Private Markets Fundraising Intelligence

Form D-style fundraising intelligence pipeline for cleaning, auditing, SQL analysis, and Tableau-ready private markets outputs.

Portfolio role

This is a niche finance and research analytics repo. It shows data cleaning, quality checks, SQL analysis, and BI-ready outputs for private markets style fundraising data. It is a strategy analytics proof point, not a founder operating-system module.

The founder problem

Private markets data is messy, fragmented, and hard to trust. The founder problem is turning filings or synthetic market data into credible fundraising intelligence with visible quality checks.

What this repo does

generates or acquires fundraising-style data
cleans and standardizes fields
audits data quality
runs SQL analysis
exports Tableau-ready outputs

What a founder gets in 10 minutes

quality scorecard
capital by asset class
fundraising by quarter
investor concentration outputs
Tableau master dataset
dashboard preview

Before and after

Before:

messy filings
unknown data quality
manual investor-market analysis
hard-to-repeat charts

After:

clean pipeline
quality audit
structured SQL analysis
BI-ready outputs

Who this is for

founders researching private markets
strategy operators
investment research teams
BizOps analysts
students learning data ops

Quick start

Run pip install -r requirements.txt.
Run python3 main.py.
Open outputs/quality_scorecard.json first.
Load outputs/tableau_master.csv into Tableau or a BI tool.

How to fork and use this for your company

Click Fork.
Rename the repo if needed.
Use the synthetic data path first.
Replace the filing source only after the demo pipeline works.
Keep API keys and database credentials in local environment variables.
Document public sources and quality caveats before sharing analysis.

Non-technical path

Run one command: python3 main.py.
Read one output first: outputs/quality_scorecard.json.
Open one BI file: outputs/tableau_master.csv.
Keep private or paid data out of public forks.

Input format

filing or synthetic fundraising records
issuer fields
offering amount
asset class
exemption type
investor count
quarter

The default sample data and examples are synthetic, anonymized, or template-only unless the repo explicitly documents a public source. Keep private customer, prospect, employee, investor, borrower, merchant, payment, or company data out of public forks.

Output files

outputs/quality_scorecard.json: quality audit summary
outputs/tableau_master.csv: BI-ready dataset
outputs/*.csv: analysis outputs
dashboard/dashboard_preview.png: dashboard preview

Example founder workflow

Monday: refresh source data.
Tuesday: run cleaning and quality audit.
Wednesday: inspect outliers and completeness.
Thursday: run SQL analysis.
Friday: turn findings into strategy memo.

Customization guide

Customize these before using the repo for a real company:

source data
quality rules
asset-class mapping
SQL questions
dashboard fields

Where this fits in the Founder OS

This is a strategy and data-ops module. It is useful alongside startup-metrics-playbook when a founder needs market intelligence rather than internal operating metrics.

Why this matters

This is not a chart-only project. It is a data-quality-first intelligence workflow.

Roadmap

SEC EDGAR ingestion hardening
source citation report
dashboard workbook export
market memo template
scheduled refresh option

Contributing

Practical improvements are welcome when they make the workflow easier to fork, run, or adapt. Keep changes focused on the operating workflow and avoid adding private data or mandatory paid dependencies.

License

MIT License. See LICENSE.

Built by

Built by Shubham Singh, a founder-facing operator focused on RevOps, GTM systems, startup metrics, AI workflows, and operating systems for early-stage teams.

Use this in your company

Fork it, replace the sample inputs with your company context, and run the workflow. Start with the main output listed in the Quick Start section. Keep private data out of public forks.

If you are a Founder's Office candidate

Use this repo to understand how a founder-facing operator turns messy inputs into decisions, cadence, and execution artifacts. Fork it, adapt it to a real company example, and write a short case note explaining what changed.

Detailed implementation notes

The founder-facing guide above is the fastest path. The original repo-specific notes are preserved below for deeper implementation context.

Problem This Solves

Private markets data is messy, fragmented, and hard to trust. The problem is not only collecting filings; it is standardising them, auditing quality, and turning them into investor-ready intelligence.

How It Helps

Provides a forkable analytics pipeline for Form D-style fundraising data: acquisition/generation, cleaning, quality checks, SQL analysis, and Tableau outputs.
Demonstrates the data-ops layer behind platforms like PitchBook or Preqin: field completeness, duplicates, outliers, exemption mix, asset-class trends, and fund-size concentration.
Keeps the default demo runnable without paid data access while documenting where synthetic assumptions are used.

When To Fork This

Fork this if you work on alternative assets research, financial data operations, fund intelligence, or regulatory-data analytics.
Fork it when you need a template for turning messy filings into quality-scored, dashboard-ready datasets.
Use the synthetic generator for demos, or wire the acquisition path to your own filing source and MySQL environment variables.

SEC Form D analytics pipeline for private-market fundraising intelligence, with a default 8,500-record synthetic dataset modeled on Form D distributions and an optional SEC EDGAR acquisition path.

Built to replicate how private markets data platforms like Preqin and PitchBook acquire, standardise, quality-assure, and analyse alternative-assets fundraising data for institutional investors.

Use This In Your Company

This repo is designed to be forked into an internal company workflow. Fork it, replace the sample inputs with your company context, and keep only the parts that match your operating cadence. No permission request or sales call is needed before using it; the repo is the handoff. Check the license if you plan to redistribute your version.

Use it as a fundraising-market intelligence pipeline for Form D, private markets, or investment research.
Keep the flow: filing data -> quality checks -> fundraising trends -> Tableau-ready outputs.
Replace sample filings or filters with the market, geography, asset class, or issuer segment you care about.

Minimum Edits To Make It Yours

Change these first:

Edit	Where	Why
Replace or connect the filing source.	`src/01_acquire_data.py` or `src/01_generate_data.py`	Determines whether the pipeline uses synthetic demo data or your own filing source.
Tune standardisation rules.	`src/02_clean_standardise.py`	Keeps issuer, exemption, fund type, state, and amount fields usable.
Adjust quality rules.	`src/03_data_quality_audit.py`	Defines what counts as missing, duplicate, outlier, or low-quality.
Update asset-class and dashboard filters.	`sql/analytics_queries.sql` and `outputs/tableau_master.csv`	Makes the intelligence layer match your investor, market, or research focus.

You can leave the SQL schema, output CSV names, Tableau-ready structure, and dashboard preview alone on the first fork. First swap the source and quality rules; then redesign analytics.

Key Metrics

Metric	Value
Filings Processed	8,500 demo filings modeled on Form D distributions, plus optional live EDGAR acquisition path
Data Quality Score	93.7 / 100 (Grade: A)
Clean Record Rate	92.96%
SQL Queries Written	12 (MySQL - trends, Pareto, exemption analysis)
Quality Rules Applied	9 (duplicates, missing fields, outliers, inconsistencies)

The Problem

Alternative assets - private equity, venture capital, private debt - don't trade on public exchanges. There's no centralised price feed or public registry.

Private markets platforms solve this by systematically acquiring data from regulatory sources, standardising inconsistent fields, and quality-assuring records before delivering them to institutional investors.

Key questions this pipeline answers:

Which asset classes are raising the most capital - and when?
How concentrated is capital across fund sizes?
What exemption types dominate private fundraising?
Where does data quality break down in regulatory filings?

Key Findings

Peak quarter: 2023Q4 - 898 filings, ~$364B in total offerings
Capital concentration (Pareto): Mega funds (>$1B) are 8% of filings but 78% of total capital raised
Dominant exemption: 506(b) accounts for ~60% of filings - standard for institutional fundraises
Investor dynamics: Mid-sized LP groups (11-50 investors) drive the highest total capital - consistent with institutional co-investment clubs
Data quality: 85.8% clean record rate before remediation; 92.96% after applying 9-rule audit framework

SQL Query Results

All 12 analytical queries executed in MySQL Workbench with real outputs.

Query	Preview
01 - Quarterly Fundraising Trends
02 - Asset Class Breakdown
03 - Exemption Breakdown
04 - Fund Size Pareto
05 - Investor Concentration
06 - YoY Trends
07 - Close Rate by Asset Class
08 - Duplicate Detection
09 - Data Quality Distribution
10 - Top Fundraises
11 - Vintage Cohort Analysis
12 - Field Completeness Check

Design Decisions

Synthetic data over direct API pull - SEC EDGAR's Form D XML API requires individual filing-level HTTP requests at scale, which is rate-limited and slow. Synthetic dataset generated using real Form D field distributions from SEC statistics and Preqin market reports - ensuring distributions, fund sizes, and exemption ratios mirror reality.

Rule-based quality flags over ML anomaly detection - data management teams in financial platforms need explainable, auditable rules, not black-box scores. Each flag maps to a specific, documentable business rule that compliance or data ops teams can act on.

Dashboard Preview

Dashboard Link

https://public.tableau.com/views/PrivateMarketsFundraisingIntelligenceSECFormDAnalytics/Dashboard1?:language=en-US&:sid=&:redirect=auth&:display_count=n&:origin=viz_share_link

Tech Stack

Python · Pandas · NumPy · MySQL · Tableau Public

Pipeline:

Form D Structure -> Data Acquisition or Synthetic Generation -> Standardisation -> Quality Audit -> SQL Analysis -> Tableau Dashboard

How to Run

# 1. Clone the repo and install dependencies
pip install -r requirements.txt

# 2. Run full pipeline
python3 src/01_generate_data.py
python3 src/02_clean_standardise.py
python3 src/03_data_quality_audit.py
python3 src/04_sql_analysis.py

# Or run all at once
python3 main.py

# 3. Load outputs/tableau_master.csv into Tableau

Optional MySQL loading uses environment variables from .env.example: MYSQL_HOST, MYSQL_PORT, MYSQL_USER, MYSQL_PASSWORD, and MYSQL_DATABASE. Do not hardcode database credentials in source files.

Real-World Application

Mirrors data infrastructure workflows used by private markets platforms to:

Acquire and standardise fundraising data from regulatory filings
Apply systematic quality frameworks to alternative assets data
Deliver clean, analysis-ready datasets to institutional investors
Surface fundraising trends across asset class, fund size, and exemption type

Applicable to: private markets platforms, financial data operations, alternative assets analytics, and institutional investment research teams.

Part of a founder/operator toolkit for people building practical startup operating systems. ← Back to Profile

Name		Name	Last commit message	Last commit date
Latest commit History 11 Commits
dashboard		dashboard
outputs		outputs
screenshots		screenshots
sql		sql
src		src
.env.example		.env.example
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
main.py		main.py
requirements.txt		requirements.txt

Folders and files

Latest commit

History

Repository files navigation