VoteFlow
Turning live scoring sessions into structured, comparable data.
ByteGrind designed and built VoteFlow, a multi-tenant web platform that captures scores from facilitated sessions, enforces a review workflow, benchmarks results against historical pools, and generates client-ready reports.
Session results lived in disconnected files.
Facilitated scoring sessions produce a large amount of structured data: who was scored, by which group, against which questions, and how the votes were distributed. Managing that by hand made results slow to consolidate and difficult to compare.
The workflow also needed controls that files cannot enforce: a review step before results count, segmented benchmarking against past sessions, minimum vote thresholds, and reports that always follow the client's template.
Project Snapshot
- Score capture
- 0–10 vote distributions per question
- Session lifecycle
- Draft → In Progress → Submitted → Approved → Archived
- User roles
- Admin, Reviewer, Captor, Viewer
- Benchmarking
- Rolling historical comparison pools
- Stack
- TypeScript, PostgreSQL, background jobs
One platform from capture through approved report
Structured Score Capture
Each session records complete 0–10 vote distributions for every question, tied to a specific client, group, question set, and session date.
Review and Approval Workflow
Sessions move from draft through submission, review, and approval or rejection, with a documented reason and full history at each step.
Historical Benchmarking Pools
Results compare against rolling historical windows across segmented comparison pools, with minimum vote thresholds before a comparison is reported.
Bulk Imports with Dry-Run Validation
Historical data imports run in validation mode first, reporting imported, skipped, duplicate, and error rows before anything is committed.
Client-Ready Report Generation
Background jobs generate downloadable reports from client-specific templates and track each job from queue through completion.
Multi-Tenancy, Roles, and Audit Logs
Each organization's data stays isolated, access follows defined roles, and administrative changes are recorded in an audit trail.
Production software that protects the integrity of the data.
VoteFlow runs as a production system on a TypeScript API and PostgreSQL database with versioned migrations, background job processing, and a release-based deployment pipeline. Administrators manage clients, groups, and question sets; captors record sessions; reviewers approve results; and reports generate on demand.