The growing number of video service platform offerings translates into increased competition when it comes to securing viewers and retaining customers. Which is why companies like Netflix, HBO, Prime Video, Youtube TV, Hulu, to name a few, invest millions in marketing campaigns, for TV series and movie licenses, and on high quality in-house TV and movie productions.
The video service catalog is key
The success of a video service primarily depends on what the streaming platform has to offer in its catalog. Consequently, and as we have just mentioned, it’s no wonder these video service businesses invest so much money developing high quality content. Examples include Netflix’s Money Heist, Amazon’s The Marvelous Mrs. Maisel, and Game of Thrones on HBO.
It’s not just about developing in-house, high-quality content. In addition, these video services are battling it out to win existing licenses for content that complements their proprietary super-productions. By doing so, they add even greater appeal to their catalogs since they are salvaging series and/or movies that may no longer be offered on television.
Neither should we overlook the big companies like Youtube TV, Hulu, or FuboTV. They are betting that live broadcasts, little by little, will also replace television. It goes even further. For example, many of the new platforms now offer sports events, which have primarily been offered exclusively on cable. Adding these events to the lineup of movies, series, and documentaries diversifies the content on offer, making it more attractive and comprehensive.
But once we have a competitive catalog…. how do we get the audience to discover, and more importantly, to consume it?
With Jump Deep Recommender, you will be able to recommend content to your viewers based on automatic learning, powered by artificial intelligence.
While it is true that a combined strategy of content curation and automatic recommendations has proven to be a successful approach for showcasing the video service catalog, recommendations based on automated learning are undeniably an effective way to successfully promote the catalog, and with obvious cost benefits.
With Jump’s video content personalization tools, you can recommend content from your catalog that is more in line with the tastes of your audience and based on similarities with the content they already watch. It is a very accurate mechanism that analyzes and tracks patterns of user content consumption.
Our tools do not only make automated recommendations, they also facilitate the definition of a personalization strategy aligned to the video platform’s personalization goals, for example:
- Stipulating which type of content recommendations should predominate (movies, series, or live events, etc.)
- Dictating which genre (adventure, comedy, etc.) should be most recommended.
- Including editorial content, interspersed with automatic recommendations.
- Generating blacklists of content we do not want to recommend.
- Creating and customizing in-player overlay recommendations.
- Creating and customizing recommendation grids in multi-device applications.
- Generating recommendations that are sent in e-mail marketing campaigns, push notifications, etc.
And it doesn’t stop there. You can send your audience proposed content depending on different scenarios such as:
- Different recommendations according to the time of day (morning, midday, afternoon, night)
- Different recommendations on the weekdays versus on the weekends
- Different recommendations depending on the device used to consume the content – Mobile, Smart TV, Web, etc.
Still, offering automated recommendations should not be an act of faith. We should be able to measure the effectiveness of our recommendation strategy so that we can understand if it is effective or if we need to run through different iterations, testing different scenarios using A/B testing, etc.
What kind of metrics should we be looking at when we want to measure the effectiveness of our recommendations?
At a minimum, we should be able to measure:
- Recommendation impressions: Impressions of recommended content, comparing the % to the total number of impressions
- Recommendation impressions CTR: Clicks on recommended content, comparing the % against the total number of clicks on our catalog
- Recommendation Plays: Recommended content playbacks, comparing the % to the total number of catalog playbacks
- Recommendation playback conversion rate: % of playbacks of recommended content over the total clicks of recommended content, and comparing the % to the catalog conversion rate
In summary, being able to rely on a tool to personalize our content offering and thereby increase content consumption – a tool powered by machine learning, like Deep Recommender – is a must for any competitive video service.