Skip to main content
Adviser Screening selects investment advisers by description rather than by name. It ranks the SEC Form ADV roster on the axes read from each firm’s Part 2A brochure, crossed with the value or growth tier revealed by the firm’s own 13F book. One row is returned per firm, keyed by CRD.
Plan: Professional and above · Credits: 1 per call; screen facets are free
“Quantitative asset managers running long/short who issue ETFs” is one call here, and is unanswerable from the directory or from a single firm’s profile.

Which Adviser Endpoint to Use

Detailed Adviser Filter Pages

By firm type

Ten primary types, subtypes, services, role flags.

By strategy & style

What it runs and how it invests, declared vs revealed.

By products & scale

Vehicles, asset classes, geography, AUM.
If you hold a name, a CRD or a CIK, use the Form ADV Adviser Directory for directory rows, or the adviser profile for everything about that one firm. If you hold a description of a firm rather than its identity, use this screen.

Start With Adviser Screen Facets

Call the adviser screen facets endpoint first. It is free, and it returns every filterable value with the number of firms carrying it, plus a totals block. Filter values are exact slugs. A guessed spelling returns an empty page rather than an error, so the facets response is what tells you whether a value exists. The 13F style tiers are the only Title Case vocabulary in this dataset: Deep Growth, Growth-Biased, Blend / Core, Value-Biased, Deep Value.

Five Adviser Screening Axes

What the firm isprimary_type and subtype, read from the brochure text. A subtype is always a child of a primary type, so filtering on a subtype already implies its parent, and three primary types have no children at all. services_offered (what it sells) and role_flags (facts that hold at the same time as the primary type) sit on this axis. What it runsinvestment_strategy. How it investsmanagement_approach, analysis_basis, implementation_style, equity_styles, declared_vg_bucket, market_cap_focus, investment_horizon, and the flags uses_short_selling, uses_leverage, uses_derivatives, uses_options and uses_esg. What it sellsproduct_types is what the firm issues (mutual fund, ETF, private fund, wrap program, separately managed account, model portfolio); asset_classes is what the money buys (equity, fixed income, real estate, private credit). Two vocabularies, never interchangeable: an ETF issuer that buys bonds is etf plus fixed_income. What its book showsstyle_tier, computed from the firm’s 13F holdings. firm_type is an independent sixth label derived from ADV Part 1 numbers for 13F filers only. When it and primary_type disagree, that is two classifications disagreeing; report both with their sources.

The Screen Nothing Else Can Do

Cross declared_vg_bucket, which is what the brochure says, with style_tier, which is what the 13F book holds, to find managers whose portfolio contradicts their own marketing — for example declared_vg_bucket=value_biased with style_tier=Growth-Biased,Deep Growth. The facets response’s totals.with_both is how many firms that screen can reach at all. Report that number alongside the hits.

Every Brochure Filter Is Also a Coverage Filter

The brochure layer covers part of the roster and is still being ingested. Every brochure filter silently drops firms whose brochure is not in the store. has_brochure=true makes that narrowing explicit. The facets response reports totals.with_brochure against totals.firms.
Never report “only N firms do X” without both of those totals. Without them, a coverage gap reads as a finding about the industry.

Empty Is Not No

The brochure tags are conservative — set only where the brochure states the thing in so many words — so silence is common and is counted separately in the facets response as brochure_silent_on_*. Roughly five brochures in six name no market-cap focus, three in four name no equity style, and three in four name no geographic coverage. market_cap_focus=large_cap therefore returns the firms that wrote it down, not the large-cap managers, and geographic_coverage=emerging_markets is not the list of emerging-market managers.
Two axes are almost always silent and cannot carry a conclusion. implementation_style is unclear on roughly 98% of brochures, and management_approach on roughly 87%. Filtering implementation_style=discretionary returns the few dozen brochures that used that word, not the discretionary managers.

How Adviser Screen Sorting Handles Unknowns

Four of the five sorts drop every exempt reporting adviser. aum_total, aum_discretionary, accounts_total and employee_count are populated for registered firms and for no ERA at all, because an ERA files an abbreviated Form ADV that does not report them. Ranking on any of the four therefore removes that entire registration class silently.crd is the only sort that keeps every firm. When rows are dropped, the response carries an unranked block saying how many matched but could not be ranked.

Query the Adviser Screen

Use the institutions screen endpoint. Filters are combined with AND; a comma-separated value inside one filter matches ANY of the listed values. No filter is required.
string
Comma-separated list of the ten primary types.
string
Comma-separated list of subtypes. Implies the parent primary type.
string
Comma-separated strategy tags.
string
The brochure’s own value/growth bucket: deep_growth, growth_biased, blend, value_biased, deep_value, unclear, not_applicable.
string
The 13F-revealed tier, Title Case and spelled exactly.
string
What the firm issues.
string
What the money buys.
boolean
Restrict to firms whose Part 2A brochure is in the store. Set it before drawing any conclusion from a brochure filter.
number
Total regulatory AUM floor in USD. max_aum sets the ceiling. This family uses min_/max_ rather than _gte/_lte, and an unknown parameter returns a 400 rather than being ignored.
string
default:"aum_total"
aum_total, aum_discretionary, accounts_total, employee_count or crd. order is asc or desc.
integer
default:"100"
Rows per page, between 1 and 1000. Page through with cursor.
The full filter set also covers q, sec_registration_type, firm_type, state, country, the private-fund flags, brochure_scope, services_offered, role_flags, management_approach, analysis_basis, implementation_style, equity_styles, market_cap_focus, investment_horizon, the five brochure usage flags, geographic_coverage and has_style. See also Form ADV Part 2A Brochure Data for the axes this screen filters on, 13F Institutional Holdings for the book behind style_tier, and ETF Screening and Company Screening for the same pattern on other datasets.