ANLY FPX 5510 Assessment 3 Comparison of Analytic Methods

Student Name

Capella University

ANLY-FPX5510 Advanced Business Analytics

Prof. Name:

Date

Comparison of Analytic Methods

The marketing department of a large financial services organization recently experienced a budget cut and reached out to the analytics team to maximize the efficient use of the budget. The analytics team was tasked with analyzing the communication strategies used with different customer groups to ensure that prospective customers are not being contacted in ways that lead to wasted resources on individuals not interested in opening accounts with the company.

Compare

The analytics team presented two possible solutions. The first solution involved performing a cluster analysis using data that indicated whether customers were being advertised to through various channels, including telephone, email, Facebook, Twitter, Instagram, Pinterest, Yahoo, Google search, and US Postal Service mail. The goal of this analysis was to identify groups of individuals with similar behaviors and use this information to make decisions on reducing the number of communication channels the financial services organization uses for marketing. Essentially, the analysis aimed to determine which communication methods customers preferred. Based on the analysis results, it was recommended to continue advertising through phone, mail, and Google while discontinuing the use of other communication forms.

The second solution involved conducting a decision tree analysis. This analysis had similar intentions to the cluster analysis, using data that showed whether customers were being advertised to through telephone, email, Facebook, Twitter, Instagram, Pinterest, Yahoo, Google search, and US Postal Service mail. Additionally, this analysis incorporated demographic information and the total amount of deposits made by customers in the previous year. The objective was to identify characteristics that could help the marketing department focus its efforts on high-deposit customers. Based on the results, it was recommended that advertising via Instagram continue for customers younger than approximately 43 years old. For customers aged 56 to 80, advertising should be discontinued as this group had a lower percentage of high deposit amounts in the last year.

Appropriateness of Analytic Methods

Cluster analysis is a set of techniques used to classify objects or cases into groups, referred to as clusters. This method is commonly applied in marketing to identify similar groups of buyers. The process of cluster analysis involves several steps: defining the problem, selecting a distance measure, choosing a clustering procedure, deciding the number of clusters, understanding the cluster profiles, and finally assessing the validity of the clustering. In the case of the business problem described here, the use of cluster analysis is appropriate, as it helped identify groups of customers with similar behavior and determine the best marketing and advertising strategies (Statistics Solutions, 2019).

Decision tree analysis is a graphic representation of various alternative solutions to a problem and is often used in organizational decision-making processes. When making a decision, management predicts alternative ideas and solutions, which are represented graphically in a decision tree to simplify the decision-making process (Mulder, 2017). In the context of the business problem discussed, decision tree analysis was a suitable method for researching data based on specific criteria. The decision tree analysis was used to gather data on particular customer groups to guide advertising efforts, allowing the marketing team to focus on high-deposit customers. The analysis resulted in a recommendation to continue advertising via Instagram for younger, high-profile customers while discontinuing this advertising for older customers.

Recommendations

The analysis solution I recommend for addressing the business problem is the first solution, the cluster analysis. This approach focuses on the type of communication rather than specific customer groups. The analysis evaluated various advertising channels, including telephone, email, Facebook, Twitter, Instagram, Pinterest, Yahoo, Google search, and US Postal Service mail, to determine which channels were most effective in reaching customers and which were not as efficient and costly for the company. In contrast, the decision tree analysis focused on customer demographics and aimed to exclude certain age groups from receiving advertising. Additionally, cluster analysis offers several advantages, such as being less expensive than other methods. It is more economical to observe clusters of units in a population than to randomly select units scattered across a state. Furthermore, cluster analysis allows for the accumulation of larger samples, enabling the company to gather information from multiple areas (Money Matters, 2019).

Implementation

Introducing analytics to an organization is not a one-time task but rather a shift towards a new business approach. As innovations become more customized to customer, product, and operational needs, analytics-driven incentives will play a crucial role in processes. To implement the results from the cluster analysis, a centralized approach will be employed. In this strategy, all associated assets, such as software, hardware, skilled resources, metadata, business relationships, and project management capabilities, are managed under one organization. This approach offers several benefits, including knowledge sharing, consistent implementation and support, visibility, and clear lines of communication. It also reduces costs through shared hardware and software across multiple projects. The first step in this strategy is introducing the process and data, specifically the cluster analysis. The associated team will then be trained on the methods and data. Following this, an engagement, project management, and development methodology will be built, ensuring the framework is established before any real projects commence (Sheikh, 2013). Once the data and processes are implemented, it is crucial to maintain and monitor the use of the cluster analysis to ensure it is applied effectively.

References

Statistics Solutions. (2019). Conduct and interpret a cluster analysis. Retrieved from https://www.statisticssolutions.com/cluster-analysis-2/

Mulder, P. (2017). Decision tree analysis. Retrieved from https://www.toolshero.com/decision-making/decision-tree-analysis/

Money Matters. (2019). Cluster sampling | Definition | Advantages & Disadvantages. Retrieved from https://accountlearning.com/cluster-sampling-definition-advantages-disadvantages/