In a generation digital marketing has evolved to become the primary way many brands run their campaigns all over the world.
But none of it would be possible without the underlying technology infrastructure that has subsequently developed around programmatic advertising.
To practitioners the language of programmatic may well be second nature, but for many in the sector the programmatic landscape can seem confusing hard to penetrate.
This is the case even for experienced traditional marketers, and even digital marketers who may understand the technology beneath owned media, but struggle with advertising technology which sustains paid media.
To help marketers build their expertise Oracle recently hosted a webinar which steps marketers wanting to understand more about the programmatic advertising landscape through all the key elements of the programmatic landscape.
In part one of this report we look at that landscape and in part two, we describe programmatic strategies available to markets.
But lets start with the basic question, what is programmatic advertising?
Put simply it is a marketing approach that delivers the most relevant message to the right person on the right device at the right time to achieve a desired action.
It is optimised in real time based on data that allows the marketer to focus on individual impressions instead of block buying advertising slots. This is what makes it very different to traditional advertising.
Instead of static inventory with analytics derived from surveys and panels, the programmatic approach allows advertisers to serve impression that are both dynamic and relevant – because they are based on who is viewing the impressions.
Importantly, as this is based on the idea of one on one advertising it also allows markets to derive one on one user insights.
Underpinning the programmatic advertising landscape are digital platforms and exchanges which enable the buying and selling of advertising inventory across mobile, desktop, search, display and video advertising.
Advertisers and publishers can transact in real time just like on the stock exchange, although the amount of transaction the ad tech sector supports each day dwarfs the volume on a financial exchange!
Advertisers interact through what is called a data management platform via a supply side platform (SSP) while marketers interacting with the DMP through demand side platforms (DSPs).
These interactions are facilitated by add exchanges the middle.
Like most other forms of digital marketing, the success of these interactions and the effectiveness of programmatic campaigns is based upon the quality of the data.
First, second and third party data
The most valuable information for any advertiser is the first party data which is basically the data that is proprietary to them and found in places like the advertiser’s web site, its customer relationship management platform or even the email data to which it already has access.
Various platforms like Eloqua, CXD and Blue Kai allow marketers to integrate this information into their programmatic activity.
Second party data on the other hand comes from when the advertiser has a direct relationship with a publisher and is able to use their data as and when required.
Sometimes, however, the reach provided by the first and second party data simply isn’t enough. Then, marketers and their agencies use third party data is important to expand the reach of a campaign.
Third party aggregators and publishers collate data they collect in the form of cookies from their 100s of web sites and sell it to advertisers.
Various vendors like Nielsen, Iota and Data Logic provide the demographic, geographic and other types of data needed to supplement the first and second party information.
The data sets can be huge and the sheer amount of data can be overwhelming. It needs to be managed which is where a data management platform comes into the conversation.
A DMP is the backbone of data driven marketing. It serves as a unifying platform to collect, organise and activate first, second, and third party audiences data from any source including online, offline, mobile and beyond.
Once the demand side data is aligned we can then use various DSPs like Dataxu, TradeDesk, or Rocketfuel to work with the various advertising exchanges to purchase relevant inventory from the supply side.
Here’s a simple example of how this all folds together.
Imagine a campaign where a brand uses its first party data but determines that much more reach is required. The brand then purchases third party data that matches its specific geographic and demographic requirements.
These two pieces of data are then combined through a DMP. And all of this comes together in a matter of milliseconds.
Imagine next that a consumer goes to a publisher’s web site. The publisher calls its web server likely before the consumer’s page has fully loaded. The ad server checks its rules and determines what ad it can serve and at what price. The ad sever then instructs the consumer’s browser to call the advertiser’s ad server, and then the publisher’s ad server counts an impression .
The advertiser’s ad server knows it can serve the specific creative and an impression is counted to that site. Placement, combination and the campaign spend is logged for that impression.
And the user sees the advertisement. This all happens in real time.
Of course, this being a programmatic campaign, another consumer viewing the same page at the same time will potentially see a completely different ad based on the persons characteristics.
Pixels and cookies
A tagging pixel is essentially a piece of code that is placed on a web site by a marketer and it generates a notice of visits to the page by a browser. Pixels often work in conjunction with cookies recording when a particular computer visits a specific page and they can be played across the site or on certain conversion pages only.
The placement is determined by what you want to measure.
It is important to understand that different types of pixels do different thing.
Conversion pixels for instance capture conversion events. This is the only way markers can record view-through conversions and post click conversions. The conversion pixels are installed on the page where the marketing goal is achieved such as a form page or a landing page.
Optimisation pixels on the other hand are installed across an entire site and used to better identify ideal targets in prospecting and site retargeting campaigns.
Finally, data collection pixels allow data collection companies to anonymously identify and classify web site visitors into various categories.
All of these pixels help programmatic vendors track the success of a campaign. And they are used to build look alike profiles so that a brand’s programmatic vendor can find more of its ideal audience online.
Pixels also provide rich insights into your audience such as the type of websites they list and their interesting certain categories
So what is a cookie? Put simply it is a mechanism specified by a http protocol that is implemented by the browser for web sites to store data locally.
For instance cookies are used to help a site remember that a visitors logged in rather than making them login every time they come back.
They are also important for saving shopping cart information and for tracking other behaviour online.
From a security perspective cookies can only be sent to the domain that originally sent them. For instance only oracle.com can set oracle.com cookies.
In part two, we will look at different programmatic strategies brands can employ in their campaigns.
About the author
Mandar Dadegaonkar is the Senior Manager, Digital Marketing, Oracle which is a corporate member of Which-50. Members provide their insights and expertise for the benefit of our senior executive audience. Membership fees apply.