> For the complete documentation index, see [llms.txt](https://help.delpha.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.delpha.io/core-concepts/trusted-sources-and-how-delpha-weighs-them.md).

# Trusted sources & how Delpha weighs them

Every Delpha resolution cross-checks trusted, verifiable sources in a strict priority order. See the source hierarchy and how each level is weighed, with Ultimate Parent as the reference example.

Every Delpha resolution is grounded in **trusted and verifiable sources**. Delpha's specialized LLMs analyze and consolidate these sources; the final verdict is reconciled deterministically against your record, with the evidence attached. You can see this hierarchy applied step by step in [Understanding the Ultimate Parent Analysis](/use-cases/ultimate-parent/how-selection-works/understanding-the-ultimate-parent-analysis.md).

Each use case defines its own source hierarchy in strict priority order. Below is the hierarchy for **Ultimate Parent**, the reference example:

1. **Company Website (Direct Digital Footprint):**
   1. *Logic:* The primary signal of commercial identity. Delpha’s agents analyze the entity’s self-declared affiliation and branding to determine if it presents itself as an independent operator or a subordinate unit.
2. **Global Brand Websites (Commercial Network Verification):**
   1. *Logic:* Validating the operational link by checking if the subsidiary is officially listed within the parent’s commercial ecosystem (e.g., location finders, brand portfolios), which confirms shared infrastructure.
3. **Press Releases (Strategic Announcements):**
   1. *Logic:* M\&A activity dictates operational change.
   2. *Rule A:* News < 3 years old is Primary Evidence (recent acquisitions often imply active integration efforts).
   3. *Rule B:* News > 3 years old is Secondary Evidence (valid only if not contradicted by Source 1 or 2, ensuring divested units are not incorrectly linked).
4. **Securities Filings (Regulatory Disclosures):**
   1. *Logic:* Used to resolve ambiguity in complex structures (e.g., 10-K filings), distinguishing between operational subsidiaries and purely financial “Variable Interest Entities.”
5. **Official Business Registries (Legal Baseline):**
   1. *Logic:* The “Fail-Safe.” Used only when commercial sources (1-4) are silent. This ensures that legal shells are only used as a parent of last resort when no commercial parent can be identified.

{% hint style="warning" %}
Delpha analyzes all the sources above to [**reflect how companies operate in the real world**, not who legally owns them.](/use-cases/ultimate-parent/how-selection-works/ultimate-parent-concepts-brands-vs.-operating-structures.md)\
This approach allows Delpha to accurately represent **franchise and multi-brand networks**, where multiple legal entities operate under a single brand identity, ensuring the Ultimate Parent recommendation aligns with **sales, marketing, and go-to-market realities** rather than pure legal ownership.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.delpha.io/core-concepts/trusted-sources-and-how-delpha-weighs-them.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.