Actionist·Retell AI
Retell AI

Retell AI

· #396 most-used

Build, deploy, and manage AI voice agents that handle real phone calls

SalesCommunicationAnalyticsAIAutomationPhone & SMS

Retell AI is a developer platform for building and deploying AI-powered voice agents that handle real phone calls — inbound, outbound, and web — at scale. It provides a REST API for provisioning phone numbers, creating and managing agents, uploading knowledge bases, configuring LLMs, placing calls, and retrieving transcripts and post-call analysis. Connect Retell AI to Actionist and your agents can fire outbound calls when leads qualify, retrieve transcripts and sentiment scores after every call, keep knowledge bases current when documentation changes, monitor concurrency before batch runs, and log all call data to your CRM — without anyone logging into the Retell dashboard.

Try Retell AI with ActionistVisit Retell AI
Average time saved
14 hours
per person · per month
≈ 2 workdays back

Eliminates manual work. Agents eliminate manual call logging, knowledge base updates, post-call CRM entry, and compliance transcript reviews that would otherwise require human time after every call.

Schedule

What your Retell AI agent runs on autopilot

A week of scheduled jobs your Actionist agent will execute on your behalf.

28Scheduled jobs
7Agents at work
24/7Always on
Agents
Tue–Thu
Tue
Wed
Thu
7a
8a
9a
10a
11a
12p
1p
2p
3p
4p
5p
6p
Multi-app workflows

Retell AI × every other app you use

End-to-end automations that span multiple apps — each one a real business outcome.

6Workflows
6Apps spanned
~33 hrsSaved / week
6Personas served
For sales
Featured2 apps

MQL-to-call: prospect connected within about a minute of qualifying

When a lead reaches MQL status in HubSpot, the agent registers the call with the prospect's context, dials the number, and once the call completes retrieves the transcript and outcome to update the HubSpot deal record — the full first-touch cycle automated without any human involvement.

~10 hrs

Time saved for your team — every week, on autopilot

The flow
Trigger·When a lead reaches MQL status in HubSpot
Trigger
Step 1
HubSpot
Detect new lead reaching MQL status in HubSpot
write
Step 2
Retell AI
Register Phone Call with lead name, company, and deal stage
write
Step 3
Retell AI
Make a Phone Call to the prospect's number
read
Step 4
Retell AI
Get a Call to retrieve transcript and sentiment after completion
write
Step 5
HubSpot
Update deal record with call outcome and transcript summary
Result
Register Phone Call with lead name, company, and deal stageMake a Phone Call to the prospect's numberUpdate deal record with call outcome and transcript summary
The win
Saved per run
20 min
Runs / week
~30×
No qualified lead waits more than a minute for first-touch outreach
Driven bySales Agent
ROI

Savings

What your team gets back — two angles: what you stop doing manually, and what that's worth.

Without Actionist

What you do manually today

With Actionist

What your agent runs for you

  • Sales
    120 min / week
    Manual dialling and call note entry

    Reps manually dial each lead, take handwritten notes during the call, and log them in the CRM afterward — adding 15–20 minutes of admin per outbound call.

    Sales Agent
    0 min
    Agent dials and logs the outcome automatically

    When a lead qualifies, the agent registers and places the call, retrieves the transcript when it ends, and writes the outcome to the CRM deal record — all without a human touching a dialler.

  • Marketing
    45 min / week
    Manual knowledge base updates

    Marketers log into the Retell dashboard, locate the knowledge base, delete outdated documents, and re-upload revised versions — a 15-minute task done manually after every content change.

    Marketing Agent
    0 min
    Agent keeps KBs current the moment docs change

    When campaign content is updated in Notion, the agent pushes the new text to the Retell AI knowledge base within about a minute — voice agents answer accurately from the very next call.

  • Customer Support
    60 min / week
    Manual post-call escalation review

    A supervisor reviews call sentiment reports daily or weekly, manually creates tickets for bad calls, and assigns human follow-ups — by which time the customer has often already churned or complained publicly.

    Customer Support Agent
    0 min
    Agent escalates negative calls within minutes

    When Watch Call fires with a negative sentiment flag, the agent creates a ticket and books a human follow-up within about a minute — the customer hears back before the hour is out.

  • Human Resources
    150 min / week
    Manual candidate phone screening

    Recruiters call each candidate individually, take notes during the call, and manually update the ATS with outcome and notes — consuming 20–30 minutes per candidate across scheduling, calling, and logging.

    Human Resources Agent
    0 min
    Agent runs first-round screens automatically

    Recruiters define the screening criteria once; the agent calls every qualified candidate, collects responses, and logs outcomes to the ATS — all before the recruiter's morning standup.

  • Finance
    30 min / week
    End-of-month billing surprise

    Finance reviews the Retell AI invoice at the end of the billing period, discovers the overspend after the fact, and spends hours tracing which campaigns caused the overrun.

    Finance Agent
    0 min
    Agent projects spend against budget daily

    The finance agent sums call minutes via List Calls daily, projects the billing-period cost, and fires a Slack alert when on track to exceed budget — before the billing period closes.

  • Operations
    60 min / week
    Ad-hoc configuration audits

    Ops manually checks each agent's configuration in the Retell dashboard before major campaigns — a scattered, time-consuming process that often misses configuration drift until a call fails.

    Operations Agent
    0 min
    Agent audits agents and numbers on a schedule

    The operations agent checks every voice agent configuration, LLM binding, and phone number assignment weekly and flags drift before campaigns run on misconfigured infrastructure.

  • Legal
    90 min / week
    Manual transcript compliance review

    Legal manually downloads call transcripts, searches each for the required disclosure phrase, and maintains a spreadsheet of exceptions — a process that often falls weeks behind the call volume.

    Legal Agent
    0 min
    Agent sweeps call transcripts for compliance weekly

    The legal agent retrieves transcripts for all calls to disclosure-required jurisdictions, checks for the required language, and logs exceptions — a full compliance sweep in minutes, not days.

+ 100s of other Retell AI automations
Average time saved
56 hrs / person / month
Calculator

Calculate what your team saves

Team size
5 people
Hourly rate
$75 / hr
Hours saved / week
18
Hours saved / year
875
Annual ROI
$65,625

Based on Retell AI's typical team usage — the visible tasks plus a few other automations the agent runs: ~3.5 hrs / person / week of admin work automated.

Connect

How to plug Retell AI into Actionist

Pick the connection method that suits your environment.

Authenticate using a Retell AI API key. All REST API requests are authorised with a Bearer token, giving Actionist full access to manage agents, calls, phone numbers, knowledge bases, and LLM configurations.

1
Open the API Keys tab

Log in to your Retell AI account and navigate to the API Keys tab in the dashboard.

2
Create and copy the key

Click Create key, give it a descriptive name (e.g. 'Actionist'), and copy the generated key. Store it securely — it will not be shown again.

3
Paste into Actionist

Paste the key into Actionist and click Test connection. Actionist runs a lightweight read call to confirm the key is valid before any actions run.

Credentials you'll need
API key*
Retell AI dashboard → API Keys tab → Create key
Actions

19 actions your agent can call

Read and write operations available to your Actionist agent.

Triggers

1 event your agent can react to

Events your agent watches for, and the actions it kicks off in response.

FAQs

Questions about Retell AI + Actionist

How does Actionist connect to Retell AI?
Go to the Apps tab, find Retell AI, and click Connect. Retell AI uses API key authentication. In your Retell AI dashboard, navigate to the API Keys tab and create a new key. Paste it into Actionist and the agent runs a lightweight test call to confirm the handshake before any actions execute.
What permissions does the Actionist agent need on my Retell AI account?
Your API key needs to have full access to the Retell AI account. The key inherits the permissions of the account that generated it, so it can create and manage agents, phone numbers, knowledge bases, LLMs, and calls. For webhook verification, Retell uses the API key together with an HMAC-SHA256 signature — make sure the key you use in Actionist also has the webhook badge in your Retell dashboard if you want inbound call-event triggers.
Can I connect Retell AI to other apps in the same Actionist workflow?
Yes. Retell AI works well alongside CRMs, scheduling tools, and communication platforms. Common patterns: when a call ends, the Watch Call trigger fires and the agent logs the outcome to HubSpot or Salesforce; a new lead in your CRM triggers an outbound Make a Phone Call action via Retell; post-call transcripts get written to Notion or Google Sheets. Any of Actionist's 500+ connected apps can send or receive data alongside Retell AI in the same workflow.
What are the most common things agents do with Retell AI?
The four patterns that come up most: (1) post-call logging — when a call ends, the agent extracts the transcript, sentiment, and task-completion status and writes structured data to a CRM or spreadsheet; (2) outbound dialling campaigns — the agent reads a prospect list and calls Make a Phone Call in sequence, scheduling callbacks on no-answer; (3) knowledge base refresh — whenever product documentation changes, the agent updates the Retell AI knowledge base so every voice agent answers with current information; (4) agent health monitoring — Get Concurrency is polled on a schedule to ensure no campaigns are hitting the concurrency ceiling before a batch call run.
What is the difference between Make a Phone Call and Register Phone Call?
Use Register Phone Call to pre-configure a call with custom parameters before it is connected. This lets you set the agent ID, metadata, call direction, and any dynamic variables the voice agent will receive at call start. Make a Phone Call is the simpler path for ad-hoc outbound calls. For production outbound campaigns, Register Phone Call gives you more control over what context the agent starts the call with.
How do I keep my Retell AI knowledge base up to date automatically?
Retell AI's knowledge bases let your voice agents answer domain-specific questions with up-to-date content. You create a knowledge base via Create Knowledge Base, upload or link your documents, then attach the knowledge base ID to a voice agent. When Actionist detects that your product docs or FAQ content has changed — for example, via a Notion or Google Drive trigger — it calls the Retell AI API to update the knowledge base automatically, so your voice agents always answer from current content without manual re-uploads.
How does the Watch Call trigger work, and what data does it deliver?
The Watch Call trigger fires within about a minute of a call completing in your Retell AI account. It delivers the call ID, agent ID, call status, duration, and summary metadata. You can then chain a Get a Call action to retrieve the full transcript and post-call analysis fields. This is the standard pattern for building post-call workflows — CRM logging, quality scoring, follow-up scheduling — that run automatically after every customer conversation.
Does Retell AI support multiple languages, and how do I set that up through Actionist?
Retell AI supports 31+ languages natively. The language is set on the voice agent configuration — you can specify the primary language when you Create an Agent or update it later. If you run multilingual campaigns, you can create separate agents per language region, each with its own knowledge base and voice profile, and route callers to the appropriate agent based on their phone number prefix or a pre-call IVR selection step.