What you need to know about Semantic Ad Targeting
The Semantic Web is a visionary concept which has been gaining traction over the last decade. According to the World Wide Web Consortium (W3C), an international community with a stated mission to lead the web to its full potential, the semantic web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries.
So how does that relate to advertising? Well, in the sense that the semantic web provides a framework for sharing data, semantic ad targeting provides a framework for classifying different sources of content (data) according to comparable categories. At the core of this technology, ads are targeted according to the semantic categorization of adjacent content.
Think about your favorite online news website. Editors usually organize articles into recognizable sections – News, Sports, Entertainment, Weather, etc. That structure makes a newspaper website easier to read, navigate, and digest. How is semantics different? Far beyond the editing possible at a sizeable newspaper, semantic technology can understand the meaning and sentiment of content (data) and then categorize it at a very precise level. Though articles about French President Sarkozy’s popularity and the environmental impact of real estate on the French Riviera will both appear in the international news section of your news website, if you were to semantically categorize these two articles, one would appear under European news and Politics while the other might be European Business, Real Estate and Travel.
This example gets to the heart of semantic ad targeting – the ability to understand the finite details of content, categorize them, and then target relevant, category-specific advertising. Why is this important? Because studies completed by companies in the semantic ad targeting space have shown that ads targeted to semantically targeted content can generate massive improvements in advertising effectiveness. At Peer39 we have seen a boost in response rates 200% greater than run of network, and even up to 600% in some cases. That translates into a lot of dollars for ad networks, yield optimizers, ad exchanges, demand-side platforms and of course, publishers.
Let’s use the aforementioned article about French President Sarkozy as an example. If, however, the article also focuses on his wife, actress/singer/model Carla Bruni Sarkozy, then we can semantically classify that article as Entertainment content, and entertainment CPMs and CTRs are much higher than international news rates. Therefore, as an entertainment article, it can generate more revenue from adjacent ads.
Beyond the increase in revenue (and in turn, enhanced targeting for advertisers), there are other reasons to consider semantic ad targeting:
Brand Safety: One of the reasons why brand advertisers have shifted a proportionately minor share of their ad budget to online media is due to a concern about brand safety. Major brand marketers are still concerned that their ads will appear next to content that is inappropriate or simply does not fit with their brand image. Peer39 provides a customizable filtration system for content which can address the brand safety concerns of even the most family-friendly brands, helping to prevent ads from running alongside unwanted content.
Get Affordable Precision at Massive Scale: Semantics can be overlaid with any targeting strategy, working in tandem with other data providers to achieve massive scale at a relative bargain. Furthermore, semantics can deliver actionable targeting data on nearly 90% of all impressions. That’s billions of impressions with a clarity that only semantics can deliver that are accessible to media buyers. Never before have marketers had such a powerful targeting tool at their fingertips.
I hope this post helps to better explain semantic ad targeting and why you should consider this tactic to meet your online advertising needs.