On Tuesday at the Cannes Lions festival, leaders of the world's four largest agency trading desks participated in a panel hosted by Rubicon Project. The discussion touched on a wide range of topics, including deal automation, technology investment and transparency.
One macro takeaway is that the so-called trading desks resemble each other less than they used to. Xaxis for instance, has become more publisher-focused and today may have more in common with a Rocket Fuel than with an Accuen.
A transcript of the panel discussion follows.
Moderator: Jay Sears, SVP at Rubicon Project
- Brian Lesser, CEO of WPP’s Xaxis
- Stephan Beringer, chief growth officer and president of Publicis Groupe’s Audience on Demand at VivaKi
- Josh Jacobs, global CEO of Omnicom’s Accuen Media
- Arun Kumar, president at IPG Mediabrands’ G-14 Mediabrands Audience Platform
JAY SEARS: I like to start these things with some quick stats and facts. Brian, we’ll start with you and run down the line here. Out of each $1 of advertising revenue spent with your advertiser, not with you but money spent overall, what percentage of that dollar is spent with automated channels?
BRIAN LESSER: I’m sure it’s well under five cents. If you count broadcast, linear, media as well.
STEPHAN BERINGER: Six.
JOSH JACOBS: It’s in the five cent range.
ARUN KUMAR: Five to seven.
JAY SEARS: What will that percentage be in 2016? How much higher are we going to go?
BRIAN LESSER: Ten percent.
STEPHAN BERINGER: Ten percent on the five.
JOSH JACOBS: Ten to 15.
ARUN KUMAR: Twenty.
JAY SEARS: The first trend we’re going to talk about is the development of the order automation budget above the auction market. We’ve seen this in our business in Rubicon where we’ve invested in order automation and it’s pushed us closer to these fine gentlemen here. Part of that is the role of first-party data. I want to talk about both publisher and advertiser first-party data. Brian, you announced this past week you’re investing $25 million to build your second-gen DMP. How important is first-party data?
BRIAN LESSER: First-party data is extremely important. The biggest problem is there’s a finite amount of data. We do a lot of business in the US. We can augment some of that first-party data with third-party data, but some of that data doesn’t perform and it’s expensive in relation to the performance it generates. We do business in 40 markets around the world and in most of those markets, you can’t get third-party data. We developed Turbine to better harness data both from clients and publishers and create what we call second-party data to model off the data and create unique segments that perform better.
STEPHAN BERINGER: I have a question for Brian. This $25 million thing I found quite impressive. Do you know how many clients are in the room? No clients. So we can talk openly. My four months' tenure, I’ve learned that’s an issue in this market.
So, the $25 million, what I’m trying to understand is: Is that a proprietary technology?
BRIAN LESSER: Thanks for that question. My technology budget last year was $50 million. I’m not sure what you invested in Publicis—
STEPHAN BERINGER: Less than that.
BRIAN LESSER: But the $25 million is the money I invested into Turbine. And it’s proprietary. It’s built from scratch from the ground up.
STEPHAN BERINGER: Omnicom cut a deal with Aggregate Knowledge? So you’re probably not spending $25 million?
JOSH JACOBS: I think Aggregate Knowledge spent more than $25 million developing that technology. All kidding aside, our strategy is very different than Brian’s. We work as a systems integrator and work with all the technology companies across the LUMASphere. And we’re tapped into a tremendous amount of innovation being funded by venture capitalists instead of our own dollars. And we pick and choose technologies that work to piece together solutions.
BRIAN LESSER: It is different, but we don’t build everything, nor do we try to build everything. We’d never be competitive. In the early days, we built a DSP and we moved away from it since we found partners very near in the space. But the DMP we specifically wanted to build ourselves. But similar to you, Josh, we try to integrate as many great third parties as possible in terms of the technology stack.
JAY SEARS: Arun, you said one of your three priorities is working with media owners to “assign value to the media owners’ first-party data.” What does that look like at Mediabrands?
ARUN KUMAR: This goes back to what you were discussing around data. Like Brian mentioned, most markets around the world don’t have third-party data sets. Advertisers have first-party data sets that are limited, so you’re going to have to rely on publishers. So in markets like Germany and France, it’s publishers that can help you get better solutions.
It’s also about helping them understand the value of what they have, and redefining what premium means. It’s not just page one or two, but it’s my audience doing an action I actually want.
JAY SEARS: Next topic is going client-direct. What’s interesting to me, and Stefan maybe you can start here, is in addition to servicing Publicis agencies, you’re now going to independent agencies with the VivaKi offering. Why was that attractive? You all have plenty of work to do to build automation into the operating agency.
STEPHAN BERINGER: The fact of the matter is that [to Brian Lesser] you guys are investing $25 million. I should be your fiercest competitor, but we probably have a similar strategy. Anyway, there’s investment behind that. There’s lots of intelligence in there. There’s a market of independent agencies who do not have the means to go down that track and invest that kind of money into knowledge and infrastructure. Nevertheless their clients need the same kind of solutions. We speak with a lot of agencies, which then elegantly takes you to the super structure in terms of thinking about the alignment of technology, data and content.
That’s something only a couple groups in the world can actually build. Thinking about how rich your data can be is obviously useful to a broad set of clients.
JAY SEARS: Does Annalect go client-direct? Have you contracted with independent agencies outside of Omnicom or are the Omnicom operating agencies your sole focus?
JOSH JACOBS: It’s definitely more of an Omnicom focus. We have worked with a few outside shops. But in general, if you look at the breath of companies under the Omnicom umbrella, we have a tremendous amount of opportunity within our own shop to bring leverage to the market, to bring new business models in to all our affiliate network agencies. And we’ve built out a stack that’s a global data platform, a global supply marketplace, a global buying platform, that we can roll out to all those different agencies.
We’re a little less focused on building a standalone agency or business and a lot more focused on the idea of the transformative impact we can have across the entire network.
I’m not out hunting for the next client as much as I am looking for how to drive this deeper into the entire business.
BRIAN LESSER: That’s an important distinction. Coming back to my problem with agency trading desks, at Xaxis, we’re not trying to be a centralized resource for programmatic within GroupM or WPP. We’re building a programmatic media platform. In the long run we’re going to be more sellers than we are buyers. You have to think about how is Xaxis different than Accuen. We’re getting to a point where you can no longer compare these companies, it’s like apples and oranges.
People like to say Publicis is different from WPP. Well, they really are different and they have much different business models and strategies. That’s an important distinction.
JAY SEARS: Arun, client-direct?
ARUN KUMAR: Most of the focus has been in-house on the agencies. There are certain markets where there are opportunities outside, which we pursue far more aggressively. Especially markets like Japan and China where there are a lot of independent shops that don’t have their own solutions, which want to work with us and we work with them. There are markets like Spain, which are performance driven, and where we go direct to clients as a performance solution. And where 70% of our business comes from outside the group. We do see ourselves as revenue drivers, but for now the focus is on our own group.
JAY SEARS: Let’s talk about overall positioning. Arun, you have a much wider definition of advertising automation. Can you talk about that and the hub-and-spoke model?
ARUN KUMAR: So the role we see for ourselves within the group is to drive the automation agenda, which is a combination of how you’re buying programmatically and creating the marketplace and driving it all through technology. That’ s not just digital, but across the entire ecosystem. It is a vision, an aggressive vision. If you look at it, it’ll happen whether we want it or not, because that ‘s the direction in which it’s headed. It’s just a question of whether you think it’ll happen in 2016 or five years later. It’s definitely gonna happen. We see ourselves as people who have to do it internally to drive it. And because we’ve been at the forefront of programmatic all these years, we’re driving that agenda.
The hub-and-spoke model is to try and see how we can cascade down some of the developments down to the smaller markets faster so they’re not left behind, and you don’t have these two markets, one set which is really advanced and the other set that isn’t getting there.
That’s why the hub-and-spoke model works, especially for the smaller markets. But all the big markets have their own definitions.
JAY SEARS: Your definition of advertising automation is much more inclusive compared to some others. You’re including linear TV and pieces that are really only workflow. When you stated this vision of 50% automation, your Magna group issued research that resized the market to include pure programmatic and also workflow.
ARUN KUMAR: Yes, because at the end of the day, our objective is to drive efficiency and speed to market. You have to differentiate between creating a programmatic marketplace vs. automation. The two of them are not one and the same and the objectives are not one and the same either. When we say 50%, that doesn’t mean 50% will be transacted in real time over an exchange. That’s definitely not the objective.
I think we started to get broadcasters, even the linear TV broadcasters, over that hurdle, which is why we formed that consortium. I don’t see this as a single holding company issue. I think this is something which all of us in the industry will drive forward with, it’s just a matter of time.
JAY SEARS: Brian, let’s go to you. Xaxis isn’t a trading desk. We know that. You are a product company. Yes?
BRIAN LESSER: Yes. If you look at a media plan that comes out of one of our partner agencies, Xaxis isn’t a line with a number next to it. You see Xaxis TV or Xaxis Mobile.
JAY SEARS: Aren’t those packaged media offerings?
BRIAN LESSER: We’re not trying to be services. Each agency develops its own programmatic buying unit, and we help them do that. We are not a trading desk. When we founded the company three years ago, we did a lot of stuff that’s trade-desk-like, but we’re migrating towards being a supplier to trading desk, not being a trading desk.
JAY SEARS: Your premise is the trading desk function is pushed out into the operating agencies, and you are really an audience-based media company.
BRIAN LESSER: To your point, eventually all media will be traded this way. If Xaxis is occupying this space of programmatic buying experts, then one of two things would happen. Either we would take over the agencies, which is clearly not going to happen, or we’d go away, which I’m not going to let happen. We’re pushing toward the supply side and the way we do that is through products.
JAY SEARS: Arun includes linear TV. If it’s not programmatic, you’re not interested, right? The wider definition of Arun, where he’s dealing with workflow, that’s not Xaxis?
BRIAN LESSER: That’s right.
JAY SEARS: Where does Accuen sit in all this? I see these two guys as opposite ends of the ring here and you guys in the middle.
JOSH JACOBS: I think we’re talking about the same thing from slightly different perspectives. What we agree about is all media will be bought programmatically, and programmatic isn’t a channel, it’s not a separate kind of media. It’s the wrong distinction to be looking at it with. When you run a huge network, and OMG is 18,000 employees worldwide, and you say, “The way we buy and trade media is going to be transformed by technology,” if you sit there thinking the way that’s going to happen is to do all of that, you’re crazy.
The model has to be how you make 18,000 people more programmatic, and thus our agency more programmatic, instead of the other way around. The trading desk served a very useful function for all the holding companies. It allowed us to stand up this separate entity to bring new kinds of people, new kinds of skill sets, different kinds of investment models, put $25 million into development that’s very hard to do in a professional services organizations, all of those kinds of things. If you look around this room here, there’s a bunch of people who’ve worked on programmatic for fifteen years it’s seems like, and it’s kind of arrived. We’re at the point where the challenge is completely different. It’s not getting these skills into the agency, it’s not how we get technology into the agency, it’s democratizing it across the whole agency.
That’s a tipping point for everybody here in the room, and a cool one.
JAY SEARS: You said we’re going from embedding to empowering. Is the premise still to put yourself out of business?
JOSH JACOBS: If I do my job well, my job doesn’t exist. You don’t need a trading desk or a programmatic person. What you have is an agency that does business differently. And that delivers, with all due respect to efficiency and workflow, the issue here is delivering a harder working advertising dollar, not an employee who can leave an hour earlier.
For me the mission is to go back to our client and show we can invest their dollars more effectively and drive bottomline business results for them using technology to do it more effectively than we ever have before.
JAY SEARS: You just answered the workflow question. The workflow piece, which Mediabrands is inclusive of, it straddles Annalect and Accuen. Is that correct?
JOSH JACOBS: We try not to talk about efficiency and try to talk about efficacy. To the extent that automation allows us to do things differently and present a different result with a better impact for our advertisers, we’re all in.
The issue for us is not reduction of labor costs as the primary function. It’s increasing impact.
ARUN KUMAR: What we should keep in mind is that when you get into workflow automation, the objective isn’t to cut the number of people, but to make sure you get the right kinds of people in and focusing them on the right thing. The need for creativity and strategic thinking is only going up. At the end of the day, you’ll have to measure everything on the impact on outcomes. It’s just a question of whether you’re going to keep focusing your attention on stuff that can be automated, but other stuff, which can’t be automated, needs a lot more resource. And how are you going to shift that resource? But ultimately, the objective is the same.
JAY SEARS: The workflow focus gets you more on the planning piece and the strategy piece and that’s where your return is?
ARUN KUMAR: That’s where the return should be.
JAY SEARS: Stephan, you said programmatic is in a perpetual puberty.
STEPHAN BERINGER: Did I say that?
JAY SEARS: You did. If you’re successful at VivaKi AOD, does your function go away?
STEPHAN BERINGER: No. I find this conversation kind of bizarre perhaps because I’m bizarre. But from my point of view, we need a different definition for programmatic. If you take a step back, it’s about data and technology that enable you to do things in a more automated way. The big problem we’re solving, from a marketer’s standpoint, if you look at commerce and communications and marketing, 15 years ago the notion of mass media started to break down but companies still needed mass market results. They started struggling and finding siloed solutions to the problem. And now we’re entering this new era where programmatic, defined in the right way, can actually deliver mass market results without the means of mass media. That’s why I’m very critical when it comes to these isolated discussions around channels and processes, because ultimately you can’t separate them anymore.
Everything data, everything process, everything media, everything audience, segmentation, needs to be orchestrated in one central place and then leveraged as a pull-through by all instances in a group or an agency. Ultimately that takes you to a complete breakdown of disciplines if you take it to the extreme. Why wouldn’t a DM agency or sales promotion agency coming in from one angle [unintelligible] pull through the whole system and basically leverage full data power of media buying to start targeting in real time, and so forth.
That’s the scenario we’re looking at in 45 years' time, as something that will be completed by then. There are just different phases on how to get there. But to answer your question, process is inherent. You cannot separate them. If you look at it from a technology standpoint, all the campaign automation solutions that are in the market already, and how these solutions will evolve, we’ll have automation platforms that let you manage complete journeys through a complete cycle of the customer, independently from the show.
JAY SEARS: I want to talk about unified negotiation for a minute. Josh, maybe I’ll start with you. When we first started years ago, and it was just the RTB business, a lot of the folks from operating agencies and holding company executives didn’t care what we were doing. But as we built the infrastructure around order automation, in addition to the auction, and as we’ve shown how to leverage first-party data, and as the finance guys at the holding companies see the profitability of the operations you guys run, there’s been increasingly a unified approach to automation. How often, when the media owner comes in to talk to the investment team or the trading team, the discussion is unified. Where they’re talking about bespoke, order automation and the auction. Is that the outcome you’re driving for?
JOSH JACOBS: When we talk to publishers, one thing we say right out of the gate is that programmatic isn’t a separate channel. You shouldn’t have a programmatic team and a direct sales team, because you’re screwing both of them. You should look at building a relationship with a holding company. We approach the publishing community with the perspective that traditionally media has been bought on a very transactional basis. Often in a holding company where you’re talking with literally hundreds of teams that might be dealing with the same publisher, you’re not sharing information about price, or sharing data across these pipes or looking to buy smarter across them. You’re just issuing an I/O and acting like the relationship goes away after that. What programmatic and platforms have done is re-envision the relationship with the publisher community and say, publishers are actually part of our platform. They’re strategic partners to us and we need relationships with those publishers that are mutually beneficial to both of us and that allow us to more effectively achieve the results we’re looking for for advertisers.
[To Brian Lesser] You had great press about this this week, but one of the things that all the articles about the 100% Xaxis kind of news is that what that misses is there are a tone of publisher benefit involved in creating separate trading pools. When you look at why are people setting up these different pools, part of it is what we’ve all done very effectively over the last decade is we’ve put a lowest common denominator standard in place that allowed us build out this ecosystem, and that’s awesome.
It sucks for publishers though because there’s no way they can differentiate themselves within that. There’s no way for them to do higher impact type of execution.
JAY SEARS: And there’s no way for your operating agencies to differentiate themselves to their clients.
JOSH JACOBS: Absolutely. So if you look at what’s starting to happen, we’re starting to see publishers as part of the overall technology stack. What we’re looking at these various private pools to do is to bring more transparency to who are the advertisers, so publishers feel more comfortable working in different ways. We’re looking to bring specific sets of publishers into the market that we know we want to work with in more strategic ways. And we’re looking to jumpstart innovation by eliminating the hops you have that force you to a lowest common denominator type of integration. Publishers are just another technology partner.
JAY SEARS: Brian, you guys made some headlines in the last couple of weeks because you said everything will be private. It’s funny because I have no idea what triggered that. You guys have said this privately, numerous times before. You said you really don’t want to be part of the auction.
BRIAN LESSER: So GroupM, late last week, said that by the end of the year, they’re going to buy zero impressions at auction. That was a blanket statement, obviously, and since we’re part of GroupM, that would apply to us.
Within Xaxis right now, 62% of our inventory is bought on a forward contract or reserved. We still do have quite a bit, 40% or so, that we buy in a marketplace. I would agree with Josh that you probably don’t want that to go to zero because there is a need for that. But I’d like to see that at 90-95% and the reason is simple. Advertisers use WPP or Xaxis because they want a competitive advantage in media buying. That means they should use their leverage to get better rates. We feel if all you’re doing is using your first party data to buy inventory in an ad exchange, you really haven’t accomplished much.
There are a lot of advertisers who talk about bringing their trading desk in-house, and that’s actually what they mean from my experience. They’re going to have a contract with a DSP and hunt for cookies in a DSP. That’s a start, but it’s not really an accomplishment.
For us, our role in the organization is to give advertisers leverage so they have better pricing on media and can better performance, that they find their audience in more engaging places. And that means moving away from an open market.
ARUN KUMAR: I think this has been seen especially in markets outside of the US, where private marketplaces have been in vogue and they’re only growing. I agree with a lot of what Josh said, which is that you have to look at publishers as partners, which we haven’t, because I don’t think we’ve yet gotten to the stage where we tell publishers I will pay you more for an audience I really want to buy because it’s of value to me. I think the discussions are still around Can you give me a discount, can you give me a rebate, etc. We need to change that conversation. Secondly, advertisers do want transparency in the types of inventory they want to buy.
Having said that, the exchange still plays a vital role because is it possible to find inventory that’s viewable, verifiable and still adds value? It does. Which is why our position is that it’ll always be fluid. The percentage of what goes into a private marketplace vs. what goes into an exchange will be fluid. But the role of the private marketplace will become bigger and bigger. It won’t decline.
BRIAN LESSER: I agree with that. But if you have scale, and the only businesses I think will survive in programmatic will be scaled businesses, then you don’t necessarily need to go into the marketplace because you’ll become the market. Within WPP, that’s our ambition.
And I’d also say publishers will make more money doing this. Our publishers make 30% more on a CPM basis than if they were to use an exchange or open auction.
Stefan, you’re an operating agency guy by background, focused on technology. That seems to be a whole lot of unified negotiation.
STEPHAN BERINGER: I find it so strange how you argue, Brian. Because everything I’m getting from our clients is they’re running away from Xaxis because of this transparency issue. And what you’re saying now, instead of a new attempt to get going into that direction again, is you’re pulling out the inventory, etc., etc. Our clients are getting very, very clever and companies like Adobe have presentations where they say let’s bypass all the media agencies and we’ll pull through Rubicon directly and screw all these guys in the middle.
I find it very dangerous as a strategy to put a blanket over this business again, and may the better win. Ultimately, it’s about connecting brands and businesses with their audiences. That’s what we’re here for, okay? And you could say, “May the better win,” but you have to do it in a way where – I don’t want to be ethical, because this isn’t an ethical business, please tweet that – but it’s a bit weird hearing this directly coming from you as an agency that has been creating some turmoil in the market among clients by rejecting the model of the clients.
I would like to have a little discussion around that because from my point of view, it’s really annoying all the time because, seriously, we are operating exactly the opposite way as you know and we’ve inherited a model. But even if that wasn’t the case, I’d probably do the exact same thing because it’s a very honest and transparent way. Our clients know what we’re buying for them. They know our fee structures, etc., etc.
BRIAN LESSER: You disclosed the price you pay for media. Congratulations. What have you accomplished? What is the performance of the campaigns you’ve run?
STEPHAN BERINGER: One thing has nothing to do with the other.
BRIAN LESSER: Well, let me finish. Let me finish. That’s not what clients are buying. They’re not buying a business model. They’re buying performance. They want to engage their clients across all channels. Xaxis goes on a media plan with a price. If that price isn’t competitive we don’t get bought. If we don’t recover that, we don’t get bought.
STEPHAN BERINGER: Right. A couple of weeks ago, we saw Rocket Fuel and Mercedes-Benz in the press. Literally yesterday I get an email from one of our clients that went direct. And basically, they’re getting screwed because, let’s face it, it’s inventory that doesn’t exist. It’s worthless. Whatever. With us, you see what it is. We have quality standards and control the stuff…
BRIAN LESSER: And in many cases, it’s not performing.
STEPHAN BERINGER: What?
BRIAN LESSER: You invested zero in technology.
STEPHAN BERINGER: That is not true. We don’t invest $25 million.
BRIAN LESSER: The reason is that you have hung your hat on disclosure and transparency.
STEPHAN BERINGER: Yeah. And that’s a fair…
BRIAN LESSER: Maurice said he’ll never be a principal in a media transaction. We don’t agree with that. Xaxis is not an agency. We never have been an agency. We’re a platform and we feel the business model is justified by the performance. We do not get bought if we feel the prices are too high or if we do not perform.
We need this business model because we do need to invest in the technology and we need to create unique, engaging products that agencies can by. So don’t confuse Xaxis with an agency. It’s not. We’re part of WPP, which is also not an agency. We’re a platform to provide programmatic media products.
And just in your comments, you’ve thrown several things my way. It’s sort of the playbook of how Publicis argues against Xaxis, and it’s all very messy. It’s around transparency, disclosure and fraud. But at the end of the day, we perform. And that’s why we get bought.
And the business model is such that we need to make investments in data, in inventory, in technology. And we can’t do that if we are running it on a—
STEPHAN BERINGER: So, what—
BRIAN LESSER: The other thing I would say is: Be careful where you throw this mud. Because you may disclose your media prices. But do you disclose all the rebates you get from technology vendors?
STEPHAN BERINGER: Whoa, whoa… You’re an angel! You’re an angel, Brian!
BRIAN LESSER: Excuse me. My clients know exactly how we make money, and they literally sign a contract—
STEPHAN BERINGER: Oh, they do? Yeah, I’ve seen your contracts. I’ve seen your contracts.
BRIAN LESSER: “I know how Xaxis makes money, and I’m comfortable with that.”
STEPHAN BERINGER: I’ve seen your contracts. I’ve seen your strategy to come up with a new transparent model. So Brian, we know everything. Don’t worry, don’t worry.
BRIAN LESSER: It’s a very messy criticism. You need to tighten it up a bit. I know you’ve only been on the job for four months.
STEPHAN BERINGER: Brian, it’s fine, it’s fine. I probably said the wrong thing. Ultimately, Brian, I’m just reflecting on what I hear from clients and other stakeholders who’ve told me the Xaxis model has created some turmoil.
BRIAN LESSER: Absolutely not. It’s created turmoil because it’s not a traditional agency model.
STEPHAN BERINGER: You’ve not done anybody a favor because the clients have become very, very suspicious and they’re questioning many things.
BRIAN LESSER: Listen, listen: I’m not an agency trading desk.
STEPHAN BERINGER: No, you’re not.
BRIAN LESSER: I’m not an agency.
STEPHAN BERINGER: You’re a technology company.
BRIAN LESSER: Clients do have an issue with this. I won’t say that no client has an issue with this. I will say that, over time, they will understand that in order to provide value, you need to invest in the business. And it’s awfully difficult to invest the way we invest with a disclosed model. We are the most transparent company in the industry in that…
STEPHAN BERINGER: [Laughs]
BRIAN LESSER: …everybody knows how we make money. Everybody knows how we make money.
STEPHAN BERINGER: Yes. I agree with—
BRIAN LESSER: I’m not sure if people know how AOD makes money, if AOD makes money.
JAY SEARS: All right, all right. It’s all good. It’s all good. We’re all advancing advertising. Let’s move to one more topic, and we’ll move to word association. Arun, explain why Google is an ad holding company. Then we’ll survey the panel.
ARUN KUMAR: If you look at Google’s approach, in many respects, I feel jealous about the way they’ve approached it. Because as a sell-side player, they’ve done a lot of things we as a buy-side player we as an industry should have done. So if you look at the way they revolutionized the way search is bought, it was their technology. It’s basically the sell side that has driven it. If you look at the way they look at their products and the way they’ve structured it and make it [unintelligible] in terms of advertising, there are a lot of similarities there.
Now, obviously, they have a lot more technology and product than we do, which is fine. Which is a distinction. But the bigger they’ve gotten, they’ve started to mimic that behavior a lot more and there’s a lot to learn from them as well. They’ve revolutionized the buy side, whereas they are largely a sell-side player. Everything they do is to monetize advertising.
JAY SEARS: Let’s survey the panel. Yes or no, is Google an ad holding company?
JOSH JACOBS: I think they’re more complex than that. They have media platforms and skills products. We tend to buy more their skill-based products, and use them to build our own capabilities. To us, they are definitely a technology vendor, not a media vendor.
JAY SEARS: Stephan?
STEPHAN BERINGER: Agree, agree.
BRIAN LESSER: No, they’re not an ad holding company.