Data Cleaning & Quality Assurance: Transforming Raw Product Data in Excel
Data Cleaning & Quality Assurance: Transforming Raw Product Data in Excel
Embedded Files
A data cleaning project using Excel to prepare dataset for further analysis
A data cleaning project using Excel to prepare dataset for further analysis
Before
Before
After
After
This project demonstrates essential data cleaning workflows using Microsoft Excel, completed as part of the DQLab Introduction to Data Analytics Mini Bootcamp (Batch 5). The primary goal was to transform a messy, raw product dataset into a structured, consistent, and analysis-ready format.
This project demonstrates essential data cleaning workflows using Microsoft Excel, completed as part of the DQLab Introduction to Data Analytics Mini Bootcamp (Batch 5). The primary goal was to transform a messy, raw product dataset into a structured, consistent, and analysis-ready format.
Key Tasks & Methodology
Key Tasks & Methodology
- Duplicate Removal
Identified and eliminated redundant entries to ensure each product record was unique.
Identified and eliminated redundant entries to ensure each product record was unique.
- Structural Cleaning
Removed unnecessary blank rows to streamline the dataset layout.
Removed unnecessary blank rows to streamline the dataset layout.
- Data Consistency
Standardized product category formatting to eliminate naming variations and ensure accurate grouping.
Standardized product category formatting to eliminate naming variations and ensure accurate grouping.
- Data Validation & Logical Verification
Enforced a predefined business rule requiring all product prices to fall within the range of Rp 50,000 to Rp 1,000,000.
Enforced a predefined business rule requiring all product prices to fall within the range of Rp 50,000 to Rp 1,000,000.
During the validation phase, I identified an anomaly that was not recorded in the video: a product price that significantly exceeded the Rp 1,000,000 threshold.
During the validation phase, I identified an anomaly that was not recorded in the video: a product price that significantly exceeded the Rp 1,000,000 threshold.
Instead of blindly filtering it out, I applied logical reasoning to investigate the root cause. By analyzing the product context, I determined it was a human typographical error (an extra zero had been appended during data entry). I successfully resolved the issue by correcting the typo, preserving a valuable data point that would have otherwise been lost or skewed the analysis.
Instead of blindly filtering it out, I applied logical reasoning to investigate the root cause. By analyzing the product context, I determined it was a human typographical error (an extra zero had been appended during data entry). I successfully resolved the issue by correcting the typo, preserving a valuable data point that would have otherwise been lost or skewed the analysis.
Key Takeaways
Key Takeaways
- Data Integrity Matters
Data cleaning is not just a technical routine; it requires keen attention to detail and logical reasoning. A single misplaced digit can severely distort subsequent analytical insights.
Data cleaning is not just a technical routine; it requires keen attention to detail and logical reasoning. A single misplaced digit can severely distort subsequent analytical insights.
- Garbage In, Garbage Out
Ensuring rigorous data quality at the foundational stage is critical to producing reliable, trustworthy data models and business conclusions.
Ensuring rigorous data quality at the foundational stage is critical to producing reliable, trustworthy data models and business conclusions.
Page updated
Google Sites
Report abuse