Survey data architecture
Total Rewards analytics
Turn inconsistent survey exports into connected views of what employees value, how satisfied they are, and where benefits priorities diverge.
Preserve the question behind the column
Survey exports mixed labels, question text and encoded values. The work focused on recovering the meaning of each field before building a shared model, including MaxDiff topics embedded in HTML-like headers.
Separate survey types, connect their meaning
Importance/satisfaction, preference ranking and other response types need different structures. The proposed Alteryx branches and Power BI relationships keep those distinctions while allowing comparison by topic and industry.
Make a benefits conversation possible
The intended output connects benefits priorities to employee experience rather than collapsing every answer into one score. The archive supports the analytical design and implementation work, with production acceptance still unverified.
Scope & evidence
Recovered designs and partial implementations cover normalization, importance versus satisfaction and MaxDiff. A complete production rollout is not established by the archive.
SourcesAbacus archive · survey data architecture · ChatGPT archive · Total Rewards
- Abacus archive · survey data architecture
- Cleaning strategy, Alteryx workflow and Power BI model templates.
- ChatGPT archive · Total Rewards
- Recovered project context for survey analysis and benefits reporting.