What are MELT metrics?
MELT metrics are the four types of telemetry behind modern observability: Metrics, Events, Logs and Traces. Metrics are numbers tracked over time, such as latency, error rate or token cost. Events record notable changes, such as a deployment or an agent escalating to a person. Logs hold the detail of each step. Traces connect every step of one request into a single chain, so you can follow cause to effect.
What it is
MELT is the standard shape of observability data in modern software — originated in traditional systems monitoring and now the foundation of AI agent telemetry as well. Metrics are aggregated numbers (latency, error rate, cost, token usage). Events capture significant state changes (a tool call failed, an agent escalated, a handoff happened). Logs record per-step contextual data (the exact prompt sent, the tool response received, the reasoning recorded). Traces link every step of a single request or agent run into one causal chain so you can see cause-and-effect from input to output.
Why it matters
When you instrument AI agents, you are emitting MELT data extended with AI-specific signals — token usage, hallucination scores, drift detection, agent decision paths. Understanding the MELT model makes every observability tool click: OpenTelemetry, Datadog, Salesforce Agent Observability, LangSmith, Arize — they all speak this language. Picking a tool without understanding MELT leads to partial instrumentation and blind spots in production.
Key components
- Metrics — aggregate numeric signals over time (p95 latency, error rate, cost per run, tokens per hour)
- Events — discrete occurrences of significance (tool failure, escalation, deploy, config change)
- Logs — structured contextual data attached to a specific moment (the full prompt, the tool response body)
- Traces — a causal chain of operations that together made up one logical request or agent run
How it works
- Instrument the system to emit MELT data at every meaningful boundary (function calls, tool invocations, LLM calls)
- Send the data to an observability backend that organizes it by trace ID, time, and service
- Use metrics for trends and alerts, traces for debugging single runs, logs for deep context, events for state awareness
- For AI agents, extend the standard MELT payload with token counts, model parameters, hallucination flags, and quality scores
How it connects
Salesforce Agent Observability and Agentforce natively emit MELT data — every time an Agentforce agent runs a topic, calls an action, or hands off to a human, that activity surfaces as metrics, events, logs, and traces you can monitor in Data Cloud or route to tools like Datadog and Splunk via OpenTelemetry.
Good to know
MELT predates the AI agent boom — it comes from the general observability world (DevOps, SRE). The AI observability tools you see today (LangSmith, Arize, Salesforce Agent Observability) are MELT implementations enriched with AI-specific signals. OpenTelemetry is the vendor-neutral open standard for emitting MELT data, so you can instrument once and send the data to whichever backend you prefer.
Frequently asked questions
What does MELT stand for in observability?
MELT stands for Metrics, Events, Logs and Traces. It extends the older "three pillars of observability" (metrics, logs and traces) with events, a term popularised by New Relic.
How does MELT apply to AI agents?
MELT is the shape of AI agent telemetry. Metrics track cost, latency and resolution rate; events capture tool-call failures and handoffs; logs keep the exact prompt, tool response and reasoning for each step; and traces tie one conversation or agent run together end to end, so you can see why the agent did what it did.
Related terms
Agent Observability
The practice of inspecting, debugging, and understanding AI agent behavior at runtime by consuming agent telemetry — traces, metrics, logs, and events — through dashboards, alerts, and evaluation tools.
Agent Telemetry
The runtime data emitted by an AI agent — every decision, tool call, input, output, latency, and cost — used to monitor reliability, quality, and spend in production.

