UCD Smurfit Seminar Series – Display Advertising
by Eamonn O Raghallaigh
This seminar, from the Smurfit Seminar Series, was split into two parts; the first part was delivered by Damien Blake, Strategic Partner Manager with Google who discussed display advertising and the second part was delivered by Brian O’Sullivan, Manager Analytics and Attribution, Google who discussed Google Analytics, attribution and conversions.
In the first talk, Damien Blake discussed marketing objectives for display campaigns and outlined that display is generally split into two main categories; direct response – where the focus of attention is on clicks, impressions and conversions and branded campaigns – where the objectives are to create awareness and to build the brand. Damien explained that display campaigns are important in driving search activity and that display is important during all stages of the product marketing cycle, which search marketing is only critical in specific stages. Furthermore he discussed how after exposure to display ads there is a 38% increase in branded search and a 5% to 25% increase in organic search.
Damien went on to discuss the different categories of display advertising products – direct buys and network buys. In the direct buy model, advertisers purchase space directly from publishers – this is more costly that network buys however there is greater control and flexibility in terms of format, assurances in terms of placement and impressions and the customer can get detailed reporting on the campaign. A good example would be the YouTube homepage banner, which can be picked up for as little as $7,000. This can guarantee the customer a minimum of 1,000,000 impressions.
Damien then discussed network buy ads; which again falls into two distinct categories – premium and large scale networks. Premium networks tend to be sold on niche networks of sites; he used the example of a network of fashion sites. These tend to be selectively chosen for inclusion and the websites need to be of a certain standard of quality for inclusion. The second category of network Damien discussed was the large scale network such as the Google Display Network or 24/7 Media. These tend to have a large audience, which particularly appeals to those seeking branding and has a wide range of features, including targeting and reporting. However some of the sites could be of lower quality. The costs associated with large scale networks tend to be lower also. Damien went on to discuss a new advertising paradigm; real-time bidding (RTB). RTB is seen as a leap forward in the way advertising is traded. It allows for purchasers to bid on specific ad impressions in real time based upon user data and other collateral data. It has the potential to give purchasers real power in the decision making process. Finally Damien discussed some of the different ad planning tools such as Google Ad Planner and Quantcast and the tracking of campaigns across multiple networks.
In the second lecture, Brian O’Sullivan from Google focussed on the topic of Google Analytics and the areas of attribution and conversion of leads. Brian opened by describing that one of the major challenges in the today’s digital marketing landscape is the fulfilment of the promise to deliver measurable results. For many years the thorn in the side of the publisher was the justification of the advertiser’s spend; and the advent of tracking technologies inherent in digital marketing made this possible. Brian went on to describe how the attribution process is much more complex that what is immediately visible and the path a consumer takes to final purchase is much longer that what could be first perceived. Attribution is no longer solely in the last click. He explained that over 50% percent of purchasers convert after 19 days; so which ad is given the attribution? Is it the first impression, the remarketing ad or the final ‘call to action’. The attribution process is continuous and attribution can be shared across the full spectrum of touchpoints on the path to conversion. Later he discussed some of the potential attribution models; last click, first click, linear and potential some other unique and yet undeveloped model.
Brian discussed the disadvantages of the ‘last click attribution’ model which only takes into account the last ad click when calculating return on investment (ROI); this gives an inaccurate calculation of ROI as it misses previous advertising spend along the path to conversion. Brian went on to discuss some of the different tracking technologies and how multiscreen consumer behaviour and device fragmentation is disrupting the efficacy of these technologies. He discussed the challenge for the future is not only to track the channel, but also what devices those channels were used on. Brian went on to discuss Google Analytics as an example and we finished with a workshop on attribution and conversion metrics in Google Analytics.
Key Take Away Points

• Display advertising is generally split into two main categories; direct response and branded campaigns
• Display campaigns are important in driving search activity and that display is important during all stages of the product marketing cycle, which search marketing is only critical in specific stages
• After exposure to display ads there is a 38% increase in branded search and a 5% to 25% increase in organic search
• There are different categories of display advertising products – direct buys and network buys
• In the direct buy model, advertisers purchase space directly from publishers – this is more however there is greater control and flexibility in terms of format, assurances in terms of placement and impressions and the customer can get detailed reporting on the campaign
• Network buy ads falls into two distinct categories – premium and large scale networks.
• Premium networks tend to be sold on niche networks of sites and tend to be selectively chosen for inclusion and the websites, need to be of a certain standard of quality for inclusion and tend to be more costly
• Large scale network such as the Google Display Network or 24/7 Media, tend to have a large audience, which particularly appeals to those seeking branding and has a wide range of features, including targeting and reporting
• RTB is seen as a leap forward in the way advertising is traded as it allows for purchasers to bid on specific ad impressions in real time based upon user data and other collateral data.
• One of the major challenges in the today’s digital marketing landscape is the fulfilment of the promise to deliver measurable results.
• The attribution process is much more complex that what is immediately visible and the path a consumer takes to final purchase is much longer that what could be first perceived
• 50% percent of purchasers convert after 19 days
• The disadvantages of the ‘last click attribution’ model is that it only takes into account the last ad click when calculating return on investment (ROI); this gives an inaccurate calculation of ROI as it misses previous advertising spend along the path to conversion
• Multiscreen consumer behaviour and device fragmentation is disrupting the efficacy of these tracking technologies

Reflective Experience
I found the talks from the speakers in Google highly informative and educational. Damien’s talk on display advertising had some very interesting metrics and opened my eyes to the power of display to drive search, the impact of which was not obvious to me previously. From a practical point of view, the addition of display to an SEO/SEM strategy in now something I would see value in, while previously I would not. The concepts of real time bidding and behavioural targeting were interesting topics also; the ability to justify spend with micro-targeted advertising will be appealing to companies with limited advertising spend.
I found Brian’s talk on attribution very intriguing. I had always thought of attribution with reference to the last click model; but I now see his point that it is flawed to calculate ROI using this methodology as other touchpoints on the journey to conversion must have an impact in the process. The complexity of the attribution process is now clear to me and I will watch this space with interest to see what solution Google comes up with to address attribution. The further complicating factors of multiscreen viewing and device fragmentation makes it dizzying to think as to how an effective solution can be found. Overall, these series of talks shed new light on online advertising and the tracking of its effectiveness; important factors when convincing a client to allocate spend.