📈 IPO Performance Tracker

An end-to-end data pipeline that collects, parses, and analyzes U.S. IPO performance from 2019–2025.

---

What It Does

1. Scrapes IPO listings (offer price, ticker, date) from stockanalysis.com
2. Finds and downloads official S-1 filings from SEC EDGAR
3. Parses structured fields from S-1 text: financials, underwriters, VC history
4. Fetches post-IPO stock price performance at 30, 90, and 180 days
5. Loads everything into a normalized SQLite database for analysis

---

Data Sources

Source	What We Collected	Method
stockanalysis.com	IPO list, offer price, ticker	Web scraping
SEC EDGAR	S-1 filings: financials, underwriters, risk factors	Web scraping
yfinance	Post-IPO stock returns at 30/90/180 days	Python library

---

Dataset Overview

Raw scrape:              2,655 IPOs
After removing SPACs:    1,723 IPOs   (932 shell companies removed)
After price filter:      1,675 IPOs   (48 with no IPO price removed)
After yfinance check:    1,198 IPOs   (477 delisted or unavailable removed)
Final with full data:      702 IPOs   (have 90-day return data)

SPACs (Special Purpose Acquisition Companies) were excluded because they are shell vehicles with no operating financials to analyze. 2020–2021 were peak SPAC years, accounting for most of the 932 removed.

---

Database Schema

5-table SQLite schema:

ipos (main table)
  ├── financials         → revenue, net income, profitability
  ├── vc_history         → founding year, funding rounds, total raised
  ├── price_performance  → 30d, 90d, 180d stock returns
  └── underwriters       → banks that underwrote the IPO (many-to-one)

Row counts after full pipeline:

Table	Rows
ipos	1,191
financials	866
underwriters	2,648
vc_history	1,036
price_performance	702

---

Key Findings

IPO Performance by Year

Year	IPOs	Avg 90-Day Return	Notes
2019	124	-0.3%
2020	159	+45.8% 🚀	Pandemic stimulus + low rates
2021	319	-23.2%	Peak SPAC era, rate hike fears begin
2022	88	-39.0% 📉	Aggressive Fed rate hikes
2023	106	-28.3%
2024	167	-22.5%
2025	228	-4.9%	Partial year

Top Underwriters by Deal Count

Bank	Deals	Avg 90-Day Return
Goldman Sachs	283	+1.2%
Morgan Stanley	280	-1.4%
J.P. Morgan	239	+1.8%
Jefferies	143	+12.0%
Barclays	131	+7.1%

Jefferies and Barclays outperformed the bulge-bracket banks on 90-day returns, which may reflect more selective deal sourcing at the mid-tier level.

---

Project Structure

ipo_tracker/
│
├── scrapers/
│   ├── master_list.py          # Phase 1: scrape IPO list from stockanalysis.com
│   ├── edgar_scraper.py        # Phase 2: find S-1 accession numbers on EDGAR
│   ├── fetch_s1_html.py        # Phase 2: download and cache S-1 HTML files
│   ├── parse_s1.py             # Phase 2: extract structured fields from S-1 text
│   ├── vc_history_from_s1.py   # Phase 3: extract VC/funding history from S-1 text
│   └── fetch_prices.py         # Phase 4: pull post-IPO stock price data
│
├── database/
│   ├── schema.sql              # table definitions
│   ├── db.py                   # connection helper
│   └── load_data.py            # loads cleaned CSVs into SQLite
│
├── notebooks/
│   ├── clean_master_list.ipynb # Phase 1: cleaning and validation
│   ├── data_quality.ipynb      # Phase 5: QA and cross-table checks
│   └── ipo_analytics.ipynb     # exploratory analysis and visualizations
│
├── data/
│   └── cleaned/                # cleaned CSVs (raw/ is gitignored due to size)
│       ├── ipo_master.csv
│       ├── ipo_master_validated.csv
│       ├── s1_accessions.csv
│       ├── s1_parsed.csv
│       ├── vc_history.csv
│       ├── price_performance_clean.csv
│       └── ipo_analysis_ready.csv   ← final dataset
│
├── .env.example                # template for API keys and config
├── requirements.txt
└── logs/                       # runtime logs (gitignored)

---

Setup

1. Clone the repo

git clone https://github.com/vedantt17/ipo_tracker.git
cd ipo_tracker

2. Create environment and install dependencies

python -m venv .venv
source .venv/bin/activate        # Mac/Linux
.venv\Scripts\activate           # Windows

pip install -r requirements.txt
playwright install chromium

3. Configure API keys

cp .env.example .env
# edit .env and add your Alpha Vantage key

4. Run the pipeline in order

python scrapers/master_list.py        # Phase 1: build IPO universe
python scrapers/edgar_scraper.py      # Phase 2: find S-1 accessions
python scrapers/fetch_s1_html.py      # Phase 2: download S-1 files
python scrapers/parse_s1.py           # Phase 2: parse structured fields
sqlite3 database/ipo_tracker.db < database/schema.sql   # initialize DB
python database/load_data.py          # Phase 2: load CSVs into SQLite
python scrapers/vc_history_from_s1.py # Phase 3: extract VC history
python scrapers/fetch_prices.py       # Phase 4: pull price data

5. Open the notebooks

jupyter notebook

---

Challenges

Challenge	What Happened	Fix
Crunchbase blocked	Bot detection blocked all scraping attempts	Extracted VC data from S-1 text instead
SEC homepage cached	Fetcher saved SEC homepage instead of actual S-1	Fixed link finder to only accept /Archives/ paths
CIK formatting bug	CIK numbers stored as floats broke EDGAR lookups	Converted with str(int(float(cik))).zfill(10)
UBS false positives	3-letter string matched in 983 wrong filings	Removed UBS from underwriter parser, deleted bad rows
stockanalysis.com range	404 errors for years before 2019	Extended scrape forward to 2025 instead

---

For academic research purposes only.