--- title: "Database design" slug: "database-design" updated: 2026-01-12T16:36:38Z published: 2026-01-12T16:36:38Z --- > ## 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. # Database design > **Who can use this**: Tech admin **Available on**: Data Cloud > [!NOTE] > Note: > > The Gong Data Cloud schema can change in any release, for example by adding new tables and columns. Design your integrations and queries to handle new fields to ensure your reports continue to work as the schema evolves. This diagram shows the relationship between conversations and other entities such as users and stats: ![](https://cdn.us.document360.io/2837c190-3ec8-4120-9116-6d427e774667/Images/Documentation/image(206).png) The following diagram displays the relationships between the tables representing conversation data including emails and calls. ![](https://cdn.us.document360.io/2837c190-3ec8-4120-9116-6d427e774667/Images/Documentation/image(207).png) ## Conversations table The `CONVERSATIONS` table is a supertype used to store details common to all conversations in Gong. A record is generated for each call, email, and meeting. Shared attributes such as workspace, conversation type (call, email, meeting), and start time are stored here. The `CONVERSATIONS` table includes two identifiers: - `CONVERSATION_KEY`: The primary key. A unique identifier for the record in the `CONVERSATIONS` table. - `CONVERATION_ID` : A foreign key to the table for the conversation type (`CALLS`, `EMAILS`, or `MEETINGS`) ## Identifiers for calls, emails, and meetings Each conversation type has its own table that stores attributes relevant to that type. For example, the `CALLS` table stores call-specific details such as recording URL or duration. Each conversation is saved once in their dedicated table and contains the following identifiers: - `CONVERSATION_KEY`: A foreign key to the `CONVERSATIONS.CONVERSATION_KEY` - `CONVERATION_ID`: The unique identifier for the conversation in the table for the call The `MEETINGS` table includes additional identifiers: - `CALL_CONVERSATION_ID`: A foreign key to the `CALLS` table so that the meeting can be associated with a call. As not all meetings have an associated call this can be NULL. - `CALL_ID`: A foreign key to the `CALL_RECORDINGS` table. This is the Id of the call recording ## Call, email, and meeting data The `CALLS` table includes call attributes including the call spotlight. Once the call has been recorded it is added to the `CALL_RECORDINGS` table, with details such as the language the call took place in and the duration of the call. Any activities done on the call, such as feedback or comments are stored in the `USER_CONVERSATION_GONG_ACTIVITIES` table. The `USER_CONVERSATION_GONG_ACTIVITIES` table holds the ID of the user who did the activity, as well as the date and time and the activity type. The `CALL_TRANSCRIPT` table contains a JSON of the transcript. The call is split up according to who spoke when during the conversation. The JSON includes a speaker ID which represents the person speaking at that time. Comments made on calls are available in the `COMMENTS` table. Email and meeting details are stored in the `EMAILS` and `MEETINGS` tables respectively. Call, email, and meeting participants are found in the `CONVERSATION_PARTICIPANTS` table. ## Gong users All users are stored in the `USERS` table which includes user details such as their first name, last name and email address. This table maintains history for the user so you can see details such as who was a manager when a specific call took place. In addition, to support the situation where a Gong user has multiple CRM IDs, the `USER_CRM_IDS` table contains the CRM IDs for each user. ## Trackers, scorecard, and insights data Trackers are tools that identify when words, phrases or concepts are mentioned in calls and emails, allowing you to know what your reps and customers are talking about. Smart trackers and keyword trackers are stored in the `TRACKERS` table and each tracker identified in a conversation is added to the `CONVERSATIONS_TRACKERS` table. You can see data on who has filled out scorecards and their assessment of the call. The `SCORECARD_QUESTIONS` table contains a row for each question in the scorecard together with the scorecard details, such as the scorecard name and when it was created. The `SCORECARD_ANSWERS` table contains a row for each question and answer given and includes the Id of the call the scorecard was given for. You can associate your CRM objects, such as opportunities, deals, or leads, with your Gong conversations using the `CONVERSATION_CONTEXTS` table. A conversation can be associated with multiple CRM objects. For example, a conversation can be associated with a deal or opportunity and an account. You can get statistics on your reps activities and how they manage calls in the `USER_CONVERSATION_GONG_ACTIVITIES` and `INTERACTION_STATS`. These tables give you details on what kind of actions a user did on a call, such as asking for feedback and statistics on metrics such as how long they speak on a call or how many questions they ask. ## Getting Forecast data The following diagram displays the relationships between the tables containing forecast data: ![](https://cdn.us.document360.io/2837c190-3ec8-4120-9116-6d427e774667/Images/Documentation/image(208).png) Forecast data includes forecast submissions and targets. In the `FORECAST_SUBMISSIONS` and `FORECAST_TARGETS` tables, you can see each user's forecast and targets for a specific period. These can be for an individual sales rep, or for a team. When the value is the team forecast or target, the `IS_TEAM` field is set to true. The period the forecast or target is for is set according to the `PERIOD_TYPE`, `PERIOD_YEAR` and `PERIOD_NUMBER`. The period type designates whether the forecast is for a month or a quarter, the period year sets the year and the period number sets which quarter or month in the year the forecast and target are for. ### Deal likelihood scores Gong assesses each deal and gives it a score indicating how likely the deal will be closed-won in its expected quarter compared to other deals in the pipeline. The `DEAL_LIKELIHOOD_SCORES` table contains each deal’s score, score category (Low, Fair, or High) and the last time the score was updated. The details of a deal can be retrieved from the CRM using the CRM_DEAL_ID. A team member who manages and configures the Gong platform and handles CRM integrations, user provisioning, permissions, and system settings. [Find my Tech admin](/v1/docs/find-your-technical-or-business-admin). Gong’s robust data sharing service that integrates Gong-enriched data with external data platforms- for deeper data analytics and advanced insights. Access to features depends on your [company’s plan and your assigned seat](/v1/docs/plans-and-seats). ## Related - [Forecast and Gong user tables](/forecast-and-gong-user-tables.md) - [Guide to analyzing Gong data with SQL queries](/guide-to-analyzing-gong-data-with-sql-queries.md)