# Understand conversation data 1/2

The <kbd>closedConversations</kbd> and the <kbd>conversation</kbd> queries allow you to retrieve detailed information about each conversation that has been closed. These are your go-to queries for raw data export, integration into external systems, or conversation-level audits.

Here is the [link](https://graphql.iadvize.dev/queries/closedConversations) to the publicly available schema.

## What data is available

For each conversation, you can retrieve:

* Unique ID
* Start and end timestamps
* Assigned agent, group, campaign
* Messages (including sender, content, timestamps)
* Tags
* Ratings (CSAT, NPS)
* Contact info and visitor details
* Transactions\\

You can also expand nested fields to retrieve message-level data, user details, and routing logic context.

## Available filters

You can narrow down the results using:

* `interval` (required): defines the period of closed conversations to fetch
* `projectIds`: filter by project (SID)
* `answered`: Filter by conversations that received a response.
* `answeredByHuman`: Filter by conversations answered by a human agent.
* `automationLevels`: Filter by automation level (FULLY\_AUTOMATED, NOT\_AUTOMATED, PARTIALLY\_AUTOMATED)
* `channels`: Filter by conversation channels (e.g., APPLE\_BUSINESS\_CHAT, CALL, CHAT, FACEBOOK, FACEBOOK\_BUSINESS\_ON\_MESSENGER, GOOGLE\_BUSINESS\_MESSAGES, INSTAGRAM, MOBILE\_APP, SMS, TWITTER, VIDEO, WHATSAPP).
* `closeReason`: Filter by reasons for conversation closure (BOT\_TRANSFER\_FAILED, END\_OF\_BOT\_SCENARIO, VISITOR\_DROPPED\_OFF\_SCENARIO)
* `conversationId`: Filter by a specific conversation ID.
* `conversationVisitorIds`: Filter by visitor IDs associated with conversations.
* `customerSatisfaction`: Filter by customer satisfaction scores (1 to 5).
* `engagementCampaignIds`: Filter by engagement campaign IDs.
* `genAiPredictedTopicIds`: Filter by generative AI predicted topic IDs.
* `groupIds`: Filter by group IDs of operators who participated in the conversation.
* `hasConversion`: Filter by presence of a conversion.
* `hasGenerativeAIAnswer`: Filter by presence of generative AI answers.
* `hasGenerativeAIUnansweredQuestion`: Filter by presence of unanswered questions by generative AI.
* `hasVideoCall`: Filter by presence of video calls in the conversation.
* `humanHandlingTimeUpperLimit`: Filter by maximum human handling time in seconds.
* `isAiAssisted`: Filter by conversations where agents were assisted by generative AI.
* `messagesSearches`: Filter by keywords in the content of messages.
* `netPromoterScore`: Filter by Net Promoter Score (0 to 10).
* `projectIds`: Filter by iAdvize project IDs (SID).
* `respondentTypes`: Filter by respondent types (e.g., Copilot, Bot, Expert, Agent).
* `roles`: Filter by roles (e.g., operator, manager, admin, expert, bot).
* `routingGroupIds`: Filter by routing group IDs.
* `tagIds`: Filter by tag IDs.
* `targetingRuleIds`: Filter by targeting rule IDs.
* `userIds`: Filter by user IDs.
* `visitorIds`: Filter by visitor profile IDs.
* `visitorMessagesSearches`: Filter by keywords in visitor messages.

You can also paginate your results using offset and limit.

More about pagination in this [article](https://docs.iadvize.dev/technologies/graphql-api/pagination).

## Fields returned by closedConversations

#### 🆔 Core identifiers

* `id`: Unique ID of the conversation
* `externalId`: Custom external identifier

#### 🕓 Timestamps

* `createdAt`: When the conversation started
* `updatedAt`: Last update timestamp
* `closedAt`: When the conversation was closed
* `duration`: Conversation duration in seconds from visitor initiation time to user closing time using iso 8601 format.

#### 👥 Participants & routing

* `users`: List of users who participated in the conversation
* `visitor`: The visitor (contact) involved in the conversation
* `project`: The iAdvize project the conversation belongs to
* `routingRule`: The routing rule that directed the conversation
* `routingGroup`: Group of operators the conversation was routed to.

#### 💬 Content & messages

* `messages`: All messages exchanged during the conversation (transcripts
* `systemMessage`:A system conversation message attachment (engagement rule triggered, transfer information, etc. full list available [here](https://graphql.iadvize.dev/types/SystemConversationMessageAttachment))

#### 🧠 AI & automation

* `automationLevel`: Level of automation used in the conversation

#### 🔖 Tags & metadata

* `tags`: Tags manually or automatically applied to the conversation
* `language`: Language of the conversation
* `channel`: Channel used (chat, call, etc.)

#### 🧪 Ratings & feedback

* `customerSatisfaction`: Customer satisfaction score (1–5)
* `netPromoterScore`: NPS value (0–10)
* `comment`: Visitor's conversation comment.

#### 📎 Context & tracking

* `targetingRule`: Targeting rule that triggered the chat
* `engagementCampaign`: Associated engagement campaign
* `customData`: A typed conversation custom data for a conversation - more info [here](https://graphql.iadvize.dev/types/ConversationConversationCustomDataEntry)
* `closingformValues`: List all closing form values answered to custom closing form plugins.

\
💶 **Transactions**\*

* `conversions`: A Conversion attributed to a conversation

{% hint style="info" %}
\*make sure you read the info available [here](/use-cases/data-and-analytics/retrieve-messages-exchanged-within-a-conversation-1/understand-transaction-data.md) regarding transaction
{% endhint %}

## Good to know

* This query is ideal for exporting chat logs, enriching your CRM, or performing detailed analysis on individual conversations.
* Use it to feed reporting pipelines or quality assurance tools.
* If you wish to catch all conversations as soon as they go through, please make sure you read our information regarding our two [conversation webhooks.](https://docs.iadvize.dev/technologies/webhooks)\\

\
\\

\
\\


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.iadvize.dev/use-cases/data-and-analytics/retrieve-messages-exchanged-within-a-conversation-1/find-contact-data-graphql.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.