Ad Verification – Content Classification at Advertiser Level
If you knew what content your ad would be served next to, would you choose to put it there?
Verification tools which specifically analyze the content of the page are also referred to as “CV tools”. CV tools operate most effectively in real time. This means that when the third party ad is served to a page; the domain name, sub domain and URL are passed into the CV tool, where-upon the tool either looks the page up in its database of ready-scraped pages or it scrapes the page in real time. The scraping process captures information on the page including words, phrases, word combinations, multimedia; including video, and digests the relationship between these elements.
These elements are then categorized so that for instance, a page-level reference to a particular celebrity mean that the URL of the page is categorized in a celebrity category. Superior CV tools can analyze the semantics of the page in multiple languages and even calculate the cultural significance of certain phrases. The outcome is a statistical analysis which places an affinity of the page towards a particular categorization. This means that if an Advertiser paid for ads to appear next to sport related content and the CV tool reported it was shown against music content, the Advertiser could take the third party report back to the Publisher in order to have future ads targeted away from the incorrect pages and onto more relevant, promised content. Publishers have access to the same level of reporting to assist them in managing their own inventory to avoid selling inventory that is not relevant to an Advertiser from the outset of the campaign.
CV tools are interesting pieces of kit because they do not utilize user cookie information in order to provide information about page classifications. This means that they do not fall subject to any limitations surrounding privacy constraints around cookie based targeting.
The trafficker would check verification reporting around content classification and compare it to the goals of the planner in agreement with the Publisher to appear on pages geared towards certain types of content. If there is a mismatch, the trafficker would alert the planner to speak with the Publisher about the targeting and delivery concerns. A process known as “remuneration”, but not necessarily meaning that a refund is collected.
Content tends to be classified into a finite number of categorizations which are defined by the CV tool. Until these are universally accepted categorizations by all parties around the table (Advertisers, Agencies and Publishers) and written in stone by regulators like the IAB, there will be disputes about the methodology.
To make negotiations more challenging, custom content categories are available for Advertisers to create and define so that reporting lays out where an Advertiser’s ads have appeared next to content geared towards a certain sentiment or alongside praise about a competitor. In such situations Publishers may not have the available technology in their Publisher ad server to avoid delivering ads towards or away from such pages in the future.
To find out more about ad verification in advertising technology across all digital channels (with clear diagrams), get a copy of: Ad Serving Technology – Understand the Marketing revelation that commercialized the Internet – available now from..
Or Amazon USA: http://www.amazon.com/dp/1484867572/ref=cm_sw_su_dp