RTB House Announces Multi-Layer Brand Safety Mechanism Powered by NLP
RTB House is a leading provider of technology powered by natural language processing that addresses the brand safety dilemma
RTB House – a global company that provides state-of-the-art retargeting built on deep learning for top brands worldwide – announced new capabilities to solve digital advertiser’s brand safety challenges powered by advanced natural language processing.
Brand safety is a critical concern for the digital advertising industry. In order to avoid consumer backlash and lost revenue brands and publishers must ensure that ads don’t appear next to inappropriate content or within an unsafe environment. The problem is exacerbated as the number of web pages that contain fake news, violence, catastrophe, or explicit content continues to rise. Furthermore, those producing the content continually develop new ways to disguise the content to monetize their business through online advertising.
At the time of this announcement, Artur Jaworski, Business Director, UK, RTB House, said, “We are the leading provider of technology powered by natural language processing that addresses the brand-safety dilemma. Our solution is a game changer. Our brand safety solution has been just tested thoroughly by one of our key UK clients using the independent monitoring mechanism of Integral Ad Science (IAS). The test results were significantly above the IAS benchmark for the UK.”
To complicate the issue further, because each brand has a unique target audience and brand safety preferences e.g. in terms of what political views are unwelcome there is no one-size-fits-all solution.
To solve the issue RTB House has developed its multi-layer brand safety mechanisms to offer brands unique solutions based on individual requirements. With real-time content analysis and natural language processing at its core, RTB House’s new brand safety platform skims tens of thousands of internet articles and websites, ensuring brand safety is protected in real-time. Unlike other solutions, this is not only blocking articles based on blacklisted keywords in the URLs of article pages but actually checks the contents of the articles. Also, this does not just block banners on websites with blacklisted keywords but actually understands what the article is about regardless of the keywords and then decides if the blocking is required.
The technology, known as the Brand Profile layer, prevents ads from being served on websites or next to content that is not offensive in general but may have negative connotations for some brands. The Brand Profile layer works in three ways:
News filters that automatically block news containing specific non-brand-safe keywords (both in URL addresses, in the content, or even in the meta page descriptions). This enables brands to exclude specific content that is inappropriate e.g. automotive brands want to avoid ad placement alongside articles about alcohol but not necessarily alongside articles about Windshield washer fluid mentioning this contains alcohol. The natural language processing algorithm developed by RTB House has already cataloged more than one million articles and is constantly auditing thousands of new articles every day.
Website category filters help brands to avoid website categories, which may be sensitive from the perspective of different brands, for example, user-generated content, file sharing, gambling, video stream, gaming, manga, and social services.
The Blacklist allows brands to prevent ads being displayed on specific, chosen URLs.
RTB House clients use the company’s solution to deliver ultra-personalized ads based on behavior tied to users search tactics, browsing history, and basket behavior. These observations and more are collected in real-time across all devices and have been proven to improve conversions and boost revenue. RTB House’s world-class retargeting platform currently handles over two million bid requests per second, over 10 billion ad views per month and over 3.5 million clicks daily.