Who can use this: Business admin
Available on: Gong Foundation
AI Data Extractor is an AI agent that turns customer conversations into structured, reliable data and reduces the need for manual data entry. It automatically pulls details from calls and emails, and saves them in Gong and can optionally write them to the CRM fields you choose, so you can rely on accurate, consistent data.
Instead of asking reps to remember every competitor, risk, or reason for churn and update it by hand, AI Data Extractor keeps fields up-to-date based on what customers actually say. This improves data quality, supports better reporting, and frees reps to focus on selling.
What are some use cases?
AI Data Extractor gives you reliable, structured data on key topics without adding work for your team. Common examples include:
Track account and deal details: Make sure details specific to the account or deal aren’t missed, such as the number of requested seats, the discount offered on multi-year deals, the name of the CRO, or the expected closing date.
Track competitors: Capture which competitors were mentioned in deals or accounts so you can analyze win rates, pricing pressure, and positioning across segments.
Measure methodology adherence: Populate fields that track whether reps followed your sales methodology, asked key questions, or covered required topics.
Identify decision makers and influencers: Extract who is involved in the deal, what their roles are, and how engaged they are, then write this into deal or account fields.
Capture product interest and use cases: Record which products, use cases, or business problems were discussed, so you can analyze demand and tailor follow-ups.
You can create an extractor for any question that can be answered from customer conversations.
How are extractors defined?
Each extractor is made up of three main parts:
Question: What you want Gong AI to answer based on conversations. For example: “Which competitors were mentioned in this deal?”
Additional instructions (optional): Extra context that helps Gong AI answer consistently. Detailed instructions significantly improve the quality of the extractors’s answers. For example, you can explain how your company defines a specific term or methodology, or what to prioritize when multiple answers appear.
Target object and data type
Target object: Either deals or accounts.
Data type: The data type of the field the answer is stored in. Options are:
Yes/No: A true or false answer
Free text: An open-ended text answer
Single-select picklist: One value from a predefined list of options
Number: A numeric value
Date: A date
Range: A numeric value with a configurable minimum and maximum
You can choose more than one data type for the same extractor.
Set the target object
The target object, either deal or account, defines which object the data extracted by the extractor is related to, and where it should be saved. If the data you are extracting is related to deals, select Deals. If the data you are extracting is related to accounts, select Accounts. The target object is used by Gong to identify the most relevant calls and emails to use to find the answer to the question posed by the extractor.
For accounts, a wider range of calls and emails may be examined, as the answer to the question may be found over calls that relate to different open deals. For deals, Gong can identify the calls that are related to the deal and extract the answer from those calls.
Filter the target object
Note:
This feature is rolling out gradually throughout July 2026 and some functionality may vary.
By default, the extractor runs on all deals or accounts that have had activity in the last 30 days. This can be heavy on resources, especially if there are deals and accounts that are not relevant for a specific extractor. You can add filters to control which deals and accounts are analyzed, for example to deals in a specific pipeline stage or accounts in a particular region. This will reduce the resources required by the data extractor.
Set the data type
The data type sets the format you want the answer to be generated in. The data types available allow you to ask questions that return long and detailed answers, quick answers, numerical data, or a date. You can also select more than one type of formatting.
For example, if you want to extract data on “Which competitors were mentioned?”, you can set the data types to be:
Yes / No: Indicates if a competitor was mentioned
Text: List the competitors mentioned
Single select picklist: Displays the primary competitor
How AI Data Extractor works
AI Data Extractor runs behind the scenes, using Gong’s understanding of conversations together with your CRM setup. AI Data Extractor runs on published extractors only. You can publish up to 20 extractors per workspace.
The AI Data Extractor works as follows:
Admin defines an extractor: A business admin (for example, enablement or RevOps) creates an extractor and configures:
The question and any additional instructions
The target object (deal or account)
One or more output data types (Yes / No, text, single picklist)
Account data saved in Gong: Extractors with an account target object are automatically saved in Gong. You can also choose to export them to an existing CRM field.
Deal data mapped to CRM fields: Extractors with a deal target object must be mapped to a selected CRM field.
You can only choose a CRM field that is already imported into Gong.
Gong does not create new fields in the CRM, you must export data to an existing CRM field.
Gong analyzes conversations: After the extractor is published, Gong checks accounts and deals with recent customer activity and analyzes their calls over a rolling historical window. Gong AI extracts answers to your question based on what was actually said as follows:
Gong selects deals or accounts that had a call in the last month or received an inbound email.
For those deals and accounts, answers are generated according to call data from the last six months. Data from other activities such as emails are not used.
AI Data Extractor updates fields automatically: Gong recalculates answers for active accounts and deals whenever a new relevant conversation takes place. When new or better information is found, the previous value is overwritten so your data stays current.
Because AI Data Extractor runs automatically, reps do not need to do anything for the fields to be updated. The agent keeps working in the background as new conversations come in.
Which conversations are included in the analysis?
AI Data Extractor may use data from any non-private call or emails that has been imported to Gong. This can include calls and emails the team member who sets up the extractor doesn’t have permission to view, for example if their permission profile restricts their access to calls or emails.
When testing, the team member will be able to see the extractor’s results, even if they don’t have access to the underlying call. They will not be able to view the call content itself.
Where you can see extracted data
AI Data Extractor saves the values in Gong as well as exporting to your CRM, so you can use the results anywhere those fields are available. For example:
In Gong:
Deal boards, CRM fields can be added as columns on your boards
Account pages, as part of the account details
Other Gong views that you can customize
In your CRM:
List views
Reports and dashboards
Automations and workflows that depend on field values
From the perspective of your team, the fields behave like any other CRM field. The difference is that they are filled and maintained by AI rather than by manual data entry.
CRM mapping
AI Data Extractor can export generated answers to CRM fields.
For extractors whose target object is Deals, CRM mapping is required. For extractors whose target object is Accounts, CRM mapping is optional because account extractor results are automatically saved in Gong.
To be available for CRM mapping, CRM fields and custom objects must be imported into Gong. To import a custom object, contact Gong Support.
In addition to standard CRM fields, extractors can be mapped to attributes in CRM custom objects associated with accounts or deals.
Custom object mapping is supported only when there is a one-to-one relationship between the account or deal and the custom object. If multiple custom object records are associated with the same account or deal, Gong can't determine which record to update, so the update won't be applied.
In addition, the custom object instance must already exist on the account or deal. Gong can update an existing object, but it doesn't create new records.
Create an extractor
To create an extractor:
From the left sidebar, click Admin center.
In the Agent Studio tab, hover over AI Data Extractor and click Settings to open the AI Data Extractor management page.
Click + Add new extractor or
> Edit to edit an existing extractor.In Field concept, enter the question you want the extractor to answer.
Click Refine to get AI suggestions of how to improve the question so that the data extractor will be able to provide the best answers
Click Replace to replace your question with the suggested wording, or click anywhere on the screen to hide the AI suggestions.
In Provide context, instructions, or examples to help the AI find the right information, add detailed instructions to improve the quality of the answers. See Tips for writing effective questions.
In Target object, select whether you want to save the answer in the Deals or Accounts object.
Click Add Filters, select the fields you want to filter by, and set a value for each filter. The available filters depend on the target object you selected.
In Sources, select whether you want to use calls or calls and emails to generate the answer for the extractor.
Select how far back to look for data in relevant conversations.
In Data type, mark the types of answers you want to receive. You can mark multiple data types. Options are:
Yes/No
Text: Select whether the text should be:
Short: Answers the question with no additional explanation
Detailed: Full answer including explanations and additional information
Bulleted list: Answer in bullet format.
Single-select picklist
Number
Date
Range: The range is set from 1 to 5 by default. Adjust the range according to your requirements.
In CRM mapping, select the CRM field to update the answer in. Imported CRM fields and custom object attributes appear in the list. If your target object is Deals, this field is mandatory. If your target object is Accounts, this field is optional.
Click Save if you are not ready to publish the extractor and want to continue making changes.
Test the extractor
Test the extractor to see that you are getting accurate answers to your question before publishing.
Click Test it.
Select the account or deal you want to test the extractor on.
Check the results. The results include a value for each data type selected, together with an explanation of why the result was chosen.
If necessary, refine the question or context and click Run test again. You may have to do this several times before publishing.
Click Change to run the test on a different account or deal.
Click Publish when you’re satisfied with the answers.
Tips for writing effective questions
AI Data Extractor is designed to extract specific, concise answers to questions that should have an objective correct answer. The types of questions you ask and the guidance that you provide directly affect the quality of the results.
General
Use natural language. Ask, don’t prompt: Write the way you’d speak to a new colleague with no prior context. Example: “Which competitors did the customer mention during this deal?”
Be specific: Ask clear questions about the conversation. Example: “Did the customer request a proof of concept or trial?”
State your intention: Explain the goal behind the question. The same question can have different intended outcomes. Example: “Which competitors were mentioned? Focus only on companies that were considered as an alternative option for conversation intelligence"
Clarify terms: Define terms that are ambiguous or jargon.
Questions
Ask one question per extractor. Results will be less accurate if the question is compound or attempts to answer multiple things at once.
Only ask questions that can be answered from the content of a conversation. Questions that rely largely on metadata won’t be answered accurately. Example: “What stage is the deal currently in?” Stage is a CRM field. It isn’t mentioned in the conversation, the AI can’t infer it reliably.
Provide context, instructions, and examples
Use examples to guide Gong AI
Include one or more clear examples in your instructions. Examples show the model exactly how you expect the answer to look and help it stay consistent. Providing examples reduces variability and improves accuracy.
For picklist fields
Provide guidance and definitions on when each option should be chosen, particularly if the options use company/industry specific language that may not be specifically found in a conversation. This helps the model differentiate between similar options and understand nuances that may not be specifically obvious.
When providing extra context, include common variations or alternate phrasing that might refer to a specific picklist option.
If there is an “Other” option you can map a text field for this extractor and then collect those outputs there. The text field is updated even when the answer isn’t Other, so your reporting logic should only consider the text field if the picklist is set to “Other”. You won’t be able to constrain the Text field output to a single/limited word output so you may not be able to aggregate this field but it can be used to identify new options to add to the picklist
For numeric fields
Clearly specify the type of value you are looking for, such as number of seats or headcount, percentages, or time units.
If the conversation includes a range, explain how the Data Extractor should handle it. For example, indicate whether it should capture the minimum value, maximum value, or an average.
For date fields
Clearly specify the type of date you want, such as a contract start date, renewal date, decision deadline, or target go-live date.
Explain how to handle relative dates mentioned in the conversation, such as “next quarter” or “end of the month.” Specify whether it should convert them to an exact calendar date and which reference date to use, for example the call date.
If multiple dates are mentioned, explain which one to capture. For example, choose the earliest confirmed date, the latest agreed date, or the final committed date.
Clarify how to handle tentative or conditional dates, such as “hopefully in March” or “if legal approves by Friday.” Indicate whether to capture them or ignore anything not confirmed.
For Boolean (Yes/No) fields
Include guidance for when each option should be chosen
Provide the appropriate default answer
For questions seeking confirmation of a completed action or a positive state for example, "Was the payment successful?", the default is false.
For questions seeking confirmation of an ongoing status or the absence of a negative state, for example, "Is the user still eligible?", the default is true.
The model will likely not adhere to any guidance given on output formatting
For range fields
Clearly describe what the minimum and maximum values represent. For example, specify whether you want a budget range, expected number of seats, or implementation timeline.
Specify the type of value you are looking for, such as currency, number of seats or headcount, percentages, or time units.
Explain how to handle ranges expressed in different ways. For example, indicate whether "between 100 and 200," "100–200," and "at least 100" should all be treated as valid ranges.
If multiple ranges are mentioned, explain which one to capture. For example, choose the final agreed range rather than an earlier estimate.
Explain what to do if only one value is mentioned. For example, specify whether the value should be used for both the minimum and maximum, or whether no range should be returned.
If you're using a range to score or rank something, define the criteria for each score. For example, explain what qualifies as a 1, 2, 3, 4, or 5 so the AI can apply the full scoring range consistently.
What you need to get started
To start using AI Data Extractor, you need:
To connect Gong to Salesforce, HubSpot or Microsoft Dynamics and import relevant fields: AI Data Extractor uses CRM fields that are already imported into Gong. Make sure the fields you want to populate exist in your CRM, are imported into Gong, and support write access from the Gong integration user. You cannot use the CRM API with the AI Data Extractor.
A business admin to configure the extractor. This includes:
Deciding which questions to ask
Choosing the target object and output types
Mapping each AI output to the correct CRM field
Once your first extractors are published, AI Data Extractor starts analyzing conversations, updating CRM fields, and giving your organization richer, more reliable data without extra work for your reps.