The most powerful tool for publications looking to build a loyal, high-value audiences just got even more powerful.

And a lot more public.

On Wednesday, Contextly’s current and new publishers will be powered by our next-generation content recommendation system.

Instead of having an algorithm that uses a one-size-fits all strategy for generating recommendations, our new machine learning system uses multiple strategies for building multiple content recommendations for each post of a publication. As readers interact with the recommendations on a post, the optimization system selects the recommendation strategy with the best performing recommendations.

Contextly helps publications build engagement and a return audience by giving readers the ability to dive deeply into a subject with recommended related content. For those in browsing mode, Contextly also shows off engaging and interesting stories from around a publication.

Co-founded by longtime journalist Ryan Singel and technologist Ben Autrey – a ranking and recommendation specialist, Contextly is the only content recommendation system that combines the power of editorial curation with state-of-the-art machine learning.

Using Contextly’s optional editorial tools, writers and editors can quickly choose related links and link back to previous work without adding additional time to the editorial workflow. In turn, Contextly captures that editorial wisdom in order to feed a sophisticated machine learning recommendation system.

Unlike most “content recommendation” companies that are actually simply advertising companies with unlabeled paid content recommendations, Contextly doesn’t tarnish publishers’ brands with low-quality outbound links to content publishers don’t control.

Contextly’s links are purely internal, turning drive-by visitors into loyal readers.

Contextly also give publishers a powerful way to promote videos, events, conferences and newsletters. Additionally, Contextly can pair product recommendations with content recommendations for sites that combine editorial and retail, such as our good friends at Adafruit.

As Chartbeat documented in its 2013 Annual Report, the largest predictive factor in whether a reader following a link from Facebook, Twitter or Google search would return to a site was whether that reader read a second article.

Contextly works with publishers that want to build, not dilute, their brands, and which are committed to the principle that the way to build a long-term, high value audience is to pair high-value content with high-quality recommendations.

Contextly’s curation, technology and mission makes it perfect for brand name publishers with high-quality content, trade publications, niche publishers, content marketers and company blogs.

Contextly honed its new recommendation system while in the second Matter.VC class and recently accepted investment from Turner’s MediaCamp.