Robotics ETF alternative to build your own robotics stocks index and backtest it against BOTZ and ROBO
One popular robotics fund keeps 10% of its money in companies that build robots. Pick the factory, warehouse, surgical and humanoid robot makers you actually want, set the weights, and see how the basket would have done next to the funds and the S&P 500. Then buy it at your own broker.
Pick 2 to 10 holdings. Stocks, ETFs and crypto mix in one index.
The backtest did not run just now. Tap Backtest it to try again.
Educational only · Never places a trade
In short
A robotics ETF alternative is a basket of robotics stocks you weight yourself instead of taking a fund's rule. It matters because "robotics" funds own very different things. On issuer holdings files dated October 9 to 12, 2026, Global X's BOTZ held 74.7% in industrial, surgical and humanoid robot makers, First Trust's ROBT held 27.5% and put 46.9% in AI software, and ARK's ARKQ held 10.5%, with Tesla and SpaceX as its two largest positions. Building your own robotics stocks index means choosing that mix, backtesting it, and buying the stocks at a broker with no expense ratio.
Summed from each fund's published holdings, October 9 to 12, 2026
How much of each robotics ETF actually builds robots?
Every holding sorted by what the company mainly sells. The first three colors are robot makers: factory and warehouse automation, surgical robots, and humanoid robots with the joints and drives inside them. Grey is chips and AI software, amber is cars and self driving. The empty part of each bar is everything else, from rockets to gene sequencers.
Themes BOTT
79.8% robots
Humanoid index since August 2025, but 40% of it trades in China
Global X BOTZ
74.7% robots
The classic robotics fund. Five names are 41% of it
Roundhill HUMN
66% robots
Most humanoid exposure, plus Tesla at 6.3%
ROBO Global ROBO
65.5% robots
Spread across 79 names, none above 2%
KraneShares KOID
53.9% robots
Equal weight, heavy on parts and chip suppliers
First Trust ROBT
27.5% robots
Mostly AI software, led by Illumina and Meta
ARK ARKQ
10.5% robots
42% in space, defense, drones and nuclear names
Start with ARKQ, because the word robotics is in its name. On October 12, its two largest positions were Tesla at 11.2% and SpaceX at 8.9%, followed by Nvidia, Palantir and Teradyne. Add the drone makers, defense contractors, eVTOL startups, Deere and two nuclear names, and 41.9% of the fund sat in businesses that are not robots, chips or software. Only 10.5% was in companies whose main business is building robots. That is a legitimate active bet on autonomy in general. It is not a robotics fund in the sense most buyers mean.
ROBT went a different way. Its index weights companies close to equally across AI and robotics categories, so its largest holdings were Illumina, Recursion, Meta, Ocado and SentinelOne, each about 2%. Almost half the fund, 46.9%, was AI and software companies. BOTZ is the one fund where the robot makers lead: Intuitive Surgical 8.9%, Fanuc 8.3%, Keyence 8.1% and ABB 7.8%, next to Nvidia at 8.0%.
There is one more thing worth knowing about Nvidia. The S&P 500 fund SPY held 8.2% of it on October 9, more than any robotics ETF. If you already own an index fund, a robotics fund adds factory and surgical robot makers, not more Nvidia. None of these rules is wrong. They are seven answers to what a robotics investment should hold, and a basket of your own lets you give yours.
From each issuer's fund page and holdings file
Which robotics ETF holds what, and how does each one pick its stocks?
| Fund | Expense ratio | Holdings | Rule | Largest positions (Oct 9 to 12) |
|---|---|---|---|---|
| Global X Robotics & AI (BOTZ) | 0.68% | 81 | Indxx Global Robotics & AI Thematic, size weighted with an 8% cap | Intuitive Surgical 8.9%, Fanuc 8.3%, Keyence 8.1%, Nvidia 8.0%, ABB 7.8% |
| ROBO Global Robotics and Automation (ROBO) | 0.95% | 79 | ROBO Global index, weighted by a 1 to 100 robotics score | Intuitive Surgical 2.0%, Fanuc 1.8%, Illumina 1.8%, Fuji 1.7%, Celestica 1.7% |
| ARK Autonomous Technology & Robotics (ARKQ) | 0.75% | 36 | Actively managed, no index | Tesla 11.2%, SpaceX 8.9%, Nvidia 5.8%, Palantir 5.2%, Teradyne 5.1% |
| First Trust Nasdaq AI and Robotics (ROBT) | 0.65% | 111 | Nasdaq CTA AI and Robotics, modified equal weight | Illumina 2.0%, Recursion 2.0%, Meta 1.9%, Ocado 1.9%, SentinelOne 1.8% |
| KraneShares Humanoid Robotics (KOID) | 0.69% | 50 | MerQube Global Humanoid Robotics and Physical AI, equal weight | Regal Rexnord 2.2%, Renesas 2.2%, Amphenol 2.2%, Cognex 2.2%, THK 2.2% |
| Roundhill Humanoid Robotics (HUMN) | 0.75% | 41 | Actively managed, no index | Tesla 6.3%, Harmonic Drive 5.5%, Unitree (swap) 5.2%, Rainbow Robotics 4.5%, Robotis 4.3% |
| Themes Humanoid Robotics (BOTT) | 0.35% | 30 | Solactive Global Humanoid Robotics, rank weighted, 4.5% cap | Yujin Robot 5.5%, Teradyne 5.5%, Nvidia 5.4%, Rainbow Robotics 5.4%, Cymechs 5.1% |
Size matters for trading costs. BOTZ ($3.3 billion), ROBO ($2.0 billion) and ARKQ ($1.9 billion) are large and trade with tight spreads. ROBT held $789 million and KOID $355 million. The humanoid funds are young and small: HUMN had $91 million and BOTT $65 million. BOTT is the cheapest robotics ETF at 0.35%, but its published median spread was 0.30% against 0.03% for BOTZ. One fund you may still see on old lists, iShares' IRBO, no longer exists as a robotics fund: iShares renamed it the Future AI & Tech ETF (ARTY) in 2024 and moved it to an AI index.
The funds overlap less than their names suggest. BOTZ and ROBO share 28 companies but only 28.4% of their money, because ROBO spreads its weight thinly. The closest pair is the two big humanoid funds: KOID and HUMN share 22 companies and 39.4% of their weight. ROBT and BOTT, both sold as robotics, share just 4.7%. The same fund families show up on our AI stocks index builder, which compares BOTZ, ROBT and ARKQ with the AI and chip ETFs.
// The universe, five groups
Which stocks belong in a robotics stocks index?
Robot makers sell to very different customers. Factory automation follows car and electronics capital spending, warehouse robots follow e-commerce, surgical robots follow hospital budgets and procedure counts, and humanoid robots are still mostly a research and pilot market. The share you give each group decides what your index actually rides on.
1
Factory robots and automation
2
Warehouse and logistics robots
3
Surgical and medical robots
4
Humanoid robots and robot brains
5
Foreign robot makers, OTC or abroad
This is the real constraint on building your own. The largest industrial robot makers are Japanese and Swiss: Fanuc, Yaskawa, Keyence, Harmonic Drive, Nabtesco and ABB. In a typical US brokerage account they trade over the counter as ADRs or foreign shares, if at all, and their dividends carry foreign withholding tax. The humanoid makers are mostly in China and Korea. So a US-listed robotics index is a choice, not a copy: Intuitive Surgical, Rockwell, Teradyne, Cognex, Symbotic and Zebra are the core, with Tesla and Nvidia as the humanoid and robot brain bets if you want them.
Weighting is the second decision. By size, Tesla, Nvidia and Intuitive Surgical would swamp the rest, so most people cap them or use an equal weight index for the robot makers. If chips are the part of the story you believe in, the semiconductor index builder handles the Nvidia cap directly, and the electricity side of the same automation boom is on the nuclear energy ETF alternative page.
From a view on automation to a basket you can buy
How to build your own robotics stocks index in four steps
1
Pick the robots
Decide how much goes to factory, warehouse, surgical and humanoid robots, and whether chips and Tesla belong at all. That one choice separates BOTZ from ARKQ.
2
Fill the groups
Search the catalog of US-listed stocks, add 10 to 25 names across the groups, and set equal, size or your own weights with a cap on the giants.
3
Backtest against the funds
Run it against the S&P 500, and build a one-line index of BOTZ or ROBO to put a fund with ten years of history on the same chart.
4
Track it, then buy it
Follow the index daily against its benchmark, export the weights, and place the basket at your own broker when you are ready.
Robotics has a useful history to test against. ROBO dates from 2013 and BOTZ from 2016, so a basket can be run through the 2018 trade war selloff in industrial stocks, the 2020 crash and the 2022 bear market, when robotics funds fell far harder than the S&P 500. The humanoid names cannot: KOID and HUMN launched in June 2025, and many of their holdings listed in the last few years. Test the established robot makers over the long window and the young names over the short one, and look at the maximum drawdown next to the S&P 500 before the return. Backtests are hypothetical and past returns do not predict future ones. If you cap a name, test a quarterly reset with the custom index rebalancing tool. BOTZ shows why: its five largest names were 41.1% of the fund against a 40% limit in its own rules summary, between reviews.
Three ways to own robotics stocks, compared honestly
Build your own robotics index vs a robotics ETF vs a basket broker
| Robotics ETF | Fidelity Basket Portfolios | Indexes, then your broker | |
|---|---|---|---|
| Who decides what counts as robotics | The fund's index or manager | You | You, and you can test it first |
| Yearly cost | 0.35% to 0.95% of the balance | $4.99 a month flat | From $24 a month for the software, billed yearly, plus your broker's cost |
| History you can test | Since 2013 for ROBO and 2016 for BOTZ, about a year for the humanoid funds | None | As far back as your youngest stock goes |
| Japanese, Swiss and Chinese robot makers | Fanuc, Keyence, ABB and the humanoid makers are held | US-listed shares only | US-listed shares and ADRs |
| Per-stock tax losses | No | Yes, done by you | Yes, at your broker |
| Where it falls short | You take the fund's answer on Tesla, Nvidia and software | No way to test a design before buying | You place and rebalance the trades, and the Japanese makers are hard to hold |
The ETF wins when you want one ticker and no upkeep, and especially when you want the Japanese robot makers, which are the heart of industrial robotics and awkward to buy directly. If that is the bet, BOTZ at 0.68% is the honest product: it cost $0.91 a year for every $100 of robot maker exposure, against $2.37 for ROBT and $7.18 for ARKQ by the same measure. Building your own pays off on cost once the balance is large enough that a flat $288 a year for the software beats the fund fee, about $42,000 against BOTZ and $30,000 against ROBO, and whenever the mix you want is one no fund offers, such as US surgical and warehouse robots with no chips and no cars.
Owning the shares has a tax side too. Robotics stocks do not move together: a warehouse robot maker can halve on one customer's order while the surgical robot makers hold steady on procedure growth. When one name in your basket falls, it can be sold for a tax loss while the rest stays in place, which a fund does not allow. Check the dates with the wash sale calculator before you buy the same stock back.
What Indexes does and does not do: it is index construction, backtesting and tracking software. It does not hold money, place trades or give investment advice. You design and test the robotics stocks index here and hold it wherever you already invest.
Robotics ETF questions
What people ask before they buy robotics stocks
What is the best robotics ETF?
For industrial and surgical robot makers, Global X's BOTZ (0.68%) held the most: 74.7% of the fund in robot and automation companies on October 9, 2026, led by Intuitive Surgical, Fanuc and Keyence. For humanoid robots, Roundhill's HUMN (0.75%) held 34.3% in humanoid makers and joint parts. ARKQ and ROBT are mostly other things: cars, rockets, chips and software.
Does Vanguard have a robotics ETF?
No. Vanguard sells no robotics or automation ETF. The robotics funds are from Global X (BOTZ), ROBO Global (ROBO), ARK (ARKQ), First Trust (ROBT) and, since 2025, the humanoid funds from KraneShares, Roundhill and Themes. A Vanguard customer can buy those ETFs, or buy the US-listed robot makers directly in a Vanguard brokerage account.
Which robotics ETF has the lowest expense ratio?
Themes' BOTT charges 0.35%, then ROBT at 0.65%, BOTZ at 0.68% and KOID at 0.69% after a fee waiver. ARKQ and HUMN charge 0.75% and ROBO 0.95%. BOTT is small, about $65 million, and its median bid and ask spread was 0.30% against 0.03% for BOTZ, so a round trip can cost more than the fee gap.
What is the difference between BOTZ and ROBO?
BOTZ weights by company size with an 8% cap, so five names held 41.1% of it on October 9, 2026. ROBO weights by a robotics score, so no stock was above 2%. BOTZ costs 0.68% and ROBO 0.95%. They shared 28 companies and 28.4% of their money, and ROBO held more software, test equipment and electronics makers.
Is Nvidia in robotics ETFs?
Yes, in most of them. Nvidia was 8.0% of BOTZ, 5.8% of ARKQ, 5.4% of BOTT and 3.2% of HUMN on October 9 to 12, 2026. The S&P 500 fund SPY held more Nvidia, 8.2%, than any robotics ETF, so a robotics fund is not a way to add Nvidia if you already own an index fund.
What is the best humanoid robot ETF?
It depends on what you count. Roundhill's HUMN held the most in humanoid makers and joint parts, 34.3%, with Tesla at 6.3%. KraneShares' KOID spreads 50 suppliers equally. Themes' BOTT is the cheapest at 0.35%. All three are mostly Asian listings: 72.8% of BOTT and 71.9% of HUMN sit in shares traded outside the US.
Can I build my own robotics ETF?
You can build the same exposure without a fund. Pick the robot makers you want, set the weights, backtest the basket against BOTZ, ROBO and the S&P 500, then buy the shares at your broker, where there is no expense ratio and each stock carries its own tax lot. The Japanese and Swiss makers trade over the counter or abroad.
Sources
Holdings, fees and assets from each issuer's fund page and daily holdings file: globalxetfs.com (BOTZ), roboglobaletfs.com (ROBO), ftportfolios.com (ROBT), kraneshares.com (KOID) and ssga.com (SPY) for the October 9, 2026 close; ark-funds.com (ARKQ), roundhillinvestments.com (HUMN) and themesetfs.com (BOTT) files dated October 12, 2026. ARKQ's expense ratio and inception date are from a secondary data source because ARK's fund page could not be read. Index rules from the published methodology documents and prospectuses of Indxx, ROBO Global, Nasdaq, MerQube and Solactive. IRBO's change to ARTY from the iShares Trust filing with the SEC in 2024. The groups are our own classification of each holding by its main business; some companies could reasonably sit in another group, most of all Teradyne (counted as robotics for Universal Robots, although most of its revenue is chip test equipment) and Tesla (counted as a car maker even where a fund holds it for its humanoid robot). Holdings and fees change, so confirm them on the fund page before you buy. Indexes is not affiliated with any fund issuer or broker named here. We make index construction and backtesting software: we do not manage money, place trades or give investment advice, and backtested results are hypothetical.
Build the robotics index you actually meant to buy
Choose the robot makers and the weights, backtest the basket against BOTZ, ROBO and the S&P 500, and track it every day. Then buy it wherever you invest.
Keep reading