specificationVersion: "1.0"
name: Bitmovin FinOps
description: >-
  FinOps Framework 1.0 FOCUS-aligned cost management guidance for Bitmovin
  video encoding and streaming platform. Bitmovin's consumption-based model
  ties spend directly to encoding minutes, impressions, and AI analysis
  consumed. Understanding resolution multipliers, codec factors, and
  pass-count multipliers is essential for cost governance.
framework: FinOps Framework 1.0
alignment: FOCUS (FinOps Open Cost & Usage Specification)
provider:
  name: Bitmovin
  website: https://bitmovin.com
  pricingPage: https://bitmovin.com/pricing
  billingDashboard: https://dashboard.bitmovin.com
costDimensions:
  - dimension: Service
    values:
      - VOD Encoding
      - Live Encoding
      - Player
      - Analytics / Observability
      - AI Scene Analysis
      - CDN (via Streams)
      - Storage (via Streams)
  - dimension: Region
    notes: >-
      Encoding can run in Bitmovin-managed cloud regions or customer-owned
      cloud accounts (Cloud Connect). Costs may vary by cloud region and
      provider (AWS, Azure, GCP).
  - dimension: ResourceType
    values:
      - Encoding Job (VOD)
      - Encoding Job (Live)
      - Player License / Impression
      - Analytics Impression
      - AI Scene Analysis Minute
  - dimension: PricingUnit
    values:
      - Per encoding minute (10-second billing increment)
      - Per impression (Player / Analytics per 1,000)
      - Per input minute (AI Scene Analysis)
      - Per hour (Live Encoding)
costDrivers:
  - driver: Resolution
    impact: high
    description: >-
      Output resolution multiplies the per-minute rate. SD=1x, HD=2x, 4K=4x,
      8K=120x. Encoding to 4K is four times more expensive than SD for the
      same wall-clock duration.
    optimization: >-
      Limit ladder rungs to resolutions your audience actually streams.
      Remove 4K/8K rungs if analytics show low 4K consumption.
  - driver: Codec
    impact: high
    description: >-
      HEVC/H.265 carries a 2x multiplier versus H.264. AV1 may carry
      additional factors. Codec selection directly doubles or more the
      per-minute cost.
    optimization: >-
      Use H.264 for broad compatibility unless HEVC savings on CDN/bandwidth
      offset the encoding cost premium. Model total cost of ownership
      (encoding + delivery) per codec.
  - driver: Pass Count
    impact: medium
    description: >-
      2-Pass encoding adds a 1.25x multiplier; 3-Pass adds a 2.0x multiplier.
      This affects all VOD encoding jobs where multi-pass is configured.
    optimization: >-
      Reserve 2-Pass and 3-Pass for premium or long-tail content where quality
      gains justify the cost. Use 1-Pass (CRF/CBR) for bulk/low-priority
      content.
  - driver: Encoding Volume
    impact: high
    description: >-
      VOD minutes and live hours directly drive spend. The monthly free
      allowance (2,000 VOD minutes; 360 live minutes) provides a buffer for
      small workloads.
    optimization: >-
      Cache encoded outputs aggressively. Avoid re-encoding unless source
      changes. Use encoding templates and profiles to prevent configuration
      drift that triggers unnecessary re-encodes.
  - driver: Player Impressions
    impact: medium
    description: >-
      Player impressions are billed at $1.50 per 1,000 above the 10,000
      free-tier monthly impressions.
    optimization: >-
      Monitor impression counts via Analytics. Consider session-scoped
      impression reduction strategies and ensure only production traffic
      sources have valid license keys.
  - driver: Analytics Impressions
    impact: low-medium
    description: >-
      Analytics billed at $0.65 per 1,000 impressions above 100,000
      free-tier monthly impressions. Capped at 3 license keys, 10 exports,
      and 30-day data retention on pay-as-you-go.
    optimization: >-
      Filter out internal/test traffic from analytics license keys. Use
      separate license keys for staging vs production to isolate spend.
budgeting:
  recommendations:
    - Set monthly spend alerts in the Bitmovin Dashboard billing statistics page.
    - Model expected encoding minutes per month broken down by resolution tier
      and codec to forecast spend before production rollout.
    - Use the Bitmovin Statistics API (/v1/encoding/statistics) to pull
      programmatic usage data into internal FinOps tooling or cloud cost
      management platforms.
    - For predictable high-volume workloads, negotiate a committed contract
      via the Bitmovin enterprise plan to access volume discounts.
    - Purchase via AWS/Azure/GCP Marketplace if consolidated cloud billing
      is preferred for chargeback or commitment draw-down.
tagging:
  notes: >-
    Bitmovin does not natively expose cost allocation tags in the same way as
    cloud providers. Implement external cost attribution by correlating
    encoding job IDs, labels, or organizational metadata from the
    X-Tenant-Org-Id header with internal project or team identifiers.
anomalyDetection:
  recommendations:
    - Monitor x-ratelimit-remaining headers and 429 response rates as a proxy
      for unexpected traffic spikes that may signal runaway encoding jobs.
    - Set alerts on the Bitmovin Statistics API for sudden jumps in daily
      encoding minutes that exceed baseline by more than 20%.
    - Watch for 8001-8007 error codes which indicate account quota exhaustion;
      these can surface unplanned usage growth before a billing statement
      arrives.