> For the complete documentation index, see [llms.txt](https://docs.seemoredata.io/external-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.seemoredata.io/external-docs/fundamentals/our-features/waste-reduction.md).

# Waste Reduction

Seemore continuously analyzes your Snowflake environment to surface concrete savings opportunities across storage, compute, and licensing. Each insight includes estimated annual savings so you can prioritize high-impact actions.

***

## Orphan Tables

Tables and views with **no pipeline connections and zero query activity** over the analysis window. Dropping them reclaims storage and reduces clutter.

{% content-ref url="/pages/hmPDjT1vTDLqMGYrjGor" %}
[Orphan Tables](/external-docs/fundamentals/our-features/waste-reduction/orphan-tables.md)
{% endcontent-ref %}

## Unused Data Flows

Tables that are **actively written to by pipelines but never read**. The upstream pipeline spend is pure waste.

{% content-ref url="/pages/Ff5kUKbxA44YOvbg0ZIR" %}
[Unused Data Flows](/external-docs/fundamentals/our-features/waste-reduction/unused-data-flows.md)
{% endcontent-ref %}

## Frequency Optimization

Tables whose **write frequency far exceeds their read frequency**. Reducing refresh cadence saves compute without affecting consumers.

{% content-ref url="/pages/O6ObPyrtrWwZpK6cOmXJ" %}
[Frequency Optimization](/external-docs/fundamentals/our-features/waste-reduction/frequency-optimization.md)
{% endcontent-ref %}

## Inactive Users

BI platform users (Tableau, PowerBI) who have **not logged in within a configurable threshold**. Reclaiming their licenses reduces subscription costs.

{% content-ref url="/pages/o6dpFF6JiTg4LXy23Yaa" %}
[Inactive Users](/external-docs/fundamentals/our-features/waste-reduction/inactive-users.md)
{% endcontent-ref %}

## Compute Idle Optimization

Warehouses kept alive by **high-frequency queries that prevent auto-suspend**. Reducing query frequency or adjusting auto-suspend settings reclaims idle compute spend.

{% content-ref url="/pages/vO825M1cBmlkaeNbOIGZ" %}
[Compute Idle Optimization](/external-docs/fundamentals/our-features/waste-reduction/compute-idle-optimization.md)
{% endcontent-ref %}

***

## Exclude Users from Usage

Service accounts, ETL bots, and monitoring tools can generate significant query activity that inflates usage metrics. The **Exclude Users from Usage** setting lets you filter these users out of usage calculations so insights reflect real human consumption patterns.

### How to Configure

1. Navigate to **Waste Reduction → Insights Configuration** (top-right button).
2. Select the Snowflake integration you want to configure.
3. Check the users you want to exclude from usage calculations.
4. Click **Save**.

<figure><img src="/files/xuPVyXwB9BDheOFTsEmo" alt="Walkthrough: excluding users from usage calculations"><figcaption><p>Walkthrough: excluding users from usage calculations</p></figcaption></figure>

### Which Insights Are Affected

| Insight                       | Affected? | Details                                                                                                              |
| ----------------------------- | --------- | -------------------------------------------------------------------------------------------------------------------- |
| **Orphan Tables**             | Yes       | Excluded users' queries are ignored when determining whether a table has usage.                                      |
| **Unused Data Flows**         | Yes       | Excluded users' reads are ignored, so tables read only by bots still appear as unused.                               |
| **Frequency Optimization**    | Yes       | The Snowflake queries that detect write-heavy / read-light patterns filter out excluded users.                       |
| **Inactive Users**            | No        | This insight queries BI platform login activity, not Snowflake usage.                                                |
| **Compute Idle Optimization** | No        | This insight evaluates warehouse-level uptime patterns; all query activity (including service accounts) is relevant. |

{% hint style="info" %}
Changes take effect the next time insights are recalculated (typically within 24 hours).
{% endhint %}