EX.CO, the publisher video platform powering successful video strategies for the world’s leading media groups, announced an advanced contextual video content recommendation engine for digital publishers. The unique machine learning-based engine uses large language models (LLM) to provide audiences with the most relevant videos from a publisher’s video content bank in real-time, enabling more scalable video integration across an entire website without needing to produce specific content for each article or manually match articles with content.
Autovia, UK’s leading automotive content and commerce company with trusted brands including Auto Express, Carbuyer, and evo, is one of the first of many publishers currently leveraging EX.CO’s upgraded contextual recommendation engine for its network of websites.
“Context and relevance are crucial elements of our business as we strive to deliver the best user experience possible,” said Ciaran Scarry, advertising director at Autovia. “EX.CO’s contextual recommendation engine enhances our UX by allowing us to tailor video recommendations to match the specific content on each page, which is crucial for our in-market audience when they’re researching their next car purchase. This seemingly minor adjustment is a major game changer for us, providing highly-relevant content that captivates our readers and keeps them engaged.”
The LLM-based engine vectorizes text, calculates the similarities between articles and available video content, and then ranks the results to deliver the fastest and highest quality recommendations possible. For publishers who need additional video content for their pages, the engine can also tap into EX.CO’s vast content marketplace which offers thousands of high-quality videos in different verticals from premium sources.
“Audiences today will only engage with content that is truly relevant to them; however, it’s challenging for publishers to produce and match such large amounts of video content,” said Tom Pachys, co-founder and CEO at EX.CO. “After having in-depth conversations with publishers and conducting our own data research, we realized that the old ‘tag/taxonomy’-based approaches were insufficient. By integrating LLM capabilities with ML optimization models, we built a new-generation recommendation engine. We were surprised by the immediate results, which surpassed our previously highly refined models still considered best practice in this field.”
The video recommendations are served ‘at the speed of news,’ leading to higher audience engagement and retention which, in turn, has the potential to boost other vital KPIs such as revenue, brand loyalty, and subscription growth.
EX.CO’s upgraded contextual recommendation engine is now deployed for EX.CO partners. Top tier publishers are achieving an 80% relevancy match rate, leading to 4X higher engagement rates with the video player than the industry benchmark. Moreover, average negative interactions with the video player have decreased by 30-40%.
The technology currently optimizes video recommendations based on criteria including media category, title, recency, sentiment, keywords, and length. EX․CO plans to soon expand this offering by adding ChatGPT-like functionality, allowing publishers to refine video recommendations by using prompts. This will help train the engine to deliver more relevant recommendations for specific sites, sections, and articles.
SOURCE: PRNewsWire