Skip to main content

Post-Call Survey

A guide to using post-call surveys in CloudTalk to collect feedback from your callers

V
Written by Valeriia Volobrinskaia

Post-call surveys, sometimes referred to as CSAT or Customer Satisfaction Scores, are consumer feedback collection methods used by companies worldwide in various forms. This article will show you how to set up your CloudTalk phone system to automatically request and collect feedback scores from your callers directly after an inbound call.

User level: Admin

Currently the main way to use post-call surveys in CloudTalk uses the HTTP Request step to collect and send survey results. Some prior knowledge of API functionality is recommended.

Prerequisites

To use post-call surveys in CloudTalk you will need:

  • A CloudTalk Essential plan or higher

  • At least 1 number to which CSAT will be applied

  • All necessary sounds uploaded to your CloudTalk Dashboard by an admin. Recommended recordings before you begin:

    • A recording instructing the contact to stay and provide feedback after the call has ended

    • A recording instructing the contact on how to provide feedback (e.g., "On a scale of 1-5, rate our service using your keypad")

    • A recording that thanks the contact for providing their feedback

For more on uploading sounds, see: Configure Sounds and Greetings

How to Build a Post-Call Survey

Open the Call Flow Designer

1. From the Dashboard, click Numbers.

2. In the My numbers tab, find the number you want to configure.

3. Click + Configure (or the flow name, if one exists) in the Call Flow column.

Set up the call flow

4. Build your pre-call flow as needed (Greeting, Ring to, etc.).

5. Find the Ring to step that your survey should be attached to. The survey collects feedback after the call is answered, so it connects to the Answered branch.

Note: Don't forget to inform your customers (either via your introductory recording or during the agent's conversation) that you'd appreciate them remaining on the line after the call to answer the survey.

6. Click + Add answered branch on the Ring to step if you haven't already.

7. On the Answered branch, click the + button and add either a Collect Input or IVR Menu step (see below for the difference). Select a pre-uploaded recording that instructs the caller to provide their rating.

8. After the caller input step, add an HTTP Request step to send the survey results to your chosen tool (see "Sending data using your HTTP Request" below).

Caller Input Steps: Collect Your Results

When collecting survey results, you have two options: Collect Input or IVR Menu.

Using Collect Input

  1. Add a Collect Input step to the Answered branch of your Ring to step.

  2. In the Collect Input config, select a sound that instructs the caller to provide feedback.

  3. In the Pattern field, define the expected value using regex syntax:

    • For a 1-5 scale: \b[1-5]{1}\b

    • For a 1-9 scale: \b[1-9]{1}\b

    • For a 1-10 scale: \b[0-9]{1}\b|\b10\b (regex doesn't recognize 10 as a single number, so it needs a separate pattern)

  4. Save this step.

  5. Add an HTTP Request step after the Collect Input step. In the HTTP Request body, reference the caller's input using: {{input[step_id].value}} where step_id is the Step ID of the Collect Input step.

Using IVR Menu

  1. Add an IVR Menu step to the Answered branch of your Ring to step.

  2. In the IVR Menu config, select a sound that instructs the caller to rate the service (e.g., "Press 1 for poor, press 5 for excellent").

  3. Set the Seconds to wait on this step.

  4. Save this step.

  5. For each branch below the IVR Menu, add an HTTP Request step. Each branch corresponds to a key the caller pressed, and the HTTP Request for that branch should send the corresponding rating value.

IVR Menu vs Collect Input

Both options work, but they have different strengths:

  • Collect Input lets you implement wider scales (1-10, 1-20, etc.) and is easier to maintain since changes only affect a single branch. However, it requires knowledge of regex for the Pattern field.

  • IVR Menu works on a 0-9 scale and is more familiar to most users. No regex knowledge is needed. IVR Menu also lets you customize post-survey actions for specific ratings (e.g., playing a different sound for poor feedback, or tagging the contact so future agents are aware of prior dissatisfaction). The downside is that each key needs its own HTTP Request, making the flow visually complex.

Sending Data Using Your HTTP Request

For more on HTTP Request configuration, see: HTTP Request Step

Storing survey data as a contact activity in CloudTalk

If you want to store CSAT information as an activity in a customer's profile, configure your HTTP Request according to the Create Activity endpoint in the CloudTalk Developers Portal.

Include:

  • Contact ID reference in the Endpoint, so CloudTalk knows which contact the activity should be assigned to.

  • Name of the activity, simple and clear so agents can easily spot it in the customer profile.

  • Inserted value within the description of the activity (the CSAT score).

This gives your agents insight into previously submitted survey scores directly within a contact's profile.

Note: CloudTalk does not currently offer an option to build reports based on activities.

Sending data to external tools

To send data to an external tool, configure your HTTP Request according to that tool's API specification. The HTTP Request requirements are:

  • Endpoint: HTTPS

  • Method: PUT, POST, or PATCH

  • Authorization: Static authorization keys (Basic Auth or API key within URL)

  • Body: JSON body

If the API of your choice is not compatible with these requirements, you will likely need to use middleware such as Zapier or Make.

Sending data to Zapier or Make

  1. Prepare your Zap or Make scenario:

    • Trigger: typically a webhook action. When set up, a webhook URL is generated. Copy this URL.

    • Action: describes how the data should be populated (e.g., adding a row in Google Sheets or Microsoft Excel).

  2. Insert the generated webhook URL into the Endpoint field of your HTTP Request step in the Call Flow Designer.

When set up correctly, all data included in the body of your HTTP Request will be sent to the middleware and forwarded to the tool of your choice.

Bonus tip: Use the Failover for no key pressed option on your IVR Menu to handle cases where callers forget to select an option or fail to hang up.

Need help or have a question? Just reach out through our Support portal — we’re here for you.

Related Articles
Did this answer your question?