Make data-backed quick decisions for business and revenue growth using the Tableau dashboard.

The Business Problem at Hand:

The client team had no means to observe multiple divisions and LOBs data in a single dashboard, so multiple LOBs and Division teams were working in silos.


How Eucloid solved the problem:

1. By creating a single Tableau dashboard (Cross-Divisional) collecting data of multiple divisions (like Traffic, CX, and E-commerce) and LOBs (like Mobile, Display, and Memory) with aggregated tables in the Vertica database.

2. Using a Cross-Divisional dashboard, we helped multiple LOB teams to see entire relevant data in a single Tableau dashboard, resulting in data-backed strategic business decisions.

3. By automating the weekly Refresh process of the Dashboard using the Airflow tool to get used updated data in the Cross-Divisional dashboard.


Results Achieved:

1. Integrating all LOBs and divisions data into one Cross-Divisional dashboard so that each LOB team can make decisions based on how metrics for other LOBs are performing at the same time.

For example: With increased focus on e-comm and assisted ecomm more and more customers are completing high worth transactions online. So, the CX team can decrease their focus from Lead Gen Form completes to Continuation Rate to help customers journey through the pages and complete ecomm transactions (Inverse relationship between CR and Lead gen.

So as CR increases, more people shop in Ecomm, and fewer people fill out the lead gen form). Hence, CX and Ecomm teams can work hand in hand using data from each division resulting in increased revenue.

2. 80% Increase in Revenue for the Display team for going ecomm focused and less focus on lead gen.


Technical Implementation:

There were some complexities (listed below) and how we resolve them while creating Data for the dashboard.

1. Huge Volume of data: There were Million plus rows for a single day hence it was not possible to transform and load data in a table with a single query. To overcome this issue, we wrote optimized queries and fetched transformed data in intermediate tables, and on top of intermediate tables, we created one aggregated table for the dashboard.

2. Excluding Support and Bot Traffic: Some KPI numbers (like sales form completion, and continuation rate) were not matching with source data, on the analysis we figured out that it was due to support and bot traffic. Hence, we modified logic and queries to exclude support and bot traffic from traffic numbers to get accurate KPI numbers.

3. Productionization of Dashboard: For productionizing of the dashboard, we wrote parameterized queries for a weekly refresh of data and a shell script to deploy it in production. With the help of the Airflow job scheduler, the weekly job is scheduled to refresh the latest data in our Tableau Dashboard.

Looking for Data solutions for your business? We can help. Reach out at

About the Author

Ashish Gupta

The author is the Lead Data Engineer of Eucloid. For any queries, reach out to him on

LinkedIn LinkedIn