> ## Documentation Index
> Fetch the complete documentation index at: https://docs.focusalpha.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Earnings Call Transcripts

> Speaker-segmented earnings call transcripts and other company documents, searchable by meaning and citable back to the source passage.

Earnings Call Transcripts are quarterly earnings call records held in the FocusAlpha document corpus, alongside SEC filings and other company documents. Each document is associated with a company, a fiscal year and quarter, and a source URL, and is stored as ordered text segments. Transcripts are reached in two ways: by [semantic retrieval](/transcripts/semantic-retrieval) across the corpus, or by listing and fetching a specific document.

<Info>
  **Plan:** Free and above · **Credits:** 1 per call
</Info>

## What You Can Use Earnings Call Transcripts For

Find what management said about a topic — pricing, demand, margins, tariffs, competition — in their own words, with a citation. Read a full call in order rather than in fragments. Ground a generated answer in source passages so every claim can be checked. Pair management's language with the [guidance](/events/guidance) and [reported figures](/events/earnings-results) extracted from the same event.

## What Is in the Document Corpus

The corpus holds earnings call transcripts and company filings and documents. Every document carries a company, a fiscal `year` and `quarter` where the document has one, a document type, and a `source_url` pointing at the original.

Documents are stored as ordered **segments**. A segment is a contiguous passage of the document, and it is the unit that retrieval returns and cites.

## How to Find Which Periods Exist

Do not assume a company's available periods from the current date. Call the document listing endpoint for a ticker first: it returns that company's documents most recent first, and it is the authoritative answer to which years and quarters exist.

A guessed period returns stale passages rather than an error, so listing first is what prevents a silently wrong answer.

## Key Document Fields

`document_id` identifies a document and is what the fetch and segment endpoints take. `ticker` identifies the company. `year` and `quarter` place the document in fiscal time. The document type distinguishes a transcript from a filing. `source_url` links the original, and page numbers are carried where the source has them.

## Earnings Call Transcript Date Convention

`year` and `quarter` are the **fiscal** period the document covers, not the calendar date it was published. A company with a September fiscal year end reports its Q1 in January, and this dataset labels that document Q1 of the following fiscal year, as the company does.

## Earnings Call Transcript Sources

Transcripts are sourced from the calls themselves and are speaker-segmented. Filings in the corpus come from SEC EDGAR. Every document retains a `source_url`, so any passage returned by retrieval can be traced to the original document.

## List and Fetch Documents

Use the document listing endpoint to find a company's documents, then fetch one by id or read its ordered segments.

<ParamField path="ticker" type="string" required>
  The company's ticker.
</ParamField>

<ParamField query="year" type="integer">
  Restrict to one fiscal year, between 2000 and 2100.
</ParamField>

<ParamField query="quarter" type="string">
  Restrict to one fiscal quarter: `Q1`, `Q2`, `Q3` or `Q4`.
</ParamField>

<ParamField query="limit" type="integer" default="20">
  Maximum documents to return, up to 100. `offset` pages further.
</ParamField>

To read one document, pass its `document_id` to the document endpoint for metadata, or to the segments endpoint for its ordered text. The segments endpoint returns up to 100 segments per call, 20 by default.

## Related Datasets

See also [Semantic Retrieval](/transcripts/semantic-retrieval) to search the corpus by meaning, [Document Corpus](/transcripts/documents) for the listing and fetch endpoints in detail, [Management Guidance](/events/guidance) for forward-looking statements extracted from these calls, and [Reported Earnings Figures](/events/earnings-results) for the numbers the release printed.

<RequestExample>
  ```bash cURL theme={null}
  curl "https://api.focusalpha.ai/v1/documents/by-ticker/NVDA?limit=3" \
    -H "Authorization: Bearer $FOCUSALPHA_API_KEY"
  ```
</RequestExample>

<ResponseExample>
  ```json Response theme={null}
  {
    "documents": [
      {
        "id": "56eaec3c-c65e-4014-94b0-5aed74314f1c",
        "documentTitle": "NVDA Q2 2027 Earnings Call",
        "documentType": "earnings_call",
        "ticker": "NVDA",
        "year": 2027,
        "quarter": "Q2",
        "filingType": null,
        "sourceUrl": null,
        "status": "indexed"
      }
    ],
    "meta": {
      "total": 14,
      "limit": 1,
      "offset": 0,
      "latencyMs": 250
    }
  }
  ```
</ResponseExample>