Finnacl Research

R11 · free executive summary

What Your Peers Actually Charge

the adviser industry, read from its own filed disclosures

What Your Peers Actually Charge

The adviser industry, read from its own filed disclosures.

One SEC firm roster of 16,621 rows funnelled to 10,418 registered advisers with at least one individual client. 10,491 structured brochure extractions and 8,377 raw brochures read by two independently written parsers, scored against thirty-six brochures read by hand. Raw Part 1 filings for 6,836 firms as the independent replication. No survey, no aggregator, no press figure anywhere in it.


The finding we did not get

We set out to measure the industry's effective fee rate — what clients pay, as against what brochures say. It cannot be built from Form ADV. The form has no field for revenue, fees billed, fees collected, fees waived, fees discounted, the billing basis or the billing frequency. The word "revenue" does not appear anywhere in a Part 1 filing.

We print that first rather than substituting a survey for it. Every "average effective fee" in circulation comes from a self-selected sample whose denominator you cannot inspect.


The finding we did get

The published price is 1.00%, and it is 1.00% at every size. Across 5,544 filed schedules the median first-band rate is 1.00%, the interquartile range 1.00% to 1.50%, and the bucket medians run 1.00 / 1.20 / 1.10 / 1.00 / 1.00 / 1.00 from the smallest firms to the largest. A third of schedules state exactly 1.00%. Big firms do not quote a lower first number — they quote a longer schedule that reaches a lower number further out, breaking at $2m where everyone else breaks at $1m and ending at $20m where the industry ends at $5m.

The schedule bites between $1m and $5m, and almost nowhere else. At $1m the median schedule is still on its opening rate, and 53.6% of schedules have not moved off the first band at all. By $5m the median has fallen to 0.825% and only 8.9% are still on the first band. The median schedule prices a $5m account at 73.2% of its own first-band rate.

A quarter of firms decline to publish a point rate. Of the 8,896 firms whose fee section could be read, 1,227 state only a range and 945 only a cap — 24.4%. And the largest firms are the least explicit, not the most.


The better finding: the denominator

Evaluate each firm's own schedule at its own average account — regulatory assets over the account count it files — and 65.5% of firms have an average sitting inside their own first band. Evaluate the same schedules at each firm's own average individual relationship — individual assets over the individual client count it files — and that share falls to 34.3%.

Same firms, same schedules, same formula. The median average account is $518,987; the median average individual relationship is $1,220,767, 2.35 times larger. Neither denominator is wrong. An account count is not a client count: a household with a joint account, two IRAs and a trust is four accounts. Nor is the client count a household count, because the form's own instruction puts a family's trusts and IRAs in the individuals line.

Any effective-rate figure quoting one of these without the other is quoting the denominator, not the rate.


The three numbers

1.00% Median stated first-band rate across 5,544 filed schedules. Four of the six size buckets sit exactly there; the other two, at 1.20% and 1.10%, sit above it
53.6% Schedules still charging their first-band rate on a $1,000,000 account
65.5% → 34.3% Firms whose own average sits inside their own first band, on accounts then on clients

A glimpse of the scorecard

Eight markers, re-scored quarterly against a fresh roster and a fresh brochure pull.

Marker Reading, 2026-09-12 What would change our mind
The headline rate 1.00%, 32.7% exactly at it It leaves 1.00%
Where the schedule bites 53.6% still on the first band at $1m Falls below 40%
The first breakpoint $1,000,000 Moves to $2,000,000
Point-rate disclosure 24.4% publish only a range or a cap Falls below 15%
The denominator gap 65.5% / 34.3% The gap narrows below 15 points
Filing quality 144 of 1,917 wrap filers file above their own assets Falls below 3%

Two things about the filings themselves

The form's numbering is a trap in at least one place. Item 5.K.(3) is the derivatives question; the custodian gate is Item 5.K.(4), and it routes to Schedule D Section 5.K.(3). A data column labelled 5K(3) is therefore the derivatives answer. This research walked up to that and prints it rather than burying it.

144 of 1,917 wrap filers report amounts that sum to more than their own reported assets, against the instruction printed on the form beneath the boxes — and 70 of them file the identical amount in all three boxes. The total by which those filings exceed their own reported assets is $533,210,384,678. It is a fact about filings, not an allegation about firms: a box repeated in a line nothing reconciles is a filing habit, and the same habit lives in fields nobody has checked.


What is in the full report

The distribution of stated schedules by size and by client mix; where each schedule bites, at seven account sizes, on both readings of a range schedule; the denominator problem worked bucket by bucket; what advisers cannot look up about each other — custodians, advisory staff, wrap posture, discretion; six segment snapshots, one per size bucket; eight markers with thresholds; and a 42-row claim-level citation register naming the document behind every claim, including six rows recording what was not verified.


What we cannot tell you

  • What any firm charges any client. There is no realised rate in Form ADV, and none is invented here. Three brochures in the hand-read sample print their own worked example of a blended rate. Those are the only effective-rate figures any adviser in the sample files.
  • How many custodians a firm uses. Two filed fields answer two narrower questions, and they disagree constantly: more than a quarter of firms naming exactly one concentrated custodian report two or more custodial persons in the other field.
  • Whether a firm runs model portfolios. There is no field. The 30.6% we can measure is a phrase count in brochures and is labelled as one.

What this research corrected in its own work

A corpus count was described as 9,839 distinct firms. It came from a name-to-file map, and an audit of 400 entries found 152 — 38% — naming a file that never mentions that firm. Read from each file's own header the true count is 5,850. A sampling rule stated as "every 155th of 5,598, giving 36 firms" actually returns 37, and the 37th was never read. A comparison between two readings of a schedule was printed upside down in the draft. Each is corrected in place, with the old figure quoted.

2,135 automated checks pass; 135 planted defects were caught, all of them.

Disclosures. Impersonal research, identical for every purchaser and tailored to no one. This report describes an industry from its own filed disclosures. It names no security, recommends no product, and makes no judgement about any individual firm. Finnacl Research is not an investment adviser. The author holds no position in any firm or security named here; no firm payment, no advertising.

That was the whole finding, not a teaser. The full report adds the evidence behind it: every figure with its source and date, the Tier-1 snapshots, a plain-language version you are licensed to hand to a client, and the quarterly re-scores for as long as the edition is live.