Reel Talk

What is Reel Talk?

Reel Talk is an online movie/TV-show community that is a combination of IMDB and Reddit, for movie/TV enthusiasts.

It provides a platform for users to:

1. Read movie/TV-show listings, ratings, and reviews
2. Discuss their favorite content

Features highlights

Onboarding

We use the onboarding process to understand users' content preferences.

Community

Users can join a community that matches their interests.

Discussion

Users can share content relevant to the community's topic to spark discussions.

Final deliverables

Established design system

Initial exploration

The founder of Reel Talk, who is a Movie/TV-show enthusiast himself, found that the existing platforms such as IMDB or Rotten Tomato cannot fully satisfy his needs. He believes that there is a gap in the market for a platform that caters to the needs of movie/TV-show enthusiasts. To validate his hypothesis and uncover specific pain points, surveys were conducted.

Key insights

The respondents primary needs on the existing platforms are:

1. Read ratings and reviews
2. Discuss the content they’ve watched

For their discussion needs,

Youtube and Reddit are their go-to platforms.

Since they're general-purpose platforms, they also express lack of confidence in existing communities (since their are pro users), and want more features dedicated to content discussions.

Based on the survey results, our team concluded that:

The movie/TV-show enthusiasts only have general-purpose platforms to discuss content, and there are no platforms that cater to their specific needs.

This insight presents an opportunity to create a platform tailored to their needs.

However, before moving forward, we need a deeper understanding of the pain points they have on current platforms and their specific expectations for a new one.

Prospective user interviews

"How are your experience on movie- & TV-show-oriented online connections and discussions on the existing platforms?"

Key pain points from interviews

Movie/TV show enthusiasts struggle to find:

Peers who share their passion

The group of movie and TV-show enthusiasts is relatively smaller compared to general viewers, and current social platforms lack effective ways for them to identify and connect with one another. They want to talk to people who are also deeply interested in the nuances and details of a movie or TV show.

In-depth discussions

On current social platforms, most conversations stem from reviews and take place in the comment sections. It does not facilitate in-depth discussions effectively. Some interviewees also expressed a desire for specialized features, such as the ability to embed custom lists and movie/TV show listings directly into conversations.

Understanding a user's taste: Key indicators to consider

Through brainstorming and interviews, we identified few potential ways to find out user's taste.

With a clear understanding of what indicate a user's taste, I shifted focus to exploring effective ways to gather this information.

Gathering taste information on the platform

Through brainstorming, I developed three integration plans.

From user's watch history and ratings

It leverages user's viewing record and ratings to understand preferences.

Answer few questions during the onboarding process

Users select their interests and preferences when first signing up.

Feed-side quick surveys

The platform would presents quick preference questions while users browse content.

Feedback

We ask our prospective users to evaluate and score the three approaches with prototypes.

We also carried out internal design reviews and feasibility assessments for the three approaches.

After considering all factors, including user scores and internal evaluations, we believe approach #2 is the clear winner.
Regarding the cons listed from internal evaluation, we believe that with thoughtful design, we can make the process more enjoyable, ultimately reducing the drop-off rate.

Validation: Usability testing and follow-up interviews

After finalizing the direction with approach #2, I proceeded to design the high-fidelity version and conducted a usability test.

Test setup

• Task: get an account on the Reel Talk platform
• 5 Movie/TV-show enthusiasts identified by how frequently they watch content
• 45 mins remote sessions
• Thematic analysis

Findings & Iteration

1. Preference questions

Positive: Our users love these questions!

Prior to testing, there were concerns about the onboarding's length. However, none of the tested users raised this issue, indicating their strong interest in preference questions.

Negative: Users struggle with decision-making

Some users find it challenging to identify their top movies/TV shows when asked because they can't decide immediately.

Iteration

I changed the word used in the title from "top" to "favorite." This removes the need for users to rank their choices.

I also adjusted the quantity restriction from a mandatory 5 choices to "up to 10," allowing greater flexibility.

Additionally, it also includes more options to choose from (trending content)

My decision-making process

1. Wording

I deliberated over the wording and recognized that using "top" might burden users with the task of mentally ranking items, thus increasing cognitive load. I then brainstormed several alternatives such as "Preferred," "Beloved," and "Highly rated." After team discussions, we concluded that "Favorite" best suited our needs in this context.

2. Quantity constraints

The adjustment from "top 5" to "up to 10" adeptly accommodates two scenarios:

1. If users can only think of a few items (less than 5)
2. If users have more items in mind than the previous limit allowed (more than 5)

2. Birth date -> Age range

Through the validation process, we found that users had privacy concerns to provide birth date, even though I made some efforts to explain why these inputs were necessary and tried to make the process more enjoyable (see on the right).

I then decided to address these privacy concerns by updating the interface from requiring users to enter their birthday to simply selecting their age range (e.g., Under 18, 18–29, 30–39, etc.).

My decision-making process

I explored 2 additional solutions to address the privacy concerns:

1. Instead of asking for the full date of birth, I considered only requesting for the birth year.
However, this approach was rejected because it's uncertain if users are over 18 based solely on their birth year.

2. Another option was simply selecting their age.
This idea was also rejected because users would face an overwhelming number of choices.

Final design

For our users, what constitutes a good environment for discussions?

I distilled insights from our research with prospective users and developed three key design principles.

Effortless connections with like-minded peers

With the information users provide, we can create an environment that makes it easy for them to connect with one another.

Format that facilitate in-depth discussions

Users should be able to engage in meaningful and in-depth discussions with ease, fostering an environment that encourages thoughtful interaction and exchange of ideas.

Control

Users have the option to choose whom they want to talk to.

Discussion formats

Following these principles of good discussion, I developed three distinct formats for conducting conversations.

Episode- or movie-specific chat rooms

Provide chat rooms dedicated to each movie, TV show, or specific episode. Users can join in on discussions with others who've watched the same content.

Community with topic-based discussion posts

Set up a community where users can post about specific themes or genres (e.g., "Top Psychological Thrillers" or "Everything Batman"). Users can engage by commenting or reacting to posts and tagging specific movies or shows.

"Discussion circles"

Enable users to create public or private groups where they can chat about a specific show or movie together. These groups can be small or large. The environment allows users to share thoughts in a more intimate, real-time environment.

Feedback

We ask our prospective users to evaluate and score the three approaches with prototypes.

We also carried out internal design reviews based on the 3 design principles.

After weighting pros and cons, I decided to implement the second approach to build a Reddit-like community where users can post and discuss topics they are interested in.

From there, I moved on to designing the high-fidelity version of this solution.

High-fidelity designs

Validation: Usability testing and follow-up interviews

We carried out the same research as previously mentioned.

Findings & Iteration

1. Individual community

Positive: Clarity

Users value the convenience of accessing linked works to the community, enabling them to promptly understand the topics discussed within.

Negative: Need additional methods to understand the applicant

When users create a private community instead of a public one, they become even more selective about who they want to include in this community. Consequently, they seek to learn more about the applicant beyond the two customizable questions.

Iteration

The community owner can enable options in settings to receive additional information about the applicants from the platform.

These options include:

1. Has he/she watched any work by [actor/director]?
2. Has he/she watched any work linked to the community?
3. Does his/her favorite genres includes [genre]?
4. Does his/her favorite works includes [work]?
5. How is his/her activity level?

When reviewing applications, the system will show how many additional criteria are met and the owners can choose to look into the details if they

Design process

We were interested in understanding what information a community owner seeks about an applicant, so we conducted a brief survey.

The responses fall into two broad categories:
1. Applicant’s knowledge of the topic being discussed
e.g. "How much do you know about Disney animations?"

2. Applicant's level of participation
e.g. “Is this person going to actively participate in the discussions or lurk around till the end of time?”

I then explored various methods for the community owner to obtain this information.

Proposal 1: Customization

Give community owner more ways to formulate questions.

Drawback: Community owners are unable to verify their answers.

Proposal 2: Auto rejection

Community owner can setup some criteria and the system auto-rejects applicants who did not meet such criteria.

Drawbacks: Potential to inadvertently auto-reject preferred applicants.

Given the drawbacks outlined above, we chose to make the additional information optional in the settings.

2. Adding three more ways to discover different communities

When users engage with a platform to discuss movies/TV shows, they anticipate not only discussing familiar content but also discovering new options to watch.

We first asked the engineers to adjust the algorithm to provide more diverse recommendations, rather than sticking to the safe zone.

I also overviewduced two additional methods for users to discover communities that match their interests:

1. Trending communities
2. Where their friends are

My decision-making process

I asked ourselves the following question:

How might we help users discover new content that they are potentially interested in?

I then considered the following reasons for adding the two sections mentioned above:

1. As the platform caters to movie/TV enthusiasts, trending communities on the platform are somewhat likely to align with user interests.
2. Users' friends are likely to share similar movie tastes and often belong to communities that the user would also find intriguing.

Final design