> For the complete documentation index, see [llms.txt](https://sealights-docs.tricentis.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://sealights-docs.tricentis.com/test-optimization/test-optimization-strategy/enabling-test-optimization/cross-app-test-stage-strategy-configuration.md).

# Cross-App Test Stage Strategy Configuration

The **Strategy Configuration** panel is the centralized interface for managing Test Optimization for cross-app test stages. It provides intuitive, level-based configuration with granular controls over execution scope (e.g., Pull Requests vs. Baseline runs) and automatic full runs.

<figure><img src="/files/btJGEuCq0EB2aL5y890f" alt=""><figcaption></figcaption></figure>

### Access and Permissions

* **Availability:** Currently available only for **Cross-App** Test Stages.
* **Navigation:** Accessed via the **"Configure Strategy"** button on the Single Test Stage Overview page.
* **Roles:** Only users with **Admin** or **DevOps** roles can modify these settings. For all other roles, the configuration button is disabled.

### Enabling Test Optimization

At the top of the panel, a master toggle controls the activation state of Test Optimization for the stage.

* **Persistence:** If optimization is turned OFF, the configuration is saved. If re-enabled later, the last saved settings are restored.
* **Historical Data:** Turning optimization OFF does not affect the status of historic test cycles that ran while it was active.

### Test Optimization Levels

SeaLights offers three predefined levels to define the "aggressiveness" of test selection:

<table><thead><tr><th width="131.1796875">Level</th><th width="358.05859375">Behavior</th><th>Best Fit For...</th></tr></thead><tbody><tr><td><strong>Conservative</strong></td><td>Basic Test Optimization. Runs every test identified as impacted. No additional pruning. Ideal for CD/frequent deployments.</td><td>Teams with CD/hourly deployments requiring guaranteed full validation daily.</td></tr><tr><td><strong>Moderate</strong></td><td><strong>Default.</strong> Uses the basic Test Optimization algorithm plus a 80% "Maximum Capacity" filter. Prioritizes high-confidence testing for standard CI cycles.</td><td>Standard weekly/bi-weekly releases where automation is the primary gate.</td></tr><tr><td><strong>Aggressive</strong></td><td>Uses the basic Test Optimization algorithm plus a 60% "Maximum Capacity" filter. Selects the most critical subset for the fastest possible feedback.</td><td>Teams with high-volume code pushes or manual QA/Staging environments.</td></tr></tbody></table>

> *Note: The specific percentages for Moderate/Aggressive are configurable manually per customer.*

{% hint style="info" %}
Learn more about how recommendations are being generated for each optimization level on [Generating Test Recommendations](/test-optimization/how-it-works/generating-test-recommendations.md).
{% endhint %}

### Periodic Full Run

To mitigate the risk of skipping tests over time, users can schedule automated full runs.

* **Scope:**
  * **No Automatic Full Run:** Only runs on-demand (Read more about on demand full run on[Test Stage Strategy Overview](/test-optimization/test-optimization-strategy/test-stage-strategy-overview.md)).
  * **Skipped Impacted Tests:** Runs only the tests that were skipped by the optimization level in previous cycles.
  * **Entire Test Suite:** Runs all tests, regardless of impact.
* **Frequency:** Options include Daily, Weekly, Monthly, or a **Custom** interval (e.g., every 3 cycles, every 2 weeks).

**Automated Risk Management:** Periodic full runs ensure that deferred risks (skipped tests) are eventually addressed automatically.

**Dynamic Reset:** Changing the Optimization Level (e.g., from Moderate to Aggressive) will automatically update the "Full Run" frequency to match system defaults for that level.

### Optimization Scope

This section defines *which* types of executions get optimized:

* **All Executions (Default):** Optimizes every run.
* **PR Executions Only:** Baseline runs are always "Full Runs"; only PRs are optimized.
* **Non-PR Executions Only:** PRs are always "Full Runs"; baseline runs are optimized.

*Important: "Full Run" settings defined in the Strategy panel are automatically ignored for PR-level executions to ensure maximum speed during code review.*

{% hint style="info" %}
Learn how to configure Test Optimization for automatic testing, manual testing and shift-left testing, with our [detailed playbooks](https://docs.sealights.io/knowledgebase/guides/test-optimization/test-optimization-playbooks).
{% endhint %}

***

### FAQ

<details>

<summary><strong>If I turn Test Optimization OFF, will I lose my custom settings?</strong></summary>

No. SeaLights persists your configuration. If you toggle it back ON, your previous levels, filters, and full-run schedules will be restored.

</details>

<details>

<summary><strong>Does the "Maximum Capacity" filter apply to PRs?</strong></summary>

Yes, if the "Optimization Scope" includes PRs, the selected Strategy Level will be applied to PR executions. However, the *Periodic Full Run* schedule is ignored for PRs.

</details>

<details>

<summary><strong>How does "Custom" frequency work for Full Runs?</strong></summary>

You can set a specific integer value and a unit (Days, Weeks, Months, or Cycles). For example, setting it to "5 Cycles" ensures that every 5th execution is a full run, regardless of time elapsed.

</details>