On Thursday I attended the Admonster’s AdOps event in London. It was set in the London Film Museum which I didn’t even know existed, so a wonderful setting. I should add that the drink reception was in the Harry Potter room, but let’s start at the beginning.
Maybe a note that this opinion of the event and my understanding and interpretation are my own views. Not that of any associated company! And, please note that I tried to be as objective as possible. Any comments are welcome of course 🙂
Donald Hamilton kicked off the event with a nice introduction on data. Data seems to be the buzzword in the industry at the moment, and everyone is talking about it. mexad‘s CTO Guido Pfister was speaking about it also. Guido in his talk said that RTB (real time bidding) can be very effective even without data, arguing against the myth that RTB is ONLY data driven. Of course data goes very well with RTB.
Anyway, data seems to be only valuable if it is qualified, and structured. Donald clearly pointed out that the successful use of any data, online or offline, depends on analysing it properly and then use it in the right context. A theme throughout the day also emerged: publishers should look into how they collect and use their data, or whether data theft is being done by having advertiser or agency tags on their site.
There was another theme ongoing: TESTING. Yes. Having data available and adding bits of information from your online conversion funnel, then adding behavioural data (almost sounds like a recipe), you still need to test it. Go out there and see how the data works and whether it is worthwhile. Does it increase your ROI, your conversions?
And Guido from mexad was speaking about RTB technologies and bidding strategies. The first impression is still the most important, but look at the conversion funnel, the attribution model, your bidding strategy and do one thing: test was works best! Hence mexad works platform agnostic, working with all technologies to find the best possible way of getting performance from any RTB/buying platform. And, they have been testing a lot!
Why is testing so important? I believe that if you work in an industry that is very young, and then in a part of it that only exists for a few years…then there are no right or wrong ways only. There is still a lot of learning to be done, and testing is the only way to find out what delivers the best results.
Google was on stage, telling us that 55% of their inventory in the UK is being sold in real time. I wonder if it is almost 100% next year, or when we don’t speak about RTB any more as that will be the only currency. As this is the way the industry is going.
Then you had Jakob Nielsen speaking from Xaxis, the new GroupM umbrella for audience buying, DSP technology, data integration etc. Looks like a clever agency solution – let’s see how much of it is actually going to be utilised and how other agency groups react to the announcement of its formation earlier this week. As Jakob said, it is not “hokus pokus” and also will be offered to other agencies, being transparent in what they do. My question would be whether any other agency group buys into a WPP product? I just cannot see any Publicis or Aegis agency using Xasis, or a smaller agency for that matter. Maybe WPP agencies outside GroupM though.
In the IAB session about the cookie directive I was hoping for more clarification on the EU cookie directive. But the summary is that the industry should get in gear to find a self directive solution rather than getting someone who doesn’t know about the industry forcing us to adopt something that will kill targeting via cookies altogether. The ABC picked up on that at a later stage. One solution is an icon being placed on the creative to inform consumers about data collection and opt out. I guess it needs to be tested 🙂 In all seriousness, there is no “one size fits all” solution, and we need to look at a pan European solution on this one.
Yahoo!’s presentation explained some look-alike modelling and the use of offline data, e.g. Nectar card users. I think I heard that case study before, e.g. there was a trial in the US with the use of offline data by tracking conversions via a store card. However, not sure when we will be properly able to integrate offline and online data…
I also went to the AppNexus session about the 3 evolutionary stages of yield management, e.g. the evolution from tags to rules to data….I think that the explanation about floor prices and jelly wasn’t at the best time of the day, as I was getting a bit tired. However, there was a great discussion around the “manipulation of the publisher market and whether or not we are creating an open RTB environment like the finance sector does”? I wasn’t quite sure if I followed the discussion correctly but isn’t that what we are doing? And, speaking to “futurists” we should wonder if every ad server turns into a DSP or SSP in 2 years time, making local publishers available and get the best yield for them…..oh yes, we had some great discussions 🙂
The final session was moderated by Peter Kirwan. Just as he asked whether the Guardian could achieve a doubling in revenue from digital, e.g. mainly selling ad space, by 2015, the news broke that the “News of the World” was going to be closed down. I am not sure if the Guardian can achieve that ambitious goal but as Nigel Gilbert of Unanimis pointed out, the market is growing, so the prediction is feasible in line with anticipated growth of the industry. We shall see.
Another question was around the future of agencies. Again a futuristic topic, and Richard Wheaton from Neo@Ogilvy says that agencies need to struck the right balance to deliver value for money in the work they produce for their clients. And, that he cannot see advertisers taking “agency work” in house as agencies have too many expert areas and bring a new and fresh insight to clients. I agree with the latter having worked agency side before.
Marco Bertozzi, VivaKi, discussed the fear of agencies to lose business to specialists like they had done when search marketing came about 10 years ago. Hence agencies hang on to technology solutions like trading desks and DSP buying as they need to prove to their clients that they can deliver quality and performance in house. He also mentioned that they pay a higher CPM for some campaigns than with networks but of course achieve better results through buying data and working on clever re-targeting. Something I am not too sure about but maybe on the occasional campaign this might happen. Generally I have seen average CPM prices being a lot lower when buying across RTB platforms in comparison to networks (that is on a mix of RON and re-targeting campaign), besides data integration doesn’t necessarily increase the ROI by the same amount as you pay for it. But I have a whole piece about that coming up in the imedia newsletter shortly 🙂
Again there was a final finger pointing at the publishers to wake up to use their data (I am really not sure how often I heard that on the day) – some already use first party ad serving and are in control as it was assured to me from a reliable source, so I am confident publishers will figure things out very shortly.
Overall, the last session, given it was late in the day and the audience has thinned out, summed up the hype of display media advertising in the industry. I am excited. Display is back, and it is back in real time with real performance! We will kick ass with rejuvenating the top of the sales funnel and offer transparency to advertisers.
I left the event on a high note. It was an eventful day, good conversations over beers and a good vibe in the industry. About data, about performance, about display.
Onwards and upwards. See you next time!