Introduction:

This was a project for my IT Data Analyst position. The data is about Salesforce and Marketing Cloud URLs that employees at the University of Arizona have visited. The main point of this analysis was to extract the keywords in these URLs to determine what users are visiting the most via Salesforce and Marketing Cloud. These keywords were mostly Salesforce objects such as Contact, Account, Case, etc; or they were Marketing Cloud pages such as Exact Target, Dashboard, Cloud, etc. It consists of 3 worksheets--the first contains 37,000+ records for Salesforce URLs, the second contains 10,000+ records for Marketing Cloud URLs, the third contains 1,500+ records for employee data.

​​

Insights:

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

• Which pages are users visiting the most on Salesforce and Marketing Cloud?

• What's the count of pages visited by type of role (staff, supervisor, student worker)?

• What pages is each user visiting?

• What's the count of pages visited over time by year, month, and day?

​​​​

Tools:

This project was analyzed using Excel and Tableau:​​

• Excel was used because it was the tool requested by my supervisor. I started by using Power Query to split the URL column based on the '/' delimiter. I also created a complex nested IFS function with ISNUMBER and SEARCH to ensure the keywords were split correctly. Then I deleted over half the columns from the split operation and used TEXTJOIN to combine multiple keywords together for each record (some records had only 1 keyword while others had upwards of 5). Then I used Power Query again to unpivot this newly created keyword column because I needed to normalize the data because currently it didn't meet the requirements for 1NF because the column didn't have atomic values (there were multiple words separated by a comma which needed to be changed). Now there was multiple records for each URL because that URL could have more than 1 keyword in it. Then I used XLOOKUP to create a new column (Role) which linked to the third sheet, the employee data. Then I used an IFS function to categorize these roles into Staff, Supervisor, Student Worker, or Unknown. Lastly, I repeated this exact same process for the Marketing Cloud URLs worksheet. All this was done so I could import this workbook into Tableau for presentation.

​

• Tableau was used because I needed to visualize the data that was processed in Excel for presentation to stakeholders. I started by importing the Excel workbook with the 3 worksheets as 1 data source. I created a total of 8 worksheets with only bar graphs. Then I created 2 dashboards and combined that into 1 story. This was presented to a couple key stakeholders interested in what users were visiting on Salesforce and Marketing Cloud.​​​​

​

Impact:

The impact from this analysis was stakeholders were able to make data driven decisions to allocate resources to the most visited pages and also determine what kinds of roles were visiting the most pages. Moreover, time-based visit trends (year, month, day) empowered management to identify seasonal patterns or usage spikes, which can inform staffing, system upgrades, or targeted communications.
​​