# Orphan Tables

## Overview

Orphan Tables are tables or views in your Snowflake environment that have **no upstream or downstream pipeline connections** and have received **zero queries** over the configured analysis window.\
They represent storage cost with no business value — dropping them frees space and reduces clutter.

***

## How It Works

### Detection Logic

Seemore evaluates every table and view across your connected Snowflake accounts by checking two conditions:

1. **No pipeline connections** — the table does not appear as a source or destination in any tracked pipeline (dbt, Airflow, Fivetran, Rivery, etc.).
2. **Zero usage** — no SELECT, INSERT, MERGE, or other DML has been executed against the table within the analysis window.

Tables that satisfy both conditions are flagged as orphans.

***

### Analysis Window

The default analysis window is **21 days**. Seemore looks at the most recent 21-day period of query history to determine whether a table has been accessed.

***

### Cost Estimation

Each orphan table insight includes:

* **Estimated annual storage savings** based on the table's current size and your Snowflake storage rate.
* **Row count** and **storage size** to help you assess impact.

***

## Remediation

Once you confirm a table is no longer needed:

```sql
DROP TABLE <database>.<schema>.<table>;
```

You can also mark the insight as **Done** in Seemore after remediation. If the table reappears in query history, the insight will not be regenerated.

***

## Configuration

| Parameter       | Default | Description                                    |
| --------------- | ------- | ---------------------------------------------- |
| Analysis window | 21 days | How far back Seemore looks for query activity. |

{% hint style="info" %}
If you have configured **Exclude Users from Usage**, excluded users' queries are ignored when determining whether a table has usage. This means tables accessed only by service accounts or bots will still be flagged as orphans.
{% endhint %}