spec: FinOps Framework 1.0
focusVersion: "1.0"
provider: Ada
providerId: ada
created: 2026-06-12
modified: 2026-06-12
url: https://www.ada.cx/blog/unpacking-ai-agent-pricing-resolution-based-vs-conversation-based-models/
notes: >
  Ada uses a consumption-based billing model tied to AI-handled customer interactions.
  Two pricing models are offered: conversation-based (charge per conversation handled,
  regardless of outcome) and resolution-based (charge only per successfully resolved
  conversation). Ada advocates conversation-based pricing for its predictability and
  scalability. Enterprise contracts are negotiated annually with usage-based overages.

billingModel: consumption
billingCycle: monthly
contractModel: annual
currency: USD

focusColumns:
  - column: BilledCost
    description: Cost charged for each conversation or resolution unit consumed in the billing period
  - column: ConsumedQuantity
    description: Number of conversations handled or resolutions achieved by the AI agent
  - column: ConsumedUnit
    description: conversation | resolution
  - column: ServiceName
    description: Ada AI Customer Service Platform
  - column: ServiceCategory
    description: AI and Machine Learning
  - column: ChargeType
    description: Usage
  - column: ResourceId
    description: Ada instance subdomain identifier

meters:
  - name: conversations
    description: Customer interactions handled by the Ada AI agent, regardless of resolution outcome
    unit: conversation
    priceLow: 0.35
    priceHigh: 1.50
    currency: USD
    model: conversation-based

  - name: resolutions
    description: Customer interactions fully resolved by the Ada AI agent without human escalation
    unit: resolution
    priceLow: 1.00
    priceHigh: 3.50
    currency: USD
    model: resolution-based

  - name: api-requests
    description: API calls made against Ada REST APIs beyond included platform quotas
    unit: request
    included: true
    note: API access included in platform contract; no separate per-request charge documented

principles:
  - name: Understand Usage
    description: Monitor conversation and resolution volumes monthly to forecast annual spend against contract minimums and overage thresholds
  - name: Quantify Business Value
    description: Track AI resolution rate (target 25-75%+) to measure cost-per-resolution versus cost-per-human-escalation ROI
  - name: Optimize Cloud Usage & Cost
    description: Prefer conversation-based pricing model for cost predictability as AI resolution rates improve over time
  - name: Align Usage to Business Needs
    description: Size annual contract volumes to expected conversation volume with 10-20% buffer to avoid mid-year renegotiation
  - name: Benchmark Performance
    description: Compare Ada cost-per-resolution against human agent cost-per-ticket to quantify automation ROI

costAllocation:
  dimensions:
    - name: Channel
      description: Allocate costs by deployment channel (web, mobile, voice, messaging)
    - name: BusinessUnit
      description: Allocate costs by internal team or product line using Ada
    - name: Language
      description: Track multilingual support costs separately as they may incur add-on fees