AI Coding Insights is built on the OpenTelemetry telemetry your coding agents emit. Understanding how that telemetry maps onto the concepts below makes the tabs easier to read and the numbers easier to trust.

How the Data Flows

Claude Code emits token usage and model calls over OpenTelemetry on its own. That base stream is enough to populate cost and adoption. The Dash0 agent plugin adds the rest: session-level enrichment, tool calls, skills, and full conversation replay. It is the recommended path because it records the complete set of spans and attributes the views below rely on, rather than only what an agent reports on its own. Every session becomes a set of spans covering:

Tool calls: Every tool the agent invokes, with its duration and any errors.
LLM invocations: The model calls behind each response, including token usage and the model identity used to compute cost.
Errors: Failures the agent encounters, including failed tool calls.

Dash0 ingests these spans through the standard OTLP endpoint, stores them natively, and the AI Coding Insights page queries them. Because the data is grouped by model rather than by vendor, agents are comparable on the same terms. To populate the Productivity pillar, Dash0 joins these sessions with branch and pull-request events from the GitHub integration. Because the data is plain OpenTelemetry, you can also explore it directly in Tracing and Metrics.

Terminology

The terminology used throughout AI Coding Insights, in alphabetical order.

Active user

A developer whose coding agent emitted telemetry within the selected time range. Active users over time and the weekly-to-monthly active ratio (WAU / MAU) are derived from this count.

Agent

A coding tool that produces sessions, such as Claude Code or Cursor. AI Coding Insights groups data by model rather than by agent vendor, so several agents can be compared in the same views.

Cost

The spend attributed to model usage, aggregated by session, user, model, and team. Cost is computed from token usage, not subscription billing, using a Dash0-maintained pricing catalog that applies API-equivalent pay-as-you-go rates to the reported token counts.

Tip

This distinction matters because many developers use subscription plans where the provider's billing dashboard shows usage as zero or heavily subsidized, hiding the actual consumption. Dash0 calculates what those same tokens would cost at public API rates, giving you visibility into the real cost of AI assistance even when your subscription absorbs it today. This view becomes critical as providers move away from unlimited subscriptions toward usage-based billing, because the costs Dash0 shows now are what will really be paid later.

For developers on pay-as-you-go plans, the cost matches their provider bill. For flat-rate or subsidized subscriptions, it is the estimated cost at market rates. Models the catalog cannot resolve land in an explicit unpriced bucket rather than counting as zero.

Cycle time

The elapsed time between the start of agent-assisted work and the moment the resulting pull request merges. Reported as an average and as a p90 on the Productivity tab. Cost without cycle time is ambiguous, because it cannot separate efficient high-volume output from expensive inefficiency.

Derived Metrics

Aggregated time-series counters and histograms that Dash0 computes from spans, such as dash0.gen_ai.tokens.input and dash0.gen_ai.conversations. Use these to query token consumption over time and build dashboards. See Derived Metrics for the complete reference.

Model

The underlying language model that produced a response, for example claude-opus-4-8 or claude-sonnet-4-6. A single session can use more than one model. The Cost tab breaks spend down per model and ranks the top models by cost, with prompt counts alongside. Read that per-model cost against the cycle time on the Productivity tab to express throughput per dollar, the unit-economic answer to which model is worth the spend.

PR assisted

A pull request that an agent contributed to. The Productivity tab reports both total assisted pull requests and the subset attributed specifically to Claude, alongside codebase contribution per team. Together these make up the productivity signal.

Prompt

A single user request within a session. The number of prompts per user feeds the usage-pattern histograms on the Adoption tab.

Session

One run of a coding agent, scoped to a user and usually a repository. A session has a start time, a duration, one or more models, a cost, a prompt count, and a full conversation. Sessions are the unit you drill into on the Sessions tab.

Skill

An encoded workflow the agent can apply during a session. Skills used are counted on the Tools & Skills tab alongside tools and MCP servers.

Span Attributes

Metadata tags on individual spans, such as gen_ai.request.model, user.name, or dash0.gen_ai.vcs.repository.name. Use these to filter sessions and group metrics. See Span Attributes for the complete reference.

Team

The grouping used to attribute usage to a part of your organization. Teams are derived from the repository or user attributes on the incoming telemetry. Telemetry that carries no team attribute is grouped under unknown. See Set Up AI Coding Insights for how to attach team attributes.

Tool call

An invocation of a tool the agent has access to, such as Bash, Read, Edit, or a tool exposed by a connected MCP server. Failed tool calls are tracked separately so you can spot broken integrations.

Key Concepts