“Recommended for you”: Why Netflix’s rating system needs to go
Netflix subscribers have grumbled about the streamer’s recommendation engine ever since the company ditched its old five-star system in 2017. The thumbs-up and thumbs-down buttons, paired with those personalized percentage-match scores, were meant to feel simpler, yet many viewers still find the suggestions off target. The phrase Netflix recommended surfaces constantly in user gripes because the current tools rarely seem to capture actual taste.
The original star ratings had their own problems. Netflix explained that the stars never reflected what other viewers thought; they were the service’s best guess at how any given subscriber would rate a title. That distinction confused plenty of people, so the company swapped the scale for the binary thumbs method and a matching percentage that promised more accurate personalization.
Evolution of the Thumbs System
Netflix did not leave the system frozen in 2017. In 2022 the company quietly added a second thumbs-up option, giving subscribers three explicit signals instead of two. Two thumbs up now stands for a strong positive reaction, one thumb registers mild approval, and a thumbs down still marks outright rejection. Reports from early 2024 also suggested the service might soon dial back the visibility of the percentage-match score, another sign that the interface keeps shifting even while the core personalization approach stays in place.
How Netflix Recommendations Actually Work Today
Explicit thumbs remain only one input among many. Netflix’s algorithm weighs viewing history, completion rates, pauses, rewinds, and how recently a title was watched. Recent activity carries heavier influence than older habits, which is why a single late-night binge can suddenly reshape an entire homepage. Roughly eighty percent of what subscribers watch still comes from these automated rows, so the quality of every signal matters more than ever.
Thumbs-Down Effectiveness and Persistent Suggestions
One long-standing complaint is that thumbs-down titles refuse to disappear. Forums in 2024 and 2025 show the same frustration: users mark a show as unwanted, yet similar titles keep surfacing days later. The addition of two thumbs up gave viewers a stronger positive lever, but it did not solve the problem of unwanted recommendations slipping through. Netflix continues to test ways to respect negative signals without over-correcting and narrowing the catalog too aggressively.
The Role of Personalized Artwork and Top 10
Netflix’s push toward deeper personalization extends beyond ratings. Around the same time the thumbs system arrived, the service began serving different thumbnail images to different accounts for the same title. Later it added the Top 10 row, another personalized list that changes by region and account. These visual and ranking tools now influence what people try more than any single rating button, yet they still rely on the same underlying data the thumbs system feeds.
Competitors took a different route. Amazon Prime Video and Hulu kept star ratings that display average scores from other viewers, giving subscribers a quick sense of broader reception rather than a private prediction. Netflix has chosen to keep its focus on individual taste, which explains why it never restored the old stars even after the 2017 petition gained traction online.
The 2017-era complaints have not vanished; they have simply evolved. Viewers still want more nuance than a binary choice, and many would welcome an option that shows how other people rated a title. Until Netflix decides to blend community data with its personalization engine, the Netflix recommended rows will keep prompting the same debate they sparked eight years ago.
