Following a record year that saw Netflix add 36 million subscribers in 2020, the world’s most popular video streaming service saw its subscriber growth slow significantly last year. According to the company’s earnings release, Netflix added 8.3 million subscribers in the last three months of 2021, bringing total subscriber growth for the year to 18.2 million. That’s the lowest annual subscriber gain since 2016, when Netflix added roughly the same number of new subscribers. Netflix blamed a “Covid-19 pull-forward” effect, meaning the pandemic accelerated its growth in 2020 while everyone was stuck at home and needed something to watch. Now that surge is taking a toll on the company’s 2021 results.
I thought of validating this data through Netflix’s subscribers, revenue and stock price details. My objective was:
- To identify the region that contributes the most to Subscribers and Revenue.
- To identify the jump in subscribers and revenue due to COVID-19.
- To identify the correlation between Subscribers, Revenue, Stock-price, and library size.
- To understand whether the growth was global or specific to the country (country-wise histogram)
- To predict the stock price by using different predictors (like Subscribers, Revenue, and library size).
Bar Chart Region-Wise Subscribers
Inferences
- Asia-Pacific region has the highest subscriber’s growth rate since 2018 Q1.
- The subscriber growth rate is almost same during 2018 Q2 to 2019 Q1. 2019 Q2 saw a decline in the subscribers’ rate w.r.t 2018.
- In 2020 Q1 (the onset of COVID-19), Netflix saw the biggest jump in its total subscribers. In this quarter, APAC region saw a jump by 22% (highest of all times).
- In 2020 Q2, all other regions except US saw a decline in the subscriber rate.
Bar Chart Region Wise Revenue
Inferences
- Asia-Pacific region saw the biggest jump in revenue for Netflix, followed by Europe, Middle East, and Africa.
- Latin America’s revenue declined by 1% in 2018 Q3 and 2020 Q2.
- There was a significant increase in revenue rate for US and APAC region after COVID-19 onset (2020 – Q2).
- 2019 Q1 was the best quarter in terms of revenue.
Pie Chart Region Wise Subscribers & Revenue
Inferences
- United States and Canada has the largest subscriber base and contribute the highest revenue for Netflix.
- Europe contributes the most in revenue and subscriber after US & Canada.
- The price of a premium Netflix plan in EU countries like France and Switzerland are between 20-25 USD, where are the same plan in US & Canada costs 18 USD.
- The price of a premium Netflix subscription is lowest in Asian countries. The same plan in India is available for 8.6 USD.
- Asia Pacific contributes the least in subscribers count and total revenue.
Scatter Plot Region Wise Subscribers & Revenue
Inferences
- Subscribers’ growth for US & Canada is steady. It has the highest number of subscribers among all the four regions.
- EU region saw a massive increase in subscribers between 2018 Q1 and 2020 Q2. During this time, its subscribers doubled from 29339000 to 61483000.
- APAC almost tripled its subscribers from 7394000 to 22492000 between 2018 Q1 and 2020 Q2.
Inferences
- Revenue took a small dip in 2020 Q1 (COVID-19 Onset) for United States and Canada.
- Revenue for Latin America fell in 2018 Q3 and 2020 Q2 to -1%.
- Europe and Asia-Pacific has steady revenue rate between 2018 Q1 and 2020 Q2.
Box Plot for Average Stock Price, Subscribers and Revenue
Inferences
- The mean of subscribers is 166444357 and median is 162712000 between 2018 Q1 and 2021 Q2. The mean and median are almost close.
- The subscriber box plot is left skewed.
Inferences
- The mean of revenue is 4015526083 and median is 3379570500 between 2018 Q1 and 2020 Q4. The mean is away from median.
- The revenue box plot is right skewed.
Histogram Region Wise Subscribers and Revenue
Inferences
- The range for subscriber is 2801000 and 63345000 for revenue.
- Outliers:
- After the onset of COVID-19 in 2020 Q1, Netflix say a spike in number of subscribers. The outlier subscribers count is 3602000.
- In the next quarter, because of spike in subscribers, the revenue also increased to 85480000 USD.
Descriptive Analysis & Interpretations
Subscribers
| Measure | Value |
|---|---|
| Mean | 37864725 |
| Standard Error | 3296889.678 |
| Median | 32626500 |
| Mode | #N/A |
| Standard Deviation | 20851361.15 |
| Sample Variance | 4.34779E+14 |
| Kurtosis | -1.334803511 |
| Skewness | 0.250790528 |
| Range | 65510000 |
| Minimum | 7394000 |
| Maximum | 72904000 |
| Sum | 1514589000 |
| Count | 40 |
Revenue
| Measure | Value |
|---|---|
| Mean | 1176951500 |
| Standard Error | 130383399 |
| Median | 840051000 |
| Mode | #N/A |
| Standard Deviation | 824617019.7 |
| Sample Variance | 6.79993E+17 |
| Kurtosis | -0.918623238 |
| Skewness | 0.674507999 |
| Range | 2640553000 |
| Minimum | 199117000 |
| Maximum | 2839670000 |
| Sum | 47078060000 |
| Count | 40 |
Correlation between Subscribers and Revenue
| Area | Years | Subscribers | Revenue (in USD) |
|---|---|---|---|
| Asia-Pacific | Q1 - 2018 | 7394000 | 199117000 |
| Asia-Pacific | Q2 - 2018 | 8372000 | 221252000 |
| Asia-Pacific | Q3 - 2018 | 9461000 | 248691000 |
| Asia-Pacific | Q4 - 2018 | 10607000 | 276756000 |
| Asia-Pacific | Q1 - 2019 | 12141000 | 319602000 |
| Asia-Pacific | Q2 - 2019 | 12942000 | 349494000 |
| Asia-Pacific | Q3 - 2019 | 14485000 | 382304000 |
| Asia-Pacific | Q4 - 2019 | 16233000 | 418121000 |
| Asia-Pacific | Q1 - 2020 | 19835000 | 483660000 |
| Asia-Pacific | Q2 - 2020 | 22492000 | 569140000 |
| Correlation | Subscribers | Revenue (in USD) |
|---|---|---|
| Subscribers | 1 | |
| Revenue (in USD) | 0.997322199 | 1 |
Inferences
- The correlation between Subscribers and Revenue is 0.997.
- We can say the association is very strong positive.
- As subscribers increase, revenue also increases.
Correlation between variables
| Average Stock Price per Quarter | Total Subscribers | Average Library size | Total Revenue |
|---|---|---|---|
| 341 | 118.902 | 2000 | 3602 |
| 363 | 124.354 | 2350 | 3814 |
| 299 | 130.422 | 2737 | 3911 |
| 347 | 139.259 | 3229 | 4102 |
| 361 | 148.863 | 3713 | 4440 |
| 313 | 151.562 | 4188 | 4847 |
| 296 | 158.334 | 4631 | 5173 |
| 354 | 167.09 | 5192 | 5399 |
| 426 | 182.856 | 5747 | 5703 |
| 498 | 192.947 | 6210 | 6087 |
Correlation
| Average Stock Price per Quarter | Total Subscribers | Average Library size | Total Revenue | |
|---|---|---|---|---|
| Average Stock Price per Quarter | 1 | |||
| Total Subscribers | 0.682114596 | 1 | ||
| Average Library size | 0.605279262 | 0.993231498 | 1 | |
| Total Revenue | 0.60775371 | 0.98673938 | 0.995272205 | 1 |
Correlation between Subscribers and Mean Stock Price
| Stock Price (Mean) | Subscribers | |
|---|---|---|
| Price (Mean) | 1 | |
| Subscribers | 0.61008787 | 1 |
Inferences
- The correlation between average stock price and subscribers is 0.61.
- We can say the association is strong positive.
- As subscribers increase, the average stock price also increases.
Inferential statistics
Simple Linear Regression
The simple linear regression analysis was done on the data set with the response variable set as revenue and predictor as Subscribers. Thus, the study aims to find the linear regression model for the dependence of revenue on the total subscribers.
- Response Variable: Subscribers
- Predictors: Revenue in Billion
Regression
| Multiple R | 0.990828263 |
|---|---|
| R Square | 0.981740646 |
| Adjusted R Square | 0.980436407 |
| Standard Error | 4.853254644 |
| Observations | 16 |
ANOVA
| df | SS | MS | F | Significance F | |
|---|---|---|---|---|---|
| Regression | 1 | 17729.87065 | 17729.87065 | 752.7303198 | 1.42897E-13 |
| Residual | 14 | 329.7571289 | 23.55408064 | ||
| Total | 15 | 18059.62778 |
| Coefficients | Standard Error | t Stat | P-value | Lower 95% | |
|---|---|---|---|---|---|
| Intercept | 35.75999832 | 5.14334641 | 6.952671563 | 6.7349E-06 | 24.72861741 |
| Revenue in Billion | 24.43620039 | 0.890664152 | 27.43593118 | 1.42897E-13 | 22.52591577 |
Inferences
- The regression model explains 98% association between Revenue and Subscribers.
- The significance F value is very low, P-value is less than alpha (5%), so the model is significant.
Multiple Linear Regression
The multiple regression analysis was done on the data set with response variable set as Average stock price of the quarter and predictors to total subscribers, library size and total revenue. Thus, this study aims to find the linear regression model for the dependence of average stock price of the quarter on the library size, total subscribers, and average revenue.
- Response Variable: Stock Price
- Predictors: Subscribers, Revenue, Library Size
Regression
| Multiple R | 0.938137242 |
|---|---|
| R Square | 0.880101485 |
| Adjusted R Square | 0.820152227 |
| Standard Error | 26.06639393 |
| Observations | 10 |
ANOVA
| df | SS | MS | F | Significance F | |
|---|---|---|---|---|---|
| Regression | 3 | 29924.85864 | 9974.952882 | 14.6807737 | 0.00359661 |
| Residual | 6 | 4076.741355 | 679.4568926 | ||
| Total | 9 | 34001.6 |
| Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | |
|---|---|---|---|---|---|---|
| Intercept | -1351.217782 | 385.6289173 | -3.503932721 | 0.012765162 | -2294.817749 | -407.6178137 |
| Total Subscribers | 15.65098752 | 3.095861837 | 5.055454135 | 0.002321083 | 8.075686505 | 23.22628854 |
| Average Library size | -0.311805853 | 0.087649489 | -3.557417801 | 0.011964389 | -0.526276427 | -0.097335279 |
| Total Revenue | 0.124828062 | 0.105013938 | 1.188680889 | 0.279474257 | -0.132131789 | 0.381787912 |
Inferences
- The regression model explains 82% association between dependence of average stock price of the quarter on the library size, total subscribers, and average revenue.
- The significance F value is less than alpha (5%), so the model is significant.
- The p-Value for all predictors except total revenue is less than alpha (0.05). So, total revenue is not a significant predictor. Other predictors are significant for predicting avg. stock price.
Concluding Remarks
- Between 2018 Q1 and 2020 Q2, EU region doubled its subscribers from 29339000 to 61483000, and APAC almost tripled its subscribers from 7394000 to 22492000.
- United States and Canada are the biggest source of subscribers and revenue for Netflix.
- There is strong correlation between ‘Subscribers and Stock price’ and ‘Subscribers and Revenue’.
- COVID-19 caused a massive jump in subscribers and revenue during 2020, due to which the stock price fell from 549 USD to 500 USD after Q1 2021.
- Library size is a significant predictor for increase stock price, so Netflix should add more movies and shows on its platform.
Note:
All data has been downloaded from Kaggle.com an open-source data platform widely available for studies. Same can be accessed from here - Dataset
If you have any comments, feedback, or requests, please feel free to connect with me on Twitter at @HighOnDataPro. And if you liked this post, don’t forget to share it with your network!