Operational Workforce Data Analysis

Analysis of worker productivity, task completion, and quality scores across AI data projects (Image Labeling, OCR Data Prep, Speech Recognition).

Data Source

Google Sheet: Operational Workforce Tracker

Dataset

data/raw_operational_data.csv - daily worker logs with:

• Date, Project, Worker ID/Name
• Task Type (Labeling, QA Review, Data Cleaning, Transcription)
• Hours Worked, Task Status, Quality Score

Analysis Method

Data was cleaned and aggregated using Google Sheets QUERY functions to compute project-level effort and worker productivity metrics.

Clean Data (filter completed tasks with quality scores):

=QUERY(Raw_Data!A1:H,
  "SELECT A, B, C, D, E, F, H
   WHERE G = 'Completed'
   AND H IS NOT NULL", 1)

Total Hours by Project:

=QUERY(Clean_Data!A1:G,
  "SELECT B, SUM(F)
   GROUP BY B
   ORDER BY SUM(F) DESC
   LABEL B 'Project Name',
         SUM(F) 'Total Hours Worked'", 1)

Total Hours by Worker:

=QUERY(Clean_Data!A1:G,
  "SELECT D, SUM(F)
   GROUP BY D
   ORDER BY SUM(F) DESC
   LABEL D 'Worker Name',
         SUM(F) 'Total Hours Worked'", 1)

Screenshots

1. Clean Data

Filtered dataset showing only completed tasks with valid quality scores.

2. KPI Summary

Aggregated metrics - total hours by project and by worker.

3. Dashboard

Visual overview of workforce productivity and project distribution.

Repository Structure

data/               → Raw CSV data
screenshots/        → Dashboard and analysis visuals
README.md           → This file