Fathom Common Errors

Error Reference

1. 401 Unauthorized

Fix: Regenerate API key at Settings > Integrations > API Access.

2. 429 Rate Limited

Limit: 60 calls per minute across all API keys.

Fix: Implement exponential backoff. Batch requests.

3. Empty Transcript

Causes: Meeting still processing, recording too short, or audio quality issues.

Fix: Wait 5-10 minutes after recording. Check recording in Fathom UI.

4. Missing Summary

Cause: AI processing not complete.

Fix: Poll the recording endpoint until summary is available.

5. Webhook Not Firing

Fix: Verify webhook URL in Settings > Integrations > Webhooks. Test with:

curl -X POST https://your-url.com/webhooks/fathom \
  -H "Content-Type: application/json" \
  -d '{"type": "test"}'

6. OAuth Token Expired

Fix: Refresh the access token using your refresh token.

Quick Diagnostics

# Test API key
curl -s -o /dev/null -w "%{http_code}" -H "X-Api-Key: ${FATHOM_API_KEY}" \
  https://api.fathom.ai/external/v1/meetings?limit=1

Resources

• Fathom Help Center
• Fathom API Docs

Next Steps

For diagnostics, see fathom-debug-bundle.

Fathom Core Workflow: Meeting Analytics

Overview

Build automated meeting analytics: extract action items, sync to project management tools, analyze meeting patterns, and create follow-up workflows.

Instructions

Step 1: Batch Meeting Export

from fathom_client import FathomClient
from datetime import datetime, timedelta

client = FathomClient()

# Get all meetings from last 7 days
week_ago = (datetime.utcnow() - timedelta(days=7)).isoformat() + "Z"
meetings = client.list_meetings(
    limit=50,
    created_after=week_ago,
    include_summary="true",
)

for meeting in meetings:
    print(f"Meeting: {meeting['title']}")
    print(f"  Date: {meeting['created_at']}")
    print(f"  Summary: {meeting.get('summary', 'N/A')[:100]}...")
    for item in meeting.get("action_items", []):
        print(f"  Action: {item['text']} -> {item.get('assignee', 'unassigned')}")
    print()

Step 2: Action Item Extraction Pipeline

def extract_action_items(meetings: list[dict]) -> list[dict]:
    items = []
    for meeting in meetings:
        for action in meeting.get("action_items", []):
            items.append({
                "meeting_title": meeting["title"],
                "meeting_date": meeting["created_at"],
                "action_text": action["text"],
                "assignee": action.get("assignee", "unassigned"),
                "meeting_id": meeting["id"],
            })
    return items

# Sync to task tracker
def sync_to_linear(items: list[dict], api_key: str):
    for item in items:
        # Create Linear issue from action item
        pass

Step 3: Meeting Pattern Analysis

def analyze_meeting_patterns(meetings: list[dict]) -> dict:
    total = len(meetings)
    total_duration = sum(m.get("duration_seconds", 0) for m in meetings)
    total_actions = sum(len(m.get("action_items", [])) for m in meetings)

    return {
        "total_meetings": total,
        "total_hours": round(total_duration / 3600, 1),
        "avg_duration_min": round(total_duration / total / 60, 1) if total else 0,
        "total_action_items": total_actions,
        "avg_actions_per_meeting": round(total_actions / total, 1) if total else 0,
    }

Resources

• Fathom API Reference

Next Steps

For CRM sync and webhook automation, see fathom-core-workflow-b.

Fathom Core Workflow: CRM Sync & Follow-Up

Overview

Automate post-meeting workflows: sync meeting notes to CRM opportunities, send follow-up emails with action items, and maintain a meeting history database.

Instructions

Meeting-to-CRM Sync

def sync_meeting_to_crm(meeting: dict, crm_client):
    summary = meeting.get("summary", "")
    action_items = meeting.get("action_items", [])
    participants = meeting.get("participants", [])

    # Find matching CRM contact/opportunity by participant email
    for email in participants:
        contact = crm_client.find_contact(email=email)
        if contact:
            crm_client.log_activity(
                contact_id=contact["id"],
                type="meeting",
                subject=meeting["title"],
                body=f"Summary: {summary}\n\nAction Items:\n" +
                     "\n".join(f"- {a['text']}" for a in action_items),
                date=meeting["created_at"],
            )

Automated Follow-Up Email

def generate_followup_email(meeting: dict) -> str:
    actions = meeting.get("action_items", [])
    action_list = "\n".join(f"- {a['text']}" for a in actions)

    return f"""Hi team,

Thanks for the meeting: {meeting['title']}

Summary:
{meeting.get('summary', 'No summary available')}

Action Items:
{action_list if action_list else '- None recorded'}

Best regards"""

Meeting History Database

CREATE TABLE fathom_meetings (
    id VARCHAR PRIMARY KEY,
    title VARCHAR NOT NULL,
    created_at TIMESTAMP NOT NULL,
    duration_seconds INTEGER,
    summary TEXT,
    participant_count INTEGER,
    action_item_count INTEGER,
    synced_to_crm BOOLEAN DEFAULT FALSE,
    synced_at TIMESTAMP
);

Resources

• Fathom API Reference
• Fathom Integrations

Next Steps

For error troubleshooting, see fathom-common-errors.

Fathom Cost Tuning

Overview

Fathom pricing scales with per-seat licensing for team features, with primary cost drivers being transcript storage volume and recording hours consumed. Every meeting generates a transcript and AI summary that persist in storage. For organizations running dozens of meetings daily, unchecked transcript accumulation and redundant API polling for meeting data create unnecessary spend. Optimizing retrieval patterns and storage lifecycle directly reduces both API costs and plan overhead.

Cost Breakdown

API Call Reduction

class FathomTranscriptCache {
  private cache = new Map<string, { transcript: string; summary: string }>();

  async getTranscript(meetingId: string, apiFn: () => Promise<any>): Promise<any> {
    // Transcripts are immutable after generation — cache permanently
    if (this.cache.has(meetingId)) return this.cache.get(meetingId);
    const result = await apiFn();
    this.cache.set(meetingId, result);
    return result;
  }

  async listMeetings(params: { include_summary: boolean }): Promise<any[]> {
    // Always use include_summary=true to avoid N+1 calls
    // Fetches summaries inline with the list response
    const response = await fetch('/api/meetings?include_summary=true');
    return response.json();
  }
}

Usage Monitoring

class FathomUsageTracker {
  private apiCalls = 0;
  private readonly rateLimit = 60; // 60 req/min
  private windowStart = Date.now();

  async throttledCall<T>(fn: () => Promise<T>): Promise<T> {
    if (Date.now() - this.windowStart > 60_000) {
      this.apiCalls = 0;
      this.windowStart = Date.now();
    }
    if (this.apiCalls >= this.rateLimit) {
      const waitMs = 60_000 - (Date.now() - this.windowStart);
      await new Promise(r => setTimeout(r, waitMs));
      this.apiCalls = 0;
      this.windowStart = Date.now();
    }
    this.apiCalls++;
    return fn();
  }

  getUsageReport(): { callsThisMinute: number; remainingCapacity: number } {
    return { callsThisMinu