Business Analytics

Individual Assignment

Media Company Case Study

Submitted By Submitted To

MONA SETH Dr. Anup Kumar

202131013
Section E
Problem Statement:
A digital media company (similar to Voot, Hotstar, Netflix, etc.) had launched a show.
Initially, the show got a good response, but then witnessed a decline in viewership. The
company wants to figure out what went wrong.

Interpretation
So basically, we have to find which factor is responsible for decline in viewership.
Type – Predictive

Data
We have been given data for the period of 1 March 2017 to 19 May 2017.
With Columns as
Views show: Number of times the show was viewed
Visitors: Number of visitors who browsed the platform, but not necessarily watched a video.
Views platform: Number of times a video was viewed on the platform
Ad impression: Proxy for marketing budget. Represents number of impressions generated by
ads
Cricket_match_India: If a cricket match was being played. 1 indicates match on a given day,
0 indicates there wasn't
Character_A: Describes presence of Character A. 1 indicates character A was in the episode,
0 indicates she/he wasn't
Solution
We will assess this dependency using SPSS software.
After Removing all the outliers.

Case Processing Summary

Cases
Valid Missing Total
N Percent N Percent N Percent
Views_show 60 100.0% 0 0.0% 60 100.0%

Descriptives
Statistic Std. Error
Views_show Mean 225427.58 9396.220
95% Confidence Interval Lower Bound 206625.79
for Mean Upper Bound 244229.38
5% Trimmed Mean 223665.54
Median 209462.00
Variance 5297336530.2
81
Std. Deviation 72782.804
Minimum 102520
Maximum 379396
Range 276876
Interquartile Range 97235
Skewness .557 .309
Kurtosis -.467 .608
 Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Views_show .119 60 .035 .949 60 .014
a. Lilliefors Significance Correction
Since there are no outliers.
So, we can apply Linear regression analysis.
Regression Analysis

Variables Entered/Removeda
Model Variables Entered Variables Removed Method
1 Character_A, Visitors, . Enter
Cricket_match_india,
Ad_impression,
Views_platformb
a. Dependent Variable: Views_show
b. All requested variables entered.

Dependent Variable is View Shows

Predictor Variable is Character A, Visitors, Cricket Match India, Ad Impression,
Platform Views.
 Model Summary
Std. Error of the
Model R R Square Adjusted R Square Estimate
a
1 .684 .468 .418 55509.732
a. Predictors: (Constant), Character_A, Visitors, Cricket_match_india, Ad_impression,
Views_platform

R square value is around 47 %.

• This shows that predictor variable Character A, Visitors, Cricket Match India,
Ad Impression, Platform Views accounts for 47 % variance in View Shows.

ANOVAa
Model Sum of Squares df Mean Square F Sig.
1 Regression 146151013943.8 5 29230202788.77 9.486 .000b
88 8
Residual 166391841342.6 54 3081330395.235
96
Total 312542855286.5 59
83
a. Dependent Variable: Views_show
b. Predictors: (Constant), Character_A, Visitors, Cricket_match_india, Ad_impression,
Views_platform
Since F> 3 so this is case of multiple regression.
The Overall Regression model was significant F(5,54) = 9.486
p < 0.001
R square is 47%.
 Coefficientsa
Standardize
Unstandardized d
Coefficients Coefficients
Model B Std. Error Beta t Sig.
1 (Constant) 40395.934 89003.626 .454 .652
Visitors .237 .138 .675 1.722 .091
Views_platform -.224 .127 -.741 -1.765 .083
Ad_impression .000 .000 .589 4.145 .000
Cricket_match_in -11549.086 23717.731 -.051 -.487 .628
dia
Character_A 43449.119 28439.524 .276 1.528 .132
a. Dependent Variable: Views_show

Here

Visitors Not Sig. (p = 0.091)

View Platform Not Sig (p = 0.083)
Ad impression Sig (p < 0.001)
Cricket Match Not Sig (p = 0.628)
Character A Not Sig (p=0.132)

Conclusion

So only Ad impression comes out to be significant factor behind decline in number of views
of show.

Final Equation

Y = 40395.934 + (0.237 * Visitors) +( (-0.224) * View Platform) + (0.001 * Ad

impressions) + ((-11549.086) * Cricket Match India) + (43449.119 * Character_A)