Coverage
What you get
| Item | Detail |
|---|---|
| Product code | STK-ETF |
| Coverage | US-listed ETFs |
| Contents | Holdings with weights, fund profiles (issuer, expense ratio, net assets, NAV, holdings count) and daily flows |
| Sources | Issuer holdings files and SEC N-PORT filings; every row says which |
| Granularity | Holdings per ETF per date; flows daily |
| History | Latest holdings on Starter; holdings history and daily flows on Pro |
| Delivery | REST and CSV or Parquet files |
| Update speed | Daily |
Owning SPY, QQQ and a tech sector fund can mean holding the same five stocks three times. Holdings data tells you that. Flow data tells you when money is piling into or out of a fund, which drives the creations and redemptions that move the underlying stocks.
Schema
Schema
Holdings record:
| Field | Type | Required | Description |
|---|---|---|---|
symbol |
string | yes | ETF ticker |
as_of_date |
string | yes | Holdings date |
holding_symbol |
string | optional | Ticker of the held security |
holding_name |
string | yes | Name of the held security |
holding_isin |
string | optional | ISIN |
asset_category |
string | optional | equity, bond, cash, derivative |
shares_held |
number | optional | Quantity held |
market_value |
number | optional | Value in fund currency |
weight_percent |
number | yes | Percent of net assets |
source |
string | yes | issuer_file or n_port |
ETF profile records carry symbol, name, issuer, inception_date, expense_ratio, net_assets, nav, shares_outstanding, dividend_yield, holdings_count, is_leveraged and daily_flow (net creations in USD).
Example record (illustrative values):
{
"symbol": "QQQ",
"as_of_date": "2026-11-18",
"holding_symbol": "NVDA",
"holding_name": "NVIDIA Corp",
"holding_isin": "US67066G1040",
"asset_category": "equity",
"shares_held": 151204882,
"market_value": 22122784280,
"weight_percent": 8.94,
"source": "issuer_file"
}
Sample data
Sample data
No free ETF holdings sample yet. A sample file is available on request: name up to a few ETFs and we'll send their latest holdings in this schema.
Use cases
Use cases
Overlap between funds. Pull holdings for two ETFs on the same as_of_date, join on holding_symbol, and sum the smaller weight_percent of each shared stock. That number is the overlap you're paying two expense ratios for.
Look-through exposure. Multiply each ETF position in your account by its holdings' weight_percent and add them up by holding_symbol. Your largest real single-stock exposure is often not what you expect.
Flow tracking. On Pro, watch daily_flow for sector ETFs. Several days of large inflows into a sector fund mean creation baskets, which means buying in the underlying stocks.
Rebalance studies. Compare holdings history before and after a rebalance date. Names that get added or dropped show up as new or missing holding_symbol rows.
Code
Code examples
Example: latest QQQ holdings.
curl "https://api.marketdata.co/v1/etf/holdings?symbol=QQQ" \
-H "Authorization: Bearer $MARKETDATA_API_KEY"
Example: Python, overlap between two ETFs.
import os
import requests
import pandas as pd
def holdings(etf):
r = requests.get(
"https://api.marketdata.co/v1/etf/holdings",
params={"symbol": etf},
headers={"Authorization": f"Bearer {os.environ['MARKETDATA_API_KEY']}"},
)
r.raise_for_status()
return pd.DataFrame(r.json()["results"]).set_index("holding_symbol")["weight_percent"]
a, b = holdings("QQQ"), holdings("SPY")
both = pd.concat([a, b], axis=1, join="inner")
print("Overlap %:", both.min(axis=1).sum().round(2))
Pricing
Pricing
| Plan | ETF data |
|---|---|
| Starter, $29/mo | Profiles and latest holdings for US ETFs |
| Pro, $79/mo | Adds holdings history and daily flows |
| Business, $249/mo | Commercial use |
Dataset: full US ETF holdings history at $99 per year of data. Among the providers we studied, ETF data costs $9.99 to $500 a month, median $99, and one gives it away.
Side by side
How we compare
15 of the 50 providers we studied sell ETF data, and the typical price clusters around $99 a month. Holdings are public through N-PORT filings and issuer files, so we start at $29. Every row says whether it came from the issuer or N-PORT, which matters because N-PORT is quarterly and lags by about 60 days. Holdings use the same tickers as our stock bars, so look-through analysis is a single join.
FAQ
Frequently asked questions
How current are the holdings?
It depends on the source. Issuer files are usually daily. N-PORT filings are quarterly and published with a delay of about 60 days. The source field and as_of_date tell you which one you are looking at.
Do you cover leveraged and inverse ETFs?
Yes. Profiles include is_leveraged, and holdings for those funds often show swaps and futures with asset_category set to derivative.
How are flows calculated?
daily_flow is the day's net creations minus redemptions, in USD. A positive number means money came into the fund, a negative one means it left.
Do you have mutual funds?
Not in this product. It covers US ETFs only.
Is it real-time?
No. Holdings and flows update daily. Intraday ETF prices are part of our stock quotes and bars.
What formats can I get?
JSON from the API, and CSV or Parquet for the history files.
Keep reading