I believe Netflix needs social features — and users agree!
I researched Netflix and its features to understand user requirements/pain points and developed a cool new feature called “Recommendations from friends”. From a strategic point of view, I researched Netflix’s previous attempts at developing a social strategy and analyzed why this new feature is a differentiated and well-timed approach to Netflix’s social strategy.
Table of Contents 📝
· Why do this??
· Company Breakdown
· Conducting User Research
· Survey results — Discovery of pain points
· User Persona
· Recommendations from friends
a. Introduction
b. How it works
c. How this feature helps Netflix and its users
· Roadmap — Way forward?
Why do this? 📌
With this project, I challenged myself to learn the different aspects of feature development. Initially, my goal was learning technical aspects such as wireframing, design and refreshed my survey design skills. My intent was learning to ask ‘why’ at every step and learn to uncover real user needs hiding behind user feedback.
What is Netflix?
Netflix’s mission statement
“Our core strategy is to grow our streaming subscription business domestically and globally. We are continuously improving the customer experience, with a focus on expanding our streaming content, enhancing our user interface and extending our streaming service to even more Internet-connected devices, while staying within the parameters of our consolidated net income and operating segment contribution profit targets.”
Vision Statement
Becoming the best global entertainment distribution service. Licensing entertainment content around the world, creating markets that are accessible to filmmakers and helping content creators around the world to find a global audience.’ The brand promise is a quest;
Values
Judgment, Communication, Curiosity, Courage, Passion, Selflessness, Innovation, Inclusion, Integrity & Impact.
Netflix by numbers (Q4 2018) 📛
Number of paying Netflix streaming subscribers (Worldwide)–139 Million
Number of paying Netflix streaming subscribers (US)–58.5 Million
Annual revenue (Worldwide)–$4 Billion
Annual revenue (US) $2 Billion
Daily content consumption on Netflix (US)- 100 Million hours
Number of new user sign-ups in 2018 (Worldwide)–29 Million
Number of new user sign-ups in 2018 (United States)–1.5 Million
We can find updated numbers at https://s22.q4cdn.com/959853165/files/doc_financials/quarterly_reports/2018/q4/FINAL-Q418-Shareholder-Letter.pdf
Conducting User Research and its goals 📄
I created a short, multiple choice, online user survey with a goal of discovering the likes and dislikes of Netflix users, user pain points, the relative importance of Netflix features and to gather supporting data. I reached out to a lot of Netflix users via Facebook (forums such as Women in Product), Netflix community on Reddit etc. and got 90 responses.
I gathered the demographic data on users through Netflix reports. Based on this, I focused on analyzing if my survey results resonated with the popular user opinion on social platforms such as Reddit, Quora etc.
What I learned from the survey
This survey gave me an insight into implicit/hidden user needs and helped me discover the underlying reasons behind the feedback. It also served as a starting point to develop a solution which solves user problems and delights the customer.
Survey results — User preferences for a streaming service 📊
What is important for the users — (Valued attributes)
A Quick look at the numbers
As a response to a question — For a streaming service, how important is the availability of recommendations of content on the digital platform? Please rate your choice from 0 to 10 (lowest to highest).
Out of the 90 respondents, most users (80%) said the importance of the availability of recommendations of content on the Streaming interface ranged from medium to high.
When asked–For a streaming service, how important is the ease of finding content on the digital platform? Please rate your choice from 0 to 10 (lowest to highest).
Out of the 90 respondents, 55% consider the ease of finding relevant content to be highly important for a streaming service.
Interestingly, the user base is divided with the preferred source of recommendations. Most of them either prefer to get recommendations on what to watch next from their friends and family (33%) or from streaming service itself (32%). Rest of the users said they relied on other sources (internet, advertisements etc.)
Discovery of User Pain points 🔺
User data about Netflix attributes helped me identify two main user pain points:
🔴 USER Pain Point #1: Netflix users are very unsatisfied with ease of finding relevant content to watch.
When asked–How do you rate the ‘ease of finding recommendations’ on Netflix on a scale to 1 to 10 (lowest to highest) a meagre 16.5% of users rated it as excellent. Over 45% of users rated it ranging from poor to okay.
Open-ended user feedback also echoed the survey findings. Many users mentioned areas of improvement for Netflix to be “Ease of discoverability”, “Better search” etc.
One user said “If I don’t already know what I want to watch it can be very challenging to find something. Also, don’t find their recommendation engine to be accurate.”
Another one stated, “Main reason I haven’t cut the cord and given up on cable is that I love the guide screen, to search what’s on… with Netflix sometimes, it’s very difficult for me to decide what to watch.”
🔴 USER Pain Point #2: Majority of Netflix users are not satisfied with the quality of Netflix recommendations.
When asked–How do you rate the ‘Quality of Netflix Content Recommendations’ on a scale to 1 to 10 (lowest to highest), most users are unsatisfied with the recommendations provided. Only 19.8% of users believe the quality of Netflix recommendations to be excellent. Over 55% of users rated it ranging from ‘poor to okay’.
Once again, open-ended feedback is in line with the consensus.
In one user’s words: “The recommendations I’m shown are usually not relevant to my interests at all. I miss the old Netflix algorithm.”
Another user says “Select types of shows you enjoy allowing for curated recommendations that fit all your interests. Ex: I recently started watching a Christmas prince last week then almost all of my recommended shows and movies have been Christmas related. Recommendations should represent all of a user’s interests not just the most recent movie binge they’ve had that week.”
Another source of discovering pain points is the Netflix Reddit community which has over 325 K subscribers.
An excerpt taken from a popular comment (upvoted 567 times) says “Surely enough people complain about categories, searching, and recommendations that Netflix is at least aware, right? Recommendations are just the same titles with different ‘categories’. If I want to re-watch something, I’ll look in history or search for it.”
Netflix’s algorithms (Fact store and Near Real-time) famously use the spark computation and have successfully established Netflix as a highly personalized platform. These algorithms consider numerous data points including watch history translated into ‘user tastes.’ It gives recommendations based on viewing patterns, but users often skip around (a lot) to find an interesting show. Multiple viewers using the same account also makes it very difficult for the algorithm to catch up with ever-changing viewer preferences. These two user pain points show that finding relevant content is the one area where happiness among Netflix members isn’t great!
But Netflix is great, Overall! 🏆
Netflix users are happy about other attributes such as the price of subscription, content quality, ability to keep track of watched content, interruption-free viewing etc which reflects in great overall satisfaction rating of Netflix.
What do these pain points mean? 👀
I analyzed the pain points along with the user’s preferred sources of recommendations and I hypothesized that
“In absence of relevant recommendations, users gravitate towards other sources of recommendations i.e. friends and family since it is their equally preferred source of recommendations.” (Survey result #3)
A user’s feedback about Netflix’s scope of improvements confirmed this hypothesis — “I want to add a social aspect. What my friends are watching.”
Relating it to my experience, I often ask my friends what they are watching and even if the recommendations do not end up being my favourites, often, it is something I like and watch. Once again, I returned to Reddit Netflix community and found multiple threads with show suggestions ex. — ‘what to watch on Netflix (HD)’ with more than 3k upvotes and 185 responses.
Drawing from the pain points and user preferences I confirmed my hypothesis and concluded:
Netflix users find it difficult to find relevant recommendations and seek recommendations from other sources with similar interests.
👦User Persona
I brainstormed for a solution to these pain points, but I wanted to define my group of users with help of a user persona.
Here is James 👋
31-year-old, single, works for a technology company, 40 plus hour work week combined with a moderately active social life and a good group of friends & family.
He loves to binge on Netflix shows but doesn’t have time to discover shows on his own.
Scenario
James has been working long hours past three weeks and is quite excited for the upcoming long weekend! He is planning a staycation, order in his favourite food and relax while watching his favourite shows. There is only one tiny problem, he finished watching the latest season of stranger things last week. He browsed through the Netflix recommendations but found he has already watched the popular binge-worthy shows! He does not want to spend any more time browsing and wants to find a show he knows he will enjoy and can watch for three days.
His solution — James asks his friend and family members to send him recommendations on Netflix with the help of ‘Recommendations from friends’.
What is ‘Recommendations from friends’ you ask?
💡 Recommendations from friends –Introduction
After discovering the pain points with the survey, I brainstormed a better way to discover content other than Netflix recommendations. My aim was to find a solution which will help the user to find recommendations which are
In line with their interests | easy to find | from trusted sources.
I figured, what better way to find relevant recommendations than to ask the other users, like asking a friend! Well, in this case, we help the users connect to other users to find relevant recommendations.
Let me explain!
Recommendations from Friends — A new feature
Recommendations from friends is a new feature which will enable users to send and receive recommendations from other Netflix users using unique email id’s linked to their personal Netflix profiles. This new feature combines the user preference for finding content from trusted resources and increases the ease of finding relevant content. It also allows users to share recommendations one on one while maintaining the privacy of their watch history.
How it works 🎯
Once you recommend a title to your contact by clicking on recommend button, your contact will receive an email and a notification in the Netflix app will inform them that you recommended a show you think they might like. When you put in the friend’s email Id and press the ‘Recommend’ button, Netflix will provide you with an option to save the email id for future use.
This feature will be available as a clickable link alongside the Home, TV Shows, Movies recently added etc. When clicked, the link will display a list of shows which your friends have recommended to you in a similar way your watch list and Netflix recommendations are displayed.
📑How to receive and send recommendations
Steps to get a recommendation
Activate your recommendation settings by adding an email id unique to your profile in your profile settings under account settings. Share the email id with your friends to receive recommendations. Here is a pictorial step by step:
- Go to Account settings, under my profile settings add an email by updating the ‘Add email for recommendations’ link at the bottom.
2. You can also update your email by the same link.
If you receive a recommendation from a friend, next time you log in, Netflix will ask you if you wish to receive a recommendation from the friend’s id (ex. abc@gmail.com). This security measure will prevent unknown sources from sending you recommendations on Netflix. Unless you allow, it will not display the recommendation in your list. If you do not take any action, you will receive a reminder email for after a few days, so you can add the person to your list if you wish.
3. When you click on the link ‘Friends’ Recommend ’Here is how your recommendations will show up!
📑 Steps to send a recommendation
Watching a series, you think your friends will enjoy? Recommend it to your friends!
You can do it in the following two ways.
From the home screen of a series:
You can press the ‘Recommend to Friends’. Here, you will it will redirect you to the series info tab.
Alternatively, you can also go to the series info and click on the ‘Recommend’ tab (Highlighted in red in the image).
- Next, you will see a recommendation screen with a search column.
2. You can either search for a previously used ID or add a new id. As you type, previously used email ids will show up in the drop-down. If you are sending a recommendation for the first time, type the complete id.
When you are sending a recommendation to a friend for the first time, the following prompt will appear
If you enter a previously used email id and press recommend, you will find the following screen.
After your recommendation is sent, your friend will receive the following email!
But Netflix had a community feature back in 2004… Wait…What?
After coming up with this feature and developing wireframes, I found out that a few years back, Netflix developed a feature called ‘Netflix Social’.
Netflix has been playing with adding a Social aspect to their platform from a long time. Netflix’s attempts at adding a social feature helped me further validate my survey findings, but I dug deeper into why Netflix scrapped the feature. My research showed that the past trials at adding a social aspect were run down by a greater need for user data privacy. Deloitte 2018 Digital Media Trends Survey also highlights the same trend. The survey stated that 69% of streaming services users said companies should provide better data privacy and 93% expressed a need to be able to delete their online data.
Clearly, users do not want to display their complete watch list but will share selected recommendations with select individuals. Although users want a social aspect, they wish to have a substantial amount of control over the sharing of their viewing habits or history.
Why users still need this feature: ☑️
· Netflix users find it difficult to find relevant Netflix recommendations! Getting recommendations from trusted sources keeps viewers engaged and helps them discover awesome content without spending too much time and energy.
· Netflix’s social aspect is currently limited to watching Netflix as a group. The addition of this feature does not threaten user privacy. A user can control the visibility of his/her watch history and can connect with other users without worrying about sharing their information to a third-party social network.
· According to research by Limelight video, In the US, smartphones are narrowly preferred over computers and TV as the primary viewing device. As of now, Netflix limits content sharing to mobile devices via texts, email and messaging apps. This feature will encourage users to stay connected and share content without being limited to one type of device.
Why this feature will work for Netflix now (while it hasn’t in the past) 🚀
· It will increase user engagement and valued hours but not at the cost of developing a social network or piggybacking on an existing one. It will be an add-on to feature to the current recommendation algorithm, which can use this network effect data for fine-tuning the “taste communities”.
· This feature does not equate one on one connect with a social network. Users do not have to worry about taking an extra step (as in the previous feature) to avoid sharing a specific title. Users also have the liberty to send/receive multiple recommendations privately which isn’t possible with social network sharing.
· Although it might bring Netflix back in the ‘social’ arena, it ensures that user controls the privacy of their playlist as the ‘Friends recommend’ feature doesn’t display names of the viewers for a title. Recommendations from friends are sent on email. In case the user missed the email, he/she will still get a quality recommendation on the interface. It also allows users to limit receiving the recommendations by providing the associated email id only to friends’ they wish to receive recommendations from and not from a generic contact list (ex. Facebook friend’s list).
· The most important component of this feature is undoubtedly the data generated from the in-network effect that can help provide more personalized and efficient (maximum happiness per dollar) recommendations. It generates quality data for Netflix’s NRT recommendations with Spark Streaming which is instrumental in promoting new launches, trending and popular content.
🔰 Metrics to consider 🔰
To Measure the impact of this feature, I considered two Key metrics:
User engagement–Showed by hours served per day–User receiving relevant recommendations would spend more hours watching the content. This increase in hours served per day will translate to User Engagement. Increased user engagement will also create a network effect.
Growth in subscriber base–Showed by newly gained paid memberships- According to the Deloitte 2018 Digital Media Trends Survey, 54% of streaming video subscribers cited original content as a reason for signing up for a paid streaming service. Netflix original TV shows and films picked up 5 Golden Globe wins in 2019 and next year they plan to launch many new original titles and returning seasons of epic shows such as stranger things!
This feature will help the ‘original content’ (which we know is a driver for paid memberships) to reach the intended audience and it will incline trial users to move to paid memberships.
In terms of numbers, assuming implementing this feature positively impacts Hours served per day and Free to paid conversions we can draw a rough estimate on key numbers.
If we assume, only 10% of Netflix users in the US are using this feature, we can estimate (with varying conversion rates.)
· the increase in hours served per day 🕑
· the revenue generated by gaining new paid memberships 💰 💰
Future vision 🔮
We can expand this feature to include the following:
· An option to turn on notifications specifically for the recommendations received from friends by adding an option with slider button to the account settings.
· Blocking or removing a specific email/friend: This will keep the user account secure by helping the user to block or remove an email id/friend when a user no longer wishes to receive a recommendation from them.
With that, I conclude my Netflix Social features– ‘To be or not to be’ Case study! 😃
Thank you so much for reading! Your comments & feedback are highly welcome.
If you enjoyed reading this, please let me know with a clap! 👏
P.S. — I am a huge Netflix fan and am very passionate about researching and analyzing the products I love. I do not work for/represent Netflix and the ideas, research and views for this case study are strictly my own. Netflix pictorials are used for research purposes only. This case study helped me understand why an idea worked or didn’t and the impact of the timing of a feature launch on its performance. I am a tech enthusiast trying to learn and this case study was an effort in the same direction. Again, thank you, everyone, who gave feedback and helped me with this project.
📝 Read this story later in Journal.
🗞 Wake up every Sunday morning to the week’s most noteworthy Tech stories, opinions, and news waiting in your inbox: Get the noteworthy newsletter >
