Netflix case study — Breaking paradox of choice

I kept on scrolling Netflix for hours yet I was not able to choose which show to watch.

I had to read reviews on other sites to check if the show really stood upto my expectations before watching over Netflix.

Not able to choose suitable content over Netflix is a common issue among users which is more of a psychological problem but might become Netflix’s major problem later on.

Why this happens at the first place and what this case study will be trying to solve is:

1. Paradox of choice
2. Analysis paralysis

Defining “Paradox of choice.”

Though we humans have a very advanced brain, we are still susceptive to the likes of “choice paradox “ when too many options are given to choose from.

If I am still not making any sense to you, here is a study demonstrated in the Columbia University known as “jam study”.

In this study supermarket shoppers were given the option of choosing from a table offering six samples of jam or one displaying 24 varieties.

While more people stopped at the table with 24 choices, only 3 per cent went on to purchase a jar, compared to a third of all shoppers who stopped at the table with just six varieties.

When we are given too many options, it becomes tough for our brain to choose from it, also might end up choosing nothing at all.

Having too many options also lead us to our second problem which is

Analysis paralysis

Analysis paralysis refers to a situation in which an individual or group is unable to move forward with a decision as a result of overanalyzing data or overthinking a problem.

On Netflix when you have 12+ options on your mobile screen that also without any straightforward parameter to compare each other, it takes a lot of active analysis and time to make a decision which then leads to analysis paralysis, resulting into 2 effects :

1. Diminished satisfaction, even if you have made your choice.
2. Suppressed decision making.

Why are these psychological problems even worth of any attention?

90 sec or Bust. “Consumer research suggests that a typical Netflix member loses interest after perhaps 60 to 90 seconds of choosing, having reviewed 10 to 20 titles (perhaps 3 in detail) on one or two screens,” they write. “The user either finds something of interest or the risk of the user abandoning our service increases substantially.”

To tackle 90 sec or Bust, Netflix built Billion dollars of algorithm engine to provide interesting content to its users in minimum time, yet smallest research such as mine speaks differently. According to my research (in detail below) users trust their friends recommendations 3x more than going with what Netflix suggests.

So, faster & longer you engage a user, more you earn profits from them. In this scenario both of these psychological problems directly impact the decision making time of the user hence loosing on money.

Moving towards Solution

Solving the “Choice paradox” is comparatively easy in comparison to the analysis paralysis, we will try to imply best UX laws to solve both of them.

To break and recreate everything, first we need to understand how does the user makes choice of a show/movie in real world scenario.

For a quick survey I created user set of 15 people who were using Netflix on active basis.

I presented them with a structured survey form which would help me construct a broad user journey about “what steps people follow before choosing a show over Netflix”.

I gave them choices based on 2 parameters -

1. Source from where they got to know about content
2. No. of validations before watching; like what all steps they follow to check if that content is suitable for them.

Even though it’s a micro survey, but the responses had higher level of similarities which helped in moving forward with the solution.

As per the responses, I could draw a broad journey which most of the users take :

On the basis of responses :

Users generally assess a new show on 3–4 parameters before finally making a choice, hence they analyse every show on these personal parameters.

Our first action in moving towards solution will be to reduce the upfront choices and then help users analyse a show asap with the parameters they care about most.

Here we will be using our knowledge about Hicks law to reduce the cognitive load on user by reducing the number of choices on the viewport :

How did I came upto this proposed solution :

1. First section is kept as it is because of common user behaviour on how people continue to watch a movie/tv series. User would prefer to carry on a series or movie until it’s finished or have lost interest before jumping to another series/movie.
2. According to the qualitative survey; Friend recommendations is the prime source how users generally go by what to watch hence it could be created as category and will also passively let users to know what their friends are upto.
3. Trending shows is a source of discovery, so if user wants to explore something new, it’s likely that they would want to know what is currently trending.
4. If the user has scrolled this much already which indicate’s his exploring journey isn’t done yet, we can offer him/her genres to explore which was the third important parameter how user chooses which show/movie to watch.

Now solving for the analysis

Users in our survey had few common set of parameters for which they analysed a movie/show before making a choice :

Source :

Friend’s recommendation or Trending over social media

Analysis Parameters :

1. Watch the trailer
2. Look for ratings
3. Genre & cast

Currently Netflix only shows how much the show matches with you (suggested by their algo) which is again a vague parameter for user to analyse and make comparisons.

On the basis of survey, I included the parameters on which users generally make their choice, this is how it turned out.

How does the proposed solution helps in better analysis?

1. You can play the trailer first for the shows which you haven’t started yet.
2. You can know what your friends are watching.
3. You see rating of the movie or how much people like the show.

The screen has reduced the amount of upfront choices and given user option to compare the shows on basis of parameters they care about most.

Following the same approach; I worked on the home page of a movie on Netflix; which again was suffering from the same problem statements.

This is how it turned out.

Takeaways

• Cognitive load by showing too many choices is not benefiting Netflix, it is only increasing the over all decision making time for the user ultimately leading to more losses as per 90 sec or bust survey.
• My survey suggested users prefer to watch shows or movies recommended by their friends 3x more in comparison to the Netflix suggestions.
• No direct parameters to compare the options over netflix is again a factor keeping users jumping from one show to another; leading to drop-offs.

Conclusion

Over the years of technological advancement and options to choose from a larger pool have led users to change their approach on consuming content and making choices, now the users prefer to compare and validate their choices which increases cognitive load at their end leading to the problems which they are not even aware about.

Catching up with these problems and to ultimately solve them might require different approaches and continuous research, maybe region wise, behaviour wise or any other parameter suitable.

As my micro survey was proposing different solutions from the current approach from Netflix has encouraged me to conduct a bigger survey regarding it, which might not only help Netflix but all other companies to build better products for its users.

Sources :

Why Netflix thinks its personalised recommendation engine is worth $1 billion per year

Hick's Law | Laws of UX