Why Do Two ETF Overlap Percentages Disagree on the Same Funds?

On August 22, 2026, four ETF overlap checkers answered the same SCHD and VOO question with 95%, 65.5%, 8% and 0%. None of them is broken. Each answered a different question and named it on the page, either in small print or in the title. Reading an overlap percentage means knowing which one yours answered.

Every third-party figure below was read from the tool's own page on August 22, 2026, and each is quoted as of that date. These tools change their numbers and sometimes their methods, so treat the figures as a snapshot of what four pages said on one day, not as fixed properties of the tools. The what is ETF overlap post covers the standard formula itself; this one is about why overlap percentages can land so far apart.

Why do two ETF overlap checkers give different answers?

Because "overlap" names at least four different measures, and a percentage on its own doesn't say which one you got. Here is the same pair, on the same day, through four public checkers.

Four numbers, four measures, one pair of funds. The spread is definitional, not a data dispute — and the tool at each extreme is the most explicit about what it did. MarketXLS prints its reverse direction, 8% of VOO, on the same screen as the 95%. Guardfolio puts "Top-10 Weight" in the page title.

What is each of these percentages actually measuring?

Four questions, each legitimate, each with a different answer.

Sum of minimums over full lists. For every stock both funds hold, take the smaller of its two weights and add those up. This answers how much of my money is invested identically in both funds. It's what ETF Research Center's headline computes, and its own rows show the working: Chevron at 0.6% of VOO and 3.9% of SCHD contributes 0.6. My Financial Freedom Tracker writes the same definition out — "counting each shared position at the smaller of its two weights."

One fund's exposure to the other. Take one fund and ask what share of its weight sits in names the other also holds. This answers how much of this fund is redundant given that one, and it's directional — it gives two different numbers depending on which fund you name first. MarketXLS reports both and puts the larger one in its headline.

Sector similarity. Compare the two funds' sector allocations rather than their holdings. This answers are these funds pointed at the same parts of the market, which is a real question and a different one. My Financial Freedom Tracker leads with it and says why: sector data is complete where holdings data often isn't. It also states the limit plainly — "weighting the same industries is not owning the same stocks."

Top-ten weight. Restrict the comparison to each fund's ten largest positions. This answers do the funds' biggest bets coincide. For SCHD and VOO the answer is genuinely none: VOO's top ten are mega-cap technology names, SCHD's are dividend-quality names, and no ticker appears in both lists.

Do the checkers disagree about the holdings data?

The three full-list sources give direct evidence that their holdings data broadly agrees. All report 46 shared holdings and similar fund sizes: SCHD at 99, 100 and 103 names; VOO at 501, 505 and 510.

The SCHD/VOO pair page defines the 7.6% as the sum of smaller shared weights, computed from 99 SCHD constituents (100% coverage) and 501 VOO constituents (99%), using issuer and SEC N-PORT data. Our methodology describes broader data sources and limitations, including each scan's look-through coverage ratio.

The percentages converge here, but the formulas differ. VOO is lighter in all 46 shared rows: UnitedHealth is 0.59% against 5.09% of SCHD, Home Depot is 0.54% against 3.36%, and VOO's largest shared position is 0.59%. When one fund is lighter in every shared name, both calculations return the same total. Where neither fund is uniformly lighter, they diverge.

That pair-specific convergence is the useful part. The 95-point gap between the highest and lowest headline says nothing about which tool has better data. The four headlines were never comparable.

Why does the direction of the comparison change the answer so much?

Because the 46 shared names carry far more weight in SCHD than in VOO. SCHD holds around 100 stocks and VOO around 500, but counts alone don't set the direction. Those names make up 95% of SCHD; each is below 0.6% of VOO and together they make up 8%.

The asymmetry is information, not noise. Put equal money into each and almost all of the SCHD position duplicates something the VOO position already covers, while the VOO position keeps the overwhelming majority of its weight in names SCHD never held. A single headline percentage throws that away whichever direction it picks. MarketXLS picks the larger of the two and says it does — its peer comparisons are scored "on the same directional measure as the verdict: the larger of the two shares."

Sum-of-minimums avoids the choice by being symmetric: swap the fund order and the number doesn't move. That's why it's the convention the published pair pages use, and it's also why it reads low for a pair like this one — 7.6% is the shared weight measured against both portfolios at once.

What does coverage do to an overlap percentage?

It bounds it. A checker computing overlap from each fund's ten largest positions is measuring inside a slice of the portfolio and telling you nothing about the rest.

My Financial Freedom Tracker publishes exactly that boundary for this pair: its top-holdings overlap of 0% is computed over disclosed positions covering 43.5% of SCHD and 38.4% of VOO, and the page says the full-portfolio figure "is at least this much." That is a lower bound, correctly labeled, on well under half of either fund's money.

The same tool's calculator page generalizes it: public feeds "disclose only each fund's largest positions (typically the top 10), which for a broad fund like VTI covers about a third of its weight." So a top-ten measure is a fundamentally different instrument from a full-list one, and the gap between them widens exactly where a fund is long-tailed. It also explains something that looks like a contradiction: the 46 stocks SCHD and VOO share are worth 7.6% of the pair by weight and 0% of their top-ten comparison. The two funds share none of their ten largest holdings, and the shared names that do exist are small inside VOO — the biggest, UnitedHealth, is 0.59% of it.

The published post already draws the general rule — a checker that doesn't publish its coverage "is asking you to trust a number you cannot audit." This pair is the worked case.

A pair percentage leaves your position sizes out. To include them, scan the combined company exposure of your holdings. The free tool accepts up to 20 tickers without an account and shows the coverage behind its result.

Why don't the published thresholds agree either?

The same reason the percentages don't: a band is only meaningful against a stated measure, and the pages publishing bands rarely say which one they have in mind. Read across the first page of results for etf overlap percentage, accessed August 22, 2026, and the advertised cutoff for "too much" ranges from 10% to 70%.

• etf-overlap.com tiers at under 15% complementary, 15 to 40% lightly redundant, 40 to 70% partially redundant, over 70% highly redundant.
• Truthifi, dated April 3, 2026, puts a "green zone" below 40%.
• Passiv, dated February 5, 2024, describes a common guideline of below 50%.
• InvestmentNews, dated March 17, 2025, reports experts suggesting 10% to 20%, others below 33%.
• My Financial Freedom Tracker puts its 60% to 70% band on sector overlap specifically, which is the one case here where the band names the measure it belongs to.

None of these is presented with a source, and applying any of them to SCHD and VOO returns a different verdict depending on which of the four percentages you feed it. The honest reading is that there's no universal threshold to look up. The overlap explainer linked at the top of this piece works through what actually changes the reading, and the question it lands on is what job the second fund was hired to do.

How do you calculate portfolio overlap and audit the result?

Start by naming the measure, direction, coverage and holdings date, then verify one shared row. Four questions put those checks in order. Each one is answerable from the page itself, or it isn't answerable at all.

1. Which measure is it? Sum of minimums, one fund's exposure to the other, sector similarity, or top-ten weight. If the page doesn't say, the number can't be compared to any other number.
2. Which direction? If it's directional, the tool chose a fund to put first. Ask for the other reading — on SCHD and VOO the two are 95% and 8%.
3. What coverage? How many constituents went in, and what share of each fund's weight they represent. A figure over the top ten of a 500-stock fund is measuring roughly a third of it.
4. As of when? Holdings files are month-end snapshots on issuers' own schedules, and the two funds rarely share a date. Of the four pages here, one printed the date of the holdings behind its figure, one printed the date it was last reviewed, and two printed neither.

Then check the arithmetic on one row. Every full-list checker that shows its shared holdings gives you enough to do it: take a shared name, read its weight in each fund, and confirm the contribution is the smaller of the two. UnitedHealth is 5.09% of SCHD and 0.59% of VOO on the PortLens page, so it contributes 0.59 — not 5.09, and not the average. If a page won't show you a row, you can't run this check at all.

All investing involves risk, including possible loss of principal, and past performance does not guarantee future results.

How do I check a pair myself?

When the two constituent lists can be compared safely, the PortLens ETF overlap checker returns the sum-of-minimums figure with the shared holdings table, the constituent count and coverage behind it, and each fund's holdings date, with the result anchored to the older of the two. It's free and needs no account. The SCHD and VOO comparison quoted throughout is at SCHD vs VOO; a pair like VOO vs VTI at 86.7%, as of June 30, 2026, shows what the same measure looks like at the other end of its range.

Pairwise checks answer a pairwise question. Overlap across three or more funds stacks in ways no pair page can see, which is what a whole-portfolio look-through is for. A free PortLens scan explodes the funds you hold into their constituents and sums your exposure to each company across all of them. Hold it to the same standard as everything above: when the issuer's own daily holdings file or the fund's SEC N-PORT filing resolves, scan look-through uses every constituent that real source publishes, while a small set of funds (chiefly UCITS/LSE listings such as CSPX.L and VUSA.L, which file no N-PORT) fall back to a curated approximate top-ten that the coverage ratio does not see past. Every scan reports its own coverage ratio, and below roughly 75% coverage the concentration figures understate reality. The portfolio x-ray post walks through a worked example.

For the next question, “how much of one company do I own across all my funds and direct stocks?”, the dated whole-portfolio calculation with downloadable inputs shows each contribution. A pair percentage cannot answer it without your allocation. You can measure combined company exposure in your own portfolio; the free scan accepts up to 20 holdings without an account and states its coverage.

Key takeaways

• Four public checkers returned 95%, 65.5%, 8% and 0% for SCHD and VOO on August 22, 2026, because they compute four different measures. The spread isn't evidence of conflicting holdings data.
• Two sum-of-minimums results and one directional result land at 7.6%, 8% and 7.64% on this pair because VOO is lighter in every shared name, not because the formulas are identical.
• Directional overlap has two answers, and the shared weights decide which is larger: for SCHD and VOO, the same 46 names make up 95% of SCHD against 8% of VOO.
• A top-ten measure covered 43.5% of SCHD and 38.4% of VOO on the one page that published its coverage, so it can only ever be a lower bound on the full-portfolio figure.
• Before comparing two overlap percentages, establish that they share a measure, a direction, a coverage level and a holdings date. Four numbers that share none of those aren't in disagreement — they were never answering the same question.

This article is for information and education only and is not investment advice. Analytics referenced are computed as described in our methodology; see our disclosures.