Technology That Puts Ads in the Right Context

Are Contextual Ad Systems Built for Today's Internet?

The short answer is no, because today's contextual matching technologies are not providing sufficient precision, adaptability and confidence in smart ad placement.

In today's Internet, tools and functionality have empowered web publisher with unseen capabilities to create, crowd-source and publish content. As a result, users are consuming more specific content, more dynamically and interactive each day. Therefore, the online consumer audience continues to fragment more and more while the sheer volume of topics grows exponentially. At the same time this is a great opportunity as the consumers' unique needs and personalities are reflected in the content of the pages they interact with daily across traditional web sites, social networks, and other user-generated content.

So, how do marketers and ad networks generate and distribute relevant ads to their target consumers with confidence in such an environment? So far, the industry is failing to match available targeting and brand protection capabilities to reflect that change and provide more specific and safe ad placements. There is a clear lack of topical adaptability and ad systems being too slow to connect the lively content changes of publisher with the right advertisers.

Users are not targeted with more specific ads and therefore ads are ignored as not helpful or even annoying. At the same time media buyers are not given the tools to reach the valuable parts of their audiences at scale beyond premium inventory. The result is that advertisers pay for irrelevant impressions, or see poor ad performance, or reach too little of the valuable parts of their audience even if they move in mid-tier non-premium inventory. This ultimately leaves valuable marketing opportunities on the table and under utilizes the Web's commercial power.

Pattern Proximity + Dynamic Category Matching = Monetization of Unstructured Content

Proximic leverages several unique proprietary technologies. At the core, a real-time matching engine associates users and unstructured text with ads. The Proximic platform applies the following steps to generate high-performing targeted ad matches:

Pattern Profiling: Extracts the contextual essence from unstructured content
Proximic quickly analyzes each page or context along with other complementary pages to generate a contextual "fingerprint" of the text. Novel, proprietary "Pattern Proximity" technology applies pattern recognition of symbols, such as characters and character sets. This approach is agnostic to intimate knowledge of language or language changes. As such, the technology is massively scalable into any market within several hours, such as the Company's roll out into China.

Dynamic Category Matching: Connects the contextual essence of pages with structured targeting data
Proximic links the contextual essence fingerprint to the structured world that advertisers target against. This unique system uses crowd-sourced data across the Web, to automatically distill category connections to become compatible with the challenges of ever-changing increasingly dynamic topical evolution. The result is a continuously evolving and self-maintaining Dynamic Category Net spanning thousands of categories - at an order of magnitude that exceeds linguistic approaches. Proximic identifies relevant connections for a profiled page among millions of indexed high-value targeting keywords and display categories. For example, a page on "Computer Waste" will trigger ad keywords such as "Green Tech" or "Sustainable Computing", as well as terms not even present on the page itself, such as "Eco-Labeling of IT Products".

Performance Optimization: Post-ranks by projected ad performance
When partners integrate Proximic into user feedback data, real-time analytics and validation of keyword-bid relationships determine the best performing ad type and keyword in terms of effective CPM (eCPM). The optimization engine continually improves through hypotheses testing of the ad match relationships with each partner's user stream.

As a Result, Proximic Is More Dynamic, More Granular and More Versatile

The effect of this matching engine is to match in real-time users with commercial content with a high degree of relevance. To find out how it differentiates from other solutions in the market go here...