WhatsApp Chat History and Export

Every WhatsApp conversation is logged as a chat under Logs → WhatsApp, separate from phone and web calls. The tab has its own filters, a transcript view styled like WhatsApp, and a CSV export.

Open WhatsApp logs

Go to Logs and choose the WhatsApp tab, next to Phone and Workflows. The table lists chats from the selected date range, newest first. The default range is the last month.

Column	What it shows
Assistant	The WhatsApp agent that handled the chat, with its version. Filter with the icon in the header.
Contact	The customer’s WhatsApp number in international format. Below it is their name: the `name` custom variable if set, otherwise their WhatsApp profile name.
Started on	When the customer’s first message arrived. Sortable.
Status	Ongoing, Completed or Failed. Filter with the icon in the header.
Duration	Time from the first message to when the chat closed.

Logs → WhatsApp, opening a completed chat and reading its transcript.

A chat shows Ongoing until it closes, which can be hours after the last message. See when a chat opens and closes.

Filter chats

Select Filters to narrow the list:

Filter	Notes
Date range	Applies to Started on.
Assistants	Only WhatsApp agents are listed.
Chat status	Ongoing, Completed, Failed.
From number	The customer’s number.
To number	Your business number.
Chat ID	The full chat UUID, for example from a webhook’s `chat_id`.

Active filters show as chips above the table, and they are kept in the page URL, so you can share a filtered view by copying the link.

Read a chat

Select a row to open the chat panel. The header shows the agent and version, the chat ID with a copy button, the status, the customer’s number and the channel. The tabs below it are:

Tab	Contents
Chat details	Chat duration, custom variables (including values from pre-call tools), and platform analysis: summary, key points, classification and call status reason.
Advanced analysis	Your Advanced Analysis fields, when configured.
Transcript	The conversation as WhatsApp bubbles with timestamps. Tool calls appear inline as small pills with their duration, for example the pre-call greeting or a template send.
Evals	Evaluation results, when the chat has been evaluated.

The ⋮ menu has Test this agent, Download analysis, Copy transcript and Download transcript.

Export to CSV

Filter first

The export uses the filters currently applied on the WhatsApp tab.

Choose the export

Select Export at the bottom of the table, then Export without analysis or Export with analysis.

Check your email

You’ll see Chat history report (with the applied filters) is being sent to your email. The email links to a zip containing chat_history_report.csv. The link is valid for 7 days.

Export offers the chat history with or without analysis columns.

The CSV has one row per chat, with these columns in order:

Column	Description
`ID`	Chat ID.
`To Number`	Your business number’s Meta phone number ID.
`From Number`	The customer’s number in E.164 format.
`Name`	The `name` custom variable, or else the customer’s WhatsApp profile name.
one column per custom variable	The values of each custom variable.
`Chat Status`	`ongoing`, `completed` or `failed`.
`Transcript`	The conversation, one `speaker: text` line per message.
`Created At`	When the chat started.
`Agent Name`, `Version`	The assistant and version that handled it.
`Chat Duration`	Length of the chat.
`Channel`	`whatsapp`.
`External Chat ID`	The ID of the chat on the messaging platform.
`Classification`	The platform analysis classification.

Export with analysis adds summary, key_points and action_items, followed by one column for every other platform and advanced analysis field. An export can hold up to 200,000 chats. If nothing matches the filters, you get an email saying the export was empty.

Fetch chats through the API

The same data is available to your backend with your workspace API key in the X-API-KEY header.

List chats
Get one chat

curl -G "https://prod-api.ringg.ai/ca/api/v0/chat/history/v2" \
  -H "X-API-KEY: your-api-key" \
  --data-urlencode "start_date=2026-09-01" \
  --data-urlencode "end_date=2026-09-24" \
  --data-urlencode "channel=whatsapp" \
  --data-urlencode "status=completed" \
  --data-urlencode "limit=20" \
  --data-urlencode "offset=0"

Parameter	Description
`start_date`, `end_date`	Date range on the chat start time.
`limit`, `offset`	Paging. `limit` defaults to 10.
`agent_id`, `version_id`, `status`	Repeatable filters.
`channel`	`whatsapp`.
`from_number`, `to_number`, `chat_id`	Exact-match filters.
`include_analysis`	`true` adds platform and advanced analysis to each row.
`sort_order`	`desc` (default) or `asc`.
`download`	`true` queues the CSV export email instead of returning rows.

The response is { "calls": [...], "limit", "offset", "count", "total" }. Each row has id, from_number, to_number, name, status, created_at, call_duration, agent, channel, external_chat_id and classification. Transcripts are not included. Fetch a single chat for those.

curl -G "https://prod-api.ringg.ai/ca/api/v0/chat/details" \
  -H "X-API-KEY: your-api-key" \
  --data-urlencode "chat_id=3c6e2a1d-5b7f-4e8a-9c21-0f4d7b8e6a53" \
  --data-urlencode "send_analysis=true"

Returns { "status": "success", "data": { ... } }. The data object holds the chat’s id, status, from_number, to_number, agent_id, agent_name, version fields, initiation_time, chat_duration, custom_args_values, classification and transcript. With send_analysis=true it also holds platform_analysis and client_analysis.

For near-real-time processing, subscribe to chat webhooks instead of polling this API. all_processing_completed carries the transcript and both analyses in one payload.

Get Started

WhatsApp

API Reference

Webhooks Setup

Telephony Integration

MCP Server

Parrot STT

External Integrations

Privacy and Compliance

WhatsApp Chat History

Open WhatsApp logs

Filter chats

Read a chat

Export to CSV

Fetch chats through the API

​Open WhatsApp logs

​Filter chats

​Read a chat

​Export to CSV

​Fetch chats through the API

Open WhatsApp logs

Filter chats

Read a chat

Export to CSV

Fetch chats through the API