Tips for Winner Video Engagement

By Gal Ofel November 14, 2018

As Director of Product, I spend a fair share of my time speaking with publishers about the pain points of their business, and understanding what holds growth back. What amazes me time after time, is the complicated relationship digital publishers have with video. It seems that for most publishers, video can be your best friend or your worst enemy.

Video is considered to be an eyeball magnet, drawing users into longer sessions and creating additional monetization opportunities. Indeed, when video engagement is done right, the session length can jump from the average 2.5 minutes it takes to read an article, to an average 10-15 minutes of short clip ‘micro binge-watching’.

But nailing user engagement is a fine art. In fact, many publishers consider video ROI to be their #1 challenge.

We have been in the business of video engagement for a long time, and have made it our mission to crack the code on optimal video engagement.

Based on years of engagement data collection and analysis, our video experts share these three video engagement best practices:

  1. Optimize Video Matching: Poor Video Allocation is Engagement Suicide

Riding the pivot to video, many publishers do their best to include at least one video unit within each page. But this sometimes backfires, as articles with unrelated (or kinda-related-but-not-really) video can become real engagement killers.

Allow me to illustrate:

A user chooses to read an article about Lady Gaga’s latest movie with Bradley Cooper, and scrolls through the article to read about the film’s production and cast. Now imagine the same user reaching a video unit which offers a completely irrelevant video clip. The chances of creating video engagement would be slim, as most users will simply skip the unit and cost the publisher several lost ad opportunities.

But what if the same unit could serve fresh footage from Lady Gaga’s latest red carpet appearance? an exclusive interview with the film’s cast, or a rare ‘behind the scene’ moment?

When choosing to read an article, users are actively selecting a ‘Content Track’, and providing data on what topics, people, brands or events they are interested in. The natural best practice, in this case, would be sticking to the track defined by the user and serving a video clip that reflects the content of the page.

But many publishers don’t have access to an unlimited video library, definitely not one that is updated daily. Even if access to such a collection was possible, manually matching video content to each article can be extremely time-consuming, nevermind completely unscalable.

This is where Publishers should be using reliable and automated video recommendation engines. These advanced video tools should have access to the freshest high-quality content on the web. They should also be able to scan each page instantly and match video clips that feature the people, brands, events and content categories found on the page.


  1. Don’t Break the Flow

Matching the first piece of video content is a crucial moment for creating engagement. But what comes next matters even more.

Remember our ‘Content Track?’ In order to increase time on site, publishers must continue the flow of the selected Content Track and finish each piece of video with a promise of additional relevant video content. This can be done by grouping video clips by the featured people, brands and categories.

Once video clips are contextually linked to each other, content can be organized into infinite ‘Content Tracks’ that offer a continuous flow of relevant engaging video content.

But contextually linking relevant clips to each other at scale depends on the publisher’s ability to extract and collect data from each piece of video content. This task can be tricky, as most content syndicators provide little or no information on the clip level.

Data is the missing piece.

In order to gain visibility into video data, publishers should demand that content syndicators transform from being dumb “video pipes”, and become reliable data providers, attaching critical metadata to each video clip and enabling Publishers to categorize video inventory.

  1. The Freedom of Choice

VOD and OTT platforms have changed the way users consume video, and have raised the UX bar. In a world of Youtube and Netflix, users expect video to be served as an indexed and searchable experience. They also expect the freedom to choose the content they find most interesting.

True, Your website may not be a Netflix or a YouTube, but with the right tools, you can offer an on-page ‘VOD-like’ video discovery experience, inviting users to browse through your video content, just like they would on YouTube.  

To do this, publishers need to follow these two steps:

Step one: Bundle-up your content

The first step towards enabling video discovery is ensuring your content in bundled into tight contextual tracks, where relevant clips are linked to each other (as described in section #2 above).

Step two: Put the user back in the driver’s seat

The next step is optimizing the way the contextual content is served, by bringing the user back into the game. In other words, don’t limit your users with the video placed on the page! Invite them to browse through more content options and extend their content journey within your site.

Engagement is Serious Business

With advertisers’ allocating nearly 60% of their digital budgets to video, video engagement is more valuable than ever.

With this in mind, we have made it our business to help publishers maximize every session, increasing time on site and creating the conditions for digital video ‘micro binge-watching’.


AnyClip’s LuminousX solution is designed to completely revolutionize the way video is maximized for publishers and their users. Contact us today to start reinventing your user engagement with a contextual video experience!