Good Harvest Business Analysis - Performance And Recommendations

Divided into a number of areas including retail, wholesale a box delivery system (not included in the data), and Harvest Kitchen.

The data is inclusive of a whole year of trading. This is the second year of business and the business is still in a start-up phase. The reported high Cost of Goods (COGS) is reportedly consistent with the Organic industry.

The main challenges in the business are revenue (i.e. lead generation/new business), Cost of Goods (COGS margins) and average sales.

The business has a 6 person team, 1 delivery van, a retail outlet and a cold store warehouse.

Task instructions:

1)Ensure the dataset is appropriately scaled (each measure in SPSS must be set correctly). Report on what you changed.

2)Produce descriptive statistics (see lecture 1) for the most appropriate variables to give an overview of how the business and its products are performing

3)Use the statistical analyses you have learned to answer all of the questions below

As guide - answer these main research questions:

1.What are the top/worst selling products in terms of sales?

a.Is there a difference in payments methods?

2.Are the differences in sales performance based on where the product is located in the shop? How does this effect both profits and revenue?

3.Is there a difference in sales and gross profits between different months of the year?

4.Are their differences in sales performance between different seasons?

a.How does this relate to rainfall and profits?

Introduce the business and its problems (see website link above for more information)

• Problem definition and business intelligence required  – Outline the research questions and list what analytics methods you have chosen to answer each one and why
• Visualise the descriptive statistics  – using what you have learned show the relevant descriptive statistics through data visualization/ graphical displays (Tables, graphs, pie charts etc.).
• Results of the selected analytics methods and technical analysis use each research question as a heading then present the analysis findings and explain them. Higher marks for including extra research questions not defined here and creating nice tables/displays opposed to copy and pasting directly from SPSS output.
• Discussion of the results and recommendations  Based on the analyses what conclusions about the business can be summarized/drawn and what recommendations can be offered to the CEO.

Good harvest is a firm based in Sunshine Coast which offers delivery services for their organic products. The company is still in a startup phase being that this is its second year in business. This means that they have to sell their goods at higher prices compared to the other business who have been in the game for longer. The other challenges faced by the business is low average sales and low revenue and low workforce. These, according to the Huffington post, are challenges faced by every other start up business out there (Nwobu 2016). This analysis seeks to determine the performance of the company and its products, and provide recommendations that could give the business a way forward.

The challenges faced by the business represents a significant financial burden to the company. Being a new business, this could not only lead to extreme financial challenges, but to the closure of the business as a whole. The cost of goods must remain high so as to generate some profit, but not so high that it scares off the customers. Finding the perfect balance between generating profits and retaining customers is one of the toughest challenges faced by new businesses (Nwobu 2016). According to Ganesan (2016), bringing the sales department to order could be the biggest breakthrough of a startup, since this will help generate a steady revenue, which can be used to run and manage the other aspects and departments of the business. To solve the current shortcomings in the business and ensure its success in this startup unfriendly environment, effective strategies must be put in place.

Two datasets are used in the analysis, the first dataset contains data for the food shop for product mix while the other dataset contains data for the food shop sales summary. The former comprises of ten variables with 1034 observations each, while the latter is made up of eighteen variables, each with 366 observations. These variables are both quantitative and categorical.

 I changed the Product Class category from Ordinal to Nominal and Product Category from Ordinal to Nominal. This is because both Product Class and Product Category are categorical variables not based on merit or order (Bagdonavicius & Nikulin, 2011).

1. Net Profit

Descriptive Statistics

	N	Minimum	Maximum	Mean	Std. Deviation
Net Profit ($)	1034	0	8703	164.74	482.106
Valid N (listwise)	1034

The average net profit is given by 164.74, the maximum is 8703, and the minimum, 0.

1. Total Sales

Descriptive Statistics

	N	Minimum	Maximum	Mean	Std. Deviation
Total Sales ($)	1034	0	17276	369.96	1014.719
Valid N (listwise)	1034

The average total sales is given by 369.96, the maximum is 17276, and the minimum, 0.

1. Cost of Goods

Descriptive Statistics

	N	Minimum	Maximum	Mean	Std. Deviation
Cost of Goods ($)	1034	0	8573	205.22	561.072
Valid N (listwise)	1034

The average cost of goods is given by 205.22, the maximum is 8573, and the minimum, 0.

The above three plots shows the histogram for the profit total, gross sales and the net sales. As can be seen, the histogram for the profit total shows that the data is right skewed while both the gross sales and the net sales appear to be normally distributed.

The boxplot presented is for the worst performing and best performing products. As can be seen, the worst performing products have total quantity sales less than 100 while the best performing have sales averaging to almost 500 in number with some products having sales up to almost 4000.

Descriptive Statistics

The top selling product is the product with the maximum sales is Bananas Cavendish, with a total sales of 17276. The worst selling product is the with the minimum sales is Scarves Small, with a total sales of 0.

For deeper insight of business performance, Analysis of Variance (ANOVA) tests will be used to test and analyse the data for various properties (Stevens, 2002). The questions to be answered by the ANOVA test include:

1. Is there a difference in payments methods?
2. Is there a difference in sales and gross profits between different months of the year?

• Are their differences in sales performance between different seasons?

1. Is there a difference in sales and gross profits between different months of the year?
2. Are their differences in sales performance between different seasons?
3. Difference in payments methods

Hypothesis

Null hypothesis (H0): There is no difference in payments methods

Alternate hypothesis (HA): There is difference in payments methods

Level of significance = 0.05

Analysis Results

The p-value of the (ANOVA table i) on page 7 is less than the alpha (0.05). This provides sufficient evidence to reject the null hypothesis; we therefore reject H0. This means that there is a difference in payment methods.

We then perform a post hoc analysis to determine where the difference exists among the four payment methods (cash, credit card, visa card and MasterCard).

The results of this analysis as per figure 2 on page 9, show that cash is the most common mode of payment, credit card and visa card are the second most used, and MasterCard, the least used. There exists a difference between cash payments and all the other modes of payment. There is no significant difference in credit card and visa card payment, and finally, there exists a significant difference between MasterCard and the other payment methods (Montgomery, 2001).

1. Differences in sales performance based on where the product is located in the shop

How does this effect both profits and revenue?

Hypothesis

Null hypothesis (H0): There is no difference in sales performance for location in shop

Alternate hypothesis (HA): There is difference in sales performance for location in shop

Level of significance = 0.05

Analysis Results

The p-value of the (ANOVA table iii) on page 8 is less than the alpha (0.05). This provides sufficient evidence to reject the null hypothesis; we therefore reject H0. This means that there is a difference in sales performance based on the location of the product in the shop (Chiang, 2003).

We then perform a post hoc analysis to determine where the difference exists among the five locations (front, left, outside front, rear, and right).

The results of this analysis as per figure (iv) on page 10 show that goods in the left location of the shop made the highest sales, followed by products in the right. Those on outside front received the lowest sales.

1. Difference in sales and gross profits between different months of the year
2. Hypothesis

Null hypothesis (H0): There is no difference in sales between different months of the year

Alternate hypothesis (HA): There is difference in sales between different months of the year Level of significance = 0.05

Analysis Results

The p-value (0.22) of the ANOVA table (v) on page 12 is greater than the alpha (0.05). This provides sufficient evidence to accept the null hypothesis; we therefore fail to reject H0. This means that there is no difference in sales between different months of the year (Gelman, Analysis of variance? why it is more important than ever, 2005).

1. Hypothesis

Null hypothesis (H0): There is no difference in gross profit between different months of the year

Box Plots

Alternate hypothesis (HA): There is difference in gross profit between different months of the year

Level of significance = 0.05

Analysis Results

The p-value of the ANOVA table (vi) on page 12 is less than the alpha (0.05). This provides sufficient evidence to reject the null hypothesis; we therefore reject H0. This means that there is a difference in gross profit between different months of the year.

The results of this analysis show that November recorded the highest profits while June recorded the lowest profits.

1. Differences in sales performance between different seasons

Hypothesis

Null hypothesis (H0): There is no difference in sales performance between different seasons

Alternate hypothesis (HA): There is difference in sales performance between different seasons Level of significance = 0.05

Analysis Results

The p-value (0.814) in the ANOVA table (vii) on page 12 is less than the alpha (0.05). This doesn’t provide sufficient evidence to reject the null hypothesis; we therefore fail to reject H0. This means that there is no difference in sales performance between different seasons.

1. Relationship between rainfall and profits

A correlation analysis is performed to determine the linear relationship between rainfall and profits (Gelman & Hill, Data Analysis Using Regression and Multilevel/Hierarchical Models, 2006). The results of the analysis according to figure (ix) on page 11 shows a correlation coefficient of 0.885. This implies a strong positive linear relationship between the two variables. Meaning that an increase in one rainfall results in an increase in profits.

Conclusions

From examining the financial status of the organic firm business, we notice that the business performs different according to different months and different seasons. There are months when sales and profit recorded are high, there are seasons (rainfall seasons) when profitability is high, and there are certain locations that guarantees sale of products more than others. Since the profits and quantity of sales are mostly based on fruit or vegetable production, the firm can take advantage of certain seasons when a particular fruit is most likely to give the best products, and plant the vegetable then. The firm can also try to maximize their sales during the months that recorded the highest profit like November. Another way the firm can develop financial advantage is to position their best selling and profitable products on the shop locations which recorded the highest sales, i.e., left and center of the shop. I believe that implementing these suggestions will lead to a better financial performance of the business and ensure its future success

1. ANOVA Table

ANOVA

Payment

	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	73567963.508	3	24522654.503	697.861	.000
Within Groups	51304001.858	1460	35139.727
Total	124871965.366	1463

1. Post Hoc Table

 Post Hoc Tests

Multiple Comparisons

Dependent Variable: Payment

Tukey HSD

(I) Payment Methods	(J) Payment Methods	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Upper Bound	Lower Bound
Cash	Credit Card	-180.519(*)	13.857	.000	-216.16	-144.88
	Visa Card	-151.552(*)	13.857	.000	-187.19	-115.91
	Mastercard	382.202(*)	13.857	.000	346.56	417.84
Credit Card	Cash	180.519(*)	13.857	.000	144.88	216.16
	Visa Card	28.967	13.857	.157	-6.67	64.61
	Mastercard	562.721(*)	13.857	.000	527.08	598.36
Visa Card	Cash	151.552(*)	13.857	.000	115.91	187.19
	Credit Card	-28.967	13.857	.157	-64.61	6.67
	Mastercard	533.754(*)	13.857	.000	498.11	569.39
Mastercard	Cash	-382.202(*)	13.857	.000	-417.84	-346.56
	Credit Card	-562.721(*)	13.857	.000	-598.36	-527.08
	Visa Card	-533.754(*)	13.857	.000	-569.39	-498.11

*  The mean difference is significant at the .05 level.

1. ANOVA Table

ANOVA

Total Sales ($)

	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	134299725.024	4	33574931.256	37.176	.000
Within Groups	929333380.817	1029	903142.255
Total	1063633105.841	1033

1. Post Hoc Table

Post Hoc Tests

Dependent Variable: Total Sales ($)

Tukey HSD

(I) Location of product in shop	(J) Location of product in shop	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Upper Bound	Lower Bound
Front	Left	354.531(*)	90.712	.001	106.65	602.41
	Outside Front	-2811.617(*)	284.761	.000	-3589.76	-2033.48
	Rear	36.679	104.135	.997	-247.88	321.24
	Right	332.860(*)	93.438	.004	77.53	588.19
Left	Front	-354.531(*)	90.712	.001	-602.41	-106.65
	Outside Front	-3166.148(*)	278.682	.000	-3927.68	-2404.62
	Rear	-317.851(*)	86.136	.002	-553.23	-82.47
	Right	-21.671	72.842	.998	-220.72	177.38
Outside Front	Front	2811.617(*)	284.761	.000	2033.48	3589.76
	Left	3166.148(*)	278.682	.000	2404.62	3927.68
	Rear	2848.297(*)	283.336	.000	2074.05	3622.55
	Right	3144.477(*)	279.582	.000	2380.49	3908.47
Rear	Front	-36.679	104.135	.997	-321.24	247.88
	Left	317.851(*)	86.136	.002	82.47	553.23
	Outside Front	-2848.297(*)	283.336	.000	-3622.55	-2074.05
	Right	296.181(*)	89.003	.008	52.97	539.39
Right	Front	-332.860(*)	93.438	.004	-588.19	-77.53
	Left	21.671	72.842	.998	-177.38	220.72
	Outside Front	-3144.477(*)	279.582	.000	-3908.47	-2380.49
	Rear	-296.181(*)	89.003	.008	-539.39	-52.97

*  The mean difference is significant at the .05 level.

1. ANOVA table

ANOVA

Gross_Sales

	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	1508892.474	11	137172.043	1.300	.222
Within Groups	37349615.455	354	105507.388
Total	38858507.929	365

1. ANOVA table

ANOVA

Profit Total

	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	35370.948	11	3215.541	3.867	.000
Within Groups	294370.006	354	831.554
Total	329740.954	365

1. Post Hoc table
2. ANOVA table

ANOVA

Average_Sale

	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	15.148	3	5.049	.316	.814
Within Groups	5654.334	354	15.973
Total	5669.483	357

1. Correlation table

Correlations

		Rainfall	Profit Total
Rainfall	Pearson Correlation	1	.008
	Sig. (2-tailed)		.885
	N	365	365
Profit Total	Pearson Correlation	.008	1
	Sig. (2-tailed)	.885
	N	365	366

References

Aldrich, J. (2005). Fisher and Regression. Statistical Science, 20(4), 401–417.

Armstrong, J. S. (2012). Illusions in Regression Analysis. International Journal of Forecasting (forthcoming), 28(3), 689.

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.

Chiang, C. L. (2003). Statistical methods of analysis, World Scientific.

Cox, D. R. (2006). Principles of statistical inference.

Ganesan, S. (2016, August 22). 6 challenges faced by early-stage startups that some effective tools can help you combat. Retrieved from https://yourstory.com/2016/08/challenges-early-stage-startups/

Gelman, A. (2005). Analysis of variance? why it is more important than ever. The Annals of Statistics, 33(1), 1–53.

Gelman, A., & Hill, J. (2006). Data Analysis Using Regression and Multilevel/Hierarchical Models. 45–46.

Hinkelmann, K., & Kempthorne, O. (2008). Design and Analysis of Experiments. I and II (Second ed.).

Howell, D. (2002). Statistical Methods for Psychology. 324–325.

Kutner, M. H., Nachtsheim, C. J., & Neter, J. (2004). Applied Linear Regression Models. 25.

Montgomery, D. C. (2001). Design and Analysis of Experiments (5th ed.).

Moore, D. S., & McCabe, G. P. (2003). Introduction to the Practice of Statistics (4th ed.). 764.

Nwobu, U. (2016, August 25). Most Common Challenges Faced By Start-Ups. Retrieved from https://www.huffingtonpost.com/ursula-nwobu/most-common-challenges-faced-by-start-ups_b_11701900.html

Rouaud, M. (2013). Probability, Statistics and Estimation. 60.

Stevens, J. P. (2002). Applied multivariate statistics for the social sciences.