What are the ARKK ETF constituents?
The ARKK ETF constituents are the securities the ARK Innovation ETF owns on a given date, each with a weight, a share count and a market value. Because the fund is actively managed, that list is a point-in-time record of portfolio manager decisions, not the output of a published index rule set.
The distinction matters for anyone building on the data. An index fund's holdings can be reconstructed from a methodology document and a rebalance calendar. ARKK's holdings can only be known by reading the fund's disclosed positions for a specific date, so the date is as much a part of the record as the ticker.
Any named position, weight or ranking you read about ARKK, including in this article, should be checked against a dated source before you use it. For that reason this article does not reproduce a table of current weights. A weight table pasted into a blog post is stale within days, and a stale table copied into a model is worse than no table.
What follows treats ARKK holdings as a dataset instead: what the fields are, why sources disagree, how to measure concentration, how to identify securities without ambiguity, and how to keep a history you can trust. The same method applies to any actively managed ETF, and the broader background is covered in ETF Constituents: How to Get Holdings Data 2026.
Why do ARKK holdings differ between sources?
Two sources can show different ARKK holdings for the same fund because they describe different dates, different share classes, or different treatments of non-equity lines. Most disagreements are not errors. They are differences in what was captured and when.
As-of dates and refresh timing
A holdings list describes the portfolio at a particular close, but it may be published the same evening, the next morning, or later. A data page may show a "last updated" stamp that reflects when the page was refreshed, not the date of the positions underneath. When you compare two sources, compare the position date first.
A static snapshot also ages quietly. A page that shows positions from many months ago describes a portfolio that may no longer exist, so it is useful as an illustration of a past state and not as a description of the current fund.
Share class and listing
The same strategy can be wrapped in more than one listing, such as a US-listed fund and a differently domiciled or accumulating share class. A reader who lands on a page for a different listing, expecting the US-listed ARKK holdings, would be looking at a different security record, even where the underlying strategy overlaps.
For a quant or a data vendor the lesson is to key every holdings file to a fund identifier and a listing, not to a fund name. Name matching across vendors is where silent mismatches begin. The comparison in ETF Global vs Morningstar: 2026 looks at how vendors differ in positioning, which is a related source of these gaps.
Rounding, cash and derivative lines
Weights are rounded differently across sources, so a column that sums to 100.00 in one file may sum to 99.98 or 100.03 in another. Cash, money market balances, receivables and any derivative exposure may appear as separate rows, be netted into a single line, or be dropped entirely. If you compute concentration from a file with the cash row removed, your top-10 weight will be slightly higher than from a file that keeps it.
Delayed refreshes
A source may refresh its holdings on a lag relative to the issuer's own disclosure. You cannot tell this from the list itself, only from the date field. Coverage and timing across providers are discussed in ETF Holdings Data Availability 2026, which is worth reading before choosing a source for anything that feeds a model.
How concentrated is ARKK, and how should you measure it?
Concentration in ARKK is best measured from a holdings file you recompute at each update: the number of holdings, the top-10 weight, and a dispersion measure such as the effective number of positions. A quoted figure from a web page is a snapshot, and for an active fund it can move materially between updates.
Any top-10 percentage you read on a web page describes the date printed on that page, not today. Treat it as evidence of how top-heavy the portfolio was on that date, and compute the current number yourself from a dated file.
The three measures worth computing
- Number of holdings: a count of distinct securities with a non-zero weight, excluding cash and similar lines. State whether you excluded them.
- Top-10 weight: sort by weight descending and sum the first ten rows. Report whether the denominator is total net assets or equity holdings only.
- Effective number of positions: divide 1 by the sum of squared weights, with weights expressed as fractions. A fund with 40 holdings but a few dominant ones can have an effective number far below 40.
Active share and why recomputation matters
Active share compares the fund's weights with a benchmark's: half the sum of absolute weight differences across all securities in either list. ARKK has no index methodology to serve as the obvious benchmark, so the result depends on the benchmark you choose, and you should record it alongside the number.
Recomputation matters because an actively managed ETF can change share counts at the manager's discretion, and prices change weights even when nothing is traded. A concentration figure stored once and reused is a stale input. Computing it from each new file keeps the measure tied to the position date.
What fields does an ARKK holdings file contain?
A usable ARKK holdings file contains, at minimum, a fund identifier, a position date, a security identifier, a name, a weight, a share count and a market value. The identifier is the field most often underspecified, and it determines whether your history survives corporate actions.
| Field | What it holds | Why it matters |
|---|---|---|
| Fund ticker | The ETF the row belongs to | Keys the row to ARKK, not a similarly named listing |
| Position date | The date the portfolio describes | Separates holdings dates from publication dates |
| Holding ticker | The security's trading symbol | Convenient but not stable over time |
| Holding name | The issuer's name as reported | Useful for review, unreliable as a join key |
| Identifier | A persistent security identifier such as a FIGI | Survives ticker changes and symbol reuse |
| Weight | Position as a share of the fund | Input to concentration and overlap measures |
| Shares | Units held on the position date | Separates trading from price drift |
| Market value | Value of the position on the position date | Reconciles weights to fund assets |
Why tickers alone break
Tickers change when companies rename themselves, merge, or move listings, and a symbol can later be reassigned to an unrelated company. A history keyed on ticker will either split one security into two series or merge two unrelated securities into one. Neither error raises a warning.
AmericanETP includes Bloomberg FIGIs in its constituent list holdings files, which gives you a persistent identifier to join on across dates. That is the practical reason to carry an identifier column from the start, instead of trying to repair a ticker history later.
What the fundamentals file adds
Holdings files tell you what a fund owns. The AmericanETP fundamentals file adds the ETF description, leverage, and the constituent count, which lets you check that the number of rows in a holdings file agrees with the stated count. Leverage is also the field you want when screening a universe, since a leveraged product and an unleveraged one can share a theme but behave differently as inputs to a model.
For a deeper treatment of formats and field conventions, see ETF Constituents Data: Sources and Formats 2026 and ETF Constituent Lists: 2026 Data Guide.
How do you get ARKK holdings as daily data?
There are three routes to ARKK holdings: the issuer's own page, a retail data site, or a delivered CSV file. They differ less in the numbers than in dating, history and machine readability, and those three traits decide which one fits a production workflow.
The issuer page
The issuer is the origin of the disclosure, which makes it the reference for what the fund reported. Its page is built for a person reading it. Whether a given file can be fetched by script, how long history is kept, and what format it arrives in are things to check on the issuer's site directly, since they can change.
A retail data site
Retail pages are convenient for a quick look, and they commonly show a top-10 table, sometimes with a login wall or a cookie banner in front. The common weakness is the date: a page may show a stamp that reflects its own refresh and not the position date. They also rarely offer a clean history you can load into a database.
A delivered CSV file
A delivered file has a fixed schema, a date in the file name or a column, and an archive you can reproduce. AmericanETP provides twice-daily updates, with a primary update at 6pm EST and a secondary update at about noon EST. Files are available through the Downloads page, and FTP access is available for automated retrieval.
The archive of holdings files goes back to July 2009, which matters for any research that needs the portfolio as it stood on a past date. ARKK is a US-traded ETF and so falls within the scope of US-traded ETFs and indexes covered, but you should confirm the ticker in the files before relying on it.
The trade-offs between pulling data through an API and loading files are laid out in ETF Holdings Data API vs CSV Feeds, and the retrieval mechanics are in Daily ETF Holdings Download: 2026 Guide. The cadence question is covered in Daily ETF Holdings Data Update 2026.
How do you track changes in ARKK holdings over time?
You track changes by comparing two consecutive dated holdings files on a persistent identifier, then classifying each row as new, exited, or changed. The key step is separating share count changes, which reflect manager action, from weight changes that come only from price movement.
The comparison structure
Load the earlier file and the later file, and join them on the fund ticker and the identifier column. Use an outer join so that rows present in only one file survive. The working table has these columns:
fund_tickeridentifierholding_namedate_prevanddate_currshares_prevandshares_currweight_prevandweight_currstatus
Set status to new when shares_prev is empty, exited when shares_curr is empty, and otherwise compare share counts. A difference in shares is a trade. No difference in shares with a difference in weight is price drift.
Why the distinction is real
Weight is a ratio, so it moves when any holding's price moves. As a piece of arithmetic, if a position is 5% of a fund and its price rises 10% while everything else is flat, the position becomes 5.5 against a total of 105.5, or about 5.21%. The share count did not change, and no decision was made. A tracker that reads weight changes as trades will produce false signals.
Why archives matter
A diff only works if the earlier file exists in its original form. If you overwrite yesterday's file, or rely on a site that shows only the current list, you lose the ability to reconstruct what the fund held on a past date. That matters for backtests, where using today's list for a past date introduces look-ahead bias, and for point-in-time research and audit. Archived files back to July 2009 allow that reconstruction.
For a prebuilt view of changes, AmericanETP also has a Constituent Change Report page. Whatever you use, keep the raw dated files, since derived reports can always be rebuilt from them and the reverse is not true.
Where does ARKK sit among thematic and global ETFs?
ARKK sits in the thematic, actively managed part of the US-traded ETF universe, and the useful way to place it is by comparing its holdings with those of other funds. The aim is to learn whether ARKK adds exposure you do not already hold, or repeats it.
The method is mechanical. Take the holdings files for ARKK and for a set of comparison funds, and compute the overlap by summing the minimum weight of each shared security. A high overlap with a broad growth fund means ARKK is mostly a concentrated tilt of something you already own. A low overlap means it carries distinct positions.
If you want to group the overlap by sector, classification is the weak point. A thematic fund's own label may not match the sector scheme applied to its holdings. Fix one classification scheme before comparing funds, and say which one you used. ETF List by Sector: 2026 Classification Guide walks through the differences.
AmericanETP covers 3878 US and global indexes and US-traded ETFs, so overlap checks can run across a wide set from one provider and one identifier scheme. The reference list in Global ETF List: Top Funds by Holdings 2026 is a starting point for choosing comparison funds. As a limit, overlap tells you about shared securities and not about shared risk, since two funds with no common holdings can still respond alike to the same factor.
What does ARKK holdings data cost to license?
AmericanETP prices by subscription: $1500 per year or $150 per month for an individual, and $2500 per year or $250 per month for firm-wide use. The subscription covers the holdings data as a whole, not ARKK alone.
| Licence | Annual | Monthly |
|---|---|---|
| Individual | $1500 | $150 |
| Firm-wide | $2500 | $250 |
The firm-wide option is the relevant one for teams where several analysts, a risk group and a engineering group all read the same files. Pricing for other vendors should be checked on their own pages on the day you evaluate them, since it changes and it is often tied to scope. As a comparison starting point without any figures, see ETF Global Pricing: 2026 Cost Guide and ETF Global Review 2026: Data Buyers.
If your question is about scope or a use not covered here, the Contact page is the route to ask.
FAQ
What stocks make up ARKK?
ARKK holds a set of equities chosen by its portfolio managers, and the set changes over time, so any list is valid only for its position date. Read the holdings from a dated file and record the date alongside it. Positions and weights quoted in articles, including a top-10 table on a retail page, should be treated as snapshots.
What new ETF did Cathie Wood launch?
That depends on the date you ask, and the answer lives in the issuer's own fund list. From a data standpoint, a new fund appears as a new ETF record, with its own description, leverage flag and holdings count in a fundamentals file. Check the issuer's current product page and compare it with your own ETF universe to see what is new.
Which ARK ETF is the best performing?
Performance rankings depend on the period and the measure, and they are not part of a holdings dataset. A holdings file shows what a fund owned, not what it returned. Take returns from a dated performance source, state the window, and be wary of any ranking that does not name one.
Can you provide a review of the ARK Innovation ETF (ARKK)?
From a data perspective, ARKK is an actively managed, concentrated fund whose holdings change at the manager's discretion. That makes recomputed concentration, share count tracking and archived history the key tools for assessing it. This article does not assess investment merit, which depends on a mandate and a time horizon you must set yourself.
Where can I download daily ETF holdings for ARKK?
You can use the issuer's page, a retail data site, or a delivered CSV file. For repeatable work, a delivered file with a stated position date and an archive is the better fit, and AmericanETP provides this with twice-daily updates and FTP access. The mechanics are covered in Daily ETF Holdings Download: 2026 Guide.
How often do ARKK holdings change?
The fund can change positions on any trading day, but not every day produces a change, and weights shift daily with prices even when shares are unchanged. The reliable method is to diff consecutive dated files and separate share count changes from price drift. A twice-daily update schedule lets you see changes close to when they are reported.
To see how twice-daily holdings files, fundamentals and archives back to July 2009 fit your workflow, visit AmericanETP.