> For the complete documentation index, see [llms.txt](https://limechatai.gitbook.io/limechat-product-guide/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://limechatai.gitbook.io/limechat-product-guide/whatsapp-marketing/flows/action-nodes/abandoned-checkout-phone-call.md).

# Abandoned Checkout Phone Call

Action Node · AI Voice Recovery Call

### What It Does

This node automatically triggers an AI-powered phone call to customers who abandoned their checkout process.

The AI voice agent contacts the customer, attempts to recover the cart, communicates offers or discounts, and helps push the customer toward completing the purchase.

This is commonly used in abandoned cart recovery flows where WhatsApp messages alone may not be sufficient to drive conversions.

***

### Example

A customer adds products to their cart but leaves without completing payment.

After a predefined delay, the workflow triggers this node.\
The AI voice agent calls the customer and says:

> “Hi Rahul, you left 2 items in your cart. We’re offering an additional 10% discount if you complete your purchase today.”

The customer can respond naturally during the conversation, and the AI agent continues the interaction based on the configured prompt.

***

## When to Use This Node

Use this node when:

1. You want to recover abandoned carts using voice calls
2. WhatsApp reminders are not driving enough conversions
3. You want to communicate time-sensitive offers verbally
4. You need personalized customer engagement at scale
5. You want AI agents to handle recovery conversations automatically

***

## Requirements

 Before configuring this node:

* A voice agent must already be configured
* Customer phone number should exist in the event data
* Required cart or checkout event data should be available
* CRM inbox setup is optional but recommended for tracking conversations

***

## This Node Is Triggered By

This node is typically triggered after:

1. Abandoned checkout events
2. Cart abandonment workflows
3. Delayed recovery campaigns
4. Manual recovery automation flows

***

## Setting It Up

### Inputs

| Input                 | Type        | Required | Description                                         |
| --------------------- | ----------- | -------- | --------------------------------------------------- |
| CRM Inbox             | Dropdown    | No       | Inbox through which the call will be managed        |
| Voice Agent           | Dropdown    | Yes      | AI voice agent used for the call                    |
| Customer Name         | Event Field | No       | Customer name used during the conversation          |
| Item Names            | Event Field | Yes      | Product/items added to cart                         |
| Offer Name            | Text        | No       | Offer communicated during the call                  |
| Discount Applied      | Text        | No       | Discount shared with the customer                   |
| Custom Prompt         | Text        | Yes      | Instructions defining how the AI agent should speak |
| Additional Event Data | Event Field | No       | Extra event data passed into the call               |

***

## Input Field Details

### CRM Inbox

Optional CRM inbox where the voice interaction and ticket activity will be tracked.

### Voice Agent

Select the AI voice agent responsible for handling the recovery conversation.

### Customer Name

Dynamic customer name variable used for personalization during the call.

Example:

* Rahul
* Priya

### Item Names

Dynamic cart item data shared with the AI agent.

Example:

* Running Shoes
* Blue T-Shirt

### Offer Name

Optional offer label communicated during the call.

Example:

* Summer Sale
* Checkout Recovery Offer

### Discount Applied

Optional discount information shared verbally with the customer.

Example:

* 10% OFF
* ₹500 discount

### Custom Prompt

Defines the AI agent’s behavior, tone, and recovery strategy.

Example:

You are a polite sales assistant helping customers complete their purchase. Mention the abandoned products, communicate available offers clearly, and encourage the customer to complete checkout today.

### Additional Event Data

Additional contextual event data passed into the voice workflow.

Example:

* Cart value
* Checkout link
* Preferred language
* Previous purchase history

***

## Outputs

| Output             | Type    | Description                                                |
| ------------------ | ------- | ---------------------------------------------------------- |
| Call Status        | Signal  | Indicates whether the call was successfully initiated      |
| Call Answered      | Boolean | Indicates whether the customer answered the call           |
| Call Completed     | Boolean | Indicates whether the conversation finished successfully   |
| Customer Intent    | Text    | Captured customer intent or outcome from the call          |
| Call Recording URL | Text    | Recording link for the completed call                      |
| Failure Reason     | Text    | Error or failure reason if the call could not be completed |

***

## What Happens After It Runs

| Signal                      | Result  | Recommended Next Step                   |
| --------------------------- | ------- | --------------------------------------- |
| Call initiated successfully | Success | Continue the recovery workflow          |
| Customer answered           | Success | Capture intent or share checkout link   |
| Customer did not answer     | Retry   | Retry later or send WhatsApp reminder   |
| Call failed                 | Failure | Trigger fallback flow or alert the team |

***

## Best Practices

 Keep the AI prompt conversational and concise.

Mention abandoned products naturally instead of sounding robotic.

Clearly communicate urgency if an offer is time-sensitive.

Pass customer-specific context wherever possible for better recovery rates.

***

## Key Reminders

1. A valid customer phone number is required
2. The Voice Agent field is mandatory
3. Custom prompts directly influence AI conversation quality
4. Ensure event fields contain valid cart data before execution
5. Use fallback workflows for unanswered or failed calls