19 January 2012 in ExchangeWire EMEA 1 Comment

The PostView: Why ‘Perfect’ Efficiency Will Not Be Good For All

The PostView is a new coulmn written by senior execs working in the European online advertising industry.

Like it or not we live in world where buying display advertising is still inherently inefficient. The convergent trend is for increased efficiency through ‘owning’ or building the entire stack – the likes of Google, Microsoft and Adobe appear to be accelerating their efforts in this space. However, when you break down what the end result could look like it actually might not be an entirely good thing for the performance industry, or at least some of the key industry players, to work in a world where there is 100% efficiency.

Currently if you were to look at most (and I’m talking 90%+) online DR plans they are at best inefficient. The move towards centralised and specialised buying and bidding is a good thing but the reality is that most Media Planners/Buyers still don’t understand the basics – consolidated retargeting, consistent rules of engagement, 100% transparency etc. The large Trading Desks and specialist independent Agencies will continue to address this but they will also rely on technology to help prove their case. The cold hard fact is that most of us who work in display have no real idea what’s working, what isn’t working and, more importantly, why… it’s the algorithm!

Take a step back and think: does it make sense for someone to claim credit for an action (eg sale) based on an impression that was shown to a user 20 days ago? Does it make sense to bombard pre-qualified buyers through RET when it’s likely they have already decided whether to purchase or not? Is it right that Agencies give their clients data away without asking how and why that data is to be used (And ‘they told us that they need site data so their system can better optimise’ is not a valid response)? My belief is that only once there is true efficiency in the system – ie when it’s possible to plan, buy and learn across multiple devices in real-time based on a range of metrics (engagement?) – will these plans start to represent true ‘value’ for money. But that might not be a good thing for some.

Inefficiency in the way we buy and trade online media has led to a bloated industry (I read somewhere that a leading DSP was taking 25% margin but quoting a sub 10% fee!) – an industry where the publisher sees a fraction of what the buyer is prepared to pay for the right impression. When the uber platforms begin to evolve and it’s possible to optimise towards engagement across multiple devices, with a single, consolidated view of the consumer it becomes clear where the inefficiencies lie.

My hypothesis is that DR will be shown to have some effect but nothing like the amount of influence currently discussed to justify the large budgets flowing through the channel. Subsequently it will need to work harder – and smarter – to prove it’s worth. As it becomes more difficult to justify massive amounts of DR spend Agencies and Networks will be forced to work in a more efficient, collaborative way to drive optimal results. The days of 50%+ Network margins will soon be over and Agencies will be increasingly held to account on where the spend is actually going.

As an industry we should be investing time and resource TODAY in educating our clients and our staff to truly understand where there is wastage and what success looks like. The likes of Google, MSFT, Yahoo! and Adobe should be facilitating this education process because the DR gravy train (the bedrock of a lot of their revenue!) will be running on rockier ground within the next 5 years. If you don’t know where your ads are running, if you are remarketing through multiple platforms and if you’re making huge margins on simple buys enjoy it while it lasts. Realise that technology driven efficiency is coming. It might not be tomorrow but get with the program today – or risk becoming insignificant in the future.



  • Media Warrior

    Good post. I think we have created an automated, margin free (for trading desks) way of gaming the system the VERY best we are able to. The networks are struggling, because for too long they rested on their laurels and didn’t reinvest the 80% they were making back into the technology or ecosystem. It serves them right that they are struggling now. 

    However, it struck me at a leading DSPs drinks, everyone sat around saying ‘the performance through this DSP is x times better than this DSP’, what can that be translated as is ‘this DSP is better at gaming the system as anyone else’. We need more investment into the measurement space (goodbye reports which can only measure 30 exposures to conversion) and badly. If we can measure the value of an impression better (even algorithms are flawed, they are optimising towards a backward metric!) which comes from the investment in measure, this will bring more investment into the supply space.The supply side will then take away ALL them ads at the bottom of your page and start giving better ad slots and formats. Instead of 500+ ads a month that a user sees online that becomes ‘wallpaper’, they start to see 200 ads and become more enthused by the formats. Metrics such as ‘awareness’, ‘recall’, #favourability’ etc. will come more prolific – even in the DR space (DR does not mean SQUASH PUBS!) AND more investment into this, will then bring the big brands into the space.Perhaps we have done it the wrong way around, but lets sort out the problems with measurement to help the supply side!