Data Analysis Of Book Sales In Academic Online Book Store

Profit Analysis

Books are the good friends of human. In the study life we used many books for our degree, certificate course, diploma etc. In the past, book purchase was the tedious work. It takes so much your valuable time. But today, you can get the desired book at your place within the stipulated time. This all is possible due to eCommerce. Online shopping is one of the major form of eCommerce.

Today we heard the terms like amazon, Flipkart, eBay etc. This all are online shopping firm. In the online store, Book section is very developed. You can get the all primary information from the website itself. In the books, academic books for different subjects are available.

Online shopping is increasing exponentially in recent decade bring new challenges to the service provider. Business competition and customer satisfaction are the most important factors in the eCommerce business.

About Data:

We have data regarding the sale of books (2760 books) in the month from Academic Online Book Store. We considered the following attributes

1. Book Name
2. Book Price (in $)

• Book Sale Price (in $)

1. Profit (in $)
2. Number of customers who bought the particular book
3. Shipping Type (Free or Paid)

• Customer Type (New or Existing)
• Geographical Region (NSW, NT, QSD, SA, TAS, VIC, WA)

1. Book Category (Engineering & Transportation, Arts & Photography, Education & Teaching, Computers & Technology, Medical Books, Science & Math)

We used Total Monthly sale amount (in $) and Total monthly profit (in $) variables for the study objectives which is defined as

Total Monthly sale amount (in $) = Book Sale Price (in $) × Number of customers

Total monthly profit (in $) = Profit (in $) × Number of customers

Project Problem:

• We studied the profit for different attributes (shipping type, customer type, geographical region and book category).
• We test whether the mean number of customers for different levels of shipping type, customer type, geographical region and book category are significantly different or not.
• We carried correlation analysis for different variables like product price, profit, sale price and number of customers.
• We develop the predictive model of total monthly sale using regression analysis. 

Data analysis without statistical tool and techniques is considered to be incomplete. There is vast literature about statistical tools and techniques. Selection of proper tools and techniques is the important aspect of analysis.

We calculated the profit percentage using total monthly sale amount (in $) and total monthly profit (in $) for shipping type, customer type, region and category. We have given summary statistics for number of customers for shipping type, customer type, region, and category.

We test the mean number of customers for different levels of attributes by two sample t test and one way ANOVA.  We studied the correlation between product price, profit, sale price, number of pages and number of customers. We develop the predictive model for total sale amount by using regression analysis. We used Python 3.6.5 IDLE and MS-Excel for the data analysis. The sample code are given in appendixes.  We used Grus (2015), McKinney (2012), and Pedregosa et al. (2011).

Profit Analysis:

We calculate the profit percentage by dividing total monthly sale amount (in $) by total monthly profit (in $). Table 1 shows the total monthly sale amount (in $) by total monthly profit (in $) and profit percentage for shipping type, customer type, region and category. We referred Berenson et al. (2012), Black (2009) and Mendenhall and Sincich (1993). 

Summary Statistics for Customer Numbers

From Table 1 we observed that

• Academic Online Book Store earns on average 5.69% profit on each book.
• Books which shipped freely gives more profit than books which shipped by customer payment.
• Existing customers gives more profit than new customers.
• There is very little difference in the profit from the region.
• Books from Computers & Technology category gives more profit than other. 

Table 1: Profit analysis according to for shipping type, customer type, region and category

Attributes	Levels	Total Monthly Sale (in $)	Total Monthly  Profit (in $)	Profit Percentage
Shipping Type	Free	74054.30	4224.21	5.70%
Paid	139298.71	7922.91	5.69%
Customer Type	Existing	83962.14	4791.55	5.71%
New	129390.87	7355.57	5.68%
Region	NSW	31871.88	1803.08	5.66%
NT	18002.11	1034.34	5.75%
QSD	35164.01	2003.66	5.70%
SA	31997.10	1828.95	5.72%
TAS	30610.50	1747.77	5.71%
VIC	31900.75	1822.79	5.71%
WA	33806.66	1906.53	5.64%
Book Category	Arts & Photography	40689.62	2304.26	5.66%
Computers & Technology	50152.30	3144.80	6.27%
Education & Teaching	11344.72	612.85	5.40%
Engineering & Transportation	16113.68	845.13	5.24%
Medical Books	74286.63	4083.62	5.50%
Science & Math	20766.06	1156.46	5.57%
Total		213353.01	12147.12	5.69%

Customer is pillar of any business. If customers attracted towards your products, sale and profit will increases automatically.  In the Table 2 we represents the descriptive statistics including size, mean, standard deviation, minimum and maximum for number of customers. We used the well-known books for this section such as Casella and Berger (2002), DeGroot and Schervish (2012), Hodges Jr and Lehmann (2005), Pillers (2002) and Ross (2014). 

Table 2: Summary statistics for numbers of customer who bought the books for shipping type, customer type, region and category

Attributes	Levels	Size	Mean	SD	Min	Max
Shipping Type	Free	822	5.00	2.17	1	11
Paid	1938	4.00	2.01	1	9
Customer Type	Existing	1094	4.30	2.14	1	11
New	1666	4.30	2.09	1	11
Region	NSW	403	4.32	2.04	1	11
NT	284	3.62	2.08	1	9
QSD	428	4.53	2.16	1	10
SA	411	4.31	2.11	1	9
TAS	407	4.23	2.05	1	10
VIC	399	4.47	2.12	1	9
WA	428	4.38	2.08	1	9
Book Category	Arts & Photography	541	4.15	2.02	1	9
Computers & Technology	654	4.19	2.17	1	9
Education & Teaching	142	4.49	2.02	1	9
Engineering & Transportation	216	4.28	2.08	1	9
Medical Books	986	4.22	2.05	1	9
Science & Math	221	5.22	2.26	1	11
Total

We can observed following from Table 2:

1. Averagely Academic Online book store get 4.3 customers for each book with standard deviation 2.11.
2. Mean number of customers who bought the books at free shipping is 5 whereas mean number of customers who bought books by paid shipping is 4.

• Mean number of new and existing customers is same.

1. In NT region, mean number of customers is less than other region.
2. For Science & Math, mean number of customers is more than mean number of customers demanding other category books.

Two Sample t-test:

We used two sample t test for testing the significant difference between mean number of customers for

1. Free Shipping and Paid Shipping
2. New Customers and Existing Customers

In Table 3, we represent the test statistic and p-value of two sample independent test assuming unequal variances.

Table 3: Two sample independent test for shipping type and customer type

Attributes	Levels	Test Statistic	p-value
Shipping Type	Free and Paid	11.35	0.000
Customer Type	New and Existing	0.04	0.965

We observed the following from Table 3:

1. There is significant difference in mean number of customers who bought the books at free shipping and who bought at paid shipping.
2. There is no significant difference in mean number of new customers and existing customers who bought the books.

One way ANOVA:

We used one way ANOVA for testing the significant difference between mean number of customers for

1. Geographical Region (NSW, NT, QSD, SA, TAS, VIC, WA)
2. Book Category (Engineering & Transportation, Arts & Photography, Education & Teaching, Computers & Technology, Medical Books, Science & Math)

Table 4 shows the value of F statistic and p-value for one way ANOVA.

Table 4: Output of one way ANOVA for Category

Attributes	Level	F Statistic	P Value
Geographical Region	NSW, NT, QSD, SA, TAS, VIC, WA	6.54	0.000
Book Category	Engineering & Transportation, Arts & Photography, Education & Teaching, Computers & Technology, Medical Books, Science & Math	10.01	0.000

 From Table 4 we observed the following

1. There is significant difference between the mean of number of customer who bought the books from different geographical region. Mean number of customers from NT region is less than other region.
2. There is significant difference between the mean of number of customer who bought the books of different category. Mean number of customers who bought the Science & Math category book is significantly more than other category.

Correlation Analysis:

Table 5 represents the correlation coefficient between book price, Sale price, profit and number of customers. Correlation coefficient tells us the relation between variables.

Table 5: Pearson’s correlation coefficient for Book Price, Sale Price, Profit and Numbers of customers

	Book Price	Sale Price	Profit	Numbers of customer
Book Price	1
Sale Price	0.999	1
Profit	0.997	0.979	1
Numbers of customer	0.018	0.019	0.020	1

From Table 5, we observed that

1. Book price is positively correlated with sale price (strong correlation), profit (strong correlation) and number of customers (weak correlation).
2. Sale price is positively related with profit (strong correlation) and number of customers (weak correlation).

• Profit is also positively correlated with number of customers (weak correlation). 

We develop the predictive model of total monthly sale using regression analysis. We develop the model for total monthly sale by using number of customers as a predictor.  We used simple linear regression model. Table 6 represents the F Statistics, P value, R² and regression coefficients of simple linear regression.

Table 7: Output of Regression Analysis

F Statistic	3655.046
P Value	0.000
R2	0.57
Intercept	0.084
Slope	17.971

We observed that P Value =0.000 suggests that there is significant relationship between total monthly sale and number of customers who bought the books. We fitted the following straight line as

Total sale (in $) = 0.084 + 17.971 × Number of Customers

If for particular book if we got 10 customers then total sale (in $) is 0.084 + 17.971 × 10.

Recommendations to the company

From the data analysis, we observed that

1. As the mean number of customers for free shipping books is significantly more than paid delivery suggest that we should give free delivery to most of the books so that total sale will increases.
2. In NT region, mean number of customers is less than other regions suggest that we should use some more marketing strategies in NT region.

• Science & Math books are highly demanded than others so we should provide desired book to each needy one.

An implementation plan based on the recommendations you have provided

1. We should appoint new staff in shipping department so that free shipping will be possible for most of the books which result in high total sale and profit.
2. We should adopt the new innovative marketing strategies in NT region to attract the customers. We can use more advertising hoardings in NT region.

• We can give some offer or facility for NT region.     

1. We should advertise each Science & Math book.
2. We should keep sufficient stock of Science & Math Book.

Conclusions

We observed that Academic Online Book Store earns on average 5.69% profit on each book. Books which shipped freely gives more profit than books which shipped by customer payment. Existing customers gives more profit than new customers. There is very little difference in the profit from the region. Books from Computers & Technology category gives more profit than other.

We observed that there is significant difference in mean number of customers who bought the books at free shipping and who bought at paid shipping and no significant difference in mean number of new customers and existing customers who bought the books.  There is significant difference between the mean of number of customer who bought the books from different geographical region and different category. From regression analysis, we observed that there is significant relationship between total monthly sale and number of customers.

We have also provided recommendations and plan for company.

List of References

1. Berenson, M., Levine, D., Szabat, K.A. and Krehbiel, T.C., 2012. Basic business statistics: Concepts and applications. Pearson higher education AU.
2. Black, K., 2009. Business statistics: Contemporary decision making. John Wiley & Sons.
3. Casella, G. and Berger, R.L., 2002. Statistical inference (Vol. 2). Pacific Grove, CA: Duxbury.
4. DeGroot, M.H. and Schervish, M.J., 2012. Probability and statistics. Pearson Education.
5. Grus, J., 2015. Data science from scratch: first principles with python. ” O’Reilly Media, Inc.”.
6. Hodges Jr, J.L. and Lehmann, E.L., 2005. Basic concepts of probability and statistics. Society for industrial and applied mathematics.
7. Lee, G.G. and Lin, H.F., 2005. Customer perceptions of e-service quality in online shopping. International Journal of Retail & Distribution Management, 33(2), pp.161-176.
8. McKinney, W., 2012. Python for data analysis: Data wrangling with Pandas, NumPy, and IPython.” O’Reilly Media, Inc.”.
9. Mendenhall, W. and Sincich, T., 1993. A second course in business statistics: Regression analysis. San Francisco: Dellen.
10. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V. and Vanderplas, J., 2011. Scikit-learn: Machine learning in Python. Journal of machine learning research, 12(Oct), pp.2825-2830.
11. Pillers Dobler, C., 2002. Mathematical statistics: Basic ideas and selected topics.
12. Ross, S.M., 2014. Introduction to probability and statistics for engineers and scientists.Academic Press.
13. Wolfinbarger, M. and Gilly, M.C., 2001. Shopping online for freedom, control, and fun. California Management Review, 43(2), pp.34-55.