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

# Explainability

> Replay a served request or inspect an item score, stage by stage.

**Insights → Explainability** shows why an item was ranked where it was: its score at each stage, the rules that matched, and the path through the pipeline.

***

## Running an explanation

There are two ways in:

| Enter | What you get |
| - | - |
| **Request ID** | A **replay** of a decision that was actually served: the list the user saw, with the evidence for each item. Copy the `request_id` from an API response, an event or [Analytics](https://docs.neuronsearchlab.com/console/analytics.md). |
| **Item ID** (with an optional **User ID** and **Context**) | A **recomputation** of that item's current score path for that user in that context. Use it to test a rule or setting before or after you change it. |

Enter one, then run the explanation.

***

## What you see

### Final score and score breakdown

The item's final score, and how it was built: the base relevance from retrieval (how close the item is to the user or seed item, from 0 to 1), the learned ranker's score where it ran, and each adjustment from discovery settings and rules.

### Applied rules

Each active rule is listed with:

* **Rule name and type** (boost, pin, filter, etc.)
* **Match status** - a green **matched** badge if the rule's conditions were satisfied for this user-item pair, or a grey **no match** badge if conditions did not match.
* **Conditions** - what the rule checks (item metadata fields, segment membership, etc.)

This tells you immediately which rules are affecting this specific recommendation and which are not.

### Pipeline stages

Each pipeline stage is shown with a status badge:

| Status | Meaning | Colour |
| - | - | - |
| **passed** | Stage ran normally | Green |
| **disabled** | Stage is turned off in the pipeline | Grey |
| **skipped** | Stage was skipped for this request | Yellow |
| **partial** | Stage ran but with limited data | Amber |

### Item metadata

The item's stored metadata, so you can check whether a rule matched on the values you expected.

***

## Common debugging scenarios

### "Why is this item ranked so low?"

**1. Check similarity**

Check the base relevance score. If it is low, the user's behaviour does not strongly align with this item.

**2. Check suppressing rules**

Check for `bury` or `filter` rules that matched. A bury rule reduces the score; a filter rule removes the item entirely.

**3. Check higher-priority pins**

Check if a higher-priority `pin` rule placed another item in the position you expected this one to occupy.

### "Why is this item appearing when it shouldn't?"

**1. Look for unexpected rules**

Look for `boost` or `pin` rules that matched unexpectedly. The rule's conditions may be broader than intended.

**2. Check segment membership**

Check whether the user belongs to a segment that triggers a promotional rule.

**3. Verify item metadata**

Verify the item's metadata, because the rule may be matching on stale or incorrect data.

### "Why are rules not applying?"

**1. Check active status**

Check the rule's **Active** toggle; it may be inactive.

**2. Check the schedule**

Check the **schedule**; the rule may have a start date in the future or an end date in the past.

**3. Check the segment condition**

Check the **segment condition**. The user may not belong to the target segment.

**4. Check pipeline configuration**

Check the [pipeline](https://docs.neuronsearchlab.com/console/pipeline.md). If the **Rules** stage is off, no rules apply regardless of their configuration.

***

## Tips

* **Use Explainability before launching a campaign.** Test your rules against real user-item pairs to verify they behave as expected.
* **Combine with Analytics.** If engagement drops after deploying a new rule, use Explainability to check whether the rule is matching more broadly than intended.
* **Test edge cases.** Try explaining items for users who are in multiple segments, or items that match conditions for multiple rules, to understand priority interactions.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.