Ask about anything in Docent Documentation

Type a question in your own words below. Answers are written from our published articles and cite the ones they came from.

Recently updated

    ↑↓ select · ↵ open · esc close
    Dashboard
    Edit Article Logout

    Exporting Chat Data


    The Export page downloads a messenger's conversations, and its tickets when ticketing is on, as CSV files you can open in Excel, Google Sheets or a BI tool. Open a messenger and click Export in its sidebar, just below Reports.

    Anyone who can see Reports can use Export: the Owner, Admin, User and Reports roles.

    Download a file

    Each file has its own card with its own period. Choose Last 7 days, Last 30 days (the default), Last 90 days or Last 12 months, then click Download CSV.

    • The conversations period runs back from now, for example the last 30 × 24 hours.
    • The tickets period counts whole days in your support hours time zone, or your own when support hours are off.
    • Times in both files are in UTC. Durations are in seconds (conversations) or hours (tickets).

    Conversations

    One row for every conversation started in the period with a question or a reply. Conversations opened without either, such as a proactive message that was opened and closed, are left out. If you've chosen to leave out your team's testing in Reports, it's left out here too.

    ColumnsWhat they contain
    conversation_id, started_utc, statusWhich conversation, when it started and whether it's open, snoozed or closed
    handled_by, assigneeWho had it (the AI, waiting, or a person) and the teammate it was assigned to
    outcomeHow it ended: ai_confirmed, ai_assumed, human_handled, team_started, missed, waiting or other
    ai_answers, ai_no_answersHow many times the AI answered, and how many questions it couldn't answer
    handed_off_utc, handoff_reasonWhen and why it was handed to your team
    first_human_reply_utc, first_response_secondsThe first reply from a person and how long after the handoff it came
    closed_utc, time_to_close_secondsWhen it closed and how long it was open
    rating, rating_commentThe customer's rating (1 to 5) and comment
    start_page, proactive_messageThe page it started on and the proactive message it replied to
    country, channel, device, visitor_typeWhere the customer was, how they reached you (website, email, WhatsApp), their device, and whether they were verified, identified or anonymous
    visitor_messages, team_repliesHow many messages the customer sent and how many replies your team sent
    follow_up_rule, follow_up_detailsThe question rule that collected follow-up details, and the details as "Label: value; Label: value"

    Tickets

    The Tickets card appears when ticketing is on. It downloads one row for every ticket opened in the period, open ones first and the longest open first, with the columns number, subject, customer, account, status, priority, type, assignee, opened, closed, open_hours, overdue and rating.

    Exporting one ticket view

    To export just the tickets in a view, with its filters, use Export on the ticket table in the inbox. See Working with Tickets.

    Opening the files

    The files are UTF-8 CSV, which Excel opens directly. A value that starts with a character a spreadsheet would read as a formula (such as =, +, - or @) begins with an apostrophe, so text a customer typed can't run as a formula. How long conversations are kept depends on your data retention setting, so older periods only include what's still stored.


    How helpful was this article?

    👍 or 👎

    Related Articles

    Markdown Version