Bit 470 Week 4

BIT-470
Howard
Analytics Life Cycle In the data collection and manipulation of Sport T’s database, there are nine questions regarding the data that would be useful in improving company standards. What percentage of buyers purchase multiple items in one checkout? How much is the average full purchase from a customer? How many customers only purchase discounted items as opposed to full retail priced items? What is our Year-to-Date profits over expenses? What is our customer satisfaction ratio to unsatisfied customers? What is our average click-to-hit ratio on our online and social media advertisements? What is our average time from purchase to processing an order? What is our average time from processing an order to shipment? What is our average

ETL is a three-step process which stands for Extract-Transform-Load. This process comprises of: extracting the desired data from a source, transforming the extracted data into a specific format, and loading the transformed data into a destination such as a data warehouse (Haag & Cummings, 2013). After the ETL process is performed, data-mining tools can be used to turn this data into useful information. For the first three questions, the database would need to capture each checkout price, how many items are purchased, the individual price of each item, and if the item is discounted or full MSRP. This specific data will likely originate from a customer oriented database that will then flow into the data warehouse for full ETL. For YTD profits, the database would need to capture all purchases, sales, profits, and expenses from the current year. Sport T’s company data will originate from an in-company database which focuses on business expenses and profits. In solving customer satisfaction, the KPIs to consider would be survey questions and answers from responding customers as well as customer opinion on what can be improved. For customer surveys, we will ask