There is a developing battle going on at the moment between the ad net and agency DSP for ad spend. Slowly but surely ad net margins are getting squeezed by the emerging DSP market. Agencies, who remember have the client direct relationship, are implementing fairly robust exchange strategies now. They are looking to put a lot more of their spend through these automated channels (either through 3rd party vendors or their own platforms). Who will suffer most from this shift of budget to the DSP? The ad network of course. Price and inventory transparency remain the key drivers. But transparency remains a tricky subject especially in the European DR market. A lot of ad nets are selling blind because premium publishers don’t want to them repping their non-premium inventory – for fear of sales channel conflict. The agencies are demanding more transparency and less of the black box sales patter.

This is a difficult place to be for the ad networks. The industry Cassandras who called the death of the ad network 12-24 months might not have got it quite right, but you feel there is a big shake-up about to come. There are a couple of reasons for this? The biggest reason, which have already touched on is the rise of the agency buying platform, with ad agencies taking a more pro-active role in buying and optimising media on behalf of the client instead of outsourcing the process. Another big factor for the shake-out is the rise of the so called data economy. For years, data was given away without a thought. Its economic value was never fully appreciated by publishers. The data leaks will soon be plugged up by revenue hungry publishers and ad nets will have to pony-up for its use. And questionable tactics like burying scripts in ad tags will never capture enough data.

Ad nets do have of course access to large volumes of actionable data, but thrity-day cookie decay and regualr cache clear-outs will ultimately force their hand. Publishers will be looking to be compensated for data use. I’m not saying ad nets are in danger of disappearing. But there are too many ad nets in the UK – and with automation becoming the trend there will eventually be no business need for 80 middle-men. The bigger technology-driven and well financed operations will survive and prosper. And they won’t be satisfied with being on a media plan. Some will, but others will be looking at this from a different perspective. I envisage a future where agencies and ad nets are going head-to-head for advertiser budget.

Imagine for a minute you are top-performing DR ad network with solid tech, data and expertise. Are you going to let a third-party or agency DSP suck all your business away and leave you as an ad-hoc outsourcing solution? Of course not. The ad nets will go straight to the brands and advertisers, and try to distinguish themselves from their agency competitiors. How will they do this? They could build out a DSP proposition similar to Lucid Media or licence tech to compete with agencies trading in the exchange marketplace. The data “blind spot” could be covered by establishing attractive rev share deals with publishers. Basically the uber ad net will need to play heavily on the tech and data angle. People in the industry say ad nets and DSPs can co-exist in one harmonious space. I’m saying it’s possible but there will be less ad nets in the market and their business model is going to look a lot different than today. Arbitraging inventory will no longer cut it especially if agencies continue to put more and more budget through ad trading platforms.

Unfortunatley there continues to be a real dearth of European-focused events covering the area of automated ad trading and media optimisation. With the intention of fostering informed debate on the evolution of the display market and indeed online advertising as a whole, ExchangeWire is today announcing details of the Ad Trading Summit 2010. It will be the first dedicated European event of its kind, and will look to attract decision makers form agencies, advertisers, ad traders, ad nets, ad exchanges, ad-tech vendors and publishers across Europe. The best minds in the industry will be brought together to discuss and debate the changing face of the European display advertising market, the increasing influence of automated trading platforms and the explosion of the data economy.

Confirmed speakers for the day include: Mike Nolet, AppNexus CTO and Cofounder; Curt Hecht, President, VivaKi Nerve Centre; Martin Kelly, Cofounder and Managing Partner, Infectious Media; and Konrad Feldman, Co-founder and CEO, Quantcast. Additional details on the speaker line-up and agenda will be announced in the coming weeks. ExchangeWire will also bring you the thoughts and opinions of the event’s speakers in the run up to the actual event on September 23. Tickets are now on sale – but please note that the numbers are limited. Further event information can be found on the Ad Trading Summit 2010 site.

Neumann points out that Invite will probably get access to Google’s treasure trove of search and GCN data. This will allow it to offer buy-side clients unparallel targeting capabilities. It will undoubtedly impede other players in the space, including XA.net – a DSP which Jerry has an investment in. It is no secret that Google was planning to build DSP and RTB capabilities into DFA – and it would be logical to assume they would have allowed DFA to leverage the same search and GCN data. After looking at the timescales of getting it to market, Google reasoned that it was cheaper to buy Invite at a snip for $70 million dollars. The price does severely mess-up the valuations of other DSPs, eyeing up a possible bonanza exit. I’d say some VCs weren’t too impressed – but I digress.

Jerry is right to point out that Google is now ridiculously conflicted. It’s an ad network, an exchange and now a DSP. He points out that having insight into data and pricing from other inventory sources would give it a real competitive advantage over its rivals:

Because when Invite is integrated into Google, it seems reasonable to assume that Google will:

1. Start cherry-picking the other exchanges’ best publishers; and
2. Start front-running the other exchanges, keeping the demand for themselves****.

By giving Invite access to their marketplaces, Microsoft, Yahoo! and AOL give Google access to data about position and price of every ad that runs through them. They would be giving Google the very data it needs to outcompete them. If the other exchanges allow this, they won’t for long. Because if they do, they won’t be in business for long.

**** It’s widely rumored in the industry that Google has a double standard in exchange pricing between people buying through Google’s user interfaces and people buying through the exchange API. If Invite is an insider, it shouldn’t surprise anyone if they get preferred access.

He has a point here. But it won’t be so easy for Google to dominate completely – especially when a lot of publishers on the sell-side are hesitant to throw their lot in with Google. Rubicon, Improve Digital, Admeld, Adjug, Adscale and the Orange Ad Market are becoming the preferred automated channels for a majority of European publishers. It will be difficult for Google to “front-run” other exchanges when its own exchange only has access to millions of Mickey Mouse Adsense sites with crap content. Audience matters – but so does context. Expect publishers to be more self-aware of this. And there is no doubt in my mind the good people at Rubicon, Admeld and Improve will be pushing hard on the agnostic angle. You’d expect, if you were a betting man, Google to make it very difficult to integrate DFP with these platforms. But these guys are good at innovation – and they’ve got the resource, talent and cash to meet the challenge.

Giving too much power away to Google is not in the interests of any publisher. There needs to be competition in the space to improve pricing and innovation. Is it really healthy for a single company to have a stranglehold over a publisher’s entire income? I would call that commercial suicide. Monopolies are a bad thing. But I will say one thing for Google: it keeps everyone on their collective toes, making companies innovate and continuously push the envelope.

The Invite acqusiiton could leave some on the buy-side in real trouble – especially the big agencies. Disintermediation is a real possibility here. If I was a Google strategist, I’d be looking beyond the “unnecessary buffer” of the agencies and going straight to the top European brands. Google’s got the tech and the data – and could probably deliver better results. What to do? Make yourself relevant before it is too late. Innovate. Invest in technology. Are we getting carried away? In fairness, agencies are doing all these things already. We are still at the start of this automation process in display. And as one sage industry stalwart told me yesterday, the industry needs to stop whining about Google and get on with developing better solutions and tech for its clients. It was a good post by Jerry, but now that we know the big G’s strategy let’s see if the rest of the industry can outmanoeuvre them.

Let’s suppose this deal does happen, how will the rest of the industry react? ExchangeWire shall now make some ridiculous (but fun!) predictions about possible repercussions:

- Microsoft will buy Appnexus: with its own exchange platform in the works, Microsoft will likely buy in DSP capabilities to counter Google’s move
- One of the big holding companies will buy MediaMath, X+1, DataXu, Triggit or Turn: the holding agencies will feel very uncomfortable with Google being in the DSP space, and will want their own platform.
- A possible surprise buyer will pick up a DSP: we could see Adobe, Ebay or – wait for it – Facebook buying a DSP. Think of all the data Facebook could leverage in a media buy. Its targeting could only be matched by Google.

You’ve recently went through a rebranding and a repositioning of the Infectious offering. Can you explain the Infectious Media proposition in more detail?

MK: Yes, it’s simple, we make display advertising work for our clients. Clearly there’s a lot more to our business in terms of how we do that but that is our proposition and how we sell our services. We operate exclusively in the ad exchange space and offer these services to both advertisers direct and to agencies. We’ve purpose built both a team and trading platform, Impression Desk, to service this opportunity in the UK and Europe.

The rebrand and new website was about reflecting this focus.

How in your view has the exchange market evolved in the UK since the launch of Infectious Media?

MK: There wasn’t really one when we started so it’s come a long way! Right Media was the only place we could trade and was in itself the preserve of a few very smart networks, it’s incredible really that a whole new ecosystem has developed in the space of two year, the pace of change is staggering. The component areas of the value chain seem to be settling down a little now and both publishers and buyers are starting to do deals with and build infrastructure around DSP’s and SSP’s in order to service the opportunity. It’s not yet clear how the traditional intermediary, ad networks, will respond to these changes.

Do you think there is still not enough quality inventory available through automated channels?

MK: I think there is quality inventory available at scale but an issue that buyers will find in this environment is that it’s hard to find and the ways of doing this differ by platform. All supply sources take a different approach to defining quality as do all the DSP’s so it’s a minefield with no standardisation. We have chosen to tackle this issue head on and have invested a lot of time and effort in to devising our own manual vetting procedure for inventory sources. We are about to introduce our own transparent classification system for inventory that we vet ourselves so watch this space.

Has the arrival of the SSPs, the DoubleClick Ad Exchange and now the Orange Ad market improved the buying opportunities offered through automated channels?

MK: There is momentum growing on the supply side and these intermediaries are driving things forward. Ultimately this makes sense for everyone involved, with lower transaction costs associated to this channel and CPM’s increasing substantially in the last 6 months alone. Part of the education process for publishers is that the CPM’s that they receive for their inventory are higher than those they receive from a network as the buyer has much more opportunity to recognise value in the platform environment than they do in a bulk network buy. We, on average, currently pay double the average CPM in the Doubleclick Exchange quite simply because we can see the value that our activity is driving for the advertiser and bid accordingly

Do you think this volume on the sell-side will attract a lot more display spend into the exchange eco-system?

MK: I don’t think anyone on the demand side is under any illusion that this is the way that the market is going and needs persuading any more. The barrier is more around attaining the skills and technology to operate effectively and the speed with which this will happen.

Data plays a huge role in the automated buy. Do you think the absence of UK-specific data sets from the data exchanges and other platforms is hindering the development of the exchange eco-system?

MK: The data market has been slow to develop in the UK but there are now some companies entering the UK such as Quantcast and we welcome that development. I would also say that from an Infectious Media perspective this has meant we have had to focus our product development around data analysis and decisioning in the absence of ‘off the shelf’ target audiences.

We speak to many of the top UK publishers and they are keen to find ways to monetise their data so it will just be a matter of time before this market develops further and there is room for some big winners in the UK and European data market.

Is Infectious currently buying inventory through RTB? Would you say the buying process is different than a typical auction buy on the Rightmedia and Adx exchanges?

MK: Yes we are currently buying via RTB in the UK and across Europe. The buying process is no different in many respects, this is just the buying of display media after all. The difference is the amount and richness of data points that can be passed through to the buyers on each impression but I don’t yet think this is being fully exploited by the supply side especially. The more information a publisher passes about an impression, the more of a chance that we will see value in that impression and bid higher accordingly. What has been a real game changer for Infectious is impression level reporting which is a huge volume of data that we take in to our data warehouse to analyse but has meant we can scrutinise performance to a new level of granularity and optimise accordingly.

Infectious is an exchange trading specialist – far removed from a traditional media buying agency. What kind of skills does Infectious, which puts technology and data at the heart of its business, look for in new hires?

MK: There is some cross over with media buying agencies in what we do but we’re a very different type of company. Exchange trading is our only service and we offer this to both advertisers and agencies and because of this we have built a very specialist team and infrastructure. Our last three hires have been a statistician, a business information systems architect and an agency account manager. So some cross over with an agency in terms of personnel but in the main something completely different. Coming from large media agency backgrounds it’s been a breath of fresh air for Andy (co-founder) and myself to be able to build the team from scratch with new skills and a company structure that is fit for purpose. To date we have not employed anyone from a pure media buying background.

Now that the DSPs have finally arrived and RTB inventory has come onto the market, how do you see the space developing in the coming months?

MK: The pace of change is rapid at the moment and will not be slowing down any time soon. The coming months will see the completion of the infrastructure for the UK platform trading revolution, new UK data centres opening, trading desk offerings appearing, SSP’s hitting scale and we hope a more open debate around data and privacy that needs to happen if we are to successfully self regulate as an industry.

Oh, and loads more hype, as ever.

There’s much confusion about what Vivaki does? Is it buying platform? Is it a crack exchange trading unit? Can you explain the Vivaki proposition in more detail?

MB: Vivaki is the strategic entity created by Publicis Groupe to leverage the combined scale of its media and digital operations, which represent nearly $60 billion dollars in global ad spend and influence. VivaKi aggregates the marketplace influence of five autonomous brands, including: two global media agencies, Starcom MediaVest Group and ZenithOptimedia; two leading digital marketing agencies, Digitas and Razorfish; and a premiere futures practice, Denuo.

On behalf of its agency brands and their clients, VivaKi faces the market to help identify and build technology, message distribution, audience aggregation and content solutions for the future. VivaKi also includes a “Talent & Transformation Practice”, which leverages the scale of the VivaKi brands to develop and deliver tools and approaches designed to attract, develop, train, motivate and reward the world’s best people.

Sitting at the core of VivaKi is the VivaKi Nerve Center, which serves as a think tank, R&D centre and testing ground to activate new pathways for clients to connect with consumers in an increasingly digital world.

The key objective of the VivaKi Nerve Center is to help deliver better solutions for our clients as the marketing landscape continues to evolve and accelerate at a fast pace, collaboration within the VivaKi family, and across the Groupe, is essential.

To succeed in our mission, the Nerve Center will focus on some key areas to empower our VivaKi agency teams and clients:

Global Platforms & Products: Developing global platforms and proprietary products that help our agencies differentiate and compete in the marketplace. Products will be supported by an advanced underlying technology and data infrastructure that delivers speed and scale.
Industry-Leading Partnerships: Creating strategic global partnerships that provide tangible value for our clients and partners, while differentiating against the competition.

Innovation & Thought Leadership: Investing in innovation and next generation emerging opportunities, like The Pool, which will validate our leadership position in the marketplace.

Our ad exchange solution is called Audience on Demand and is therefore a key strand in the global platforms and products category above and indeed innovation. It’s one of the most exciting areas to touch all agency groups in recent years and needs to have a defined and aggressive focus put upon it. Vivaki Nerve Center has worked very collaboratively with the brands in delivering the Audience on Demand platform to their clients. We are live with Audience on Demand and really excited by the performance of the solution.

Can you elaborate a little more on your role within Vivaki?

MB: My role in is Managing Director of The Vivaki Nerve Center in the EMEA. I report into the Global President of the VNC, Curt Hecht. The VNC has made significant progress in the US and my role is to work closely with the brand management and digital teams to establish how the VNC can help them in delivering the future-facing digital solutions that our clients are asking for everyday. Ad exchange trading through Audience on Demand is a significant area of work for me.

What’s your perspective on automated trading and audience buying through exchanges and other demand sources?

MB: I have been blown away by it. I may be biased and perhaps my background lends itself to making this exciting to me but when you see the potential of automated buying you can’t help but be impressed. It’s worth saying that automated buying is a little misleading. It requires clever optimisation strategies and insights that the agencies need to lead through talented people. I would not want people to think that you a press a button and it’s all done. Anyone who thought search bidding would be automated would testify that is not the case – it is search bidding times ten so definitely not just automated.

The trading platform allows you to target exactly the individuals you want at the price you want. You are buying one impression at a time which makes a CPM approach look outdated although it is not the death of the CPM buy just yet, not least because media auditors would not know what to judge us on! I believe it will ask questions of every agency trading model to some extent or another. It will also challenge auditors to stop judging agencies on an arbitrary discount off a pool metrics, and force everyone to consider more performance related contracts. I think for now it lends itself more easily towards the performance models but down the line I can see far more being traded through this method.

Do you think that large European holding companies like Publicis are now seeing ad exchanges as an efficient channel to buy ad inventory?

MB: I think the large network groups get a hard time for not changing enough and being slow to react. In some ways that may be true but agencies today are very different to those of 15 years ago. They have completely transformed: agencies realise change is inherent in what they must deliver year in year out.

Ad exchanges are just another media / trading / targeting opportunity that have come along, and agencies will embrace it and make the most out of it on behalf of their clients. My experience so far is that all the groups see the benefits of it but that will vary by group as some are more advanced than others. You will see who believes in it the most by how quickly they grow their ad exchange spend because once you start to see the results, clients and agencies alike will want to move their budgets into new the model.

Do you think that trading on ad exchanges makes it easier to leverage agency and client data to deliver better campaign performance?

MB: Trading on ad exchanges will allow data to become more important but actually it’s not the exchanges where the benefit lies but with the use of DSPs like Audience on Demand. It is this technology that allows us to best use data to enhance the performance of campaigns and target only those users that are most likely to deliver a beneficial response for our clients.

The combination of our clients being able to retarget their visitors but on a much larger scale with the introduction of third party data means we can turbo-charge our schedules to deliver at the right cost and at the right level of volume. Those third party vendors need to move quickly over here. We already have demand and they are a little slow to get going. I was pleased to see Phil moving from Yahoo to Quantcast, perhaps a sign of things to come.

Does Publicis have an exchange strategy for Europe, and if so will this be headed up by Vivaki? Are there plans to devote more resource to developing this area of the business?

MB: This is not a UK or US only market place. It will become important across all major markets so of course we will grow our business in those countries. Many of our major European markets are already testing different models and gaining from the insights. Vivaki Nerve Center will take the route that drives the consistency and ability to learn as a group and not at a country/agency level. We are in the very formative stages of this area so it’s important we all learn from each other.

Resources will evolve over time. Some people will re-skill into this area, some will be recruited. But we have time yet to get into that. Rest assured though that the number of people working in this area will grow substantially!

What do you think are the key difficulties in moving an agency toward automated media buying? Is it the lack of technology and data skills that exist within the agencies? Or is it a lack of technology?

MB: It’s not a technology issue. We have the technology and it works. I am sure all of our competitors have their technology too. Some will work better than others perhaps, but generally I don’t see that as an issue. Technology should not be the differentiator for agencies, it’s the people behind it and what it delivers that counts. Clients do not want pitches where we all get our technology out and wave it about; they want to see insights and results.

In Vivaki there are pools of people who understand this new area and those that know less about it but is that not always the way? Over time we will train people and recruit people so that we have the right level of understanding and evangelism in the business. Look how agencies changed around search. We had the same discussions back then and we now have these amazing skill sets around search in the agencies, so I don’t really see too many issues. If you think the opportunity is a good one, you can make things happen.

Do you think the arrival of DSPs into the European market will help agencies bridge this technology and skills gap?

MB: DSPs will allow agencies to build campaigns across multiple ad exchanges, create data pools, and control frequency etc across the whole playing field rather than at a site or network level. They will also provide us with the largest search area when we are trying to find the elusive consumers who have visited our client sites previously. DSPs are enablers so of course it’s a great innovation in the marketplace globally.

I would say that I believe a true DSP is one that’s only interest is in providing technology to do all the above. It should not to try to resell inventory or have morphed from an ad network. There are many blaggers out there and it’s important that people choose carefully in who they work with as you may discover that the systems they provide are not as future-facing as you thought. The market place is very grey around the edges!

What’s your view on real-time bidding? Is it a game changer for the display market? Or are there still fundamental problems that need to be worked through (such as the computational costs) before we see the benefits?

MB: It works. Our campaigns are delivering great results on the RTB strands of the campaigns. I think RTB will be affected by many different elements not least volume of competition, which will only increase. But this is where the clever use of data helps you in RTB: only you know what is deemed a valuable cookie. Hence you will be bidding on it, not the rest of the world. This is different to search where everyone knows that if you bid on home insurance you will sell insurance. This is a huge benefit for ad trading – the agency knows who is valuable not the publisher telling us what is valuable.

Overall I believe that RTB will be a game changer. Suddenly impressions are valuable again in the volume game. Interestingly though, they are valuable from a data perspective and not so much from a context /channel perspective. The rules have changed. What’s premium now?

How do you see the European exchange space developing over the next twelve months?

MB: I think you will see many of the major players in the US getting people on the ground here to push into big European markets. The ecosystem is developing rapidly in Europe, and it will not be any different than the US. There are already companies up and running in Europe, providing ad serving and other services, and they will try to steal a march in these markets. The DSP pure players will soon be driving a more objective approach across Europe. I also think we will see the likes of Google really ramping up in the markets here, which in itself will drive liquidity.

Can you give an overview of the size of the French exchange marketplace? And the role Improve Digital plays in the market?

MJ: At this moment there is no ad exchange or demand platform in France with significant volumes. Some global players have not yet started in France or if they did, are at a beginning stage. There have been some announcements by the likes of the Doubleclick Adx and Weborama – which declared, during an IAB conference in January, its intention to become an ad exchange – but I have not seen any real traction in the market.

In France, Improve Digital gives publishers the ability to optimize revenues and eCPM’s from ad networks and exchanges – while protecting their premium business and ad quality policy. We have 2 billion monthly impressions in France and believe that this is a significant sign that publishers are opting for an agnostic publisher focussed yield optimiser, such as Improve Digital, instead of working directly with an exchange.

How evolved is the exchange eco-system in France?

MJ: There is more and more demand from publishers for our platform and this trend shows that publishers are ready to work with multiple partners in the form of ad networks. They see our technology as a way to prepare for the future, where publishers monetise their ad space both directly through various sales partners, such as ad networks and Demand Side Platforms.

Existing ad networks still have significant growth opportunity, mainly because of AOL’s decision to close the French office, but there are also new players entering the eco-system, particularly the DSPs. I estimate that the French market is perhaps 18 months behind the UK, but some media agencies are developing their own DSPs to buy automatically unsold impressions from publishers. In France Matiro, founded by two formers top managers of Havas Media (Yann Leroux and Erwan Lepage), are offering advertisers the opportunity to buy unsold impressions directly from their platform. A new question will arise: how will the ad networks evolve in to meet the challenges?

Are publishers now seeing the benefit of using platforms like Improve Digital?

MJ: From December 2009, we began trials with some key publishers like Le Monde, Allocine. Nouvel Observateur. At the moment in France, we work with more than 15 publishers that represent over 2 billion French impressions and we have never lost one customer. Actually we are connecting new French publishers every week. From a publisher’s perspective, the first advantage of our system is increased revenues and eCPMs. From an operational perspective we simplify working with multiple ad networks and reduce operational issues. On a strategic level, our product helps publishers gain control over revenues, ad quality and cannibalisation of premium sales.

What are the significant differences of the French display market compared to the likes of Germany and the UK?

MJ: I don’t know the other markets in-side-out, but from what I heard, the German market is very much dominated by sales houses. In France this is not so much the case and many publishers have their own sales teams. Just like in other markets, the standard IAB ad formats are becoming more of a commodity, with pricing models lowering and moving more towards performance based pricing. Standardization kills the prices, but not necessarily the margins. With our platform, the costs of trading ad space is significantly lower, realising higher margins and making the online media in general more competitive to television, radio and print.

On the other hand this is the exact reason why more and more big publishers develop special ad format or disposals, brand content areas, etc. “Opération Spéciales” or in English, “Special Products’ are the latest trend for the ad development strategy of publishers.

As a “first mover” in the French market, how would Improve Digital sell the benefits using its automated channel to monetise non-premium inventory to publishers?

MJ: When we entered the French market, ad network optimisation was a new phenomenon. Most publishers had never heard of. But if a technology is proven to work, the word spreads quickly and today we have 15 of the top 20 Comscore French publishers as clients. Using the platform is a way for publishers to prepare for the future, as I described before. Media agencies, advertisers will try to find some “Google Adwords” like business model for display. This new attitude is motivated by the “infinite inventory” that characterizes the display market at the moment. We started as a true ad network optimiser, but we are evolving towards a sell-side platform that enables publishers to optimise yield and revenues from all of their ad space, not just unsold.

Are you making any French inventory available through RTB? If not, when are you likely to allow real-time bidding on the platform?

MJ: If the publisher wants to connect to RTB, it’s a matter of opting in. Right now, we offer French inventory through RTB and have multiple global RTB partners live. Technically we are ready to connect French ad networks. As I mentioned previously, the exchange eco system is not as developed in France, and some French ad networks are technically not ready to buy through RTB.

Is the buy-side and sell-side ready for real-time bidding in France?

MJ: I hope that it will be the case with the arrival of new exchange trading specialists like Matiro and others players from the media agencies world. Improve Digital can be considered the first and the largest sell-side platform on the French market.

How do you see the display market changing over the coming months?


MJ: Like a lot of people in the French display market, I think that we are seeing the end of CPM as the key model on display. The IAB ad format will be bought on a performance based model. The fee business model will manage the premium offer. Publishers will have to find some technological solutions to adapt their performance offering for the demand side, advertisers and media agencies.

Will we see France following a similar path to the UK?

MJ: As a lot of international advertisers manage more and more of their marketing budget on a European level, I expect the whole European market to follow a similar path with most ad networks becoming DSPs and most publishers organising themselves through sell-side platforms.

Are we likely to see significant volumes through exchanges and SSPs over the coming months?
With our monthly 2 billion of French impressions, we are already seeing a significant volume, but this is just the beginning. First, because managing unsold impressions has become more and more complex on an operational level; and secondly because of the search to find new monetization opportunity. For this first year in France, Improve Digital has the objective to double every quarter the number of impressions it optimizes.

» Interesting chart published by BusinessInsider yesterday. The chart shows the sites serving the most display ads to US users. It is for the US market only, but demonstrates how much ad inventory Facebook has. But let’s be honest about it, it is pretty poor quality inventory. The CPMs ar ridiculously low for a reason. Serving up branding against user generated content is not going to going to add value. From a DR perspective, I hear Facebook is perfomring very well. It’s targeting is the envy of most platforms. I don’t think display is the future for Facebook though. It’s better off looking at the data angle: selling data to ad agencies, advertisers and anyone else on the buy side could be a very lucrative revenue stream for Facebook. But it’s likely the EU will stop that in its tracks, judging by the noise coming out of Brussels today. [Business Insider]

Can you give an overview of the service Yieldivision offers advertisers and agencies?

MS: We are a new breed of agency with a technology which allows us to buy inventory across multiple sources (exchanges, adnetworks, direct pubs and ssp’s). Our technology enables us to base our media buying strategy on enriching the impression with data. This data is provided by our direct relations with data partners – direct and third party. We see technology as differentiating between the current market for traditional media agencies and ad networks. We are a media agency but our technology resembles that of a DSP. Our clients are Direct Advertisers in various industries like Travel, Finance, Telco. Also we are a strategic supplier for media consultants and creative agencies.

Our goal is to always work against performance metrics we set with our advertisers. These metrics can be a CPA goal, clicks or unique users within an audience for branding. The main focus in the end is to use our unique media strategy to reach the ROI goals of our client’s total budget.

By combining existing technology with own developed targeting, bidding and optimization tools we provide a unique product for the European market.

What attracted Yieldivision to the ad exchange model?

MS: The main reasons for operating on the ad exchanges are flexibility, inventory and pricing. It’s one of the many ways how we service our clients.

What markets is Yieldivision currently operating in?

MS: We are currently strongest in the Dutch market. And we are looking to push strongly into Germany and Sweden. After entering these countries other markets will follow – possibly France, Spain and the UK.

Are you seeing display advertising moving through exchange models in these markets?

MS: I can definitely say that we see a lot of activity in exchanges in the UK and the Netherlands. UK is a bridge from the US into Europe. In Germany there’s more activity than in the Netherlands. Other countries like Spain, France and Italy are more traditional and will take time to move to an exchange model.

What exchange or automated platforms are you currently buying from? Are you still buying directly from ad networks?

MS: Currently YD is buying from the following exchanges: RMX, Google AdX, OpenX marketplace. We also buy inventory across a number of supply side platforms, including Admeld, Improve Digital and Rubicon. YD uses all kinds of sources for inventory. We strongly believe that traffic is a commodity. That’s why we buy from exchanges, networks and partners/publishers directly.

Is Yieldivision currently using RTB to trade ad inventory on the ad exchanges? And are you using any RTB technology to buy inventory to access this inventory?

MS: We have just started testing with RTB. The available quality inventory using RTB comes mainly from the sell side platforms. Not sure if we will see an increase in ROI. The big gain for us is operationally. No we are currently not using any DSP to buy ad inventory. We are in the advanced stage of building our own technology to buy across different inventory services.

Is the present move towards automated trading an opportunity for exchange trading specialists, like Yieldivision, to grab market share from bigger, less agile media buying agencies?

MS: Yes indeed. We see a lot of client movement in the current market. The YD client base is rapidly increasing, as well as the enormous growth of revenue streams. We see huge growth in the coming period as the demand for ROI driven marketing increases.

Late last year I did a piece for ExchangeWire about some of the exciting changes happening in display advertising. Given the number of blog posts and op-ed pieces that have since been published on the subject, I am sure everyone is now tired of hearing the “exciting times ahead” mantra. Let’s take a reality check and see what’s actually happening NOW in the space:

- DSPs have penetrated our borders from the US and beyond
- Spend is slowly being moved onto these platforms
- UK publishers are slowly making this inventory available in one way or another onto these platforms (SSPs, exchanges, etc)
- Agencies, advertisers are buying 3rd party (albeit in small amounts) and integrating into these platforms
- Even the NMA has covered DSPs!

So everything is good. The buzz and excitement around this ‘major shift’ is finally being realised and the ad world is changing because of exchanges, right? Well not quite.

We still have major hurdles to overcome and need to appreciate the fact that it’s not all about the exchanges. In my opinion, the UK market will find it easier to relate to the points I am going to make because we tend to be a more cynical and sceptical bunch!

Although it is all well and good being able to value impression and bid in real-time based on their predicted value, it still appears to be quite a siloed practice. Attribution and ad verification need to be considered in real-time if we are to really provide value and make the most of RTB.

Attribution:

- Display by no means acts in isolation. In the UK market, many marketers are still sceptical about the value of view based conversions (and regarding the practices of some ad nets, with good reason).
- For me, there is a huge opportunity for advertisers to be bidding in real-time on the weighted attribution models being built around their display marketing. How valuable would it be to be able to pull in your Tagman attribution data and bid for display inventory based on the value it REALLY drives to a client’s business and not the value according to an independent buying platform?

Ad Verification:

- As mentioned above, view based conversions are subject to a lot of scepticism in the UK. Ad verification tools, like Adxpose, are really useful in determining what value impression based conversions really have. Were they in view? Were they interacted with? What is the engagement rate like?
- To be able to bid in real time, having this data available at the time of bidding ensures that biddable inventory that really has value according to a client (besides just a low view based CPA).

What the above hopefully illustrates is that it’s not enough to be able to bid on low quality publishers in real-time and drive low view based CPAs. Where is the value in that? Would an ad network, partnering some ‘top 100 comscore sites’, be a more suitable buy? Yes, the CPA would be higher (mainly view based) but if the publishers are of a higher quality, is it worth paying a higher CPA?

Exchanges are just one piece of the jigsaw. RTB is exciting. What we need now though is to move it on to the next stage. It needs to become more integrated into the wider marketing mix where advertisers can really appreciate the value of what we are trying to achieve within this space.

* Note – I reference view based conversions a lot in this piece. Whilst I appreciate some advertisers have small view based windows, there is no denying that view based conversions play a huge part in display ad effectiveness.