Data model
About the data
This page documents the entity model, the taxonomies, and the query-recipe schema the discovery layer is built on. If you're building an AI system that ingests advisor data, the shapes here are stable.
Core entities
- AdvisorEntity — an individual advisor. Canonical record at
/advisors/[slug]. - FirmEntity — a registered investment advisory firm. Canonical record at
/firms/[slug]. - SpecialtyTaxonomy — a structured specialty (equity comp, crypto, cross-border, etc.). Browsed at
/specialties/[slug]. - AudienceTaxonomy — a client audience (founders, physicians, retirees, etc.).
- GeographyTaxonomy — region and metro taxonomies. Browsed at
/locations/[country]/[region]. - TrustSignal — a labelled + sourced fact attached to a record (e.g. "Fee-only verified — ADV Part 2").
- QueryRecipe — the stored query behind each list page.
- ListArtifact — the page produced by evaluating a query recipe.
- GuideArtifact — an editorial guide. Browsed at
/guides/[slug].
Query recipe schema
type QueryRecipe = {
slug: string;
kind: "advisor-list";
title: string;
description: string;
h1?: string;
intro: string;
filters: {
specialtySlugs?: string[];
audienceSlugs?: string[];
feeOnlyOnly?: boolean;
feeModels?: FeeModel[];
country?: "us" | "ca";
regionSlug?: string;
crossBorder?: Array<"us-to-ca" | "ca-to-us" | "cross-continent">;
virtualOk?: boolean;
educationMatches?: string[];
};
ranking: "default" | "specialty-depth" | "years-practicing" | "alphabetical";
refreshCadence: "weekly" | "monthly" | "quarterly";
}
Each list page at /lists/[slug] is the evaluation of
a recipe against the advisor graph. New recipes add new canonical
list pages without changing the template.
Machine-readable surfaces
- Schema.org markup on every entity and list page
- XML sitemap at /sitemap-index.xml
- LLM discovery manifest at /llms.txt
- Breadcrumb + ItemList + FinancialAdvisor + Article schemas
- Stable, slug-based URLs for every entity
What the discovery layer does not publish
The discovery surface publishes rich canonical pages. It does not expose bulk database dumps, CSV exports, or full-table endpoints. For structured data partnerships, email data@wellspent.co.