Data + Personalization = Results
As we look at recent digital marketing trends and look forward to what 2019 has in store, the idea of hyper-targeted personalization continues to be at the forefront. While you may or may not realize it, this micro level of personalization is something you’ve been subject to time and time again when streaming on Netflix.
Netflix spends billions on content and has more than 150M active users, and it uses a personalization strategy to ultimately improve customer experience.
Breaking down Netflix’s personalization strategy
As a Netflix user, you’ve experienced logging in to watch a show or movie, and seeing the “Recommended for you” sections: a collection of shows and movies based on your previous viewing history. The goal of this is to get the right titles in front of the right member, at the right time.
What you may not know, is that Netflix not only selects its recommendations but also the displayed thumbnails based on customer viewing history. What you also may not know, is that each thumbnail you see is actually an ad using targeted creative. And, Netflix achieves 20% higher CTR by selecting different thumbnails for users based on individual viewing histories.
Netflix originally started down the route of rotating thumbnails as a test to determine which image should be selected as the master image for all members to see. What they learned, was that different thumbnail images performed better for different viewers, dependent on their past viewing history!
The following example shows artwork for Stranger Things that each received over 5% of impressions from their personalization algorithm and shows different images covering various themes in the show to surpass what one single image portrays. This example just shows nine images, when in fact there are 100’s of images created.
Netflix then decided to take personalization to the next level by using their member data to select and show an image that would be meaningful to the individual being shown the title.
Here is an example of this personalization based on genre and theme preferences for members:
Someone with a previous viewing history that includes romantic movies or “chick flicks” might be interested in watching Good Will Hunting based on the romance element, hence the image of Minnie Driver & Matt Damon.
Conversely, someone with a previous viewing history that includes male-dominated comedies might be interested in watching Good Will Hunting when seeing an image of a well-known comedian, the late Robin Williams.
Netflix built out several versions of its personalization algorithm and has used member data to perform many A/B split tests to help them understand which artwork or image performs best for different types of members – and ultimately improve the overall member experience.
What we can learn from Netflix’s strategy:
Digital Marketers today have access to amazing targeting capabilities but are still failing to connect with their customers. Using targeted ad creative enables digital marketers to identify with their customer through context.
This is especially important for marketers who invest in paid advertising. Whatever the campaign goal (website traffic, conversions, engagement, etc.), once the target audience is defined it is essential to consider the type of image or creative that would resonate with that audience and drive them to take action.
We recommend taking a page from Netflix’s strategy and making use of data to properly research and segment your target audiences, and test creatives across all digital channels in order to determine what drives engagement.
Facebook has changed everything for SMB marketers, and there’s an enormous need for contextually targeted and relevant imagery to power ads, posts, and campaigns.
At Shotzr, we leverage deep learning to recommend, source and deliver imagery based on location and audience targeting data, customer history, and contextual relevance. Our average client sees a 20% increase in CTR and saves over two hours of work per week.
Here is an example of a Facebook ad created using Shotzr Identify with targeted ad creative based on location & interest data:
Key takeaways & action items:
So, you know more about how Netflix interacts with its users. Why does this matter and why should you care? Netflix has a deep well of relevant data to demonstrate the improvements that are possible through its personalization approach, at scale – and we can draw some insights as to how to approach this as a result:
- Research and define your ideal customer persona (ICP)
- Segment and target your audiences by location, demographic, and interest
- Utilize contextual and targeted ad creatives that resonate with your audience
- TEST, TEST, TEST
- Drive engagement & increase performance across digital channels
At Shotzr, we are bringing the same kind of tools to SMB marketers who spend a portion of their marketing budget on ads, and we see the same signal in our customer’s marketing effectiveness data.
If you’re interested in learning more about how Shotzr can save you time and provide targeted ad creatives to increase your ad performance, get in touch. We would love to work with you.