Guided question and answer for Contact Center flow | Community
Skip to main content
Newcomer
January 29, 2026
Question

Guided question and answer for Contact Center flow

  • January 29, 2026
  • 1 reply
  • 31 views

For one of our groups in Contact Center, we need to collect information from our callers and then have that information sent to an Inbox so it can be listened to by the team. What we’re planning is that when the callers hit the flow, they will be presented with a question → answer the question; and then the system will repeat this for as many questions as there are. When all questions have been recorded, all of these answers should then be forwarded to an Inbox where the interaction can be reviewed.

This is the configuration that we’re trying to get to work but we’re having some issues getting the CollectInput widget to capture responses from the callers. And we do have “Voice input” enabled.

This is the basic flow but not sure if this will get us what we want:

Question 1 (Send Media widget) → Collect verbal response to question 1 (Collect Input widget)

Question 2 (Send Media widget) → Collect verbal response to question 2 (Collect Input widget)

Thank you message → Route collected responses to Inbox for review

    1 reply

    ArvinG
    Employee
    Employee
    January 29, 2026

    Hi there! This is an interesting use case. Just to clarify:

    • After the caller answers the questions, do you want the caller to still speak to a live agent, or is this an after hours voicemail style intake?

    • When you say “forwarded to an Inbox,” do you mean a ZCC queue / agent inbox, where the agent will handle the live call, or do you mean a message for later review even if no agent answers?

    • Do you need the answers stored as text fields (best for intake), or do you need audio playback of each answer?

    asmith2Author
    Newcomer
    January 30, 2026
    • After the caller answers the questions, do you want the caller to still speak to a live agent, or is this an after hours voicemail style intake?

             After the caller answers the questions, there is no need to speak to a live agent.

    • When you say “forwarded to an Inbox,” do you mean a ZCC queue / agent inbox, where the agent will handle the live call, or do you mean a message for later review even if no agent answers?

             The idea is to collect all of the answers that the caller leaves and have them forwarded to an Inbox that exists in              Contact Center for later review by agents.

    • Do you need the answers stored as text fields (best for intake), or do you need audio playback of each answer?

              Audio playback will work for this.

    ArvinG
    Employee
    Employee
    February 5, 2026

    Appreciate you sharing the extra context here. I did some testing, and it looks like this isn’t natively supported in Zoom Contact Center, but here are a few possible workarounds:

    1. Speech-to-text intake flow using the Collect Input widget: Start by sending a media message with a question — for example, “Please state your full name.” The customer speaks their response, which gets converted into text (not recorded audio). The flow then moves to the next question using another Send Media widget > Send Media confirmation that thanks the caller and an End widget to close the call. From the agent or reviewer’s side, the responses appear as text variables in the engagement record.
    2. Voicemail-style recording flow: Start with a Start widget, followed by Send Media instructions and another Send Media widget that plays all your questions sequentially. Then, route the call to a queue configured to accept voicemails, so the caller can record their message. Add an End widget to finish the call. From the agent or reviewer’s perspective, they’ll receive one continuous audio file containing all responses — accessible in the Inbox or Engagements tab, where it can be played, downloaded, or marked as resolved.
    3. Custom integration option: The Script widget could call an external API that captures and stores each audio response individually.

    Hope this helps. Thanks!