specificationVersion: "1.0"
id: recall-ai-finops
name: Recall.ai FinOps Framework
description: >
  FOCUS-aligned cost and usage model for Recall.ai meeting bot infrastructure. All costs are usage-based with no committed spend or platform fee on the Pay As You Go plan. Charges accrue per second of active bot time, transcription time, and extended media storage. Volume tiers are available on Launch and Enterprise plans.
url: https://www.recall.ai/pricing
framework: FOCUS 1.0
billingModel: usage-based
currency: USD
commitmentOptions:
  - type: PayAsYouGo
    description: No commitment, no platform fee. Pay only for active usage billed per second.
    minimumSpend: 0
  - type: Launch
    description: Volume-committed plan with discounted per-hour rates. Custom pricing negotiated with sales.
    minimumSpend: custom
  - type: Enterprise
    description: Custom contract with SLAs, HIPAA BAA, SSO, and white-glove support.
    minimumSpend: custom
skus:
  - id: recall-bot-standard
    name: Meeting Bot - Standard
    description: Standard meeting bot recording. Billed from joining_call event to terminal event, prorated to the nearest second.
    unit: recording-hour
    unitPrice: 0.50
    currency: USD
    billingPeriod: per-second
    platform:
      - Zoom
      - Google Meet
      - Microsoft Teams
      - Webex
      - Slack Huddles
      - GoTo Meeting
  - id: recall-bot-web-4core
    name: Meeting Bot - Web 4-Core Premium
    description: Premium 4-core web bot variant. Adds $0.10/hr to the standard recording rate.
    unit: recording-hour
    unitPrice: 0.10
    currency: USD
    billingPeriod: per-second
    notes: Charged in addition to the standard recording rate.
  - id: recall-bot-web-gpu
    name: Meeting Bot - Web GPU Premium
    description: GPU-accelerated web bot variant for high-compute workloads. Adds $1.00/hr to the standard recording rate.
    unit: recording-hour
    unitPrice: 1.00
    currency: USD
    billingPeriod: per-second
    notes: Charged in addition to the standard recording rate.
  - id: recall-transcription-builtin
    name: Transcription - Recall.ai Built-in
    description: Built-in AI transcription engine. Billed per second of transcription duration.
    unit: recording-hour
    unitPrice: 0.15
    currency: USD
    billingPeriod: per-second
    notes: Third-party transcription providers (e.g. Deepgram, AssemblyAI) are billed directly by those providers.
  - id: recall-storage-extended
    name: Extended Media Storage
    description: Storage for recording media beyond the 7-day free tier. Charged per media hour per 30-day rolling period.
    unit: media-hour-per-30-days
    unitPrice: 0.05
    currency: USD
    billingPeriod: per-30-days
    freeAllowance: 7 days per recording
  - id: recall-api-requests
    name: API Requests
    description: REST API calls to bot management, retrieval, and calendar endpoints. No additional charge.
    unit: request
    unitPrice: 0.00
    currency: USD
    notes: No rate-limit charges; no per-request fees.
costOptimizationRecommendations:
  - Use webhooks instead of polling the bot retrieval endpoint to avoid unnecessary API calls.
  - Terminate bots promptly after meeting ends to avoid charges for idle waiting_room time.
  - Use the standard bot variant unless GPU or 4-core resources are required; premium variants significantly increase costs.
  - Use third-party transcription providers if they offer lower per-minute costs than the Recall.ai built-in at $0.15/hr.
  - Monitor extended storage usage; retrieve and store recordings locally within the 7-day free window to avoid $0.05/media-hour/30-days charges.
  - Negotiate a Launch plan volume discount once recording hours are predictable to reduce the $0.50/hr base rate.
  - Choose the closest regional deployment to reduce latency and potential re-processing costs.