Problem definition and business intelligence required

This report is about an online retail company called, Retail Surge. The company has its business divided into several areas including Boy’s, Men’s, Girl’s, Women’s and Customisation. The company’s product range includes clothing, shoes, sporting equipment and accessories. This report seeks to analyse and understand the product categories that generate more income to the company. It also sought to understand the product categories that had the largest cost of goods. Lastly, the study looked at the association between gender/website user groups and customer attitudes.

This study sought to answer the following research questions.

• Which product categories are making the most profit?

To answer this research question, analysis of variance (ANOVA) was employed (Hinkelmann & Kempthorne, 2008). ANOVA is used to analyse variation in the means of groups that are more than 2. Since the product categories were more than 2, ANOVA was the most ideal test to be used.

• Which product category costs the most (COGS)?

Again to answer this research question, analysis of variance (ANOVA) was employed (Hinkelmann & Kempthorne, 2008). ANOVA is used to analyse variation in the means of groups that are more than 2 (Gelman, 2005). Since the product categories were more than 2, ANOVA was the most ideal test to be used.

• Is there a difference in payments methods?

Answering this research question required the use of t-test is that test that helps compare the means of two groups (Sawilowsky, 2005). Since there are only two groups (PayPal ad Credit Card), t-test became the most ideal test.

• Are there any differences in the user groups on all of the customer attitudes?

To answer this research question, Chi-Square test of association was used. Chi-Square test of association helps to identify whether there exists any kind of relationship/association between two categorical/nominal variables (Bagdonavicius & Nikulin, 2011). The research question to be tested involved two variables with nominal data values hence Chi-Square was the most ideal test.

• Are there any differences in gender on all of the customer attitudes? (6 outcomes)

This is the last research question that the study sought to answer. Just like the immediate previous question, this research question was answered by performing a Chi-Square test of association. The research question to be tested involved two variables with nominal data values hence Chi-Square was the most ideal test.

For this section, the study sought to test the following hypothesis.

H₀: There is no significant difference in the average profit for the different product categories

H_A: There is significant difference in the average profit for the different product categories for at least one of the product categories

This was tested at 5% level of significance. To test this, analysis of variance (ANOVA) was used. 

Table 1: Descriptive Statistics

From the descriptive table above, it can be seen that the product with the highest profit to be the customized items (M = 25.00, SD = 0.00). The product with the least profit was the boy’s shoes (M = 3.30, SD = 0.00).

Results of the selected analytics methods and technical analysis

Table 2: Test of Homogeneity of Variances

Before running the ANOVA, we checked for the homogeneity of variances. Levene’s test of homogeneity showed that the variances are not homogenous (not equal).

Table 3: ANOVA

A one-way ANOVA was performed to check whether there are significant differences in the profit made. The p-value was found to be 0.000 (a value less than 5% level of significance), this leads to rejection of the null hypothesis and hence we conclude that there is significant difference in the average profit for the different product categories for at least one of the product categories. Post-hoc using Games-Howell showed that all the products were significantly different in terms of the average profit. 

For this section, the study sought to test the following hypothesis.

H₀: There is no significant difference in the average cost of goods for the different product categories

H_A: There is significant difference in the average cost of goods for the different product categories for at least one of the product categories

This was tested at 5% level of significance. To test this, analysis of variance (ANOVA) was used. 

Table 4: Descriptive Statistics

From the descriptive table above, it can be seen that the product with the highest cost of goods to be the customized items (M = 9.80.00, SD = 0.00). The product with the least average cost of goods was the men’s clothing (M = 1.00, SD = 0.00).

Table 5: ANOVA

A one-way ANOVA was performed to check whether there are significant differences in the cost of goods (COGs). The p-value was found to be 0.000 (a value less than 5% level of significance), this leads to rejection of the null hypothesis and hence we conclude that there is significant difference in the average cost of goods for the different product categories for at least one of the product categories. Post-hoc using Games-Howell showed that all the products were significantly different in terms average cost of goods.

Next, we sought to find out whether there is significant difference in payment methods. To test this, the following hypothesis was tested at 5% level;

H₀: There is no significant difference average total purchases paid with PayPal and Credit Card

H₀: There is significant difference average total purchases paid with PayPal and Credit Card

The results are given below;

Table 6: t-Test: Two-Sample Assuming Equal Variances

We performed an independent t-test in order to compare the average total purchases paid with PayPal and Credit Card. Results showed that the average total purchases paid with PayPal (M = 3.42, SD = 3.61, N = 612) did not significantly different with the average total purchases paid with Credit Card (M = 3.63, SD = 4.40, N = 612), t (1222) = -0.896, p > .05, two-tailed. Essentially the results showed that the payment method does not in any way (significantly) influence the total purchases made.

For this, we sought to find out whether there is significant association between the user groups and the customer attitudes. The null hypothesis was that there is no significant association between the user group and the customer attitude. A Chi-Square test of association was performed and the results are given below; 

Table 7: Chi-Square test of association (user group and customer attitudes)

The above table shows that there is significant association between the website user groups and all the customer attitudes (p < 0.05).

Lastly, in this section, just like section 4 above, we sought to find out whether there is significant association between the gender and the customer attitudes. The null hypothesis was that there is no significant association between the gender and the customer attitude. A Chi-Square test of association was performed and the results are given below;

Table 8: Chi-Square test of association (gender and customer attitudes)

Clearly, the above results shows that significant association exists between gender of the customer and five of the customer attitudes (p < 0.05). Results showed that there was no association between gender of the customer and whether they would recommend company to a friend (p = 0.578).

This study sought to analyse and understand the product categories that generate more income to the company. It also sought to understand the product categories that had the largest cost of goods. Lastly, the study looked at the association between gender/website user groups and customer attitudes. Results showed that customized items generated more profit than any other product. Also, the same customized products had the highest cost of goods. There was no significant difference in the average purchases made from the two different payment methods.

Based on the above findings and conclusions, the following recommendations are made to the Company’s CEO;

• The management (CEO) should come up with ways of reducing the cost of goods so as to maximize on the net profits.
• More focus should be put of customer attitudes among the different groups of customers. Results showed that different customer groups had varied customer attitude either towards the company or towards the product. 

References

Bagdonavicius, V., & Nikulin, M. S. (2011). Chi-squared goodness-of-fit test for right censored data. The International Journal of Applied Mathematics and Statistics, 30–50.

Gelman, A. (2005). Analysis of variance? Why it is more important than ever. The Annals of Statistics, 33(5), 1–53. doi:10.1214/009053604000001048

Hinkelmann, K., & Kempthorne, O. (2008). Design and Analysis of Experiments. Journal of the Royal Statistical Society, 251 (5), 251–276.

Sawilowsky, S. (2005). Misconceptions Leading to Choosing the t Test Over The Wilcoxon Mann–Whitney Test for Shift in Location Parameter. Journal of Modern Applied Statistical Methods, 4(2), 598–600.