> This page is for version Generative (default).
> For other versions, use one of these documentation indexes:
> - Generative (default): https://developers.ada.cx/generative/llms.txt
> - Scripted: https://developers.ada.cx/scripted/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://developers.ada.cx/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://developers.ada.cx/_mcp/server.

# Greetings and Handoffs

When your AI Agent answers a phone call, it will start off the conversation with a **[Greeting](/generative/docs/automation/greetings/greeting-management)**.

If it needs to escalate the conversation to a human agent, it will do so using a **[Handoff](/generative/docs/handoffs/zendesk/zendesk-chat/zendesk-chat-configuration)**.

You can use our structured flow builder to create your Greeting and your Handoff-which uses modular blocks to define the logic that you want your AI Agent to follow and exactly what it should say.

## Build and adapt your Greeting and Handoff(s) \[#greeting-and-handoff]

Are you ready to create your Voice AI Agent Greeting and Handoff? While there are
some differences between how Messaging and Voice content work in Ada,
you'll notice that there are a lot of similarities between the two.

### Understand how content is organized \[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4569185542878433638916263866]

In AI Agents that have Voice enabled, each Greeting and Handoff is divided into two
tabs: **Messaging** and **Voice**. Under each tab, you can tweak the content for
that specific channel.

### Create a Greeting for Voice\[#greeting]

In addition to greeting end users with a friendly and helpful AI Agent persona, make two things clear in your greeting:

* An AI Agent is speaking, not a human
* The call is being recorded

Follow these best practices to maximize caller engagement and prevent abandonment or premature escalation:

* Keep your Greeting short and to-the-point
* Avoid capturing information from the end user upfront unless absolutely critical
* Encourage the end user to ask an open ended, descriptive question to kick off the conversation

### Block differences in Voice versus Messaging \[#block-differences]

This section details the differences between how some blocks work in Voice versus Messaging.

#### Capture block \[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4596625883974433648988007292]

When you're capturing information using the Capture block, it's important to think
about the most accurate way of getting that information from the end user,
and to communicate clearly what you want them to do.

There are four ways you can validate end user information. For all of
them, make sure you mention the input you're expecting so the end user
knows what to do, along with any other guidelines they might need.

| Data type    | Input methods | Example prompt                                                                                                                  |
| ------------ | ------------- | ------------------------------------------------------------------------------------------------------------------------------- |
| Text         | Speak         | Using a **complete date, such as January 1, 2001**, please **say** your date of birth.                                          |
| Number       | Dialpad       | **Using your dialpad**, please enter the **9-digit number** after S M, **followed by the pound key**.                           |
| Phone number | Dialpad       | **Please use your dialpad** to enter the **10-digit** phone number associated with your account, **followed by the pound key**. |
| Yes or No    | Speak         | The phone number you entered was 636-555-3226. Is that correct? **Please say yes or no**.                                       |

When end users are entering numbers into the dialpad, they can use the `#`
key to indicate that they're done. Otherwise, your AI Agent assumes they're
done five seconds after the last digit the end user entered.

#### Smart Capture block \[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4596625883974433648988007292]

In some cases, Voice AI Agents can struggle to accurately pick out responses
from what end users say. You might find that your AI Agent saves a whole phrase
and says something like "hello, my name is Rob" instead of "hello, Rob,"
or that it has trouble accurately understanding the difference between a
"b" or "p" when you're trying to capture a booking reference number.

To increase the chances that your AI Agent can accurately understand how to
pick relevant information out of what the end user says, you can use the
Smart Capture block, which uses generative AI to understand what they said and save a
more accurate output.

When you use the Smart Capture block, your AI Agent becomes better at picking
out information from the end user's response. When a end user says something
like "'a,' as in 'apple,'" it's able to pick out the correct letter, and
if they ask the AI Agent to repeat something, it's smart enough to repeat it
rather than try to parse their question as a response.

The Smart Capture block only proceeds when the end user has verified that
what it captured is correct; if it isn't, it tries again. Alternatively,
if the end user wants to exit the capture state, they can ask an unrelated
question to do something else.

> **Note**
>
> The Smart Capture block is only available for Voice Greetings and Handoffs.

Here's how to use the Smart Capture block:

1. From the block drawer in your Greeting or Handoff find the **Smart Capture** block and drag it
   into your response.

2. Under **Question**, enter the question your AI Agent will ask to prompt
   the end user to provide information.

3. Under **Capture Type**, select the format you want to capture the
   data in, so your AI Agent can better validate the end user's response.

   Depending on your selection, additional options to configure may
   appear:

   * If you selected **Choose from a list** as the capture type, the
     **List options** section appears. For each option the end user can
     choose from, click **Add option** and enter the value that they
     can choose.

     When end users encounter this block, your AI Agent will read the entire
     list of available options to them before they make their
     selection. Unlike the List Option block, you can't choose to only
     read out the options if the capture is unsuccessful the first
     time.

   * For some capture types, you can specify how you want the end user's
     response to be formatted. When you select those capture types, the
     **Output the captured information in a specific format** toggle
     appears. You can toggle the setting on, then add instructions in
     the box that appears.

     You can specify a format using existing conventions (e.g.,
     "MMMM/DD/YYYY"), or you can use plain language (e.g., "replace all
     spaces with underscores").

   * If you capture a **Number**, **Numeric String**, or **Phone Number** from an end user, your end users will be able to provide the value by speaking it *or* by entering it into their phone's dial pad. Before capturing a value, your AI Agent will automatically tell end users that they can either say the number or enter it into their dial pad.

4. Under **Save Capture as Variable**, select a variable to save the
   end user's response in.

5. Choose to toggle the **Skip capture if the variable already has a
   value** setting on or off. That way, if the variable you chose
   already has a value associated with it, your AI Agent won't ask the
   end user to provide that information again.

6. Click **Save**.

#### Transfer Call block \[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4516056577880033652597058235]

With the Transfer Call block, you can transfer the phone call outside of the Ada. Most of the time, this means handing off the
call to a human agent.

Note that the call transfer takes place whenever the Transfer Call block appears in a flow. So if you have any blocks below the Transfer Call block, your AI Agent won't execute those blocks.

When configuring the Transfer Call block, you will need to select where you are transferring the call to. Your options include **Phone number**, **SIP address**, **Amazon Connect**, **Aircall**, or **Twilio**. Which of these you select will depend on which contact center platform or technology you are integrating Ada into. For more information, see [**Integrate your Voice AI Agent with your contact center technology**](https://docs.ada.cx/docs/generative/deploy-ada-across-channels/voice/integrate-your-voice-ai-agent-with-your-contact-center-technology)

1. If you selected **Phone number** from the **Transfer to** drop down, you will see a **Phone number** field. Enter a phone number or insert the
   variable containing the phone number you want to transfer the call to.
2. If you selected **SIP address** from the **Transfer to** drop down, you will see a **SIP address** field. Enter a SIP address or insert the
   variable containing the SIP address you want to transfer the call to.
* The **Include a User-to-User header when transferring to the SIP address** checkbox will also appear. You can select it to provide additional information with the SIP transfer, so the human agent receiving the call has additional context, or to use when you're routing the call to the appropriate department or group.

* In the **Key** fields that appear, enter names for the data fields that you want to include when transferring the phone call.

* In the **Value** fields, enter the value that you want to send for each Key. You can use variables to dynamically populate these for an end user.

* Click **Add another row** to add more key,value pairs to the transfer, or hover over a pair and click **Delete** ![](/_fern-files/_img/0a157919e7d474dfc5dd172698eb0e3c0c25ee947f5d6bf8e8fe6a1dfa68048e.webp) to remove it.
> **Note**
>
> The user-to-user header that contains these key,value pairs shouldn't exceed 400 characters.
5. If you selected **Amazon Connect** from the **Transfer to** drop down, you will see an **Amazon Connect destination queue** field. Enter the queue ID of the queue in Amazon Connect that you want to transfer the phone call to. For more information on integrating with Amazon Connect, see **Integrate Voice with Amazon Connect** in [**Integrate your Voice AI Agent with your contact center technology**](https://docs.ada.cx/docs/generative/deploy-ada-across-channels/voice/integrate-your-voice-ai-agent-with-your-contact-center-technology) at Ada's Help Docs.
6. If you selected **Aircall** from the **Transfer to** drop down, you will see a **Destination type** field. Select if you want to transfer to a **Number**, **Team**, or **User** in Aircall. Or, if you just want the call to continue in the Aircall Smartflow that was used to transfer the call to Ada, select **None**. For more information on integrating with Aircall, see **Integrate with Aircall** in [**Integrate your Voice AI Agent with your contact center technology**](https://docs.ada.cx/docs/generative/deploy-ada-across-channels/voice/integrate-your-voice-ai-agent-with-your-contact-center-technology) at Ada's Help Docs.
7. If you selected **Twilio** from the **Transfer to** drop down, Ada will drop from the call and continue in Twilio. For more information on integrating with Twilio, see **Integrate Voice with Twilio** in [**Integrate your Voice AI Agent with your contact center technology**](https://docs.ada.cx/docs/generative/deploy-ada-across-channels/voice/integrate-your-voice-ai-agent-with-your-contact-center-technology) at Ada's Help Docs.
* The **Pass additional data to Twilio when transferring the call** checkbox will also appear. You can select it to provide additional information back to Twilio when disconnect from the call, so the human agent receiving the call has additional context, or to use when you're routing the call to the appropriate department or group.
* In the **Key** fields that appear, enter names for the data fields that you want to include when transferring the phone call.
* In the **Value** fields, enter the value that you want to send for each Key. You can use variables to dynamically populate these for an end user.

#### End Call block \[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4636489368681633652597220595]

The End Call block is a simple one - if you put it into a flow,
it disconnects the call. Generally, it's a good idea to let the end user
be the one to end the call, but this block can be helpful at the end of
your content if you confirm with the end user that they don't need any
further assistance.

Note that your AI Agent ends the call when the flow gets to the End
Call block. If you have any blocks below it in a flow, they will not be executed.

### Minimize pauses while your AI Agent performs backend actions in a Greeting or Handoff \[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4512127993027233836099277747]

Some blocks, such as the HTTP Request block, perform backend actions
that take some time to complete. In some cases, this can cause pauses
while Ada is performing that action, during which a end user might wonder
if your AI Agent had heard the last thing they said or become frustrated.

You can easily mask this latency: immediately before blocks that perform backend actions, place one or more Speak blocks. That way, the backend actions can start processing while your AI Agent is speaking.

### Test your AI Agent's Voice content\[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4635816428822433714706372475]

While building out Voice content, it's a good idea to test how your
flows will sound to end users.

> **Note**
>
> Location metavariables depend on the end user initiating the
> phone call, and don't work if Twilio calls them instead. As a result, if
> your structured content depends on metavariables being populated, you may have to
> hardcode them using a [Set Variable block](/generative/docs/handoffs/handoff-management#set-variable-block)
> so you can test it properly.

1. In the top right corner of the flow that you want to test:

* **Greetings**: Click **Test answer**
* **Handoffs**: Click **Test handoff**
  A test chat window appears.

2. At the top of the test chat window, click the **Ada Web Chat**
   dropdown and click **Voice**.

3. Enter your phone number and click **Call**. Your AI Agent calls you and
   serves you your structured content. From there, you can hang up, or continue
   testing your AI Agent.

   ![](/_fern-files/_img/f9649bb9e6b22c173ecda4f21693d3d571ad0f98ce2964afafc5195e6902749a.webp)

#### Test your end user experience\[#UUID-a5baa6c5-e3da-e273-08e3-2da0b460c955_section-idm4581792249728033714762599342]

You can test your AI Agent end-to-end by clicking on the **Test AI Agent** button at the bottom of the side navigation.

---

Have any questions? Contact your Ada team, or email us at [](<mailto:help@ada.support%20%E2%80%9CHelp Docs inquiry%E2%80%9D>).