The City and State
Executive Summary
The present report analyses the data provided as part of the ongoing efforts to arrive at data driven decision making for the coffee roaster and supplier firm –‘Sublime Delights.’ The report takes into account the BADIRTM methodology in order to develop an analysis strategy and derive meaningful and actionable intelligence for the management of the company to take effective decisions. The report starts with a brief introduction of the company, proceeds further with the problem statement and the methodology used. The report then goes on to process and analyze the data as collected from various sources towards successful completion. The report uses various statistical tools to measure the variability, central tendency as well as dispersion of the data. The measurements help us determine the sales trends, supply mix and the marketing mix for the firm over a period of time. The report concludes by putting forth various recommendations regarding the improvements in quality, reduction in costs, potential growth areas and the restructuring of the marketing mix.
Index
Executive Summary 2
Index 3
Introduction 4
Background 4
Discussion 5
Task1 5
Summary 15
References 16
Sublime Delight Case Analysis within the BADIR Framework: A Business Report
Introduction
Statistics can be used in a very effective and focused manner to gain valuable insights into business processes and events, so as to take effective and timely business decisions.
Among the various approaches that are used, the BADIR methodology is one of the most effective analytical tools to yield actionable information from the overwhelmingly large pool of data available to businesses nowadays. BADIR is in fact an acronym for the framework that can be used to apply data analysis to decision making. The steps of the framework are (1) Business Question (2) Analysis Plan (3) Data Collection (4) Derive Insights (5) Recommendations.
‘Sublime Delight’ is a company that produces various varieties of coffee roasts. The three different varieties are Espresso, Mocha and Sublime Delight-which is a specialized product of the company. It is a small yet rapidly growing company which began 5 years ago.
About eighteen months ago, the company management had decided to add take away outlets and three mobile units. There are seven takeaway sites and the three mobile units cater to the building sites, commercial sites and offices and two of the mobile units are also available for catering.
Background
Over a period of time, Geoff the founder and owner of Sublime Delights’ had been finding it increasingly difficult to give time to his interest as a connoisseur of coffee. He had been finding himself increasingly engrossed in day to day business activities. Thus he hired a general manager, who was very much into data collection and analysis as a means of making effective business decisions. He started a trend of collecting data on almost every big and small aspects of the business on a wide variety of parameters within the firm. Although, he was yet to take any meaningful decisions, he left the firm for a better opportunity and this created a mess. I worked in the dispatch and delivery section and have been allowed by the owner to draw meaningful inferences from the data left by the ex-general manager and this report is based on a careful analysis of the data selected from the same for presentation to the board of directors.
Discussion
The paper is based on three datasets collected from the company records with reference to three tasks pertaining to presenting actionable information to the board of directors. The tasks, with respect to the data sets and the processing done in consonance with the BADIR methodology is presented as mentioned below.
Task1
Business Question
Analysis Plan
In order to determine how the sales and profits are distributed, the following assumptions and metrics were arrived at.
Hypotheses/assumptions
The task 1 of the study assumes the following and seeks to validate the same
The sales are evenly distributed among the various outlets (H1).
The sales vary in an even manner over a period of a year between the three outlets (H2).
The Profits vary evenly among the three outlets (H3).
The profits vary evenly with respect to time over a period of one year (H4).
Metrics
As per dataset1 for Task 1, it is seen that a month by month sales report, showing the amount of beans sold in Kgs is available. Thus as per the business question above, we need to measure the central tendency as well as the variability aspects of the sales from different channels of sales namely external customer, internal carts and mobile units.
Data Collection
The data for the three outlets including the sales data has been collated from the company accounting systems. Apart from this the costs data was available from company sources.
Derive Insights
Define the Insight metrics for each hypothesis
Based on the above metrics and hypotheses, the following insight metrics were developed as mentioned below.
In order to validate H1 through H4, the following steps were taken.
The Average-External Customer orders (AECO), The Average Internal Cart Orders (AICO) and Average Mobiles Orders (AMO) were worked out to give the central tendency of the sales for the three outlets/channels as shown in the table and chart below.
Average sales (Kgs) per customer/outlet per month channel wise
Descriptive statistics for average sales in kgs per customer outlet were calculated as follows
Total sales in kg. per month channel wise
Based on the Prices and costs data provided, the monetary profits were calculated as below.
The same are graphically represented as below.
Pearson’s correlation coefficient was calculated between AECO, AICO and AEMO as shown below.
Validation of assumptions
H1
As per (1) and (2) above, it is evident that the average sales are uneven with respect to the outlets. The difference in means of AECO, AICO and AMO show that the sales are at completely different level. However, as shown in (3), the total sales as opposed to average sales are poised differently. The Total Sales levels of AECO and AICO above show a comparable level of sales for the External and Internal customer channels, whereas the Mobile orders channel is lagging behind at a very low level.
Thus H1 cannot be accepted.
H2
As per (1) and (2) above, although the sales levels are varying for a given channel with respect to time, the level of variance (22,21) as well as SD (5,5) for AECO and AMO respectively shows that the sales for these two outlets vary in a similar fashion with time. It cannot however, be considered an even variation for the sales levels with respect to time. Thus H2 cannot be accepted.
H3
As per (4), it is apparent that the Total profit per month is very unevenly distributed among the channels. Thus H3 cannot be accepted.
H4
As per (3), the profits are not evenly distributed with respect to time. Thus H4 cannot be accepted. Thus Assumptions H1 through H4 are rejected. Thus sales are unevenly distributed with respect to channels and time.
Recommendations
The difference in relative average sales per outlet unit and the total sales for the Mobile Units shows that the outlet has good potential and additional units sold can really add to the bottom line of the company.
Although the sales levels are comparable, the profit levels have a marked difference for the external customers and internal cart outlets, the latter showing a lower profit at similar sales levels. Thus the expenses for the latter need to be minimized to achieve a higher profit from the channel.
Business Question
Is there a significant difference between the outlets in terms of service and coffee quality.
Analysis Plan
As per the business question above we need to measure the variability in the outlet in terms of service quality and coffee quality.
The task is thus designed to validate the following two hypotheses/Assumptions.
H1: There is a significant difference between the outlets in terms of Coffee and Service Quality
The same can be restated as null (Ho1) and alternative (Ha1) hypotheses as follows.
Ho1: There is no significant difference between the outlets in terms of Coffee and service Quality
Ha1: There is a significant difference between the outlets in terms of Service Quality
Data Collection
In order to collect the relevant data for the task, the customers were asked to rate the service (1 Poor – 5 excellent), quality of coffee (1-poor, 5 – excellent).
The customers were also asked about the frequency with which they purchased from a given outlet (A less than one per week, B, Once or more per week, C Daily, D more than once per day.)
The data collection was done through survey cards filled on a particular day.
Derive Insights
In order to arrive at the desired results for the analysis, the relevant statistical tool that needs to be used is one way ANOVA. However, prior to that the non numeric data for the Frequency of visit was converted to numeric data as follows.
The assumptions made were as follows.
The less than once per week frequency was assumed at level 0.
The once or more than once per week was assumed to be at once per week only at 0.14.
The more than once per day was assumed as twice per day=2
Thus the data based on the above assumptions is shown as below.
The results of one way ANOVA are as tabulated below.
As is evident from the above results, the F-value is greater than F-critical value. Therefore, the null hypothesis is rejected. This means that there is a significant difference in the outlets in terms of coffee and service quality.
Recommendations
Since there is a significant difference in terms of service and coffee quality, thus the management needs to make efforts to improve quality across the outlets and make it more consistent with the standards.
Business Question
As per market analysis prior to the period in question, the supply mix to the various external outlets was supposed to be 30% Espresso, 10% Mocha and 60% Sublime.
The board is interested in knowing with respect to the above prediction, whether
The prediction is true for January 2015?
The prediction is true for December 2015?
Has the mix changed overall between January and December 2015?
What would be the new market mix based on December data?
Analysis Plan
Based on the Business questions posed for task 3, the following assumptions are made for validation through analysis of the dataset 3.
Ha: The supply mix prediction for January 2015 is true.
Hb: The supply mix prediction is true for December 2015.
Hc: There is a considerable change overall between January and December 2015.
Hd : The new marketing mix based on the December data shall be more uniform among the different product markets.
Data Collection
The data collected for the task is the bean mix supplied to the external outlets in January and December 2015. The data has been collected through company supply records and a market analysis has predicted a market mix for the same as 30% Espresso, 10% Mocha and 60% sublime.
Derive Insights
In order to arrive at the results as sought per the task and underlying assumptions, the dataset 3 was processed to further summarize the Supply mix (E%,M%, S%) and give a glance at the overall picture as shown below for the same.
Based on the above table, it is evident that predicted mix almost holds true for the month of January to a considerable extent as the variance mix (-3, 5 and-2) percentage points is pretty miniscule from a prediction point of view. Thus Ha cannot be rejected. Further, the December mix has variance mix (7, 18, -25) percentage points is quite considerable from the predicted values. Thus Hb is rejected. Furthermore, the change mix (10,13,-23) percentage points from January to December 2015 is a considerable difference, thus Hc is accepted. Finally, the December mix (37, 28, 35) is much more uniform than the January mix(27, 15,58) and thus shows that the future mix shall remain more uniform.
Recommendations
Although the predicted mix held true to some extent at the beginning of the period in question, the mix changed over a period of time to be more uniform. Thus the factors behind the prediction need to be reassessed for a fresh evaluation of the supply mix and the underlying marketing mix.
Summary
There is a significant difference in terms of service and coffee quality. The management should make efforts to improve quality across the outlets and make it more fall in line with the standards for quality.
The difference in relative average sales per outlet unit and the total sales for the Mobile Units shows that the outlet has good potential and additional units sold can really add to profits. Since the sales levels are comparable, the profit levels have a marked difference for the external customers and internal cart outlets the latter is showing a lower profit Thus the expenses for the latter need to be minimized to achieve a higher profit from the internal channel.
Thus the factors behind the prediction of supply mix need to be reassessed for a fresh evaluation of the supply mix and the underlying marketing mix. Also, the uniformity in the supply mix predicted is precursor to the uniformity in sales for the different product markets.
References
Armstrong, R., Eperjesi, F. & Gilmartin, B., 2001. The application of analysis of variance (ANOVA) to different experimental designs in optometry. The College Optometrists, pp. 248-256.
Piyanka Jain, 2013. BADIR: Framework to get from Data to Decisions. [Online] Available at: http://www.aryng.com/whitepaper/bgft/BADIR_Framework_Overview.pdf[Accessed 05 May 2016].