> For the complete documentation index, see [llms.txt](https://knowledgebase.fabricdata.com/insights/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://knowledgebase.fabricdata.com/insights/origin-insights-mcp/best-practices.md).

# Best practices

Origin Insights MCP answers business questions in natural language about content strategy, distribution, and market analysis. You ask for the outcome you want; the assistant selects the right domain tool and `action` behind the scenes. These are effective ways to query it.

### Sample queries

| Question                                                               | What Origin Insights returns                                                                |
| ---------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- |
| *"Which titles have high demand but low availability in this market?"* | Titles ranked by demand with the catalog presence gap — an editorial/licensing opportunity. |
| *"How do platforms in Brazil compare on catalog and pricing?"*         | Side-by-side comparison: catalog size, content mix, prices, and tiers.                      |
| *"Where is this title available and under which business model?"*      | Country-by-country breakdown: platform, model (SVOD/TVOD/AVOD), and price.                  |
| *"What are the top sci-fi shows on Netflix in 2024?"*                  | Catalog exploration filtered by genre, platform, and year, with metadata.                   |
| *"What moved in the market this week?"*                                | A market radar: demand risers/fallers, catalog entries/exits, and plan repricings.          |
| *"Should we renew the license for this title in Brazil?"*              | A keep-or-drop verdict with confidence and supporting demand + reach evidence.              |

#### By role

* **Content buyer:** *"What should I acquire for Latin America — high-demand titles my market can't stream?"* · *"Which titles are leaving Netflix US soon, and where else do they stream?"* · *"Which upcoming releases have the strongest pre-release buzz?"*
* **Pricing analyst:** *"Compare subscription plans and prices across services in Argentina."* · *"How has the rent/buy price of this title changed over time?"* · *"Which services run free trials or promos in Mexico right now?"*
* **Market researcher:** *"Compare the catalogs of Netflix and Max in Spain — how much overlaps?"* · *"Which genres are trending in Brazil, and which titles drive that demand?"*

### Recommendations

* **Narrow the context:** specify country, platform, and time window where they apply — the answers are more precise and faster.
* **Ask a business question, not a query:** ask for the outcome you want ("where should we launch X?"), not for tables or columns.
* **Ask for reports:** for a full visual report on a title use `generate_title_report`; `add_executive_insights` enriches any report with executive analysis. Market-wide reports (radar, trends, calendar, plans, promos) live under the `reports_market` tool.
* **Lean on playbooks for decisions:** `compare_titles`, `evaluate_license_renewal`, `build_acquisition_brief`, and `calculate_platform_value` run several lookups and return one decision-ready answer.
* **SQL as a last resort:** `execute_sql` / `query_raw_data` exist for bespoke exploration, but the domain tools cover most cases.

### Example agent session

> **Goal:** decide whether a title is worth licensing in a market — check its demand, see where it already streams, and produce a shareable report.

```
1. titles { action: "find_title", keyword: "..." }             → resolve the title's uid
2. demand { action: "measure_title_popularity", "..." }        → how it trends in the market vs the average
3. availability { action: "find_where_to_watch", uid: "..." }  → where it currently streams + business model
4. reports { action: "generate_title_report", uid: "..." }     → one-call visual report (demand + availability + awards + pricing)
```

Each step is a single tool call with one `action`. You never chain actions inside a call — pick one action per call, feed the result (usually a `uid`) into the next.

### Errors, retries & safety

* **Read-only:** all tools read from the catalog; nothing writes back. Any `get_*` / `find_*` / `search_*` / `measure_*` action is safe to retry.
* **Title resolution first:** most title-level actions need a `uid`. Use `titles → find_title` (or pass `title_search`) to resolve it. If several catalog entries match, the tool returns a numbered candidate list instead of data — pick one, then call again with that `uid`.
* **Ambiguous matches:** when a title or platform name is ambiguous, the response lists options rather than guessing. Confirm the choice and re-issue the call.
* **Misses are explicit:** if something isn't in the catalog, the answer says so verbatim ("Not available in the Origin Insights catalog") rather than inventing data.
* **Validate before advanced queries:** for `raw_data → execute_sql`, use `list_tables` / `get_table_schema` first, and prefer the domain tools whenever one fits.

### Governance & data

* **Catalog-scoped:** every answer is grounded in the Origin Insights catalog for the turn — no external sources.
* **Freshness:** ask *"when was this data last updated?"* (`data_guide → check_data_freshness`) to see per-domain refresh timestamps before relying on a number.
* **Methodology:** ask *"how is the demand score built?"* (`data_guide → explain_demand_scores`) or any how-it-works question (`data_guide → ask_knowledge_base`).
* **Reports render in English** regardless of the query language.

{% hint style="info" %}
Tools update automatically: if a query wasn't supported before, it may be now without you needing to do anything.
{% endhint %}

### See also

* Getting started — data domains and the full tool catalog.
* MCP access — connection methods.

***

<details>

<summary>Related Articles</summary>

* [Getting started](/insights/origin-insights-mcp/getting-started.md)
* [MCP access](/insights/origin-insights-mcp/mcp-access.md)

</details>

> We hope you found this article helpful. If you have a question this article doesn't address, reach out on the [Service Desk](https://fabric.atlassian.net/servicedesk/customer/portal/336).


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://knowledgebase.fabricdata.com/insights/origin-insights-mcp/best-practices.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.