Token Economics

Full definition

Per-action cost equals (input_tokens × input_price) + (output_tokens × output_price), summed across every model call in a workflow. Levers include: prompt compression, caching, smaller routing for easy queries, distillation, batching, and choosing context windows tightly. AIDOLS uses token-economics modeling on every engagement under ai-strategy-consulting to translate model choices into P&L.

Why it matters

Most failed GenAI products fail not on quality but on margin. A copilot that costs $4 per active user per day and is sold for $30 per user per month is a structurally unprofitable product no matter how loved it is.

Example

A SaaS rebuilds its support agent: prompt compression cuts input tokens 38%, caching reduces effective input by another 60%, and easy queries route to a smaller model. Net per-conversation cost: $0.14 → $0.022.

Related terms

• Inference Cost
  Inference cost is the dollar cost of running a trained AI model in production — per request, per user, or per business outcome — and the operating expense that determines whether an AI feature has positive unit economics at scale.
• AI Total Cost of Ownership (TCO)
  AI Total Cost of Ownership (TCO) is the total cost of an AI system over its full lifecycle — including model and inference costs, infrastructure, integration, data preparation, governance, monitoring, retraining, talent, and exit costs — usually expressed as 3-year fully loaded.
• AI ROI
  AI ROI is the financial return generated by an AI investment relative to its total cost — including build, inference, MLOps, governance, and change-management cost — and the metric that ultimately determines whether an AI program survives the next budget cycle.
• Token / Tokenization
  A token is the basic unit a language model reads or writes — usually a sub-word fragment (about 4 characters of English text) — and the unit by which API pricing, context-window limits, and inference cost are all measured.