Netflix is not just a successful streaming platform or producer of original video content — in many ways, it behaves like a customer data platform (CDP). Its algorithms collect massive volumes of data on what subscribers watch, enjoy, and search for before spitting out recommendations for other shows they might like. These insights are even used to automate and personalize the UI experience, right down to changing content thumbnails and cover art.
We may not think about Netflix’s algorithms and user profiles often, but they’re a major factor in the company’s success. The streaming service calls its data platform the “core foundation in driving all of our product decisions that directly impact our customer experience,” and constantly seeks new ways to innovate and effectively use this data. It’s working — the company has a user retention rate of 93%, an almost unbelievable figure among marketers. Data capabilities are helping Netflix sustain its market advantage, even as competitors like Disney+ and Amazon Prime rise to challenge it.
For an intro to CDPs and how they work, go here and check out our other articles on CDPs:
Navigate the Fluff: A Best Practices Guide for CDPs and Customer Retention
Asking the Experts: Is It the Age of CDPs?
Why CDPs Fail: A Tale of Three Unfulfilled Expectations
Focus on specific data goals
While Netflix knows a great deal about our viewing habits, those insights are based on a surprisingly narrow data range. The platform’s core data process is to gather information exclusively from within the service. It doesn’t collect data from outside the platform, nor does it track the habits of non-subscribers. Netflix focuses on the ways specific users utilize the platform and extrapolates trends from there.
In other words, Netflix is a prime example of the ways small, local data can lead to strong business success. Instead of struggling to obtain every detail about its subscribers, the platform tries to understand everything about how they use it — which is ultimately the most valuable information. Netflix has followed this core business strategy since it first launched, serving as an excellent lesson for marketers in other spaces.
Watch precisely how users engage with your platform
While Netflix’s data range is relatively narrow, it squeezes every ounce of information it possibly can from what it has. The platform is constantly monitoring and collating every action a user takes. We’re talking everything that could matter, from the videos you watched to the ways you watched them: completion rate, time of day, when you paused, fast-forwarded and rewound, when you started, and when you stopped halfway through then resumed. Netflix takes this information both in aggregate and individually to build a portrait of how specific users consume content. With these insights at hand, the algorithm compares, contrasts, and makes suggestions.
Personalize product recommendations with expertly-assigned content tags
When marketers consider personalization, they usually build campaigns based on their biggest market segments and personalize within those groups. Netflix, however, focuses on the “market of one” with a constant stream of unique personalization elements. Its CDP is trained to identify core audiences for new products based on a highly detailed tagging system that goes beyond genre to include dates, actors, and other traits within the content. This means Netflix can generate data-backed recommendations that will be a near-perfect fit in terms of shared product traits. It works, too: approximately 75% of Netflix’s watch numbers come from their recommendations algorithm.
These tags aren’t created by the algorithm itself, however. There are hundreds of tags within Netflix’s system, painstakingly applied by content experts. Those tags are processed within the recommendation algorithm, which measures any similarities and provides recommendations to users. Netflix’s CDP can even be used to project which audiences will enjoy greenlit original content — Orange Is the New Black was expected to be a smash hit among Weeds fans, for example. But that insight doesn’t happen without a manual helping hand, because the CDP only knows how products are used — not what the product is.
Use CDP features to save your marketing budget
When Netflix was advertising House of Cards to its audience, it had enough customer information to create ten different versions of the trailer for various audience segments. While this may seem excessive, it actually optimized Netflix’s marketing budget. Instead of trying to mass-produce trailers that don’t account for market segments, Netflix saved money by delivering ad content to the most receptive audience. Thanks to this highly efficient data use, the company now considers marketing a multiplier on content spend.
In many ways, this summarizes the power of Netflix’s CDP: while a great deal of work is required on the backend, the end result brings increased performance and efficiency, while staying well within budget. For marketers, that means you don’t have to look far to enhance the power of your CDP. If Netflix can do it, so can you.