Data Analysis with KPMG

Sprocket Central Pty Ltd., a fictional, medium size bikes & cycling accessories organization, had approached Tony Smith (Partner) in KPMG’s Lighthouse & Innovation Team. Sprocket Central Pty Ltd was keen to learn more about KPMG’s expertise in its Analytics, Information & Modelling team.

Smith discussed KPMG’s expertise in this space. In particular, he spoke about how the team can effectively analyze the datasets to help Sprocket Central Pty Ltd grow its business.

Primarily, Sprocket Central Pty Ltd. needed help with its customer and transactions data. The organization had a large dataset relating to its customers, but their team was unsure how to effectively analyze it to help optimize its marketing strategy.

The client provided KPMG with 3 datasets:

🟦 Customer Demographic

🟦 Customer Addresses

🟦 Transactions data in the past 3 months

I decided to start the preliminary data exploration and identify ways to improve the quality of their data.

Sprocket Central Pty Ltd then gave us a new list of 1000 potential customers with their demographics and attributes. However, these customers did not have prior transaction history with the organization. The marketing team at Sprocket Central Pty Ltd was sure that, if correctly analyzed, the data would reveal useful customer insights which could help optimize resource allocation for targeted marketing. Hence, improve performance by focusing on high value customers.

Using the existing 3 datasets (Customer demographic, customer address and transactions) as a labelled dataset, my role was to recommend which of these 1000 new customers should be targeted to drive the most value for the organization. In building this recommendation, we needed to start with a PowerPoint presentation which outlined the approach which we decided to take.

I prepared a detailed approach for completing the analysis including activities – i.e. understanding the data distributions, feature engineering, data transformations, modelling, results interpretation and reporting. This detailed plan needed to be presented to the client to get a sign-off.

The client was happy with the analysis plan. After building the model we needed to present our results back to the client. Visualizations such as interactive dashboards often help us highlight key findings and convey our ideas in a more succinct manner. A list of customers or algorithm won’t cut it with the client, we needed to support our results with the use of visualizations.

This Project was done as a part of Forage's Virtual Experience Program called KPMG Data Analytics Consulting Virtual Internship. All of the datasets used are not real, and the company Sprocket is fictional as well.