Three ETFs feel diversified. They can be one bet on the same ten stocks.

Here is a portfolio that looks responsible: 40% in VOO, 40% in QQQ, 20% in Apple, because you like Apple. Three tickers, two of them broad index funds. It reads as diversified in every brokerage app you own.

Now do the math the app refuses to do. Apple is roughly 6% of the S&P 500 and roughly 8% of the Nasdaq-100 — the exact weights drift month to month, the arithmetic does not. Your effective Apple position is 20% held directly, plus 40% × 6% through VOO, plus 40% × 8% through QQQ. That is about 26% of your portfolio in one company. You did not buy a diversified portfolio with a small Apple tilt. You bought a quarter-Apple portfolio with index wrapping.

This is not an Apple problem. Run the same math on Microsoft or Nvidia and you get the same shape, because the same handful of mega-caps sit on top of both indices. Overlap tools put the weight overlap between VOO and QQQ somewhere around half, and the top ten names alone are more than a third of the S&P 500 right now. Buying both funds is not two decisions. It is one decision, twice, with different expense ratios.

The way every brokerage and most trackers show concentration — percent of portfolio per ticker — is wrong twice, in opposite directions.

It calls concentrated portfolios diversified. The example above: three tickers, none above 40%, looks fine. The 26% single-company exposure is invisible because it is smeared across three lines.

It calls diversified portfolios concentrated. Hold nothing but VOO and a naive per-ticker view reports 100% concentration in "one position" — a reading that would be true of 100% in one biotech and is meaningless for a fund holding five hundred companies. Anyone who has computed a Herfindahl index over raw tickers has produced this false alarm.

Both failures come from the same place: the ticker is the wrong unit. The unit that matters is the company your money ends up in, regardless of how many wrappers it passed through on the way.

Look-through analysis decomposes each fund into its constituents and merges them with your direct holdings. Effective Apple = direct Apple + Apple reached through every fund that holds it. Done honestly, it needs two guardrails that naive implementations skip:

The remainder of a decomposed fund is diversified, not concentrated. If you expand a fund's top ten constituents, the remaining few hundred names should count toward your total as spread-out money — not get squared into a concentration index as if the fund were one stock, and not silently disappear.

A fund you cannot decompose stays a fund. If constituent data is not available, the honest treatment is a "diversified fund" floor — keep it whole and flag it — rather than guessing or, worse, treating it as a single concentrated name.

With those in place, a concentration index over effective exposures finally means something: a portfolio of one broad ETF reads as diversified (correct), and the VOO/QQQ/Apple portfolio above reads as roughly one quarter one company (also correct, and alarming in the useful way).

Fund-vs-fund overlap tools exist and are genuinely useful — ETF Research Center's checker will tell you the weight overlap between any two funds in seconds, free. If you want to know whether VOO and VTI are redundant before you buy, use one.

What they answer is a question about two funds in the abstract. What they cannot answer is the question about you: your dollar weights, your direct positions merged in, your cost basis, on today's constituents, without re-entering everything every time you check. Overlap between VOO and QQQ is a fact about two funds. "26% of my money is in Apple" is a fact about your portfolio, and it only falls out when the decomposition runs over what you actually hold.

Opula's concentration tool runs this look-through over your holdings and returns it as data: your effective top exposures with the path each one arrived by (AAPL via direct + VOO + QQQ), how much of each fund the decomposition covered, which funds could not be decomposed and were left whole, and a concentration index computed the honest way — fund remainders dilute it rather than inflate it.

Two disclosures, because they are the difference between a number and a marketing number. The decomposition uses each fund's top ten constituents, so it is an approximation that understates single-name exposure slightly rather than overstating it. And this is current exposure only — what you hold today. Forward-looking math over it is a separate, paid question.

Ask Claude "how concentrated am I really?" with Opula connected and the effective exposures come back computed, not estimated — the same answer on the second ask as the first. What to do about a 26% position is then an actual conversation, which is the part a model is good at.

Do VOO and QQQ overlap? Substantially. Every stock in QQQ that is also in the S&P 500 — which is nearly all of them — is in both funds, and because both indices are market-cap weighted, the biggest overlapping names are the biggest positions in each. Weight overlap is commonly measured around half.

Is holding multiple ETFs automatically diversified? No. Diversification is a property of the underlying companies, not the number of wrappers. Three funds tracking mega-cap-heavy US indices can be less diversified than one fund tracking a broader one.

How do I find out how much of a single stock I actually own through ETFs? Multiply each fund's weight in that stock by the fund's weight in your portfolio, sum across funds, and add your direct position. That is the look-through exposure. Tools can do this continuously against current constituent data; by hand it goes stale the month you compute it.

Why does my tracker say I'm 100% concentrated when I only hold one index fund? It is computing concentration over tickers, not underlying companies. One ticker equals total concentration by that measure, even when the ticker contains five hundred stocks. A look-through concentration index fixes this by treating the fund's contents, not its symbol, as the unit.