Voiceflow
Visual AI agent builder for customer service across chat, voice and phone
Overview
Voiceflow is a Toronto-built platform for designing, deploying and monitoring customer-service AI agents across web chat, WhatsApp, SMS and voice. Its V4 release (February 2026) combines autonomous Playbooks with deterministic Workflows on a collaborative canvas. Billing is a monthly plan plus usage credits, with plan prices shown only inside the app.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 14 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Limited features
- One-time credit grant that does not renew
- Not suitable for production, per the docs
- For individual builders
- Access to all LLM models
- Higher agent limit than Free
- Monthly credit allotment that resets each cycle
- For growing teams
- LLM fallback models
- Priority support
- Unlimited product usage
- Single sign-on
- Private cloud hosting
- Custom credit allotments and an account manager
- Free trial, no credit card required
- Multi-client workspace management
- White-labeling and client handoff tools
- Access to all major model providers
Use cases
What you can produce with Voiceflow
- Customer-service AI agents deployed on web, in apps, on WhatsApp, by SMS and over the phone
- Autonomous Playbooks combined with deterministic Workflows on a shared visual canvas
- A different LLM assigned to each step, with fallback models on Business
- LLM-scored evaluations and analytics on live conversations
- Development, staging and production environments for agent releases
- Multi-client, white-labelled workspaces for agencies
Voiceflow's collaborative canvas is still its strength: designers, CX leads and engineers build the same agent, mixing autonomous Playbooks with locked-down Workflows, picking models per step and checking results with built-in evaluations. Channel reach (web, WhatsApp, SMS, voice) and SOC 2, ISO 27001 and HIPAA coverage suit mid-size and enterprise CX teams.
The catch is price visibility: no figures on the public site, plan prices only inside the app, and credit-based usage on top. Auto top-ups now stop agents going dark at the cap, but set a spending limit and model voice usage before launch.
What we found
Voiceflow began in Toronto as Storyflow, a team making interactive children's stories for Amazon Alexa, and took the Voiceflow name in November 2018 when it turned its internal drag-and-drop skill builder into the product. It spent the next few years as a collaborative design tool for voice and chat assistants, the "Figma for conversational interfaces" in its own words, and is now an AI agent platform for customer service. Voiceflow says more than 3,000 customers run over 10,000 live agents on web, in apps, on WhatsApp, by SMS and on the phone.
The current generation, V4, was unveiled on 18 February 2026. Instead of drawing every path by hand, a builder defines Playbooks for open-ended tasks and Workflows for fixed processes on a visual canvas, and the agent routes between them. A Context Engine limits what the model sees at each step to cut token costs, different models can be assigned to different steps, and evaluations, analytics and separate development, staging and production environments sit around it.
Pricing is the weak spot for anyone comparing tools. The public pricing page shows no numbers, only a self-serve track for agencies (free trial, no card, usage-based billing, white-labelling) and a demo track for businesses. The billing docs describe four plans, Free, Pro, Business and Enterprise, each a monthly fee plus editor seats, phone numbers and usage credits, with prices shown only inside the app. The Free plan's credits are a one-time grant. Earlier versions of this page said agents simply stopped when credits ran out; the docs now offer automatic top-ups and larger credit bundles.
Voiceflow has raised more than $35 million. The latest disclosed round, $15 million led by OpenView at a $105 million valuation, closed in April 2023. Felicis, True Ventures and Amazon are also backers. For CX and product teams who want a shared canvas with real engineering hooks, it remains a strong choice.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
Notable facts
- Voiceflow was originally called Storyflow and built exclusively for Alexa skills before pivoting to a multi-channel platform as voice assistants faded in importance.
- The platform's Figma-like real-time collaboration was a first for conversational AI design tools, allowing design-developer handoff without exporting documentation.
- Voiceflow's template library includes pre-built conversation flows for over 50 industries, reducing average deployment time from weeks to days.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including Voiceflow's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Voiceflow unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Voiceflow directly →
Spotted something out of date? Suggest an update →
More in Agents & Automation
