Regal
Regal is an enterprise voice AI agent platform for contact centers. Teams build no-code agents once, deploy them across phone, SMS, chat and WebRTC, then orchestrate them alongside human staff and rehearse everything before launch.
What is Regal?
Regal is a voice AI agent platform built for enterprise contact centers. Its own summary is blunt: Regal helps businesses build, improve and manage voice AI agents that, in the company's words, sound human and scale like software.
The product splits into three layers. First the agents themselves, covering phone, SMS, chat and WebRTC voice, built once and deployed across every channel. They are claimed to speak more than thirty languages with native fluency, and their tone, persona and interruption handling can be tuned across the major text-to-speech providers. Second the building layer: a no-code AI Agent Builder aimed at teams without developers, alongside Regal Copilot, AI that builds AI and promises the trip from idea to deployment in days. Third the orchestration layer, where Regal differs most from a plain voice-bot vendor. It brings a Unified Customer Profile that consolidates data in real time, an event-driven Journey Builder, IVR and intelligent routing, Conversation Intelligence with live transcription and automatic summaries, a Sales Dialer offering power, preview and progressive modes, plus analytics, custom objects and configurable dashboards.
Two capabilities are unusual on this market. Simulations and A/B testing let teams validate an agent before it touches a real customer, and the phone-number tooling, Branded Caller ID and Spam Remediation, tackles the answer-rate problem that quietly undermines outbound programs. Human agents are no afterthought either: the Unified Agent Desktop gives them a 360-degree customer view and lets AI transfer calls with full context.
Regal was founded by Alex Levin and Rebecca Greene after running a contact center at Angi, where legacy software forced a low-touch, low-cost strategy they judged bad for customers. The company is based in New York and sells into regulated, trust-sensitive sectors: insurance, financial services, healthcare, Medicare, education, local services, eCommerce, legal services, loan collections and BPO.
It is unmistakably an enterprise product. Regal sets the bar at twenty-five agents and 100,000 call minutes a month, there is no self-serve signup, and no price appears anywhere on the site.
What it does
- Answer and place customer calls around the clock with voice agents that hand off to humans in context
- Qualify, score and route inbound leads the moment interest appears
- Book, confirm and chase appointments to cut no-shows
- Run automated collections and payment follow-up campaigns
- Build one agent and deploy it across phone, SMS, chat and WebRTC without writing code
- Rehearse agents in simulations and A/B test variants before customers ever hear them
- Brand outbound caller ID and clear spam flags to protect answer rates
When to use Regal / When not to
A quick filter to help you decide if Regal is the right fit.
When to use Regal
- Enterprise contact center leaders running at least 25 agents and 100,000 call minutes a month
- Insurance and financial services teams handling qualification, quotes, claims and collections at volume
- Healthcare and Medicare operations built on appointment reminders, enrollment calls and patient follow-up
- Revenue teams that need instant speed-to-lead on inbound interest and outbound qualification
- CX and support organizations wanting 24/7 coverage with clean, context-carrying escalation to humans
When not to use Regal
- Small businesses and startups below the 25-agent, 100,000-minute floor Regal sets for itself
- Buyers who want to sign up online: there is no self-serve account, no published price and no free tier
- European organizations bound by EU data residency, since servers and data centers sit only in the United States
- Workloads touching social security numbers, HIPAA health records, biometrics or card data, which the DPA excludes outright
- Teams expecting a mobile app, as none is published for either iOS or Android
How to use Regal
A typical end-to-end flow, from setup to results.
- Hear it first: call the AI agent advertised on the homepage on 212-500-6000, no account required
- Book an enterprise demo through the request-demo form, as there is no self-serve signup
- Sign in to the web platform once your account has been provisioned
- Connect your systems through the 40-plus prebuilt integrations, the API or webhooks
- Consolidate customer data into the Unified Customer Profile and custom objects
- Build your first agent in the no-code AI Agent Builder, or let Regal Copilot draft it
- Design the surrounding logic: journeys, segments, IVR routing and human escalation paths
- Rehearse the agent in Simulations, then A/B test variants before going live
- Deploy the same agent across phone, SMS, chat and WebRTC
- Monitor with Conversation Intelligence, dashboards and QA, and iterate with your Forward Deployed Engineer
Pros & Cons
Pros
- Genuine multichannel reach, phone, SMS, chat and WebRTC, from a single agent build
- Simulations and A/B testing offer a rare way to validate an agent before customers meet it
- Contractual no-training commitments on both sides, Regal's own and its LLM providers'
- Twelve subprocessors named outright in the published DPA, which few vendors bother to do
- Telecom-grade extras, Branded Caller ID and spam remediation, that protect answer rates
- White-glove guidance included in every plan, with Forward Deployed Engineers on setup and tuning
- More than thirty languages claimed, with fine control over voice, persona and interruptions
Cons
- No public pricing whatsoever: a sales conversation is the only route to a number
- A hard enterprise floor of 25 agents and 100,000 minutes a month shuts out smaller teams
- Neither a permanent free plan nor a free trial is advertised anywhere on the site
- Data is hosted only in the United States, with no EU residency option on offer
- The DPA covers the CCPA alone though GDPR is claimed in marketing, and no Article 27 representative exists
- No postal address is published, and the legal entity name differs between the terms and the privacy policy
- No mobile app, and the trust center renders only in JavaScript so its contents cannot be read plainly
Pricing & Plans
Regal publishes no price. The pricing page carries only a request form, the company stating that its model depends on factors specific to each team. Billing is described as paying for the features you use, with discounts as committed spend and contract length increase. No permanent free plan and no free trial are advertised, and no entry-level amount or currency is disclosed anywhere on the site. The practical barrier is qualification rather than money: Regal states it works with enterprises of at least twenty-five agents and 100,000 call minutes a month.
Data, GDPR & hosting
A consolidated view of how Regal handles your data.
GDPR overview
Regal asserts GDPR alignment in marketing terms: the security page says it adheres to standards "including GDPR and CCPA", and the product page calls the platform CCPA and GDPR compliant. The contractual reality is narrower. Regal publishes a Data Processing Addendum, last updated 26 February 2025, but that document defines Applicable Data Protection Laws as the CCPA alone, with no GDPR articles, no standard contractual clauses and no transfer mechanism. No Article 27 EU representative is designated and no DPO is named. Servers and data centers are in the United States only, so any European use is an outbound transfer. Concrete positives remain: twelve subprocessors named in the open, contractual no-training commitments, SOC 2 certification, and a privacy contact at privacy@regal.io, with access and deletion requests routed to support@regal.io.
Who owns the data?
Regal's terms are explicit: as between the parties, the customer owns all Data sent to or through the platform. The customer grants Regal only a non-exclusive, non-transferable license to host, process and transmit it. Regal keeps its own side: the platform, its underlying works, and the de-identified Diagnostic Data it gathers on how the service performs. One asymmetry deserves attention. Any feedback or suggestion a customer volunteers becomes Regal's to use perpetually and irrevocably, with no credit and no compensation. Under the Data Processing Addendum, Regal Voice acts as a CCPA service provider and never as a seller of personal information.
Reuse rights
Customers keep their data and may reuse it freely: they own it, and Regal holds nothing more than a service license over it. Regal may access, host, copy, process and transmit that data for three purposes only, namely running and improving the service, providing support and fixing technical faults, and complying with law or the confidentiality clause. Data may pass through AI tools, but Regal commits contractually not to use it to train, refine or improve those tools for the benefit of anyone but the customer, and states it trains no proprietary models of its own. Twelve subprocessors are named, among them Amazon Web Services, Twilio, Snowflake, OpenAI, Anthropic, Google and ElevenLabs. Regal does not sell personal information.
Data retention & training
Hosting summary
Regal's privacy policy is unambiguous: its servers and data centers are located in the United States. Anyone using the service from elsewhere is transferring personal information out of their own region and into the US for storage and processing, and Regal reserves the right to move data onward from the United States to other countries for storage, processing and service delivery. The policy warns explicitly that receiving regions may have weaker privacy and data-security laws than the user's own. No European or other regional residency option is offered anywhere on the site. Amazon Web Services and Snowflake appear among the twelve subprocessors named in the Data Processing Addendum, consistent with US cloud infrastructure. One technical caveat: the public IP resolved for the domain belongs to a Cloudflare anycast node and says nothing about where the application actually runs. The privacy policy, not DNS, is the reliable source here.
Things to keep in mind
Risks and trade-offs to weigh before adopting Regal.
- Automated outbound calling in the United States requires prior TCPA consent, and Regal notes most companies lack it for past customers
- Agents that sound human can mislead callers who never realize they are speaking to software, since US law imposes no duty to disclose it
- GDPR compliance is asserted in marketing while the contract covers only the CCPA, a gap easy to miss and expensive to discover late
- HIPAA is used as a selling argument, yet the DPA requires customers to warrant their data holds no HIPAA-protected health information
- Retention is discretionary: data is kept as long as Regal deems needed, with no published duration and no purge schedule
- Handing frontline conversations to AI can quietly erode a team's own listening and de-escalation skills, and the knowledge of why customers really call
- Opaque pricing and a hard enterprise floor make it easy to sink evaluation time before discovering the product was never meant for you
Setup & Integrations
Technical difficulty
Moderate, and front-loaded. The agent builder is genuinely no-code and Regal Copilot claims to carry an agent from idea to deployment in days, so writing the conversation needs no developer. The work sits around it: connecting data systems, defining telephony flows, configuring human escalation and ensuring the agent extracts the right fields. Regal answers this with Forward Deployed Engineers who stay through setup, launch and tuning, and white-glove guidance is included in every plan. There is no self-serve path, so onboarding starts with a sales conversation rather than a signup form.
Deployment
Integrations
Behind Regal
Fundraising
Social
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
How much does Regal cost?
Is Regal a fit for a small team?
Which channels do the agents cover?
Will my customer data be used to train AI models?
Where is the data hosted?
Is there a DPA, and does it cover GDPR?
Does Regal offer an API?
What certifications does Regal claim?
Can I try it without talking to sales?
How old must a user be?
Should you pick Regal?
Regal is one of the more complete voice AI platforms aimed at large contact centers, and its strengths are structural rather than cosmetic. Building an agent once and running it across phone, SMS, chat and WebRTC is genuinely useful; being able to rehearse that agent in simulations and A/B test it before a customer ever hears it is rarer still. The telecom layer, branded caller ID, spam remediation and number management, solves a problem most AI-agent vendors ignore. On data, Regal is more forthcoming than most: twelve subprocessors named in the open, and contractual commitments that neither Regal nor its LLM providers will train on customer data.
The reservations are equally clear. Nothing about the commercial terms is public. There is no price, no free plan, no trial and no way to open an account without a sales call, and Regal states plainly that it wants organizations with at least twenty-five agents and 100,000 call minutes a month. Below that threshold you cannot evaluate this product on your own.
The legal picture deserves a careful read. GDPR compliance is claimed in marketing copy, yet the published DPA is framed on the CCPA alone, no Article 27 European representative is designated, and servers sit exclusively in the United States. A European buyer with data-residency obligations should treat that as a blocking question, not a detail. The same caution applies to healthcare, where HIPAA is used as a selling point while the DPA asks customers to warrant their data contains no HIPAA-protected health information.
For a well-funded US enterprise with high call volumes and a compliance team to read the contract, Regal is a serious candidate. For everyone else, the door is largely closed.
- Choosing a selection results in a full page refresh.
- Opens in a new window.