Introduction:

This was a project for my IT Data Analyst position. The data is about Salesforce and Whatfix users. The main point of this analysis was to determine the number of Whatfix vs. Non-Whatfix users for things like Department, Role, Primary Organization, etc. There are 2 worksheets--the first contains 1,400+ records for only Salesforce users and the second contains 1,200+ records for only Whatfix users.

​​

Insights:

There were many important insights to uncover from analyzing this data:

• What's the number of Whatfix vs. Non-Whatfix users by department?

• What's the number of Whatfix vs. Non-Whatfix users by date created?

• What's the number of Whatfix vs. Non-Whatfix users by role?

• What's the email, department, parent organization, and role of only Whatfix Users?

​​​​

Tools:

This project was analyzed using Excel, SQL, Tableau, and Python:​​

• Excel was used because it was requested by my supervisor. I started by creating a column called 'Whatfix User?' in the Salesforce worksheet to identify whether a user was a Whatfix user or not. I did this by using XLOOKUP to look up the Salesforce user's email in the Whatfix worksheet; and if the email was found, then that Salesforce user would also be a Whatfix user. Then I created 7 pivot tables to answer key questions and compiled them in a dashboard on Excel.

• SQL was used because I wanted to analyze the data further to derive more insights. I started by cleaning the 2 worksheets which included changing column names, dropping columns, changing data types, adding new columns, updating blank values, and standardizing the data. Then I outlined a total of 11 questions and wrote queries for each one. The clauses used included: SELECT, FROM, WHERE, GROUP BY, SUM, COUNT, HAVING, ORDER BY, SUBQUEIRES, WINDOW FUNCTIONS, INNER JOIN.​

• Tableau was used because I wanted to visualize my analysis from the SQL queries and explore the data even more to identify trends and patterns not discovered using Excel and SQL. I started by importing the 2 Excel worksheets. Then I created 3 worksheets with visuals comprised of pie charts and bar graphs. One of these visuals used a parameter and calculated field to alternate between visuals; this was very useful because instead of creating 5 different bar graphs, the user could just use a drop-down menu to switch between these 5 visuals. Then I compiled these 3 worksheets into a dashboard.

• Python was used because it was requested by my supervisor for further analysis. The data was already cleaned, so I started analyzing it. I outlined a total of 10 questions and used Pandas to answer them and Matplotlib to visualize them. The main Pandas syntax used were groupby, sort_values, and date. The main Matplotlib visuals used were bar graphs and stacked column charts.​​​​

​

Impact:

The impact from this analysis was stakeholders were able to identify the distribution of Whatfix vs. Non-Whatfix users across departments, roles, and organizations. This made it possible for stakeholders to potentially tailor training programs and increase adoption where gaps exist. Understanding user trends over time allowed for better forecasting and planning of resources. The detailed breakdown of Whatfix user attributes (email, role, department, etc.) equipped teams to target specific users for feedback or feature rollouts, enhancing user satisfaction and retention. Furthermore, the visualizations and dashboards offered a clear, interactive way to monitor adoption metrics and refine strategies for improving Salesforce and Whatfix integration effectiveness.​​