07 · Research automation & commercial intelligence
Private-Company Intelligence Workflow
A practicum workflow that structures fragmented private-company research into a validated database, scoring model, and decision-ready outputs.
Executive summary
A UC Davis MSBA practicum engagement focused on making private-company research more systematic: collect and standardize public and commercially available signals, store them in a relational structure, and support qualification review with transparent scoring.
Problem
Private-company information is fragmented across company, project, geography, licensing, and qualification sources, making consistent research and comparison time-intensive.
Approach
- 01Define a shared company schema across business identity, projects, geography, licensing, and qualification signals.
- 02Build Python collection and standardization workflows with validation checkpoints for incomplete and inconsistent records.
- 03Design a SQL database that keeps company, project, location, licensing, and scoring entities reviewable.
- 04Create an Excel scoring model and stakeholder outputs that support prioritization while preserving analyst review.
Data & methodology
Data sources
- Public and commercially available private-company research signals
- Company, project, geography, licensing, and qualification fields
- Analyst-reviewed validation and scoring inputs
Methods
- Python collection and standardization
- Relational SQL schema design
- Data validation and exception review
- Transparent Excel scoring logic
Public case-study language is limited to verified résumé evidence and excludes all client-specific data and confidential deliverables.
Outputs & findings
Key outputs
- Structured research workflow covering 1,000+ private companies
- Relational company and project database design
- Excel-based qualification scoring model
- Decision-ready research and stakeholder outputs
What the work demonstrates
- Expanded a repeatable research workflow to 1,000+ private construction companies.
- The current résumé reports an estimated approximately 70% reduction in manual research effort.
Limitations
- No client records, company-level scores, contacts, or confidential work product are included in this portfolio.
- The efficiency figure is a workflow estimate, not a controlled productivity study.
- Scoring supports research prioritization and does not replace analyst diligence or professional judgment.