Database Driven Decisions for Business (Research Paper Sample)

Module Name Database Driven Decisions for Business Programme Name Student Reference Number Assessment Title Data Driven Decisions for Business (CW1) report Declaration of Original Work: I hereby declare that I have read and understood BPP’s regulations on plagiarism and that this is my original work, researched, undertaken, completed and submitted in accordance with the requirements of BPP School of Business and Technology. The word count, excluding contents table, bibliography and appendices, is ___2465___ words. Student Reference Number: __________ Date: ______ Contents TOC \o "1-3" \h \z \u 1.0.Introduction and Project Plan PAGEREF _Toc195571179 \h 3 1.2. Value of Data Analytics to S&S Austen Bank PAGEREF _Toc195571180 \h 4 2.0. Data Quality Issues and Data Analysis PAGEREF _Toc195571181 \h 5 2.1. Data Quality Issues and Remedies PAGEREF _Toc195571182 \h 5 2.2. Exploratory Data Analysis Summary PAGEREF _Toc195571183 \h 6 2.3. Data Tables and Commentary PAGEREF _Toc195571184 \h 7 3.0. Data Charting and Commentary PAGEREF _Toc195571185 \h 10 3.1. Chart 1: Deposit Value Trends by City PAGEREF _Toc195571186 \h 10 3.2. Chart 2: Customer Volume & Deposit Value by Service Type PAGEREF _Toc195571187 \h 11 3.3. Chart 3: Oxford Branch Before vs After Renovation (Trend Chart) PAGEREF _Toc195571188 \h 12 4.0. Conclusions and Recommendations PAGEREF _Toc195571189 \h 13 4.1. Response to Management Questions PAGEREF _Toc195571191 \h 13 4.2. Strategic Recommendations PAGEREF _Toc195571192 \h 14 4.3. Future Use of Data Analytics PAGEREF _Toc195571193 \h 14 Reference List PAGEREF _Toc195571194 \h 16 Introduction and Project Plan As a newly appointed data analyst within the Retail Banking division of S&S Austen Bank, my role is to support the organisation’s strategic decision-making with data analytics. S&S Austen Bank, a long-standing institution in the UK’s financial landscape, is facing a rapidly evolving banking environment where digital services are gaining popularity, yet traditional high-street branches still play a critical role—especially among the older population (Ballou, Heitger and Stoel, 2018). In light of this, the bank’s top management seeks to determine whether the current retail branch model can support its future growth ambitions. My specific responsibility is to analyse the performance of three major branches—located in Cardiff, Oxford, and Edinburgh—to provide evidence-based insights and guide future decisions. This report aims to examine the trends in deposit values and customer volumes, assess the performance of different savings services, and evaluate the impact of a recent renovation of the Oxford branch. Ultimately, the goal is to determine the best location for opening new branches and inform marketing strategy decisions that focus on high-performing services. The paper comprises four primary analysis areas. Section 1 explains project goals and introduces the analytics framework to guide analysis. Section 2 lists data quality issues and provides extensive data tables that summarize significant performance patterns across cities and service categories. Section 3 shows consumer and deposit patterns throughout time, especially branch Oxford before and after its restoration. Section 4 concludes with a summary of the full report's key findings and bank-strategic recommendations. CRISP-DM (Cross Industry Standard Process for Data Mining) will guide this project through the following phases: Business Understanding (defining project goals and requirements), Data Understanding (analysing raw data), Data Preparation (cleaning and structuring data), Modelling (applying descriptive analytics techniques), Evaluation (evaluating results for bank goals), and Deployment. CRISP-DM is a structured and repeatable method that helps S & S Austen Bank use data to improve customer service, branch operations, and expansion decisions. 1.2. Value of Data Analytics to S&S Austen Bank S&S Austen Bank relies heavily on data analytics to improve decision-making and strategic planning. In a highly competitive and digitising financial world, insights gleaned from customer and operational data allow the bank to make educated, timely, and evidence-driven choices (Yu et al., 2024). With comprehensive data analysis, the bank is able to identify and recognize patterns in customer behaviour, service consumption, and regional performance variances, allowing it to better match expansion goals with genuine market needs. For example, how client deposits fluctuate over time and branches differ in scoring performance may inform judgments when expanding, promoting services, or whether remodelling in a newer branch yielded returns on investment (Khan and Saha, 2025). Thus, it enables the bank to transition from reactive decision-making to more proactive and predictive approaches, which can be critical for long-term viability and client happiness. To assess and improve performance, S&S Austen Bank can implement a few key performance indicators (KPIs). These include average deposit value per customer, which measures the depth of the customer relationship; customer volume trends, which indicate the effectiveness of customer acquisition and retention strategies; and top-performing savings services based on deposit value, which guide resources and targeted marketing (Yu et al., 2024). Other factors include quarterly deposit growth by branch, service utilization rates, and branch-specific income contribution. With the correct data analytics tools and frameworks in place, such KPIs can provide fine-grained insights into operational performance (Engel et al., 2022). These measures, when examined on a regular basis, can also provide early warning signals of potential issues or opportunities, allowing S&S Austen Bank's leadership to pivot their initiatives as necessary. 2.0. Data Quality Issues and Data Analysis 2.1. Data Quality Issues and Remedies The quality of data is the most crucial for the success of any analytic endeavour. Some common generic errors with data in business datasets include missing values, duplicate records, inconsistent formats for data, and outliers. Missing values occur due to data entry errors, system processing limitations, or incomplete interface with a customer, leading to some Analysts not being able to analyse data (Keskar, Yadav and Kumar, 2022). Duplicate records usually arise in situations where data from multiple sources are merged. They inflate results and distort averages. Inconsistent formats, like date structures or currency symbols, can disrupt time-series analysis and automated reporting. These errors should be identified and corrected as they can lead to the invalidation of results and hence the investigation, especially when giving weight to strategic decisions. In the course of the preliminary data review performed in relation to the S&S Austen Bank dataset, certain specific problems were noted. Included were some null values in deposit amount fields, especially with respect to children's savings and ISAs; there were also inconsistencies with respect to date format across entries, some being formatted as 'MM/DD/YYYY' and others as 'DD-MM-YYYY'.. All these issues underwent a systematic data cleaning procedure (Chaimaa, Najib and Rachid, 2020). For small gaps, null values for deposits were substituted using mean imputation, while some records with excessive missing data were omitted from the analysis so that the integrity of the dataset remains intact. Duplicates were then identified based on combinations of unique customer ID and timestamp and removed. Date fields were harmonized into a consistent format across records using a date parser. Outliers were placed under context assessment—these were kept so long as they looked to be valid (e.g., seasonal peaks), while extreme anomalies were either capped or excluded (Villar and Khan, 2021). Equipped with these cleansing tools, we can get a reliable dataset of very high quality for the final analytical stages of the project. 2.2. Exploratory Data Analysis Summary There are trends and statistics that underpin the very basic customer behaviour and branch performance. General depends mostly on the descriptive statistics calculated on the entire dataset. The average monthly customer traffic across all branches has been a bit over 1,200, with some seasonal variability (Wong and Wong, 2021). Monthly deposits averaged £3.8 million, with a standard deviation of £600,000, which indicates the degree of variation in deposits is more on the moderate side. Among the savings services, individual savings accounts (ISAs) and bonds savings accounts had the highest average deposit values, whereas children savings accounts logged the lowest, indicating that these services have differing effects on the bank's financial aspirations (Villar and Khan, 2021). Some of this information might serve as a rough benchmark for performance assessment across different branches or services. Trend analysis during this three-year period (2021-2023) showed several performance characteristics. The timing of the peak months as far as deposit values are concerned turned out to be March and October, as historically they are noted to coincide with seasonal financial events like tax planning and year-end savings (Wong and Wong, 2021). Conversely, January and August showed lower deposit volumes, possibly due to holiday-related expenditure and decreased financial activity. On an annual basis, 2023 demonstrated the strongest deposit growth, particularly in the Oxford branch, likely reflecting the positive impact of its mid-year renovation. The Edinburgh branch maintained steady performance throughout, while Cardiff displayed moderate volatility, especially in 2022. These trends provide valuable context for identifying high-performing branches and services, as well as assessing the effectiveness of operational changes like branc...