Ad Platform Uses Natural Language Processing To Safeguard Brands’ Image

Both brands and publishers have faced a crisis in recent years as online ads have appeared next to a range of inappropriate content, from terrorist propaganda and extreme political views to violence and body shaming. These challenges have been exacerbated with increases in hacking that have added more malware and phishing content on the Internet.

RTB House, a global ad retargeting platform with clients including adidas, Sephora, Walmart and, has added natural language processing (NLP) to its brand safety solution. The proprietary algorithm uses NLP in a new brand profile layer offering comprehensive inspection at the page level — including URL, content and metadata analysis — before allowing ads to be placed there. The brand profile layer can be customized for each client, and it’s available now for customers with language support for eight major global markets including Australia, Canada, India, the UK and the U.S.

Since activating the NLP-powered algorithm, RTB House has catalogued more than one million articles and is auditing 16,000+ of the most popular global articles each day. The company’s goal is to more than triple these daily audits to 50,000 articles.


The core RTB technology prevents ads from being served on web sites or next to content that’s likely to have negative effects on the brands, with technologies that can be tailored to each advertiser’s brand safety policies. The platform includes three independent features:

• A category filter to avoid sensitive web site categories like politically controversial or user-generated content;

• A news filter that automatically blocks news content containing specific non-brand-safe keywords; and

• A blacklist that lets clients block ads from specific URLs they choose.

“Our brand safety solution analyzes the actual content on web pages before placing ads through its ability to understand unstructured data,” said Gary Burtka, VP of U.S. Operations at RTB House in a statement. “It then prevents ads from being served on them through offensive-content blocking. We also provide a combination of robust security to prevent ad placements in unsafe environments.”



Access The Media Kit


Access Our Editorial Calendar

If you are downloading this on behalf of a client, please provide the company name and website information below: