Healthy Poke aimed to reduce manual tasks associated with extracting and consolidating data from various sources, including Revo, Gstock, Deliverex, and Excel files.
Prior to implementing the solution, the company spent significant time centralizing and transforming sales data to generate informative pivot tables.
The proposed solution automated these processes, creating an information schema in a cloud warehouse and facilitating the development of an interactive dashboard for agile monitoring of key business metrics. This resulted in a significant reduction in data preparation times and manual errors while enabling easy incorporation of new data sources.
Before Implementation
The company dedicated considerable time to the manual collection and transformation of data, hindering agile analysis.
Handling Excel files and manually entering data increased the risk of errors.
The lack of a centralized system made it difficult to incorporate new data sources and create new metrics, limiting opportunities.
Key Improvements
An automated system was implemented for extracting, transforming, and loading (ETL) data, drastically reducing times.
The creation of a centralized database allows for storing information from different sources.
The development of a data portal provides intuitive and accessible visualizations of key business metrics, improving decision-making.
Project Overview
1.
Process Review
Held meetings with Healthy Poke to understand their manual data processes and key reporting needs, focusing on reducing time spent on manual data consolidation from Revo, Gstock, and Excel.
2.
Solution Design
Designed an automated ETL system using Google Cloud Functions, Python, and BigQuery to centralize data from multiple sources, ensuring scalability for future needs.
3.
ETL Implementation
Automated data extraction and transformation from Revo and Gstock, using Python for data cleaning, adding columns, and transposing data, then ingested into BigQuery for further analysis.
4.
Portal implementation
Built an interactive data portal to provide real-time insights, enabling flexible analysis through filters for period, location, cluster, and category, replacing manual reporting with agile, automated access to business metrics.
Python
BigQuery
Google Cloud Platform
Looker
Technologies used
Juan Uribe
CEO - Healthy Poke
"This project has transformed how we manage our data. Tasks that used to take hours are now automated, saving us time and improving accuracy. The interactive dashboard gives us real-time insights into our operations, making decision-making faster and more informed. The scalability of the solution also allows us to grow effortlessly, adding new data sources and metrics as needed."
