ElevenLabs Batch Calling is an outbound telephony feature built on top of ElevenLabs Conversational AI. Where standard conversational AI deployment handles inbound calls — customers call a number and the AI agent answers — batch calling reverses this: the AI agent initiates calls to a list of recipients and conducts the conversation.
The feature was announced by ElevenLabs in its blog as a new capability for ElevenAgents and is accessible through the ElevenAgents dashboard under the Batch Calls section, as well as through the API. It requires a phone number configured in ElevenLabs (via Twilio or SIP trunk integration), a conversational AI agent configured with the appropriate system prompt and voice, and a contact list with recipient phone numbers and any personalisation data.
How Batch Calling Works
Agent Configuration
The first step is configuring the conversational AI agent that will conduct the batch calls. This is identical to configuring any ElevenAgents voice agent: set the system prompt (the agent’s instructions, persona, and conversation objectives), select the voice (any ElevenLabs voice or cloned voice), configure the LLM (GPT-4o, Claude, Gemini, or ElevenLabs’ own conversational model), and set up any tools the agent needs (CRM lookups, calendar integrations, data collection forms). The agent should be tested in a simulated conversation before batch deployment — ElevenLabs provides conversation simulation tools for this.
Contact List Preparation
Prepare a contact list containing at minimum: recipient phone numbers in E.164 format (+1XXXXXXXXXX for US numbers). Optional personalisation variables can be included for each contact — name, appointment date, account balance, order number, or any other data the agent prompt references. These variables are injected into the system prompt or first message for each individual call, allowing the AI to personalise the conversation without requiring a separate agent configuration per recipient.
Batch Scheduling
Batch calls can be executed immediately or scheduled for a future time. The timezone parameter allows specification of the local time for call execution — critical for outbound calls where legal regulations and etiquette standards prohibit calls outside business hours in the recipient’s location. Setting call_timezone: ‘America/New_York’ with an execution time of 10:00 AM ensures recipients in New York receive calls at 10:00 AM their local time, regardless of the server’s timezone.
Execution and Monitoring
Once a batch is initiated, ElevenLabs handles the telephony infrastructure — dialing each number, managing connection state, handling voicemail detection (where configured), and logging call outcomes. The ElevenAgents dashboard provides real-time monitoring of batch execution: calls completed, calls in progress, calls failed, and calls queued. Full conversation transcripts and analysis are available for each completed call through the conversations API endpoint.
Setting Up Batch Calling via the API
Create a Batch Call
API endpoint: POST /v1/convai/batch-calls. Request body includes: agent_id (required — the conversational AI agent to use), recipients (required — array of objects each containing phone_number and optional custom_llm_extra_body for personalisation variables), scheduled_time (optional ISO 8601 datetime for scheduled execution; omit for immediate execution), call_timezone (optional — recipient local timezone for scheduling), and call_outcome_webhook_url (optional — webhook to receive call outcome data after each call completes).
Retrieve Batch Call Status
API endpoint: GET /v1/convai/batch-calls/{batch_call_id}. Returns: batch status (pending, running, completed, failed), total_calls count, calls_completed count, calls_failed count, and a summary of outcome categories from data collection if configured in the agent.
Individual Call Outcomes
Each call in a batch generates a conversation record accessible through the conversations API. GET /v1/convai/conversations?batch_call_id={batch_call_id} returns all conversations from a specific batch. Each conversation includes: call duration, transcript, analysis results (if the agent has evaluation criteria configured), and data_collection results (any structured data the agent was instructed to collect during the call).
Use Cases
Appointment Reminders — Healthcare and Services
Healthcare providers, dental practices, hair salons, and any service business with scheduled appointments face a consistent challenge: no-shows reduce revenue and waste capacity. AI batch calling for appointment reminders outperforms SMS and email reminders on response rate — a spoken, personalised reminder that the patient can respond to (confirm, cancel, or request a reschedule) achieves higher engagement than a text message. The agent prompt personalises each call with patient name, appointment date, time, and provider: ‘Hello [Name], this is a reminder from [Practice] about your appointment with [Provider] on [Date] at [Time]. Press 1 to confirm, say ‘cancel’ to cancel, or say ‘reschedule’ to speak with our team about a new time.’
Sales Outreach — Initial Lead Contact
Sales development teams use batch calling to make initial outbound contact with inbound leads before human SDR follow-up. The AI agent’s role in this workflow is not to close deals — it is to qualify interest, set appointment expectations, and handle basic questions before passing engaged leads to human sales. A healthcare software company running batch calls to conference leads can qualify 500 leads per day at a cost equivalent to a fraction of human SDR time, with every conversation transcript immediately available for the human sales team’s review.
Customer Satisfaction Surveys
Post-interaction NPS and CSAT surveys via voice achieve significantly higher completion rates than email surveys for certain demographics. Batch calling allows a business to survey every customer interaction within 24 hours — the conversational AI agent conducts a brief structured interview, collects the rating and feedback, and writes the results to a data collection store for analysis. The conversation transcript provides qualitative context alongside the quantitative rating.
Payment and Account Notifications
Financial services, utilities, and subscription businesses use batch calling for proactive account management: payment reminders for overdue accounts, renewal notifications for expiring subscriptions, and fraud alert confirmations. Voice outreach for payment reminders achieves higher response rates than email for overdue accounts — a spoken reminder creates a different psychological impact than a notification in an inbox. ElevenLabs batch calling handles these at scale with consistent, professional delivery regardless of call volume.
Batch Calling Best Practices
Respect time zone regulations
In the United States, the Telephone Consumer Protection Act (TCPA) restricts outbound calls to between 8:00 AM and 9:00 PM in the recipient’s local timezone. Similar regulations exist in the EU, UK, Australia, and most markets. Always configure call_timezone for the recipient’s local timezone and schedule batch calls within compliant hours. Non-compliance with TCPA and equivalent regulations carries significant financial penalties — legal review of outbound calling compliance is mandatory before deploying batch calling at scale.
Test with small batches first
Before executing a batch of 10,000 calls, test with a batch of 10-20 calls to real recipients or internal testers. Verify that personalisation variables inject correctly, the agent handles common response variations gracefully, voicemail handling works as expected, and call outcomes are logging correctly. Small test batches catch configuration problems before they affect large contact lists.
Configure data collection for structured outcomes
Batch calling generates significant volumes of conversation data. Configure data_collection fields in your agent to capture structured outcomes — appointment confirmation (boolean), lead qualification score (1-5), survey rating (1-10), payment confirmed (boolean). Structured data collection makes batch call results immediately actionable in downstream systems without requiring manual transcript review for every call.
Three Insights Most Batch Calling Guides Miss
1. Voicemail Handling Is the Most Important Configuration Decision
In outbound batch calling, a significant proportion of calls — often 40-60% — reach voicemail rather than a live person. How the agent handles voicemail determines the ROI of batch calling more than almost any other configuration choice. Options include: leaving a pre-recorded voicemail message (the agent detects voicemail and plays a recorded message), conducting a simulated voicemail conversation (the agent speaks a message as if leaving a voicemail), or logging the call as unanswered and marking for follow-up. The correct voicemail strategy depends on use case — appointment reminders benefit from clear voicemail messages; sales outreach may benefit from no voicemail to preserve future contact opportunities.
2. Call Outcome Data Is the Asset, Not the Calls Themselves
Businesses deploying batch calling for the first time focus on call completion rates and connection rates. The more valuable output is the structured data generated from every conversation — qualification scores, survey results, confirmed appointments, payment confirmations, and conversation transcripts. This data, aggregated across thousands of calls, provides insights that manual calling at equivalent scale would never produce. Design the agent’s data collection schema and conversation structure with the downstream data analysis in mind, not just the individual call experience.
3. Batch Calling and Inbound Agent Infrastructure Are Shared
ElevenLabs Conversational AI agents are configured once and can serve both inbound and outbound use cases. An agent configured for inbound customer service can be deployed in batch calling for outbound surveys using the same voice, personality, and knowledge base — the agent just initiates the call rather than receiving it. This means the investment in building a high-quality conversational agent amortises across both inbound and outbound use cases, rather than requiring separate agent configurations for each deployment scenario.
Key Takeaways
- ElevenLabs Batch Calling automates outbound AI voice calls at scale — schedule thousands of calls with personalisation, timezone control, and structured outcome collection.
- Top use cases: appointment reminders, sales outreach, NPS/CSAT surveys, payment reminders, event notifications.
- Configure timezone (call_timezone parameter) to comply with TCPA and equivalent regulations restricting outbound call hours.
- Voicemail handling configuration is the highest-impact batch calling decision — 40-60% of outbound calls typically reach voicemail.
- Structured data collection transforms batch calling from a communication tool into a data collection infrastructure — design the conversation for the downstream data, not just the individual call experience.
Conclusion
ElevenLabs Batch Calling transforms what has historically been one of the most time-intensive business communication activities — outbound phone outreach at scale — into an automated, AI-powered operation that runs without human intervention, logs every conversation, and produces structured data at a cost and scale that human calling teams cannot match. For businesses with legitimate outbound communication needs — healthcare reminders, sales development, customer success, payment operations — batch calling represents a meaningful operational efficiency opportunity. The configuration investment is a few hours; the operational impact is measurable in reduced no-show rates, higher lead qualification throughput, and structured customer feedback data that previously required expensive survey operations to collect.
Frequently Asked Questions
What is ElevenLabs Batch Calling?
A feature of ElevenLabs Conversational AI (ElevenAgents) that automates outbound voice calls at scale. Upload a contact list, configure a conversational AI agent, schedule the batch, and ElevenLabs calls each recipient, conducts the AI conversation, and logs outcomes automatically.
How many calls can I make with ElevenLabs Batch Calling?
ElevenLabs does not publish a hard limit on batch call volume in public documentation. Practical limits are determined by your phone number capacity (concurrent call limits per Twilio number), your ElevenLabs plan’s concurrent agent capacity, and applicable regulatory limits in your market. Contact ElevenLabs enterprise sales for high-volume batch calling requirements.
Does ElevenLabs Batch Calling comply with TCPA?
ElevenLabs provides the technical infrastructure for batch calling. Compliance with TCPA (US), GDPR (EU), and equivalent regulations is the deploying business’s responsibility — not ElevenLabs’. Ensure your use of batch calling complies with applicable regulations in every market you call into. Legal review is mandatory before deploying batch calling at commercial scale.
What phone system does ElevenLabs Batch Calling use?
ElevenLabs Conversational AI integrates with Twilio for telephony infrastructure. You need a Twilio account with phone numbers configured in ElevenLabs, or a SIP trunk integration for existing telephony infrastructure.
Can batch calls be personalised per recipient?
Yes — the recipients array in the batch call API accepts personalisation variables per contact that are injected into the agent’s system prompt or first message. Each recipient can receive a call personalised with their name, appointment details, account information, or any other data available in your contact list.
Methodology
Batch calling features from ElevenLabs official blog post ‘Introducing batch calling for ElevenLabs Conversational AI’. API parameters from ElevenLabs changelog (January 2026) documenting batch call timezone support and batch call enhancements. Use case context from ElevenLabs blog page 4 batch calling announcement. Compliance information from TCPA official FCC documentation. This article was drafted with AI assistance and reviewed by the editorial team at ElevenLabsMagazine.com.
References
ElevenLabs. (2026). Introducing batch calling for ElevenLabs Conversational AI. https://elevenlabs.io/blog
ElevenLabs. (2026). ElevenAgents documentation. https://elevenlabs.io/docs/eleven-agents
ElevenLabs. (January 2026). Changelog — batch call enhancements. https://elevenlabs.io/docs/changelog
