Updated June 28, 2026. Clear Money Guide aims to keep advisor-fee comparisons, calculators, local guides, and educational planning pages accurate, current, and easy to verify.
How to report a correction
- Use the Contact page and include the page URL, the issue you found, and any source or context that helps us verify it.
- For calculator or fee-example issues, include the balance, fee model, quote, or table row that appears incorrect.
- For local guide issues, include the city, county, state, or regional page involved.
What happens next
- We review correction requests as promptly as practical, usually within 5 business days.
- If a correction is needed, we update the affected page and, when material, refresh the visible update date or page context.
- If a request involves a judgment call rather than a factual error, we may update the wording for clarity without changing the page conclusion.
Editorial standards
Corrections are handled separately from affiliate relationships. See the Editorial Policy, Affiliate Disclosure, and Disclaimer.
Correction log
Newest first. We log substantive corrections — a number that moved, a claim withdrawn, a source that turned out not to say what we said it did. Typos and formatting fixes are not logged. Where a correction affects more than one page, every affected page is updated the same day and the entry says so.
August 20, 2026
Datalign matches up to three firms, not one — six pages corrected
We told readers that Datalign Advisory introduces exactly one advisory firm, and we made that the reason to prefer it over SmartAsset. Datalign’s own Form ADV Part 2A, dated May 21, 2026, says up to three: “A Consumer is then introduced to up to three suitable Participating Adviser” and “The Consumer will be contacted by all three Participating Advisers.” Its Form CRS says the same, and the brochure’s own material-changes note dates the amendment to May 21, 2026 — two months before we published the claim. Corrected on our Datalign review, Datalign vs SmartAsset, Datalign vs Zoe (including its title), SmartAsset alternatives, the matching-services comparison and the guides index, plus the meta descriptions that carried it.
August 20, 2026
Correcting a correction: Datalign does disclose what it screens for
On August 1, 2026 we withdrew a description of Datalign’s screening as “limited” on the grounds that no first-party source stated a vetting standard in either direction, and we published that negative here. That was wrong, and the search behind it was not exhaustive. Datalign’s Form ADV Part 2A, Item 4, states participating advisers “are screened for current state or federal registration status, disciplinary history, and advisory services offered” and are then interviewed about their expertise; Item 7 requires them to be SEC- or state-registered investment advisers. Its disclosures page uses the withdrawn word itself: “The screening criteria used by Datalign are limited and do not substitute for your own due diligence.” Every page that carried the withdrawn negative now states what the brochure actually discloses, and what it still does not (no credential bar, no fee-only requirement).
August 13, 2026
Six states were missing a property-tax credit they actually run: Arizona, Colorado, the District of Columbia, New Mexico, New York and Oklahoma
Getting Oklahoma’s row wrong was a reason to re-check the rest of the column, so every state that property tax relief for seniors by state did not classify as running a circuit-breaker credit was read again at source. Six were wrong the same way. Arizona runs a property tax refund credit on Form 140PTC; Colorado runs the PTC rebate; the District of Columbia runs the Schedule H credit under D.C. Code § 47-1806.06; New Mexico runs a property tax rebate for people 65 or older on Form PIT-RC; New York runs the refundable real property tax credit on Form IT-214; and Oklahoma runs the Form 538-H credit described in the entry above. The cause was the same in every case: each row had been built from a single source per state, and that source was normally the assessor-side exemption page, which has no reason to mention a credit claimed on the income-tax return. All six rows and all six state pages were corrected the same day. New Hampshire is a seventh candidate we could not resolve — its revenue department’s site refused our requests and the statute site would not resolve — so its row is unchanged and the hub says so rather than guessing.
August 13, 2026
Oklahoma’s additional homestead exemption is not keyed to age — but this entry then said Oklahoma runs “a freeze and nothing else”, and that was false
Property tax relief for seniors by state classified Oklahoma’s programme types as “Freeze, Exemption”. Read against the statute this session, Oklahoma’s only age-keyed property-tax programme is the senior valuation limitation (Okla. Const. art. X, § 8C; 68 O.S. § 2890.1), which requires the head of household to be 65 or older. The additional homestead exemption at 68 O.S. § 2890 turns on gross household income of $20,000 or less and names no age; age appears only in subsection D, which excuses an already-qualified claimant aged 65 or over from re-applying annually. The table therefore counted, as a senior programme, one that a low-income Oklahoma homeowner of any age can claim — and it contradicted our own Oklahoma page, which says in terms that the additional exemption is not an age programme. The row was changed to read “Freeze”. Amended 13 August 2026, the same day. The headline of this entry originally read “so Oklahoma runs a freeze and nothing else”, and that was wrong — in the one place on this site where being wrong is least excusable. Oklahoma does run a second age-keyed programme: 68 O.S. §§ 2904-2911, about twenty-five pages further into the same statute than the sections read the first time, let a head of household aged 65 or older, or totally disabled, with gross household income of $12,000 or less, claim the property tax paid above 1% of household income, capped at $200 and refundable under § 2911. It is claimed on Form 538-H. For a few hours a low-income Oklahoman aged 65 or over could read here that no such credit existed for them. The row now reads “Freeze, Circuit-breaker credit” and the Oklahoma page sets the credit out in full. The finding that the § 2890 exemption is not age-keyed still stands.
August 13, 2026
New Mexico taxes Social Security above an income cliff, so the count is eight, not seven
The state retirement scorecard said “only 7 states tax Social Security benefits at all” and named Colorado, Connecticut, Minnesota, Montana, Rhode Island, Utah and Vermont, while its state table recorded New Mexico as “not taxed”. New Mexico exempts Social Security only where adjusted gross income is below $100,000 (single), $150,000 (married filing jointly, head of household or surviving spouse) or $75,000 (married filing separately); above those figures the benefits are taxable. Read this session at the New Mexico Taxation and Revenue Department’s Social Security Income Tax Exemption page and its personal income tax overview. The scorecard had compressed a conditional exemption into an unconditional one. Our own retirement taxes by state table and New Mexico retirement taxes guide already stated the cliff correctly and are unchanged. One page affected.
August 13, 2026
Nevada: we published a $300,000 ceiling the statute sets at $500,000, and yesterday’s Nevada correction never reached the dataset file
Twenty pages said Nevada’s summary administration under NRS ch. 145 reaches estates up to $300,000. It reaches estates up to $500,000: NRS 145.040 says the court may order summary administration where the gross value of the estate, after deducting encumbrances, does not exceed $500,000 — as amended in 2025. Read in the statute’s own text at leg.state.nv.us on 13 August 2026. A reader with a $400,000 Nevada estate was told the shortened route was closed to them when it is open. All twenty pages were corrected the same day.
Separately, the correction we made on 12 August — Nevada’s affidavit amount for a surviving spouse and its set-aside ceiling are $150,000, not $100,000 (NRS 146.080(7) and NRS 146.070(1)(a), both as amended in 2025) — reached our pages but not our downloadable dataset. The corrected file we published on 13 August still carried the two $100,000 figures and the $300,000 one. That file is superseded by a reissue dated the same day, and every download link and dataset pointer in our page content now resolves to the reissue.
Two surfaces we cannot write still carry the superseded Nevada figures: /data/estate-probate-2026.csv, which a site plugin generates, and the /embed/probate-cost/ widget, which reads that file at runtime. We are saying so here rather than leaving a citable file wrong and unmarked. Found by our own audit of the surfaces our page-level checks cannot see, not reported to us.
August 13, 2026
A deposited dataset under-counted families by a factor of five — in the archive, not on the site
Version 1.0.0 of our open dataset Retirement Account Balances by Age, United States, 2022 published a weighted_families count for each age cohort at exactly one fifth of the true number. The Survey of Consumer Finances is released as five imputation replicates of every household, and the Federal Reserve’s own macro divides the sample weight by five so that totals over the five come out right; our build summed that weight inside a single replicate, so it estimated a fifth of the population.
Corrected in version 1.1.0 (DOI 10.5281/zenodo.21924736), published 13 August 2026. That deposit was itself reissued the next day, 14 August 2026, as 10.5281/zenodo.21928514, which adds the changelog file the 13 August record was missing; the values are the same. The concept DOI 10.5281/zenodo.21762536 resolves to the reissue, and every dataset pointer in our page content now resolves to the concept DOI. The six corrected counts sum to 131,306,389 US families and were cross-checked against a direct sum of the weight over all 22,975 records of the source archive, re-downloaded that day, grouped by the same cohorts. The computation script is corrected in the same version, so a re-run reproduces the corrected file.
Nothing published on this website changed, and no conclusion moves. Every other figure in that dataset — and every figure on retirement savings statistics — is a median, mean, percentile or share, all scale-invariant in the weights, which is why the dataset’s reconciliation against the Board’s own published table passed before the correction and passes unchanged after it. The affected field appears only in the JSON file of the deposit; it is not in the CSV and was never on any page here. Found by our own audit, not reported to us.
August 13, 2026
We withdrew a correct Iowa figure and published a wrong one in its place
Earlier today we withdrew Iowa’s $100,000 small-estate affidavit limit as unsourced and published
$50,000 instead. That was wrong, and it understated what a reader is entitled to do. The withdrawal rested on two documents that cannot settle the question: the official Iowa Code 2026 compile, which is dated December 12, 2025 and therefore cannot contain an Act of the 2026 session, and 2025 Iowa Acts ch. 148, which amends Iowa Code §633.356 in four places and leaves the dollar figure at fifty thousand. The Act that raised it is
2026 Iowa Acts, House File 2660, approved by the Governor on April 16, 2026 and — carrying no effective-date clause — effective
July 1, 2026 under Iowa Code §3.7(1). Iowa’s very-small-estate affidavit ceiling is
$100,000. HF 2660 also narrowed the test in the reader’s other direction: the probate assets must now consist only of personal property, with no interest in real estate. A reader with a $75,000 Iowa estate was told the affidavit route was closed to them when it is open. We read the enrolled Act and the Governor’s transmittal letter at legis.iowa.gov on August 13, 2026. Because the figure feeds published statistics, we recomputed rather than restated them: the national median of the 50 stated limits moves from $64,000 to
$75,000, so the comparison sentence on every state probate-cost page was recomputed, and the ranked table on
states with the highest small-estate limits was re-ranked from its own cells — its stated median moves from $62,500 to $75,000 and Iowa moves from 15th to joint 5th. Separately, our machine-readable dataset was still publishing three figures we withdrew yesterday. The file at
/data/estate-probate-2026.csv is generated by a site plugin our publishing toolchain cannot write, so we have published a corrected, dated replacement —
cmg-estate-probate-2026-corrected-2026-08-13b.csv — and moved every download link and every machine-readable dataset pointer in our page content to it (two plugin-emitted surfaces we cannot write,
/llms.txt and the
/embed/probate-cost/ widget, still point at the legacy path). The same three corrections were applied to the cost-of-dying index dataset.
69 pages were corrected today in this pass.August 13, 2026
Yesterday’s small-estate correction was incomplete, and a fifth figure was wrong
The entry below said four small-estate limits had been corrected across 43 pages. Both halves of that were wrong, and both are fixed today. That repair was scoped to figures printed beside their own statute citation, so it missed the same numbers written in ordinary prose and in table cells — which left three pages contradicting themselves in public:
Virginia probate cost showed $75,000 and $50,000,
North Dakota $100,000 and $50,000,
Iowa $50,000 and $100,000, and the
cost of dying index still carried the withdrawn figures in its small-estate column. Re-enumerating the class by the
figures rather than by their citations then found a fifth wrong number no earlier pass had seen.
South Carolina’s small-estate affidavit limit is
$45,000, not the $25,000 we were publishing on 21 pages: 2025 Act No. 26 (H.3472) raised it from twenty-five thousand dollars, signed and effective May 8, 2025 (S.C. Code §62-3-1201(a)(1)). A reader with a $40,000 estate would have been told the affidavit route was closed to them when it was open. We also recomputed every statistic derived from these figures, which the first repair did not: the national median of the 50 stated limits moves from $51,500 to
$64,000, and each state’s rank was recomputed on all 50 state pages that publish one. We also withdrew a claim on the same table that no official source supports: it said Iowa’s affidavit limit “doubles to $100,000 for affidavits presented on or after July 1, 2026”, three paragraphs from its own row saying the limit is fifty thousand dollars. 2025 Iowa Acts ch. 148 §40 amends Iowa Code §633.356(1) to read “fifty thousand dollars or less” and the Iowa Code 2026 text prints that figure with no future-effective note; the legislature’s bill-tracking and search pages could not be read at all, so we publish only what the statute we could read says.
[Annotated August 13, 2026: this withdrawal was itself wrong. The $100,000 figure is correct — it was raised by 2026 Iowa Acts, House File 2660, an Act of the 2026 session that neither document read here could contain. See the entry above.] 77 pages were corrected today. Every figure was read in the statute’s own text on August 13, 2026 — Va. Code §64.2-601(A)(1), N.D.C.C. §30.1-23-01(1)(a), Iowa Code §633.356(1), and S.C. Code §62-3-1201(a)(1) with 2025 Act No. 26.
August 12, 2026
Four small-estate limits were wrong on our state table
Building a ranked study of small-estate limits, we re-read each state’s statute at its official source and found four figures on our
small estate limits by state table that the statutes do not support.
Virginia is $75,000, not the $50,000 we published (Va. Code §64.2-601(A)(1)).
North Dakota is one hundred thousand dollars, not $50,000 (N.D.C.C. §30.1-23-01(1)(a)) — a row our own table had flagged as resting on a secondary source.
Nevada raised two figures to $150,000 in 2025: the surviving-spouse affidavit amount under NRS §146.080(7) and the court set-aside under NRS §146.070(1)(a); we showed $100,000 for both.
Iowa we published at $100,000 on the authority of 2025 Iowa Acts ch. 148; that act amends Iowa Code §633.356 in four places and leaves the figure at fifty thousand dollars, which is what the official Iowa Code 2026 text says. All four were corrected the same day. The correction reached
43 published pages, not one: besides the state table, the four figures were asserted with their own statute cites on the state retirement-tax pages, the state-pair comparison pages, the state probate-cost pages, the per-state small-estate-affidavit pages,
where you can skip probate, and the data behind the
probate cost calculator and the
estate settlement roadmap, whose numeric fields were corrected with their prose so the tools cannot compute against a figure their own text contradicts. The North Dakota, Pennsylvania and Rhode Island rows also lost the dagger that marked them as resting on secondary sources. The corrected figures are ranked and sourced in our new
study of small estate limits.
[Annotated August 13, 2026: this correction was incomplete — it reached the cited rows only, and a fifth figure was wrong. See the August 13 entry above.]August 12, 2026
Revocable living trust package: we published the study’s single-document row
Our pages priced an attorney-drafted revocable living trust
package at a median of
$2,475, middle half $1,600–$3,000, citing the LegalTemplates 2026 study of 909 law firms. That figure is in the study’s table, but it is the
Single document row. The study prices a trust package separately: a median of
$2,700, middle half
$2,500–$3,500, from the 221 firms that priced one. Sixty published pages carried the single-document figure under the word “package” and all were corrected the same day — the 51 state pages of our
living trust cost series among them, plus the
will-vs-trust breakeven calculator and the
probate cost statistics page. Two derived figures moved with it. The premium of a trust package over a will package is
$1,700, not the $1,850 we published, which had compared a trust package against a single will rather than against a will package; the seven state pages that net that premium out of a statutory probate fee were recomputed. The single-document medians are still published, now labelled as single documents rather than packages.
August 12, 2026
Attorney will-package price: six pages, two different figures, and the study says neither
Six of our pages priced an attorney-drafted will package — a will, a financial power of attorney and a healthcare document — beside a citation to the LegalTemplates 2026 study of 909 law firms. Four said
commonly $1,000–$2,000; two said
commonly $500–$2,000. Read at the source, the study says neither: its will-package row is a
median of $1,000, with a middle half of $750–$1,500, from the 319 firms that priced a package. The first version had turned the study’s median into the bottom of a range; the second was an older 2018 survey figure sitting next to 2026 citations. All six pages were corrected the same day —
how much a will costs,
estate planning package cost,
complete estate plan cost, the power of attorney and advance directive cost guides, and the estate planning cost guide — and every other figure those pages take from the same study (the $625 will median, the $300 power of attorney median, the $625 single-will median, the $300 power of attorney median and their middle halves) was re-checked against the study’s own table and is unchanged.
Later the same day: the $2,475 trust median named in the original version of this entry did match a cell in the study’s table, but the wrong one — see the trust-package correction above.August 1, 2026
Advisor fee benchmark: sample basis superseded (137 → 176 firms)
Our 2026 advisor fee benchmark was recomputed on a larger sample. Two published figures moved as a result: the combined share of firms at $250,000 that publish only a ceiling or no computable price at all went from
47.1% to 42.5%, and the share blocked by an account minimum went from
14.9% to 18.9%. The measure did not change — the sample grew. Eight pages citing the older basis were updated the same day, and the full report at
State of Advisor Fees 2026 carries the current numbers.
August 1, 2026
Advisor fee benchmark: the “entrants” rate could not be reproduced
A figure of about 0.90% (elsewhere 0.93%) had circulated for the median cost charged by firms that will not price a $250,000 household but will price a $1,000,000 one. Neither value could be reproduced under any definition we could compute. The report now uses a named definition and states it inline: 1.00%, n = 11 firms, measured at $1,000,000 on firms with no priceable value at $250,000. The older figures are not carried forward.
August 1, 2026
Zoe Financial: account minimum removed; credential list corrected
Several of our pages stated that Zoe Financial has a practical minimum of roughly $150,000 in investable assets. Zoe’s own Form CRS, dated June 29, 2026, says the opposite — “We do not impose a minimum account size for the Wealth Advisory Program.” The claim has been removed everywhere it appeared. Separately, we described its credential requirement as “CFP, CFA, or CPA”; its current vetting page lists CFP or CFA, and our pages now match.
August 1, 2026
NerdWallet referral pricing: figures corrected to the current disclosure
We stated that NerdWallet’s advisor-matching product is powered by Zoe and that Zoe pays it $100–$300 per qualified lead. We read NerdWallet Advisory’s own endorsement disclosure, effective November 13, 2025, directly: it names no network firm, and states lead-generation fees of “$13 to $1,150 per referral”. Both elements of our claim were wrong on the current document and have been replaced with what it actually says.
August 1, 2026
Wealthramp: fee-only requirement withdrawn as unverifiable
We described Wealthramp as the one matching network requiring its advisers to be fee-only. We could not source that to any document we can reach — its Form ADV Part 2A returns an access error from the SEC’s own document server, and its website returns one too. The claim has been withdrawn from every page that carried it, and our review says plainly that we cannot verify it rather than repeating it.
August 1, 2026
Datalign: screening characterisation withdrawn as unsourced
We stated that Datalign’s disclosures describe its screening criteria as limited. No first-party Datalign source states a vetting standard in either direction, so the characterisation was not ours to make. Our pages now say exactly that: its public materials do not disclose a specific advisor-vetting standard. Superseded August 20, 2026: this entry was wrong, and the search behind it was not exhaustive — see the August 20, 2026 entry above. Datalign’s Form ADV Part 2A states what it screens for, and its disclosures page uses the withdrawn word verbatim.
August 1, 2026
WiserAdvisor review: independence claim replaced with a partner disclosure
Our WiserAdvisor review carried a line describing itself as an independent analysis that WiserAdvisor did not pay for or see.
WiserAdvisor is a referral partner of Clear Money Guide, and the page now says so prominently, in the same wording as our
affiliate disclosure, rather than asserting independence. Partner compensation does not change our calculator math or fee figures.
Spotted something we have got wrong? Tell us — corrections are genuinely welcome, and this page is where they end up.
If you found the error while citing us, this is the fastest route to it being fixed — and the newsroom explains how we handle figures that go stale.
Reusing any of this? One licence covers all of it — free to reuse, adapt and republish, including commercially, with attribution. No permission needed. Figures we quote from the IRS, SSA, BLS or a state agency belong to their publishers and should be cited to them, with the data year.
