A Python toolkit for analyzing Microsoft Viva Glint survey data exports.
Version: 0.1.0 | Last Updated: March 2026 | License: MIT
vivaglint simplifies the analysis of Microsoft Viva Glint survey exports by providing a complete toolkit for data import, validation, statistical analysis, and reporting. It handles the repetitive data wrangling that Glint's native UI doesn't support — multi-cycle trend analysis, manager roll-ups, demographic segmentation, attrition risk scoring, and comment search.
All processing happens locally within your Python environment. No employee data is transmitted to any external service, including Microsoft.
Ported from R? This is a direct Python port of the vivaglint R package. All 14 exported functions have identical names, parameter names, and default values. R code translates to Python with minimal changes — see PACKAGE_USAGE.md for a side-by-side comparison.
pip install vivaglintDependencies: pandas, numpy, scipy, factor_analyzer, rapidfuzz, matplotlib, seaborn
from vivaglint import read_glint_survey, summarize_survey
# Load and validate your Glint CSV export
survey = read_glint_survey("survey_export.csv", emp_id_col="Employee ID")
# Summarise all questions
summary = summarize_survey(survey, scale_points=5)
print(summary)| Capability | Functions |
|---|---|
| Import & Validate | read_glint_survey, extract_questions, join_attributes |
| Reshape | pivot_long, split_survey_data |
| Core Analytics | summarize_survey, get_response_dist, compare_cycles |
| Advanced Analytics | get_correlations, extract_survey_factors, search_comments |
| Segmentation | analyze_by_attributes, analyze_attrition |
| Hierarchy | aggregate_by_manager |
- QUICK_REFERENCE.md — function signatures and common patterns
- PACKAGE_USAGE.md — detailed usage guide with R-to-Python translation notes
- FUNCTION_OUTPUTS.md — output column reference for every function
- vignettes/vivaglint-intro.ipynb — end-to-end Jupyter notebook walkthrough
- man/ — per-function reference pages (build with Sphinx)
See CONTRIBUTING.md. Please report security issues via SECURITY.md.
