Latest / Marketing Made Simple with Fexingo: Branding, SEO, and Digital Strategy for Beginners / How Netflix Uses Data to Create Hit Shows
Transcript
- Lucas: Netflix spent around seventeen billion dollars on content last year. That is more than the GDP of about thirty countries. And the question nobody asks is — how do they decide which shows to make? Luna: I mean, we assume it's data. But is it really? Because some of their biggest hits — 'Stranger Things', 'Squid Game' — feel like creative leaps, not spreadsheet decisions. Lucas: Right, and that's the tension. But the truth is, Netflix is probably the most data-obsessed company in Hollywood. They just hide it well. They don't make shows in a vacuum — they use viewer behavior to inform literally every stage of production. Luna: Give me an example. How early does the data come in? Lucas: Before a script is even written. Netflix looks at what genres, actors, and themes their subscribers already binge. They call these 'taste clusters'. So if they see that people who love horror also love period dramas and also watch a lot of foreign-language content, that's a signal. They might greenlight a Spanish-language gothic horror series. Luna: So 'The Haunting of Hill House' was born from that kind of pattern matching? Lucas: Exactly. Mike Flanagan's shows are a perfect case. Netflix knew horror fans were underserved in serialized TV. They also knew those fans tend to finish series quickly. So they invested in a high-quality horror anthology — and it paid off. The data didn't write the script, but it created the conditions for the show to exist. Luna: But doesn't that risk making everything feel samey? If you're always feeding the algorithm what it already likes, you end up with a library of derivative content. Lucas: That is the critique. And Netflix knows it. So they actually do something counterintuitive — they reserve a portion of their budget for 'data-defying' bets. Shows that don't fit any existing cluster. 'Squid Game' was one of those. No famous actors, a brutal premise, Korean language — every conventional metric said no. But the data also showed that their audience had high tolerance for subtitled content and that competitive survival stories had growing interest. Luna: So even the outliers are somewhat data-informed. Lucas: Precisely. They use data to reduce risk, not eliminate creativity. The real magic happens after a show is made. Netflix A/B tests everything — the thumbnail, the trailer, the description, even the order of episodes. For 'Stranger Things', they tested dozens of thumbnail variations. The one that performed best featured the kids on their bikes against a red sky. That image alone drove a measurable increase in click-through rate. Luna: I've read about that. They test up to six different thumbnails per show per user, right? And the algorithm learns which visual style you respond to. Lucas: Yeah, it's personalization at scale. If you tend to click on thumbnails with a single character's face, you'll see more of those. If you respond to action shots, the system adapts. That's why two people scrolling Netflix at the same time see completely different artwork for the same show. Luna: Which is brilliant marketing — but also a little creepy. Lucas: It's the definition of targeted marketing. And it works. Netflix estimates that their recommendation engine saves them over a billion dollars a year in customer retention. Because the more tailored the experience, the less likely you are to cancel. Luna: Let's talk about cancellations though. Netflix is famous for axing shows after one or two seasons — even critically acclaimed ones. 'The OA', 'Santa Clarita Diet', '1899'. Is that also data-driven? Lucas: Yes, and it's brutal. Netflix looks at a metric they call 'cost per completed view'. If a show costs a hundred million dollars but only fifty million people finish it, that's two dollars per completed view. But a cheaper show with the same finish rate might be four dollars. They compare that to the likelihood of the show attracting new subscribers. If a show isn't driving sign-ups, it's gone. Luna: So they don't care about prestige or awards if the numbers don't add up. Lucas: Not really. And that's the downside of pure data-driven content. It can be ruthless. But it's also why Netflix has the highest customer satisfaction in streaming. They're giving people what they actually watch, not what critics want. Luna: I wonder if that approach will change now that streaming is maturing. With subscriber growth slowing, maybe they'll focus more on retention and less on constant new content. Lucas: That's a great point. In fact, we're already seeing it. Netflix has started investing in live events and interactive content — like the 'Black Mirror: Bandersnatch' experiment or the upcoming live sports specials. Those are harder to replicate with an algorithm, but they create buzz and community. Luna: If these marketing conversations have sparked something you've actually used — maybe a new way you think about your own content or brand — that's exactly the kind of reaction that keeps this show going. We keep the show ad-free because of listeners who find value in it and want to support that. If that's you, you can buy us a coffee at buy me a coffee dot com slash fexingo. Lucas: And we really mean it when we say that support lets us keep diving into these deep cases. No sponsors, no mid-roll interruptions — just the real strategy behind how the smartest companies market themselves. Luna: So back to Netflix — you mentioned interactive content. That feels like a shift from pure data optimization toward experience design. Is that the next frontier? Lucas: I think so. Because the data will always be there, but the competitive advantage now is emotional connection. Netflix's algorithm can predict what you'll watch, but it can't make you love a show. That still requires great storytelling. The best marketing strategy in the world won't save a bad product. Luna: And that's the lesson for any brand. Use data to understand your audience, but don't let it replace intuition or creativity. Lucas: Exactly. Netflix succeeds because they marry the two. They let data inform the strategy and creativity execute it. That combination is hard to beat.