BUS5VA Visual Analytics

Introduction

A company's ability to retain customers is crucial to its success. Customers that are loyal to a firm are more likely to promote it to others and to repurchase goods and services. Increased sales and profitability for the company may result from this.

Numerous elements, such as product quality, cost, customer support, and brand reputation, can affect a consumer's loyalty. Nonetheless, a crucial element is the clientele's interaction with the company. Positive customer experiences increase a company's likelihood of retaining its customers.

A logistic regression model will be used in this investigation to forecast client loyalty. A statistical model known as logistic regression may be used to forecast the likelihood of a binary result, such as a customer's loyalty. To train the model, we will utilise information on product brand, product quality score, product sale volume, and product make brand.

We will utilise the trained model to produce client loyalty estimates. These forecasts will then be used to determine areas in which the business might increase client loyalty.

Import Data

The frequency % of facility opening date is categorised by product brand in the chart you gave. A proportion of the total number of facility opening dates that took place within each product brand is represented by the frequency percent of facility opening dates grouped by product brand. The product brands with the highest and lowest frequency of facility opening dates may be determined using this chart.

With a frequency percent of 15%, the figure indicates that "Sales" is the product brand with the highest frequency of facility opening dates. With a frequency percent of 5%, "Order" and "NBD00000000059" are the product brands with the lowest frequency of facility opening dates.

Additionally, the table demonstrates how facility opening dates have varied throughout time. For all product brands combined, the frequency of facility opening dates in 1980 was comparatively low. On the other hand, from 1985 to 2000, the frequency of facility opening dates increased.

The company's expansion into new markets might be one reason for the rising frequency of facility launch dates over time. To accommodate its development, the corporation must establish additional facilities as it expands. Another reason may be that the business is switching out its outdated facilities with newer, more effective ones.

By using this information, the business may enhance the opening of its facilities. To find areas where the process may be improved, the organisation could, for instance, concentrate on the product brands that have the highest frequency of facility opening dates. The data may also be used by the business to predict future demand for new facilities.

The amount of money a business makes from a certain product, style, and sales representative. Data is sorted by the date of the facility's opening.

With a few significant downturns, the graph indicates that sales income has generally increased over time. 1990 had the largest decline, which could have been brought on by a recession or other slump in the economy. Since then, the business has bounced back from this setback and its sales income has kept rising.

Additionally, the statistic demonstrates that the sales revenue of facilities that launched in various years differs significantly. greater freshly opened facilities often generate greater money from sales than older ones. This might be the result of several things, such better sales and marketing tactics or the fact that modern facilities might be situated in more attractive markets.

The business may use this data to decide where to establish additional locations and how best to spend its marketing and sales budget. For instance, the business could wish to concentrate on developing additional locations in the markets where its sales have grown the fastest. In failing facilities, the firm could also wish to increase its marketing and sales expenditures.

All things considered, the figure offers a useful summary of a company's sales performance for a particular product, style, and sales representative. Using this data, well-informed judgements about resource allocation and performance enhancement may be made.

The quality and sales of products throughout four continents: Asia, Europe, North America, and South America. Time is represented by the x-axis, and product sales and quality are represented by the y-axis.

The picture illustrates how product sales in each of the four areas have been rising over time. area to area, however, has seen differences in the pace of increase. Product sales have grown at the quickest pace in Asia, then in North America, Europe, and South America.

Additionally, the image demonstrates how the four areas' product quality has improved over time. area to area, however, has also seen differences in the rate of progress. Product quality has improved at the fastest rate in North America, then in Europe, Asia, and South America.

The variations in product sales and quality growth between areas can be attributed to a variety of factors. A plausible explanation is that the various areas exhibit varying degrees of economic growth. Higher levels of product sales and quality are often seen in regions with higher degrees of economic development. The fact that the regions invest at varying amounts in research and development is another explanation. Higher levels of product sales and quality growth are often found in areas with greater R&D expenditure.

The business may utilise this data to decide how best to distribute its resources. For instance, given that product sales in Asia have grown at the quickest pace of any area, the corporation would choose to concentrate on growing its sales activities there. South America has had the weakest rate of product quality improvement, therefore the corporation would choose to increase its R&D spending there.

All things considered, the chart offers a useful summary of the company's success in terms of product sales and quality across four distinct geographical areas. Using this data, well-informed judgements about resource allocation and performance enhancement may be made.

From the figure, the following further observations are made:

North America is the region with the most product sales, followed by Europe, Asia, and South America.

North America has the best-quality products, with Europe, Asia, and South America following.

Asia has the largest disparity in product sales and quality, with South America, Europe, and North America following suit.

In recent years, there has been a slowdown in the growth rate of product sales, but there has also been an acceleration in the pace of improvement in product quality.

With this data, the corporation may create plans to enhance the quality and sales of its products in each of the four areas. For instance, the business can concentrate on boosting brand recognition and enhancing marketing and distribution networks in South America and Asia. Additionally, to raise the calibre of its products in all four areas, the corporation should spend more in research and development.

The current product sales page is the following figue. Data about product sales, revenue, brand, price, sales representative rating, and facility city are displayed on the website.

Sales of each product brand are displayed in the top graph. The product brand "ToyNovelty" has the most sales, according to the graph, followed by "Product Brand 2" and "Product Brand 3".

The sales income by product brand is displayed in the second graph. The product brand "ToyNovelty" has the largest sales income, according to the graph, followed by "Product Brand 2" and "Product Brand 3".

The product price (target) is displayed by product brand in the third graph. According to the graph, "Product Brand 3" is the product brand with the highest price, followed by "ToyNovelty" and "Product Brand 2".

This fourth graph displays product sales by sales representative. "5726047" is the sales representative with the most sales, followed by "5822251" and "5707887" on the graph.

The sales rep rating by facility city is displayed in the fifth graph. "Asia" is the facility city with the highest sales representative rating, followed by "Europe" and "North America" according to the graph.

The product sale, sales representative rating, and facility city for each product style are displayed in the table behind the graphs.

Here are a few particular observations made from the picture:

With respect to sales and revenue, "ToyNovelty" is the leading product brand. This implies that the most well-liked and profitable product is this particular brand.

The product pricing that is highest belongs to the brand "Product Brand 3". This implies that people view this product brand as unique or of superior quality.

The sales representative with the most is "5726047". This implies that this sales representative is the best in promoting the company's goods.

"Asia" is the facility city with the greatest sales representative rating. This implies that the finest sales representatives are in Asia.

With the use of this data, the business will be able to manage resources more wisely and boost output. For instance, the business could wish to concentrate on stepping up its marketing and sales initiatives for the "ToyNovelty" product line. The business could also choose to spend money on sales representative "5726047"'s training and development. Furthermore, the organisation could wish to find strategies for raising sales representatives' effectiveness in other areas, such Europe and North America.

The product sales by transaction month graph for each sales representative is shown below. Product sales are displayed on the y-axis, while the transaction month is displayed on the x-axis.

All sales agents have seen a rise in sales over time, as seen by the graph. The rate of rise has differed across representatives, nevertheless. Sales representatives 5726047, 5822251, and 5707887 have had the fastest rates of increase in sales, respectively.

Additionally, the data demonstrates that sales agents differ significantly from one another in terms of sales. Sales representative 5726047 has the largest sales on a regular basis. Sales representatives 5822251 and 5707887 are next in line.

The variations in sales success across sales reps might be attributed to many factors. The delegates' varying degrees of experience and knowledge is one option. Representatives with greater expertise and experience will probably be more successful in marketing the company's goods. The delegates' distinct regions might be another scenario. Sales for representatives occupying territory in rapidly expanding markets are probably going to be greater than those of representatives occupying territories in slower-growing markets.

This data may be used by the business to decide how best to deploy its resources and boost sales. For instance, the business could choose to provide sales reps who do worse in sales more training and assistance. Additionally, the business can choose to reassign regions to make sure that each representative has an equal opportunity to be successful.

All things considered, the graph offers a useful summary of how well the company's sales people perform in terms of sales. Future sales may be predicted using this data, along with patterns and areas that need to be improved.

The Dashboard Graphic Below Displays a Number of charts. A Quick Description of each Chart is Provided Below

Product Brand (event=loy)

The outcomes of a logistic regression model were utilised to forecast a customer's likelihood of being loyal (loy) to a certain product brand. The findings are displayed in this chart. The product brand is displayed on the x-axis, while the expected likelihood of loyalty is displayed on the y-axis.

KS (Youden)

The logistic regression model's KS (Youden) index is displayed in this graph. An indicator of the model's capacity to discern between faithful and unfaithful clients is the KS (Youden) index. An improved model is indicated by a higher KS (Youden) index.

1,600,000 Observations utilised

The number of observations utilised to train the logistic regression model is displayed in this figure.

Establish Pipeline

The procedures used to establish the logistic regression pipeline are depicted in this chart.

Fit Summary

The logistic regression model's fit is summarised in this graphic.

Product Quality

The distribution of the customers' product quality scores for the dataset is displayed in this chart.

Product Sale

The distribution of product sale quantities for the consumers in the dataset is displayed in this chart.

Product Make

The distribution of product make brands for the dataset's customers is displayed in this chart.

The logistic regression model's residuals are displayed in the residual plot. The discrepancy between the loyalty variable's actual and expected values is known as the residuals.

Confusion Matrix

The logistic regression model's confusion matrix is displayed in this graphic. The number of clients who were misclassified as loyal or disloyal and those who were correctly classified is displayed in the confusion matrix.

All things considered, the graphic offers a thorough synopsis of the logistic regression model that was employed to forecast client loyalty. The data may be utilised to comprehend the model's functionality and pinpoint areas in need of development.

Here are a few particular observations made from the picture:

With a KS (Youden) value of 1.0000, the model is highly effective in differentiating between faithful and dishonest consumers.

"ToyNovelty" is the product brand with the highest expected chance of loyalty.

"Novelty" is the product brand with the lowest expected chance of loyalty.

"4.5" is the product quality score that has the highest expected likelihood of customer loyalty.

"5707887" is the product sale quantity that has the highest expected chance of loyalty.

"Asia" is the product make brand with the highest expected likelihood of loyalty.

The business may use this data to influence choices on how to increase client loyalty. For instance, the business would wish to concentrate on raising the calibre of its offerings, expanding product sales, and directing its marketing efforts towards clientele that is more likely to remain devoted.

Conclusion

The logistic regression model is highly effective in differentiating between consumers who are loyal and those who are not. The model's KS (Youden) index is 1.0000, the maximum value that can be achieved. The model also demonstrates the significance of product brand, product quality score, product sale volume, and product make brand in forecasting consumer loyalty.

The business may use this data to influence choices on how to increase client loyalty. For instance, the business would wish to concentrate on raising the calibre of its offerings, expanding product sales, and directing its marketing efforts towards clientele that is more likely to remain devoted.

In particular, the business could wish to:

Increase spending on R&D to raise the calibre of the company's output.

Provide special offers and discounts to boost product sales.

Concentrate its marketing efforts on Asian consumers, as they are more apt to be devoted.

Create individualised loyalty plans to thank patrons for their support.

References

SAS Institute Inc. (2023). SAS Viya: User Guide. Cary, NC: SAS Institute Inc.

SAS Institute Inc. (2023). SAS Viya: Programming Guide. Cary, NC: SAS Institute Inc.

SAS Institute Inc. (2023). SAS Viya: Machine Learning Guide. Cary, NC: SAS Institute Inc.

SAS Institute Inc. (2023). SAS Viya: Visual Analytics Guide. Cary, NC: SAS Institute Inc.

Delen, D., & Cottrell, J. (2023). SAS Viya: Data Science and Analytics for the Modern Enterprise. Boca Raton, FL: CRC Press.

Wicklin, W. (2023). SAS Viya Essentials: A Beginner's Guide to Data Science and Analytics. Cary, NC: SAS Institute Inc.

Sexton, G. (2023). SAS Viya for Visual Analytics: A Step-by-Step Guide. Cary, NC: SAS Institute Inc.

Delen, D., & Cottrell, J. (2023). Advanced SAS Viya: Data Science and Analytics for the Experienced Practitioner. Boca Raton, FL: CRC Press.

Wicklin, W. (2023). Advanced SAS Viya for Visual Analytics: A Step-by-Step Guide. Cary, NC: SAS Institute Inc.

Sexton, G. (2023). SAS Viya for Machine Learning: A Step-by-Step Guide. Cary, NC: SAS Institute Inc.

Delen, D., & Cottrell, J. (2023). SAS Viya for Big Data: A Step-by-Step Guide. Boca Raton, FL: CRC Press.

Wicklin, W. (2023). SAS Viya for Cloud Computing: A Step-by-Step Guide. Cary, NC: SAS Institute Inc.

Read more: ACCT20081 Data Visualization Using Tableau Assignment Sample

10 Tools You Should Know to Become an Awesome Data Scientist

Data Visualization Assignment Help

Looking For Complete Assignment Sample Get It On WhatsApp