Movie selection mechanic
Movie selection mechanic
A concept for a new movie selection mechanic that helps users overcome choice paralysis. It first takes mood into account, then narrows the list through friends’ recommendations, and then helps users quickly assess films through a short video feed and tags.
- Brief
- Kinopoisk
- Role
- project leadership, research, interface design
- project leadership
- research
- interface design
- Format
Not an official Kinopoisk product.
Context
The project task was to propose a new movie selection mechanic for Kinopoisk. Streaming services have learned to recommend content, but that has not made choosing easier. On the contrary, the more suitable options there are, the harder it becomes to choose one. As a result, the service does not help the user reach the core task – watching a movie.
Research
Research context
The project began with research: 24 in‑depth interviews with people who pay for entertainment content. Participants were aged 20 to 50. On average, they watched movies about twice a week.
We studied how direct competitors solve this problem – Netflix, Amazon Prime, Apple TV, Okko, and IVI. We also looked at UX mechanics from indirect competitors: Spotify for adapting content to mood, Pinterest for working with tags, TikTok for a fast video feed, and social networks for friends’ recommendations.
All of this later helped us create the new mechanic: mood‑based selection, a social filter, a video feed, and tags became part of the concept.
Persona
The focus was on people who pay for entertainment content. They do not arrive at the service by chance: watching movies is part of their leisure time. They have expectations for content quality, but no desire to spend the evening searching.
Key insights from interviews
A new experience needs support
A person wants to try something new, but without support they step back. They need a recommendation; sometimes it is enough to know that someone else has already tried it first.
People often look for guaranteed emotions
Choosing a film rarely comes down to genre or rating. What matters more is the expectation from watching: people want to get exactly the emotions they came to the service for.
Choice also works as a social scenario
The social scenario is a common selection mechanism. People want to keep up with the films being discussed. The main reason for this scenario is the ability to join the conversation and stay within the current cultural context.
Insight synthesis
Using JTBD, we identified 5 main jobs that users “hire” the service to do. We grouped the interview insights by the four Forces of Progress: what pushes users away from the current choice habit, what pulls them toward the new mechanic, what creates anxiety, and what keeps the existing habit in place.
Hypothesis
The research showed that the choice problem is not only about the number of options, but also about the fact that the service does not take into account a person’s mood and expectations at the moment of searching for and choosing content.
This led to the hypothesis: if the service identifies the user’s mood at the moment of choice, and then shows only relevant options supported by trust, choosing will become easier and more accurate.
Choice scenario map
The choice scenario map helped us break the user journey into steps and see where the new mechanic could create real value, and where it could easily lose the user’s attention.
Two moments turned out to be the most vulnerable.
The first was explaining the new mechanic. Everything new is perceived critically, so it is important to explain the approach: not to promise the impossible and to rely on facts.
The second was the first ten minutes of watching. If the choice does not match the expectation, the user may leave, and trust in the mechanic will decrease.
Product idea
In response to the hypothesis, a new mechanic emerged that works step by step. Each step removes part of the choice load: first, the service identifies the user’s mood, then narrows the choice, adds trust through friends’ experience, and quickly shows the atmosphere of the film.
Mood‑based selection
First, the user shares their current mood with the service. The service identifies films that may fit the request. The concept used a matrix of emotional states – one possible method.
Friends’ recommendations
Next, the social filter is activated: from the suitable content, the system shows what friends have watched or recommended. The choice is already narrowed down, and an element of trust appears.
A film feed instead of posters
After two steps of narrowing the choice, the user receives films in small portions. Each film in the list is a short video clip that makes it possible to quickly read the film’s atmosphere, rhythm, and visuals.
Personal tags
In addition to the short video feed, tags are added on top of the film. They can describe the film itself and the user’s preferences from viewing history.
Tags explain why this film may be a good fit and highlight what matters.
Development and product value
The strength of the new mechanic is that it does not end with one successful recommendation. It can become a way to answer the question “what should I watch right now” and a working tool for content selection.
For the user, this saves effort when choosing, makes it possible to try new content, and helps avoid disappointment when watching.
For Kinopoisk, this creates a new differentiating mechanic and a new approach to solving the problem of choice paralysis. The service gets more data about the user’s state at the moment of search and can select content more accurately for a specific viewing scenario.