Ahead of ATS Paris next week, ExchangeWire looks at the development of programmatic in the French market. Here, Julien Gardes (pictured below), MD southern Europe & MENA, Rubicon Project, talks to ExchangeWire about the growth of programmatic over the past few years and airs his frustrations about the handling of header bidding in the French market.

ExchangeWire: Programmatic monetisation covers a very broad spectrum. What does it mean to you?

Julien Gardes: First and foremost, we need to acknowledge that the programmatic industry has made giant steps since its inception in 2009, and its launch in the French market in 2010, when Rubicon Project opened an office in Paris. Today, programmatic monetisation accounts for 40% of display budgets in France and is growing at a significant rate. While we still have some hurdles to overcome, and are still well behind our English neighbours, I expect to see staggering growth for many years to come – we have only seen the tip of the iceberg.

The market has successfully pivoted from a volume-oriented approach to a value-oriented one; and this shift can be attributed to the mass adoption of the programmatic protocol by all parties in the ecosystem and the rise of 'Deal ID'-powered trading. The fact that buyers and sellers have both moved in the same direction has resulted in sellers opening up new, high-quality inventory and buyers bringing high-value branding campaigns to the table.

For us, the next obvious steps are to address these two areas:

– Ensure all publishers understand the value proposition behind 'holistic yield' and the need for buyers to see all of their available impressions. In this sense, our recent launches of our Exchange API (xAPI) and Header Bidding solutions are great milestones towards this goal.

– Adopt a smarter, wider approach to our Automated Guaranteed business, which will drive the cross-media capabilities of automation to new territories like IPTV, Radio, DOOH, and also Print.

From the sell side, what challenges do publishers face in effective yield monetisation?

Well, this is a secret to no-one, but most publishers in France are facing the same challenges as their colleagues overseas. Obviously, the rise of ad-blocking technology is on top of everybody’s mind; and that threat needs to be addressed by industry-wide initiatives. However, we can always extract positives out of a perceived negative; and, as a company, we believe this new phenomena is an opportunity for publishers to embrace and to optimise their users’ experiences with advertising. And, on the other side, the targeting and customisation features that programmatic brings to buyers could be easily leveraged to work towards this goal.

Fraud and non-human traffic are also things that Rubicon Project pay a lot of attention to. Our wide scope of technologies block such issues and promote a clean and well-lit marketplace. I would say that our recent rankings on the AdQuality index, provided by Pixalate, show the great efforts we have produced so far.

Lastly, I am happy to see the incredible growth of programmatic mobile. Most news-oriented publishers now have a higher consumption of their content on mobile devices than on regular PCs; and, until now, the percentage of revenues coming from mobile devices were minimal. This has drastically changed over the last 6-12 months; and we are now seeing some similar CPMs on mobile as on desktop. Given the growth forecast of both mobile usage and mobile spend, we finally see a bright future for the mobile channel.

Full technology stacks have clear benefits for the publisher. However, some would argue that technology fragmentation promotes diversification – how can this be a benefit to publishers?

Well, I would firstly argue that full-stack platforms are still in their very early days; and we are somewhat missing a proven track record of concrete benefits, even though, theoretically, they seem to be logical. It’s obvious that ad tech needs consolidation – one look at the Lumascape is enough to know that it is extremely fragmented. My vision on this is that the ‘traditional’ and static waterfall system, which ad servers have been supporting since day one, is going through a tough time given the rise of programmatic and the dynamism it implies – however the ship has not sunk yet.

Ad servers are more and more becoming a commodity for the sellers; and there’s no surprise in their willingness to become full-stack platforms in order to help publishers make more money and stop being a simple source of costs. Rubicon Project’s philosophy is to work hand-in-hand with the ad servers in order to create a virtuous circle where sellers can benefit from the best available platforms, on which they have been used to working. This is the reason why we have developed our Exchange API (xAPI) product, which has been widely adopted by more than 40 ad servers around the world, and provide in a very robust and scalable way, this holistic yield promise; which is the number one reason behind the sellers’ need for a full-stack solution.

Header bidding is being touted as the cleanest way to maximise yield. Is it scalable; and does it provide all the answers to publisher yield challenges?

Yes. Header bidding is scalable and brings greater efficiency and results for publishers. However, without being too controversial, I am a bit annoyed by the way header bidding has been defined in the French market of late; where it is being touted as a way to have different SSPs compete for the same impression… This couldn’t be a bigger mistake!

Rubicon Project has been offering a header bidding solution for quite some time and has recently been announced as the first header-bidding solution to be declared as fully compliant with Google’s AMP initiative. However, when a publisher asks us to set it up, we always want to know the valid reasons, if any, behind this query, to make sure we’re aligned on what the publisher is actually trying to achieve. I believe header bidding is not the perfect solution for all sellers and needs to be considered as one possible solution amongst others.

We believe the main benefit of header bidding for a publisher is intelligent inventory allocation. Having all the information to optimally allocate an impression to the right buyer is integral to publishers wanting to maximise yield. There are further benefits in making available impressions to programmatic buyers that previously were unable to access a significant amount of volume sold direct.

However, there’s an urban legend saying that sellers need to run multiple SSPs in parallel to optimise their revenue. While this may result in quick revenue gains, we believe this is a very short-term vision that won’t provide consistent positive results in the long term. This approach adds operational complexity in managing separate platforms. It falsely inflates the supply of inventory into the market and increases listening costs for DSPs who will start to penalise publishers using this tactic.

Another important consideration that needs to be considered is budget depletion. Forcing buyers to pay higher CPMs does not directly translate into increased revenue. Advertisers do not have unlimited budgets, and by forcing them to pay more for your inventory, you may simply be depleting this budget faster. Advertisers may respond by deeming your inventory is too expensive and simply blacklisting your URLs from future campaigns.

We want our sellers to consider header bidding as a way to optimise their CPMs and yield by dynamically allocating their impressions between direct and indirect buyers, rather than piling indirect demand, which is limited by definition. I 100% agree that header bidding should be the star of the show at the beginning of 2016. But, as with anything, it needs to be understood and activated correctly in order to be fully beneficial for the publishers.

Gardes will be participating in a panel discussion on ‘Programmatic Monetisation – From the Complete Stack to Header Bidding’ at ATS Paris on 13 April. Find more information here.