How JW Player’s Recommendations Engine Enhances Audience Engagement
You’ve seen them everywhere: shopping at Amazon, watching shows on Netflix, or discovering “people you may know” on LinkedIn. Recommendations engines allow you to view related content and products based on your online consumption behavior.
For publishers, recommendations can be an especially powerful tool in executing your video strategy. They keep eyes on the page and increase click-throughs, strengthening retention when competition for user attention is fierce. These results came through loud and clear with the A/B test of JW Player’s Recommendations engine.
We got here by combining the following two approaches in our algorithm:
- Content-based filtering – Matches the content of the item with media that’s similar to what the user is watching, based on titles and descriptions
- Collaborative filtering – Draws from the collective intelligence of the entire user network
To put this algorithm to the test, we measured it against a control variant across 15 different media properties.
Compared to the control, we found that on average, JW Recommendations produced:
- Higher engagement
49% increase in the click-through rate
68% growth in the percentage of unique viewers who watched at least 30 seconds
80% rise in incremental time watched per embed session
- Higher retention
13% jump in the percentage of viewers who returned the day after entering the experiment
11% boost in the percentage who returned 7 days after entering the experiment
At this point, you may be thinking, “Okay, this is all fine and good. But I have an editor on staff who knows our audience and could come up with recommendations too.” While that may be true, there’s a tradeoff: The process is often cumbersome, and the opportunity costs from using those editorial resources could be significant.
JW automates recommendations so that they are not only easier to implement, but are able to incisively target audiences based on our advanced data analytics. Our team of data scientists, designers, and engineers continuously update the algorithm and user interface to drive performance improvements. To learn more about how this works, join our webinar on Thursday, Nov. 2, at 1 p.m. EDT.
As the media landscape grows increasingly crowded, take it from us: When it comes to keeping your viewers engaged, revving up a recommendations engine is simply . . . highly recommended.