Problem Definition And Business Intelligence Required

Business that are involved in sales and retails have in recent times faced major challenges brought about by the digitization of almost everything. The rise of e-commerce and digitization proved to be swift and unexpected having been fueled by developments in the internet technology in terms of both speeds and scope plus advances in digital technology (Laudon & Guercio, 2014). The e-commerce and digital revolution only spared business entities that foresaw it, adapted to it or those that were too differentiated to be affected (Kotler, 2009).

The ABZ Corporation is a company that deals with the sales of DVDs, CDs, and books which include Classical, Novel, Politics, and bestsellers. The choice of products being sold places ABZ Corporation at a disadvantage in this era of digitization. With most of the music and films produced today being accessible in softcopy format from online music and film stores, the DVDs and CDs are less likely to attract customers (Maurer, 2008). The numerous digital devices and online stores have attracted many individuals to purchase their music and films from the online store (Harvey, 2014).

However, unlike the DVDs and CDs, books do have some aesthetic appeal (Rainee, Zickuhr, Purcell, Madden, & Brenner, 2012). This would be an ideal product with a significantly large market niche. The advent of online book stores and e-books, thus cannot cause the fall in prices of actual, physical books (Tim, 2017). This only serves to drive the aesthetic value of the physical books up, as well as the actual price. This relationship between the increase in price of actual book with increase in number of online book stores resembles the effect that digitization has had on art, and specifically paintings (Howard, 2014).

This paper, analyses the sales and retails sector with ABZ Corporation being a case study. The research analyses the sales of the various products categories offered by ABZ Corporation, customer trends as well as customer shopping profile. The paper aims at evaluating the sales and profits of ABZ Corporation by applying statistical analysis.

The ABZ Corporation is faced with a problem of low profits occasioned by decreasing sales. This necessitates analysis of the sales of the corporation (Total Gross Sales), as well as the profits of the corporation (Gross Profit Margins). This is with the aim of determining where the losses are being made in the business entity and advice on the appropriate measures to remedy the crisis.

The analysis in this research paper applies a number of business intelligence techniques. These techniques are: Data Visualization and Descriptive Statistics.

In the data visualization, the analysis will involve the generation of graphs, tables and charts. These visualization tools will be important in creating a picture of the nature of the different variables of the data.

The graphs will visualize the data by presenting comparisons between the categories of the various variables of the data (Roles, Baeten, & Signer, 2016). In addition, the graphs will also visualize the nature of the relationship between the various variables of the data.

The tables will simplify the data by providing more generalised data by either combining the various variables of the data or singling out a specific variable of the data.

Data Structure

The charts, like the graphs, will also visualize the data by comparing the various variables as well as the various categories within the variables of the data.

In the descriptive statistics, this paper will focus on two main measures. These measures are; measures of location and measures of variation.

The measures of location are statistics that provide information on the central value in any given data set (O’Neil & Schutt, 2013).

In this research focus will be in determining the central value for variables such as the profit (Gross Profits Margins), the Sales (Total Gross Sales), the Opening Gross Sales, the Price or Earnings Ratio, the Revenue and the age in the given data.

The measures of locations include the mean, median and mode of the given variable in the dataset provided (Theus & Urbanek, 2008).

The measures of variation are statistics that inform on the nature of the distribution of the data points of a variable in a dataset (Barbara & Susan, 2014). Measures of variation answer the question; how is the data distributed? In other terms the measures of variation evaluate the spread or the dispersion of the data points of the given data variable in the provided dataset.

This paper will focus on determining the dispersion of variables such as the profit (Gross Profits Margins), the Sales (Total Gross Sales), the Opening Gross Sales, the Price or Earnings Ratio and the Revenue in the given data. The dispersion of the age variable will also be investigated.

The measures of variation include the range, quartiles, interquartile range (IQR), the variance and the standard deviation (Martinez, Martinez, & Solka, 2010).

The provided data, ABZ Customer and Sales data, is split into two sections. The first section has the consumer attributes while the second attribute has the sales. In general, the data has a total of 16 variables.

The consumer attributes section has a total of 10 variables with seven categorical data variables, two numerical data variables and one interval data variable. The categorical data variables describe the qualities in a given case (Nguyen, Julien, & James, 2009), in this case the customers. The numerical and interval data variables are quantifier variables that provide measures in a given case (Shaffer, 2011). The table below describe the nature of the consumer attribute data variables.

Data Variable Name	Data Variable Type	Scales of Measurements
Customer ID	Interval	Interval
Gender	Categorical	Nominal
Marital Status	Categorical	Nominal
Age	Numerical	Numerical
Type of Customer	Categorical	Nominal
Region	Categorical	Nominal
Customer Payment Method	Categorical	Nominal
Source	Categorical	Nominal
Amount	Numerical	Numerical
Product	Categorical	Nominal

Table 1

The sales section has a total of 6 variables. The table below describe the nature of the sales data variables.

Data Variable Name	Data Variable Type	Scales of Measurements
Days of Transaction	Interval	Interval
Opening Gross Sales	Numerical	Numerical
Total Gross Sales	Numerical	Numerical
Price Over Earnings Ratio	Numerical	Numerical
Gross Profit Margin	Numerical	Numerical
Revenue	Numerical	Numerical

Table 2

The analysis provided the following information on the measures of location.

In the sales section of the data for the four hundred days in which the data was collected, the average Opening Gross Sales was $27.51 with a median value equivalent to $19.08. The modal Opening Gross Sale was $169.19. The Total Gross Sales had a mean value of $90.47 while the median Total Gross Sale was equal to $72.40. The modal Total Gross Sale was $37.3. For the Price over Earning Ratio, the average value was 23.6. The median ratio was 20.6 while the modal Price over Earnings ratio was 8.4. The Gross Profit Margin had a mean of 26.3, with a median value equivalent to 22.9. The modal Gross Profit Margin Value was 8.4. The mean revenue value for ABZ in thousands of dollars was 286.455 with the median value being 269. The modal Revenue value was 270. The analysis also found that the average age of the customers in the customer attribute section of the data was 43.1, rounding off to 43 years. The median age was determined as 42 years while the modal age was 46 years. The data is represented in the two tables below:

	Opening Gross Sales ($)	Total Gross Sales ($)	Price/Earnings Ratio	Gross Profit Margin (%)	Revenue ($ 000)
MEAN	27.51	90.47	23.6	26.3	286.455
MEDIAN	19.08	72.40	20.6	22.9	269
MODE	169.19	37.3	8.4	8.4	270

Descriptive Statistics

Table 3

	Age
MEAN	43.1
MEDIAN	42.0
MODE	46

Table 4

The percentiles are categories into which a population is grouped, normally four groups for the quartiles (Everitt & A, 2010). On the other hand, the main purpose for the variance is to assist in determining the standard deviation of the data (Hastie Trevor & Tibshir ni robert, 2012). The analysis provided the following information on the measures of location.

In the sales section of the data for the four hundred days in which the data was collected, the maximum value (100^th percentile) of the Opening Gross Sales was $169.19 while the minimum value (0 percentile) was $0.07, this produced a range of $0.07 to $169.19 equal to 169.12 for the Opening Gross Sales. The 1^st quartile, 2^nd quartile and 3^rd quartile were: $13.0025, $27.51 and $31.7525 respectively. The resulting interquartile range was equal to 18.75. The Opening Gross Sales had a variance of 697.7905 (4dp) and a standard deviation of 26.4157 (4dp).

The Total Gross Sales had a maximum value of $381.01 and a minimum value of $29.14. Hence, the range for the Total Gross Sales was $29.14 to $381.01 equal to 351.87. The 1^st, 2^nd and 3^rd quartiles were $39.9575, $90.47 and $105.045 respectively. The resulting interquartile range was equal to 65.0875. The Total Gross Sales had a variance of 4606.0750 (4dp) and a standard deviation of 67.8681 (4dp).

The Price over Earnings Ratio had a maximum value of 68.2 and a minimum value of 3.6. Hence, the range for the Price over Earnings Ratio was 3.6 to 68.2 equal to 64.6. The 1^st, 2^nd and 3^rd quartiles were 8.8, 23.6 and 35.8 respectively. The resulting interquartile range was equal to 27. The Price over Earnings Ratio had a variance of 262.5865 (4dp) and a standard deviation of 16.2045 (4dp).

The Gross Profit Margin had a maximum value of 74.2 and a minimum value of 3.6. Hence, the range for the Gross Profit Margin was 3.6 to 74.2 equal to 70.6. The 1^st, 2^nd and 3^rd quartiles were 9.7, 26.3 and 36.4 respectively. The resulting interquartile range was equal to 26.7. The Gross Profit Margin had a variance of 317.5363 (4dp) and a standard deviation of 17.8196 (4dp).

The Revenue in thousands of dollars, had a maximum value of 539 and a minimum value of 229. Hence, the range for the Revenue was 229 to 539 equal to 310. The 1^st, 2^nd and 3^rd quartiles were 252.75, 286.455 and 298 respectively. The resulting interquartile range was equal to 45.25. The Gross Profit Margin had a variance of 3552.81 and a standard deviation of 59.6055 (4dp).

The analysis also found for the Age of the customers in the customer attribute section of the data, the maximum age was 78 and the minimum age was 20. Hence, the range for the Age of the customers was 20 to 78 equal to 58. The 1^st, 2^nd and 3^rd quartiles were 32, 43 and 50 respectively. The resulting interquartile range was equal to 18. The Age had a variance of 152.3344 (4dp) and a standard deviation of 12.3424 (4dp).

The data above is represented in the two tables below;

	Opening Gross Sales ($)	Total Gross Sales ($)	Price/Earnings Ratio	Gross Profit Margin (%)	Revenue ($ 000)
MAX	169.19	381.01	68.2	74.2	539
MIN	0.07	29.14	3.6	3.6	229
1st QUARTILE	13.0025	39.9575	8.8	9.7	252.75
3rd QUARTILE	31.7525	105.045	35.8	36.4	298
INTERQUARTILE RANGE	18.75	65.0875	27	26.7	45.25
VARIANCE	697.7905173	4606.075009	262.5864932	317.5362952	3552.81
STANDARD DEVIATION	26.41572481	67.86807061	16.20452076	17.81954812	59.60545277

Measures Of Location

Table 5

	Age
MAX	78.0
MIN	20.0
1st QUARTILE	32
3rd QUARTILE	50
INTERQUARTILE RANGE	18
VARIANCE	152.3344361
STANDARD DEVIATION	12.34238373

Table 6

The data summary consists of the data visualization results from the analysis of the various variables and/or their categories in the data set provided.

The table below represent data on the total sales for each of the product categories in the consumer attributes section of the dataset.

PRODUCT CATEGORIES	TOTAL SALES
DVD	6012.23
CD	1522.37
NOVEL	2263.59
CLASSICAL	1774.89
POLITICS	2001.32
BEST SELLER	2340.11

Table 7

The data in the table above produces the pie chart below

Figure 1

From the table and the pie chart above, we observe that top selling product category is the DVD at 38% equivalent to $6012.23, while the worst selling product category is the CD at 9% equivalent to $1522.37.

The pie chart above also shows that there exist significant differences in the product categories in terms of the Total Gross Sales. The chart shows that the sales for the categories differ from each other.

The chart below represent data on the comparison between the amount of purchase for male and female customers.

Figure 2

The chart indicates that up to 95% of total amount of purchase is from women while the remaining 5% is from men.

The graph below shows the distribution of ages of the male customers.

Figure 3

The graph below shows the distribution of ages of the female customers.

Figure 4

The graph on the distribution of the ages of male customers show that many of the data points lie between the ages of 30 and 50 years, with similar observations made for the graph on the distribution of ages of the female customers.

Conclusion

From the data analysis, we conclude that although DVDs and CDs are of similar technology, they find themselves at opposite extremes of the sales with DVDs being the top selling product category. Given that this is a technology that is rapidly being faced out (Farris & Neil, 2010), ABZ Corporation should be concerned that their most selling product is one that has a technology that is quickly coming off the market. We also conclude that the ABZ Corporation marketing is more applying to female customers that to male customers as shown from the amounts of purchase. The age of the customers, however cannot be specified, with the products appealing to anyone between 20 years and 78 years, and a bulk of them being between 30 and 50 years.

The main recommendation would be ABZ Corporation urgently finding a way of making the content they have on DVDs and CDs available in the online platforms. However, the marketing should also be rethought in order to advertise the other product categories as much as the DVDs. Also creating a targeted marketing strategy focused on attracting more male customers would help in boosting the sales.

More and broader research also needs to be undertaken. For instance finding out which age group buys which product or which gender buys which product would provide better information to assist in providing more accurate recommendations.

References

Barbara, I., & Susan, D. (2014). Introductory Statistics. OpenStax CNX.

Everitt, B. S., & A, S. (2010). Cambridge Dictionary of Statistics. Cambridge University Press.

Farris, P. W., & Neil, B. T. (2010). Marketing Metrics. New Jersey: Pearson Education.

Harvey, E. (2014). Station to Station: The Past, Present and Future of Streaming Music. New York: Pitchfork.

Hastie Trevor, & Tibshir ni robert. (2012). The Elements of Statistical Learning. 

Howard, J. (2014). The Varieties of Aesthetic Experience. London: Pearson.

Kotler, P. (2009). Marketting Management. Pearson: Prentice-Hall.

Laudon, K. C., & Guercio, T. C. (2014). E-commerce. Business. Technology. Society. Pearson.

Martinez, W. L., Martinez, A. R., & Solka, J. (2010). Exploratory Data Analysis With MATLAB, 2nd Edition. CRC/Chapmann & Hall.

Maurer, W. (2008). The Dynamics Of Music Distribution. New York: The Original.

Nguyen, X. V., Julien, E., & James, B. (2009). Information Theoretic Measures of Clustering Comparison. Chicago: The Original.

O’Neil, C., & Schutt, R. (2013). Doing Data Science. London: O’Reily.

Rainee, L., Zickuhr, K., Purcell, K., Madden, M., & Brenner, J. (2012). The Rise of e-reading. Pew Internet Library.

Roles, R., Baeten, Y., & Signer, B. (2016). Interactive and Narrative Data Visualization for Presentation-Based Knowledge Transfer. Communication in Computer and Information Science, 739.

Shaffer, C. A. (2011). Data Structures and Algorithms Analysis. Mineola: Dover.

Theus, M., & Urbanek, S. (2008). Inteactive Graphics For Data Analysis. Boca Raton: CRC Press.

Tim, P. (2017). The Books We Don’t Understand. New York: New York Review of Books.