> ## Documentation Index
> Fetch the complete documentation index at: https://agents.concierca.ch/llms.txt
> Use this file to discover all available pages before exploring further.

# Insights: Analytics & Economics

> What the Analytics and Economics panels show, and an honest note on which are still rolling out.

The **Insights** tab answers two questions about a mission: *is it working well?* (Analytics) and *is it worth it?* (Economics). Both are built from the mission's own run history.

## Analytics — is it working?

Analytics measures the mission's **reliability and behavior** over its runs:

| Metric                    | What it tells you                                                          |
| ------------------------- | -------------------------------------------------------------------------- |
| **Success rate**          | The share of runs that finished successfully, and how it trends over time. |
| **Per-step failure**      | Which steps fail most — where the mission actually breaks.                 |
| **Per-step duration**     | How long each step takes, so you can see the slow points.                  |
| **Per-step cost**         | Credits spent per step across runs.                                        |
| **Retries**               | How often steps have to be retried.                                        |
| **Average approval wait** | How long human approval steps sit before someone decides.                  |

Read together, these point you straight at the weak link — a step that fails often, costs the most, or waits the longest for a human — so you know exactly what to fix in the [Flow](/guide/mission-flow).

## Economics — is it worth it?

Economics puts a **business value** on the mission:

| Metric               | What it tells you                                            |
| -------------------- | ------------------------------------------------------------ |
| **Cost**             | What the mission's runs have cost.                           |
| **Realized savings** | The value it has saved, based on the estimate you configure. |
| **Net value**        | Savings minus cost.                                          |
| **ROI**              | Return on investment for the mission.                        |
| **Profitable runs**  | How many runs came out net-positive.                         |

<Note>
  **Savings are your estimate, not a guess by the system.** Economics multiplies your configured value-per-run by the runs that happened — it never invents a number. If the figures look off, adjust the value estimate in the mission's configuration.
</Note>

## An honest note on what's live

Some Insights panels are still **rolling out**. Where a panel doesn't yet have enough data — or isn't fully wired to your workspace — it shows **zeros or a "Soon" marker** rather than inventing numbers.

<Warning>
  A **zero** or a **"Soon"** badge on an Insights panel means *"not enough data / still being wired up yet"*, **not** *"this mission earns nothing"*. As runs accumulate and the remaining panels ship, the real figures fill in. Trust a number only once the panel is live and has run history behind it.
</Warning>

Everything grounded in actual run history — success rate, per-step failures, durations, retries, cost — reflects your real runs today. The value-and-ROI side of Economics fills in as you configure savings estimates and as the last panels come online.

<Note>
  From an AI client, pull these figures with `mission_analytics_get` and `mission_economics_get`. See the [Missions API reference](/for-agents/missions).
</Note>