Thought Efficiency Index (TEI)
Definition
A composite score that captures how efficiently an application converts LLM calls into completed work. Most useful for relative comparison between apps over time, not as an absolute benchmark connecting spend to value for Leadership, Finance, and FinOps Practitioner Personas .
Formula, as published
TEI = Traces / ( LLM Calls x Avg Input Tokens )
- Where:
- A “trace” is a record that captures a complete timeline of an agent’s behavior, including timestamps, errors, inputs/outputs, and contextual metadata.
- Data sources:
- Cloud Cost Billing Data Usage Telemetry (eg. Azure Cognitive Services metrics) LLM Observability/Application Telemetry
- Example:
- Copilot Assistant: Good call-to-trace ratio 0.057 TEI Score = 95K Traces / (410K LLM Calls x 4,100 Avg Input Tokens) Order Processing Agent: Complex agent 0.009 TEI Score = 45K Traces / (620K LLM Calls x 8,500 Avg Input Tokens)
Calculate it
Example values to start with; replace them with yours.
From the FinOps Foundation
More Budgeting KPIs
Definition and formula: Thought Efficiency Index (TEI), FinOps KPI library, FinOps Foundation, licensed CC BY 4.0, read 16 Oct 2026. Changed: HTML removed, dashes and British spellings changed, formula split from its data sources and examples. The calculator and the notes are Finitizer’s.
Measured once, a KPI is a number. Tracked daily, it is a practice.
Finitizer computes the KPIs your billing data supports every day, by team and product, across AWS and Google Cloud.
See how Finitizer works