clicking your ads. Nobody cares if people click your ads or not. Like an uplift in actual sales in that area, right? So it's, you know, it's getting there. I think that's the future.
I think the future is, you know, one unified platform that can do pretty much everything. Not one platform for open web, one platform for YouTube, one platform for Meta, one platform for TikTok. I think that's a mess. You need one unified platform. You need to be able to measure actual outcomes, business outcomes with your investments. You need to be able to speak both languages, the CMO language and the CFO language, because that's the only way you're going to get more and more budgets into your platform, right?
Okay. Your contention is a CFO speaks sales only.
CFO needs to see.
They have to have sales in this.
Right. CFO looks at business outcomes, right? How do we bridge the gap? How do we bridge saying, okay, great, you guys have great viewability. You guys have an amazing CTR in ads. How do we see that this was actually translated into sales, into more people coming into the stores, into better brand awareness? I do not know, whatever the CFO cares about, right? How do we get more people into car dealerships? How do we get more people?
Traffic into dealerships.
Right. So every company has their own metrics, but how do we need to build that bridge in between the CMO and the CFO?
Okay. So in that answer, you said we need to have one unified platform across walled gardens and into the open internet. By the way, I heard the exact same thing from the DoubleVerify CEO and CFO. They were talking about there has to be unification across walled gardens. That is where the open web is going. Or said another way, that is what is defendable with pricing power, is something that looks across walled gardens and the open internet, because the walled gardens create their own homework. And even within the walled gardens, Meta's outcomes are not the same. Measurements are not the same as Google's, are not the same as Amazon's. So what is valuable is somebody that can go across walled gardens into the open internet. Okay. And that is what you are saying too. My question is, one unified platform that does not have connected television?
We do have.
Okay.
CTV is our numbers showed 31% lift quarter- after- quarter in our CTV, right? That goes directly into Perion One platform. We are working on more and more advancements on our CTV features and deliverability, connecting more channels, obviously connected that to our DCO. Our algorithm actually builds the creative on for CTV. CTV has to be, you know, it's a crucial part. We are happy to see that our CTV is 31% growing year- over- year.
Right. It makes sense if you're going to do a unified platform for the CMO, you must have a CTV solution.
Absolutely.
Somebody asked The Trade Desk about this just now, and they were saying they're going to create a marketplace where they put all these content creators, because there's a lot of them, through Gen AI content creators on their platform, and they'll charge a fee if somebody wants to use that creator to create custom creative for different target markets, which I think is probably where the world's going to go. Is that the kind of thing where you guys create a marketplace, or will you just use Gen AI so people can build creative through your platform and you'll have a creation engine that's sort of captive to your platform? What's your thinking there?
Yeah. We do not call it a marketplace, but that is exactly what we are aiming for, right? With our technology, you can build your own creative. You can ask our algorithm to change that creative based on so many different factors, right? If it is raining, if it is in the middle of the night, if it is on an iPhone versus an Android, right? We have so many different ways to tweak that through algorithm. That is already part of the system, right? I think what is so important, because you talked about CTV, as an industry, I think we are always thinking, you know, does CTV grow? Does out-of-home grow, right? For marketers, for CMOs, I do not think they should care. I think if I was the CFO of Nike, would I care if my investment is on CTV, or would I care that I actually sold more shoes?
By us having a unified algorithm that can shift between channels to push actual outcomes, maybe it's CTV, maybe it's not, maybe it's TikTok.
I'll push on that a little bit, because I think we all know that if you do performance only, bottom of funnel might be more performative in the current quarter, but if you do that for three years in a row, you suddenly have no one at the top of the funnel. You might decide to have, and in fact, we know this for a fact, that the returns on capital demanded by top of funnel CTV, or even for performance advertisers, are like one- to- one. If they get a dollar of spend to something they can attribute to a sale downstream, they'll take one- to- one.
Whereas at the bottom of funnel, they're looking for four or five- to- one, because they understand that this one- to- one that sort of paid for the ad dollar, they're getting brand value and reach, and they're getting awareness that's feeding the rest of the funnel over the next three years, maybe.
No, absolutely. I absolutely agree. For Perion, historically, we were very much focused on the brand awareness parts. Now with Greenbids, we have all that performance part. I think once the CMO defines, you know what, I need people to go to physical Nike stores, but I also need people to buy, right? Those are maybe two different metrics. I am going to put $5 million towards that and maybe $7 million towards this. Those campaigns can run separately in parallel, having two different metrics, right? Maybe one would have more CTV and out-of-home, while the other is going to have more TikTok, Instagram, like really push people to buy, right? It is fine. All I am saying is, when you have in front of you the outcomes and not the channels, it is a different story, right?
CMOs, in general, good CMOs do not really care if that is CTV or not. They care about, did I drive real outcome to my company?
I guess I would just say outcome over time. Outcome is not only this quarter. I mean, even though that might be the highest and best you want to divide, you want to build outcomes for next quarter and quarter.
Right. You can define what outcomes means this quarter, right?
Right.
Better brand awareness, better sales, better.
It's true.
Right? It can change. It's dynamic.
Yeah. Okay. Are you finding that that's the reason that your open web numbers have improved so dramatically, quarter- over- quarter, and now you're going to grow in the second half like you thought you might? Is that because you're getting new clients, Tal, or because these new capabilities are getting you more money from old clients? Where's the, why is it getting better so much faster in the open web?
Right. It's actually three things. Yes. Existing customers now spend more with us. You know, I think with Steve coming on board and some very talented people with him, we're seeing new customers as well. The Greenbids company that we bought, we started from a partnership. We've used them with a few customers saying, you know what, we have a few customers, they want to try out YouTube on steroids. Can you do that with us? They said, fine. It worked so well, customers came back, right? We actually saw synergies before we bought them. Through that, they said, oh, you know what, you guys are doing great YouTube. Can you do this as well? Can you do web, right? It's already starting. The synergies are only pushing our revenues.
We just got started, you know, again, we've announced Perion One and the new management team beginning of February. So they're only three months in, but we're already steering good numbers. I think, you know, Q1 looked good, but what really encouraged us is not the Q1 numbers. Q1 numbers are great, but the actual pipeline.
Okay. That sort of...
Yeah. It's actually stronger than what we anticipated. It's stronger than what we saw.
Because of Perion One.
Yes.
Because of this strategy you think is working and taking hold. Okay.
Yes. Again, we just got started, so.
Right. I was just going to say, what other, I know you've said for a long time acquisitions are a key part, and you've been sort of steadily making acquisitions every one or two quarters.
Yeah.
What's missing? Do you have to do more with the walled gardens? Is that the big piece that's missing, or is there another channel I'm not thinking of? Because , let's answer that question first, because I have another one. What's missing? What else do you have to build to make this platform, Perion One, have more pricing power or the ability to bring the next customer faster?
Yeah. I think AI, you know, is evolving so fast. We need to evolve even faster. And you can see that, you know, you saw PubMatic just launched their own AI platform. Nexon launched their own. It's a race, right?
Yeah, it's a race.
We need to be faster. Greenbids is a great addition. It is a quantum leap for us on AI. We are constantly looking for more and more companies to advance what we have and to make it faster, to make it better, to make it unified. That is our focus.
Okay. Okay. So more technology forward, using technology.
Yeah.
Okay. Direct out-of-home was the big, the Hivestack acquisition, which I want to say is two years ago. Does that give you a big competitive advantage? Because a lot of guys do not have as leadership a position in DOOH. When you talk about Perion One, one of the big pieces that you have that a lot of people have not emphasized to date is digital out-of-home.
Yeah. I think only us and now T-Mobile with Vistar, we're the only companies that have that. If we actually want to be the best friend of the CMO to do the entire journey, to manage all their, to orchestrate all their activity, you can't do this without out-of-home, right? You can't do this without CTV. You can't do this without closed gardens. I think we have an advantage that other companies don't.
Okay. One of the questions I have, Wall Street believes we're moving more toward performance. Sounds like you believe that too. Digital out-of-home has a real problem with performance. How many people walk by? How many people notice the billboard? Even if they all went into the store directly below the billboard, did they all buy something there because of the billboard? How do you speak to that? If the world's moving more performative, which I think you believe it is, how does digital out-of-home fit with that?
Yeah. Listen, I think the issue with out-of-home was always, out-of-home is a great format. I think everybody in the advertising industry believes out-of-home is great. The fact that agencies of out-of-home are so separate from all the other parts of the agency, that's the disconnect that we have. If you have one algorithm that can say, okay, what are the outcomes you need? I'm going to push out-of-home around the store, but actually also push audio for Spotify when you're getting near to the store and CTV for people that live around it, right? When you look at it on a holistic level, it makes a lot of sense. When you buy, like agencies have out-of-home separate, nobody talks with them, right?
You have CTV, and then you have social, and then you have AdWords, like the fact that it's so fragmented doesn't help provide outcomes, right? That's where the waste comes from.
Okay. Okay. So I totally hear your point, and I hadn't thought of it at all. When I think of digital out-of-home, of course, I think of Times Square, right? Which is best thing, and they had Times Square. I mean, Times Square is sort of Hivestack's anchor tenant, right?
Yeah.
You're right. Within the stores, that's a smart way. Like if you like these products, go to floor two in the back and buy bathing suits on sale or whatever. I totally get the out-of-home moving people through the store. That's smart. When we talk about big, I was going to say, you know, premium placements on digital billboards, how do you make that performative? Or do you just not make it performative and people who buy it just buy it at a low CPM for reach and they don't demand performance metrics?
Yeah. It depends, right? We have, you know, when Meta launched their AI glasses with Ray-Ban, right? They ran a huge campaign with us nationwide here on all the billboards.
With people wearing glasses?
Yeah. So, and they had video ads across all subways, Times Square. I mean, it's a great.
It's a reach. They're using it as a reach medium to say, hey, we have this new product.
Yeah.
It's substitutable to TV, except it's in your face at eye level or can't miss. Okay.
Yeah.
It substitutes to TV, linear TV in a way, not mappable.
It is. It is. The only thing is we're actually telling the advertiser, listen, you have an amazing budget with us on CTV. We can actually sell you the video ads out of home, just no audio, so different price. Again, non-skippable, a lot more eyeballs for every ad. So it's, you know, it works.
Yeah. It's sort of complementary to CTV. I totally get that. The top of the funnel. Like you say, lower price point and probably different reach, probably incremental reach to CTV, because by definition it's out-of-home. They're not home watching a CTV. To me, it's just another touchpoint where somebody can say, oh, I got to remember to ask about that or go into the store or something and buy that. Yeah, that makes a lot of sense.
Somebody actually, somebody like three months ago told me he saw on the subway ads for The Trade Desk.
What?
I said that's.
What? It's an enterprise company.
You saw it as well?
Yeah.
It's on Metro- North.
What?
Yeah. I have advertisers on digital out-of-home billboards on Metro-North. Yeah. I mean, if The Trade Desk, and those are smart people.
Smart people. Totally enterprise people.
If they understand the value in out-of-home, then it does bring value.
Maybe there's a lot of ad buyers in local in New York.
Yeah.
I mean, maybe that's why. Yeah. Okay. One of the things you did recently is you upgraded your talent team by bringing Stephen in. On duty, you have to hand him the microphone. The question on the floor, I'm going to ask this, not of Tal. I'm handing my microphone to Stephen, who just came from 17 years at Google. Why don't you tell the audience why you came to Perion?
I can do it from here?
Yeah, yeah.
It's just my voice. Thank you. Yeah. Between DoubleClick and Google, I spent 25 years, and during that time, just helped build their ad tech stacks across the globe. You know, stemming from a conversation with Tal and talking about the recent acquisition of Hivestack, actually, and just kind of my belief and how that fits into the digital ecosystem and kind of bringing less or bringing more efficiency to the industry and a lot less waste, just in terms of how people plan and execute and deliver their ad campaigns. We got into a conversation, and you told me all about Perion and kind of the vision and what they were really kind of driving towards. You know, what it reminded me of was like an old DoubleClick, like back in the day, right?
When obviously the pioneers of ad tech with the launch of their ad server, because they had so many great fundamental products that were foundational in kind of building the industry and that how they wanted to kind of pull them all together to make a more efficient, less wasteful industry. It was something that I was very much a firm believer in across the industry that we just, we needed to do better. Based on the vision, that aligned perfectly with kind of where I thought the industry needed to go. Here I am.
Okay. Great. Tal, one of the questions I'll ask you, staying on the issue of Google, is there's a real controversy as to if the DOJ in September has a hearing and says, look, Google has to spin off its ad tech, ad server and its SSP to get it out of the monopoly. Google will appeal, but I think there's a consensus, growing consensus of opinion in D.C. that she will put behavioral remedies on them from the day she decides that in September, even while they're fighting to not have it structurally spun off, but that the behavioral remedies will actually change the balance of power in the ad tech open internet. Do you have a point of view as to how much that helps Perion or doesn't help Perion?
Like I get ratings from two, which means doesn't matter to us, to 11 out of 10. PubMatic's like it's an 11 out of 10. It affects everything we do. Do you have a point of view if she effectively de facto spins off in September the Google ad server and SSP, even if she just does it behaviorally?
Yeah. First of all, I think that's going to take time. Whatever they're going to decide, that's going to take a lot of time. Second, okay, let's say it spins off, right? Now what? It's a different company. Still, people would need to buy YouTube through a unified platform like Google, even though the, let's say, DV360 is not owned by Google, by another company. How would that help other companies? Why would that? Like at the end of the day, it's not about, I think a lot of time we think that Google, by the way they act, is forcing everybody to buy your thing through them. I'm not sure that's the case. I think it's about user behavior. People are using YouTube, right? And there's a platform to buy YouTube. People are using TikTok. Whether we like TikTok or not, people are using that.
As long as they use that, they do not use other things, right? You need to buy their ads. Again, going back to CMOs, they do not care if you are buying that through PubMatic or through us or through whoever. They just need to get those audiences to buy their shoes, right? I am not sure that would actually matter to any of us in the industry.
Okay. Like I said, real controversy. Two to 11, depending on which CEO you ask. So yours is a two. Got it. All good.
Yeah.
Like you say, it probably takes a long time. Google has a lot of resources, as does the federal government. It might take a long time. Okay. Let's talk about SORT. Remember Sort? Okay. SORT used to be a big advantage of yours because you had all the great search data, which is first-party data, and used to like apply it to the open web. Is SORT, is that still a thing? Is it not really a thing? Tal, update us on what's happening with SORT and how it adds pricing power and competitive advantage to Perion today and its current, you know, sort of mix of businesses.
Right. So SORT started.
Better tell them what it is too.
Okay. SORT started about three or four years ago when everybody thought cookies are going to go away. Again, see, things take time and do not go anywhere. What we've done, it's funny, that was deep learning back then. Now everybody calls it AI, but we've done it four years ago. How do we recognize users without knowing anything about them? We've done that for web. That was very successful. In the past, I think the past year, we've extended that to CTV. Now we're using, you know, again, I think when you look at the bigger picture, cookies is no longer, it's not as big of an issue as it was before because, again, we're going to go into closed garden. YouTube doesn't use cookies. Google, yeah, Facebook, Instagram are not about cookies. We just integrated the UID 2.0.
We have that and we have SORT. It is not really about fighting our way to live without cookies. Now it is more of what a lot of companies call curation, right? How do I get inventory on CTV and web in a way that is not wasteful, right? We are using that to increase efficiency, reduce waste. Again, it all goes back to how do I provide actual outcomes? SORT actually evolved so much further than just a cookie list.
It used to just be an ID solution, and now it's actually going all the way to outcomes.
Yeah.
Does it give you pricing power and competitive advantage in getting new business?
It does.
Okay. Still, that was the original role for it, right? Was to have an ID, a unique ID solution. You're saying you've kept the uniqueness of it as a product offering. It's just a lot bigger now.
Right.
It's now about outcomes, not just an ID solution. I assume people are still using cookies, right? Since they haven't gone away or not really?
Yeah. No, people can use cookies. I mean, some of our clients are saying, you know what, just give me half of it on SORT and half of it in cookies. Let's see what works better. That's great. We can use, we have Experian, we have UID 2.0. We have it all. As I said, we're not trying to replace anyone. You know, whatever you're used to, whatever you want, you can find it on our platform. Just use it. We want to be as flexible as probably Salesforce, right? I mean, we're not going to replace all the other parts, but you're able to connect those parts into us.
Speaking of this subject specifically, doesn't scale matter? Because I mean, I don't hear different words out of The Trade Desk. They want to be everything to all people except for they're doing it at the holding company level, right?
Yeah.
Don't they have a huge competitive advantage because they have $12 billion of gross, you know, of ad spending going through their platform in terms of data advantage, in terms of algorithm updating? How does that position somebody like Perion? Like what's your competitive advantage versus a guy that's really big, has a lot more data?
Yeah. I think The Trade Desk are doing an amazing job on everything they do. Our difference is, one, we have a DCO, so creative really matters to us through technology.
DCO.
Dynamic creative optimization.
Yep. They definitely don't have that.
Yeah. They do not focus on creative. We're connected to a lot of data sources. The Trade Desk is great and we're connected to them, but we also have SORT. We also have other data partners. We have our digital out-of-home. We now have closed gardens, which a lot of the other companies do not have. You know, even Google is serving Google inventory. They're not serving Facebook inventory, right?
That's true.
We are.
That's true.
Again, I think The Trade Desk are doing an amazing job at what they're focusing on. We're looking at things a bit different.
Yeah. That is the way to create differentiation and pricing powers. Do not do what other people are doing. Does anyone have any questions from the audience? Yes, Andy.
Now that you have so much more comprehensive platform, the sales cycle a little longer, what does that look like? No, it's okay.
Sales cycle length, is it elongating or shortening?
Do you want to answer the sales cycle? Because you're the person who's in charge of sales cycle.
I am. Yeah. Right now, we haven't seen an elongated sales cycle. That is primarily because when you look at things like the Greenbids technology in particular, the application of those onto kind of existing campaigns is relatively easy, right? With their kind of deep integrations with DV360 and the likes of The Trade Desk and Meta and YouTube, it's one of those things that we can implement within a matter of days versus a matter of months. We don't anticipate right now. Say again?
More affordable.
Yeah.
It's incremental costs.
I don't think I understand that
People are paying more for it or whatever. It's costing you more. You're buying more value.
Yeah. For performative solutions, right? Like they are just for any kind of performative solution in the market, you know, the amount that they invest within the tool in terms of what they buy, they get obviously exponential return as a result of it. Yes, you pay more for it, but the dynamics of those types of programs, right, where you're paying for the performance of it, it actually, that's not really what inhibits or slows down the sales cycle. It's usually technology implementations.
Are you saying that I apologize that they're sort of indifferent to the return on investment?
No, no. They're very much concerned about the return on the investment, but because it is, they're implementing a performative solution. If I, a great example is if I have a campaign with YouTube and I expect 1.5x return on my ROI from my return on ad spend, and I know historically that when I apply these performative solutions, I get 3x ROAS, then as long as I'm covering my cost, which we obviously, we do a whole analysis in terms of, hey, this is what we can bring you. We run tests with them because that's how actually the custom algos work through kind of the testing methodology.
It doesn't necessarily elongate.
No. No, not materially.
Any other questions for Tal or for Steve? Or we actually have the CFO here too. So we got lots of people in the audience. I want to talk about search. So search has gone from 50% of your revenue. It's on its way to 5% or zero. My question is, was there, and what was interesting about that to me, Tal, we sort of saw that visibly because of changes that Microsoft was making that disrupted almost every player they've been partnered with. It wasn't just you. It sort of negatively injured a lot of different people. My question is, we also saw your open web, like advertising line item segment of revenue fall at the same moment. Was that because the advertising segment was benefiting from search data that as the search sort of disappeared? Or was it just coincidental they were going down at the same time?
I know that we're about to grow in open web, but was there four months, four quarters there where our search data, as it came out of the benefiting the advertising business, we saw negative comps in open web?
Yeah. It has nothing to do with data.
Okay.
It has a lot to do with market conditions. What happened is, beginning of last year, you know, I guess that took a bit of time for us to, for the market, the industry to realize that there's now a lot more inventory than there used to be, right? Historically, Perion made a lot of money out of managing inventory, right? We did not call ourselves an SSP, but you know, our search business was basically an SSP for Microsoft. We did not call ourselves an SSP on web, but you know, Vidazoo was basically, Vidazoo, the company that we bought, was basically kind of an SSP for websites. What we saw is a shift in the market where, you know, TikTok came and flooded the market with inventory. Then Reddit really grew and Pinterest grew.
All of a sudden, there was way too much inventory and not enough marketing budgets to fill them in. We saw those two drops because, you know, search, Microsoft did not need so much inventory as they used to in the past. I guess that is why they have changed their mechanism. We also saw that in web. To be honest, you know, we are speaking with, you know, all the major publishers here on Manhattan Island and they all saw the same thing, right? They have less inventory. I mean, they are not getting as much organic traffic from Google as they used to. All the websites. With that, you know, our web activity started to go down. It is not a coincidence. It is unfortunately for us a perfect storm because we were so heavily dependent on managing inventory.
That is why we actually made the shift on focusing on CMOs and on budgets and not on inventory. You know, we're trying to get to a place where we're agnostic to whatever inventory they need. We don't want to play that role of just managing inventory. We want to play the role of providing value to the people that actually have the money to spend on media. That was the major shift. I think through that value, we were able to get more campaigns to run on web now. We just came to it from a different angle, not from managing inventory, but to managing demand.
One of the things that sort of Jeff Green and I from The Trade Desk disagreed on stage is he thinks all the power is with buyers. I see this whole notion of curation shifting power back to the sell side, shifting back to sellers. He disagrees, no problem. What you just argued is that you agree with him, that because of the rise of TikTok and because of the rise of Amazon Prime coming into the ad market and Netflix coming into the ad market, and you gave another example that I'm forgetting, that there's been this explosion of inventory, let's call it high quality inventory, explosion of ad units that are now available, which your opinion is a shifted power back to the demand side and away from the supply side. Is that what I heard you say?
I think it's always better to be closer to the person that holds the money than the inventory, right? I think curation, even though curation helps publishers, at the end of the day, the way it helps publishers is by providing better value to the advertisers, right? You don't need to buy all my inventory. You just need to buy this part, which is, I don't know what, women that love yoga in Manhattan at whatever, right? I'm just going to sell you that segment. Again, it all goes back to what does the advertiser need? They're holding the money. They're controlling that. On that part, you know, I think I agree with them.
Is this strategy you're executing require more employees than the olden days?
No.
No?
Because we're focusing on AI so dramatically, also for internal efficiencies.
Okay.
So no.
For efficiency too, yeah. I do not know if you have seen this. I do not know how close you follow Mark Zuckerberg, but he is saying that by middle of 2026, he believes that the code writing ability of Llama, which is his large language model, will replace every mid-level software engineer. He is a software-first platform. Can you imagine?
Yeah.
I mean, that's amazing if true, because they're on the cutting edge of doing a lot of stuff. So you can replace this middle-level software engineer at Meta. How many software engineers can you replace at somebody that isn't staying at the very cutting edge?
Yeah.
That is super interesting. Okay, we have about one minute left. Any questions from the audience? Okay, then I will call it there. Thank you very much.
Thank you very much. Thank you.