# Data Overview

CircleCI exposes data across your entire CI/CD pipeline — from high-level project and pipeline records down to individual job runs and branch-level performance insights. Here's what you can access through Coupler.io.

## Available entities

| Entity              | What it contains                                          |
| ------------------- | --------------------------------------------------------- |
| Pipelines           | Trigger source, status, creation time, VCS metadata       |
| Workflows           | Workflow name, status, duration, pipeline association     |
| Jobs                | Job name, number, status, start/stop times, executor type |
| Workflow jobs       | Jobs scoped to a specific workflow run                    |
| Insights metrics    | Success rate, throughput, mean duration, p95 duration     |
| Insights branches   | Per-branch pipeline metrics and failure counts            |
| Projects            | Project name, slug, VCS type, organization                |
| Specific projects   | Extended project details for a targeted project ID        |
| Contexts            | Context name, created date, environment variable keys     |
| Self IDs            | Authenticated user ID and external identifiers            |
| Self collaborations | Orgs and projects the user collaborates on                |
| Mes                 | User profile, login, analytics tracking ID                |

## Metrics and dimensions

#### Pipeline metrics

| Field         | Type      | Description                                       |
| ------------- | --------- | ------------------------------------------------- |
| id            | String    | Unique pipeline identifier                        |
| state         | String    | Pipeline state (created, errored, etc.)           |
| number        | Integer   | Sequential pipeline number                        |
| created\_at   | Timestamp | When the pipeline was triggered                   |
| trigger\_type | String    | What triggered the pipeline (push, API, schedule) |
| vcs\_branch   | String    | Branch that triggered the pipeline                |
| vcs\_revision | String    | Commit SHA                                        |

#### Workflow metrics

| Field         | Type      | Description                                 |
| ------------- | --------- | ------------------------------------------- |
| id            | String    | Unique workflow identifier                  |
| name          | String    | Workflow name                               |
| status        | String    | Status (success, failed, running, on\_hold) |
| created\_at   | Timestamp | Workflow start time                         |
| stopped\_at   | Timestamp | Workflow end time                           |
| pipeline\_id  | String    | Parent pipeline ID                          |
| project\_slug | String    | Associated project                          |

#### Job metrics

| Field          | Type      | Description                            |
| -------------- | --------- | -------------------------------------- |
| job\_number    | Integer   | Job number within the project          |
| name           | String    | Job name                               |
| status         | String    | Job status (success, failed, canceled) |
| started\_at    | Timestamp | Job start time                         |
| stopped\_at    | Timestamp | Job completion time                    |
| executor\_type | String    | Executor (docker, machine, macos)      |
| dependencies   | Array     | Upstream job dependencies              |

#### Insights metrics

| Field                 | Type    | Description                            |
| --------------------- | ------- | -------------------------------------- |
| success\_rate         | Float   | Percentage of successful pipeline runs |
| throughput            | Float   | Average number of runs per day         |
| mean\_duration\_sec   | Integer | Average run duration in seconds        |
| median\_duration\_sec | Integer | Median run duration                    |
| p95\_duration\_sec    | Integer | 95th percentile duration               |
| total\_runs           | Integer | Total number of runs in the period     |
| failed\_runs          | Integer | Count of failed runs                   |

#### Insights branches

| Field               | Type    | Description                        |
| ------------------- | ------- | ---------------------------------- |
| branch              | String  | Branch name                        |
| success\_rate       | Float   | Success rate for this branch       |
| total\_runs         | Integer | Total pipeline runs on this branch |
| mean\_duration\_sec | Integer | Average duration for this branch   |

## Common metric combinations

* **Workflow + Jobs** — join on `workflow_id` to see which jobs are dragging down workflow duration
* **Pipelines + Insights metrics** — combine to correlate pipeline volume with success rate trends
* **Insights branches + Projects** — spot which branches and projects are failing most often
* **Jobs + Workflows** — use Aggregate to calculate average job duration grouped by workflow name

## Use cases by role

{% tabs %}
{% tab title="Engineering managers" %}

* Track deployment frequency and change failure rate across teams
* Monitor pipeline success rates over time to measure CI/CD stability
* Compare workflow duration trends before and after infrastructure changes
* Use Insights metrics to build a DORA-style dashboard in Looker Studio or BigQuery
  {% endtab %}

{% tab title="DevOps / Platform teams" %}

* Identify the slowest jobs using p95 duration data and target them for optimization
* Analyze executor usage across jobs to right-size compute resources
* Pull context data to audit which environment variables are defined per context
* Use Append transformation to combine pipeline data from multiple projects into one report
  {% endtab %}

{% tab title="Development teams" %}

* Review branch-level insights to see which feature branches have the highest failure rates
* Track job failure patterns to catch flaky tests early
* Send pipeline status data to ChatGPT or Claude to auto-generate weekly CI health summaries
  {% endtab %}
  {% endtabs %}

## Platform-specific notes

* **Project ID vs. project slug** — some entities require a `project_id` (UUID from project settings), while others use a `project_slug` (e.g., `gh/myorg/myrepo`). Check the entity's parameter field carefully.
* **Workflow IDs** — you can supply multiple workflow IDs (one per line) to pull data for several workflows in the same data flow source.
* **Job number auto-fetch** — if you don't specify a job number, Coupler.io will auto-fetch it from Workflow jobs, so you only need to set this manually for specific targeted pulls.
* **Start date filtering** — the start date applies to pipeline-based entities. Insights entities have their own aggregation window on the CircleCI side.
* **Rate limits** — CircleCI's API is rate-limited; very large orgs with many pipelines may see slower syncs on first historical load.