--- title: "Guide: Maintain AI Data Extractor" slug: "guide-maintain-ai-data-extractor" updated: 2026-08-13T11:37:43Z published: 2026-08-13T11:37:43Z canonical: "help.gong.io/guide-maintain-ai-data-extractor" --- > ## Documentation Index > Fetch the complete documentation index at: https://help.gong.io/llms.txt > Use this file to discover all available pages before exploring further. # Guide: Maintain AI Data Extractor > **Who can use this**: Business admin > > **Available on**: Gong Foundation > > **Ideal for**: RevOps [AI Data Extractor](/v1/docs/ai-data-extractor) continuously analyzes relevant conversations as new activity occurs on matching deals and accounts. Review your published Extractors regularly to make sure they still support a current business need and are scoped to the relevant deals, accounts, and conversations. Focus your review on two areas: Extractors you no longer need and the scope of Extractors you want to keep. ## Review your published Extractors As your business changes, an Extractor that was useful when it was created may no longer need to keep running. Start by identifying Extractors that no longer support a current business need. Consider unpublishing an Extractor if it: - Supported a temporary project or campaign that has ended. - Tracks products, competitors, messaging, or other concepts that are no longer relevant. - Duplicates information captured by another extractor. - Captures data that your organization no longer uses. [Unpublish an Extractor](https://help.gong.io/docs/ai-data-extractor#unpublish-or-delete-an-extractor) when you want it to stop updating but keep the data it has already generated in Gong. Existing data in Gong remains available, and data previously synced to your CRM remains in the CRM. > [!NOTE] > Note: > > [Deleting an Extractor](https://help.gong.io/docs/ai-data-extractor#unpublish-or-delete-an-extractor) removes the Extractor and its extracted data from Gong. Data previously synced to your CRM remains in the CRM. If you want to stop future updates while keeping existing data in Gong, unpublish the extractor instead. ## Adjust the scope of Extractors you keep An Extractor’s scope determines which records it runs on and which conversations Gong analyzes for those records. Set the scope to match the Extractor's purpose. Narrow it enough to exclude deals, accounts, and conversations that are not relevant, without excluding information the Extractor needs to answer the question. ![Settings for an AI data extractor, including target objects and data sources.](https://cdn.us.document360.io/2837c190-3ec8-4120-9116-6d427e774667/Images/Documentation/Data extractor maintain.png) For each Extractor you keep, check both parts of its scope: - **Target object scope**: The deals or accounts the extractor runs on. - **Sources**: The conversations Gong analyzes for each matching deal or account. ### Review the target deals or accounts [Use filters](https://help.gong.io/docs/ai-data-extractor#create-an-ai-field) to keep the Extractor focused on deals or accounts where its question is relevant. Match the filters to the Extractor's purpose. For example: - Limit a qualification Extractor to the deal stages where qualification is still relevant. - Limit a segment-specific extractor to the relevant account segment or region. - Limit an Extractor for a particular sales motion to deals where that motion applies. Combine filters when more than one condition defines the relevant population. For example, an Extractor that identifies pain points for enterprise customers could run only on enterprise accounts in a particular region. ### Review the conversation sources For each matching deal or account, check whether the Extractor needs **Calls** or **Calls & emails**, and how much conversation history it needs to answer the question. - Select **Calls** when calls provide enough information. - Select **Calls & emails** when relevant information may also appear in email. - Use the shortest conversation window that provides enough context to answer the question. A longer conversation window analyzes more historical conversations for each matching record. Account Extractors may analyze a wider set of conversations because relevant information can appear across multiple deals on the account. ## Validate scope changes before expanding them When you change an Extractor's scope, start with a narrow group of deals or accounts and [review the results](https://help.gong.io/docs/ai-data-extractor#test-the-ai-field) before broadening the filters. **Test it** checks one deal or account at a time. To validate the Extractor across several records, publish it with filters that limit it to a small segment. After reviewing the results, broaden the filters as needed. For example, limit the Extractor to deals owned by a specific team, review the results with the relevant team members, then broaden the filters if the results meet your needs. Check that: - The Extractor runs on the deals or accounts where its question is relevant. - The selected sources provide enough information to answer the question. - Irrelevant deals or accounts are excluded. - Relevant deals or accounts are not excluded. A team member who manages the business entities in Gong, such as deal and forecast boards. [Find my Business admin](/v1/docs/find-your-technical-or-business-admin). Gong’s core platform for gaining advanced visibility into your revenue operations. Also the prerequisite license needed for Gong’s add-on packages. Access to features depends on your [company’s plan and your assigned seat](/v1/docs/plans-and-seats). A team member who manages and optimizes revenue-related workflows by managing business entities such as metrics, targets, and dashboards.