My broker says I hold 16% technology. Look through the funds and it's closer to 26% in five companies.

Thesis

My broker classifies 76.5% of my portfolio as "Broad" and reports my technology exposure at 16.6%.¹ Both figures are a result of how the platform classifies funds, not descriptions of what I actually own. On a look-through basis, five companies account for roughly 26% of the book, and one of them, NVIDIA, accounts for 14.6% on its own.

I am not arguing the platform is broken. I am suggesting that position-level reporting stops telling you much from the moment a portfolio holds funds. Most retail investors never find nor investigate this because nothing in the interface tells them the number is incomplete, not wrong, just incomplete.

Analysis

The book is nine positions, all US-listed, roughly 99% equity. Here is what it looks like at the level my broker reports it:

Nine holdings, six of them are funds spread across the US market, Nasdaq 100, dividend equity, covered calls, international dividend, energy, and private credit. It reads like a diversified book, right? My broker agrees: it reports the sector breakdown as 76.5% "Broad", 16.6% Technology, 3.1% Energy, 3.1% Financials.

The problem is that "Broad" is not a sector. It is the label applied to every fund the platform does not decompose. Three quarters of my portfolio sits in a bucket that essentially means "not classified." The 16.6% technology figure is just NVDA + AVGO, the two positions the platform can see through, because they are individual shares.

So, I rebuilt the sector view by hand, using each fund's published holdings.

What the look-through actually shows

Four of my six funds hold NVIDIA. QQQM holds it at 8.54%, VTI at 6.70%, and JEPQ at 6.7%.²³⁴ None of those three funds is marketed as a technology fund. One of them is a total market index and another is an income strategy.

Broken out by company, the gap is almost entirely inherited rather than chosen:

NVIDIA is a position worth sitting with. I hold 10.55% directly, which was a deliberate decision I can defend. The funds add another 4.07% that I never sized and never reviewed. My actual exposure is 14.6%, significantly larger than the position I think I hold.

Apple, Alphabet, Microsoft, and Amazon together come to 11.5%, and not one of them appears on my statement. If NVIDIA fell 30% tomorrow, the loss against NAV would be about 4.4%, not the 3.2% my position list implies. This is the real problem, especially for everyday investors.

The geography figure is wrong too

My broker reports the portfolio as 100% United States.¹ IDV is 8.69% of the book and tracks 100 dividend payers across developed Europe, Asia Pacific, and Canada, excluding the US by design.⁵

The classification is not a data error. IDV is a US-listed, US-domiciled fund, so by listing the domicile, the label is defensible. It just answers a question nobody is asking. When an investor looks at a country breakdown, they want to know where the economic exposure sits, not where the wrapper is registered. This is a bigger blind spot than most give it credit for.

It is worth being precise about where the failure actually sits, because it is not a disclosure failure. SEC Rule 6c-11 has required US-listed ETFs to publish full portfolio holdings daily on their websites since 2019.⁷ Every number in this piece came from data that funds are legally obliged to publish, and I assembled it in an afternoon. The raw material is mandated, free and public. What is missing is anyone joining it to your actual position list and doing the arithmetic. That is a product gap, not a regulatory one, which is the more interesting kind, because a product gap is something you can close.

Where the diversification is real

Not everything here is illusory, and it would be dishonest to imply otherwise. SCHD, at 13.5% the third-largest position, holds none of the five companies mentioned before. Its largest positions are Abbott, Amgen, Merck, Coca-Cola, and UnitedHealth, and its sector weights run 21% healthcare and 20% consumer defensive.⁶ It is doing the job it was bought for. So are XLE, ARCC, and IDV.

The concentration is not spread evenly across the book. It is specific to the overlap between QQQM, VTI, and JEPQ, three funds that sound different and behave similarly.

Risks

Several things could make this analysis wrong, and I would rather name them than let a reader find them.

The underlying weights are not all measured on the same date.

The JEPQ figures come from JPMorgan's own 30 June 2026 fact sheet, a primary source. The QQQM and VTI figures come from a data aggregator reporting more recent holdings. Mega-cap weights usually move a percentage point or more between rebalances, so treat the 26.09% as accurate to roughly half a point rather than to two decimals. This doesn't materially affect the analysis, but it's good to be aware of it for precision.

Look-through overlap is not the same as correlated risk.

Owning NVIDIA three times over is not automatically three times the risk, because the funds hold it alongside different baskets that dampen or amplify the move. A proper treatment needs a correlation matrix and a factor decomposition, not just a sum of weights. What the weights sum does establish is that the exposure exists and is unmeasured, which is the narrower claim I am making.

Concentration is not automatically a mistake.

A 14.6% position in the best-performing semiconductor company of the cycle has been the reason this book performed. The argument here is not that the exposure is wrong, but rather that it should be consciously chosen rather than discovered later.

I have not verified AVGO's look-through.

Broadcom is 6.04% directly and almost certainly appears in QQQM and VTI as well, which would push the true figure higher. I did not have same-date weights I trusted, so I left it out entirely rather than make a guess at them. That omission makes my 26.09% rather conservative.

Conclusion

I am holding the position. Now that I can see the full 14.6% rather than the 10.55% on my statement, that is a decision rather than an accident, which is the only change I actually needed to make today.

What I am watching is not the share price. It is the supplying industries: foundry, memory, networking and semiconductor capital equipment. Demand for what NVIDIA sells shows up in what its suppliers are being asked to build well before it shows up in NVIDIA's own results, so a broad deterioration in orders across those industries is my trigger to reevaluate the position size. A drawdown in the stock on its own is not, because that tells me about sentiment rather than about the demand underneath it. Until that supply chain signal turns, the exposure stays where it is and I review the look-through number quarterly rather than annually.

Any portfolio holding more than one broad equity fund has a look-through exposure that its statement does not show, and the error runs in one direction: the statement always understates concentration, never overstates it. This is an important consideration, as some advisors would not even think to investigate the places that I did for this information. While it may seem rather simple once laid out, the thinking needs to be framed around real exposure. Sadly, the investor(s) who need to know this the most are simultaneously the least likely to calculate it.

That gap is the reason I built FuturePort®, and Concentration Check exists specifically to do this calculation for a portfolio in a few seconds rather than an afternoon. I would rather show the working than assert the conclusion, which is why I wrote this piece.

AI disclosure

AI was used in this research in the following ways:

• Pulling my position data from my broker and collecting published fund holdings from the sources listed below
• Calculating the look-through weights from that data
• Drafting the initial prose from those findings, which I then rewrote

The analysis, the judgment, and the conclusion are mine. Every figure has been checked against the sources cited, and where I could not verify something on a consistent basis I left it out and said so rather than estimating it.

Like most people, I firmly believe that AI is an inevitable tool that will vastly improve efficiency within our work. However, with that comes the responsibility to ensure that we disclose and use it fairly. Hence I will ensure that I disclose how I use AI and be open that it is an indispensable tool for my research and analysis.