AI Agent Index
ByHeather MacAvelia·Last verified Jun 24, 2026
Voiceflow logo

Voiceflow

4.2/ 5

by Voiceflow

✓ VerifiedMCP
Visit voiceflow.com

Enterprise AI agent builder for CX teams. Deploy across voice, chat, and WhatsApp with visual workflows, Voiceflow Core model, and 300+ integrations. Usage-based and custom pricing.

From

Custom

custom

GitHub

Stars

G2

4.6 / 5

110 reviews ↗

MCP

⚡ Yes

Compatible

Voiceflow is an enterprise AI agent builder, positioning itself as the operating system for AI customer experience. Founded in 2019, Voiceflow has evolved from a voice assistant design tool into a full platform for building, testing, launching, and iterating AI agents across chat, voice, and messaging channels. The platform combines visual flow design with agentic playbooks, deterministic workflows, version control, prototype testing, and production pipeline management. Named a G2 2026 Best Software Award winner, Voiceflow powers AI agents for Turo, StubHub International, Sanlam Studios, Trilogy, and eSnipe. Published case studies show measurable results: Trilogy automated approximately 60% of customer support across 90 products in 12 weeks, eSnipe automated 70% of support interactions, StubHub International built and launched a customer support agent in 90 days with a non-technical team, and Sanlam Studios built an AI financial coach that converts 45% of conversations into leads. These results position Voiceflow in the enterprise CX automation category alongside Decagon, Sierra, and Ada. In May 2026, Voiceflow launched Voiceflow Core, its first proprietary model optimized for the tasks agents actually perform: tool calling, multi-turn reasoning, and instruction following inside playbooks. Teams can also choose from major LLM providers including GPT-4, Claude, and Gemini, or bring their own model on qualifying plans. Integration depth includes 300+ apps and software integrations. Key native integrations include Zendesk (bidirectional ticket creation and lookup), Twilio (outbound voice calls and SMS), Vonage (IVR and voice channels), WhatsApp Business API, Shopify, HubSpot, Discord, Linear, Notion, Manychat, and SendGrid. The platform supports Slack and Microsoft Teams embed, and any external service accessible via REST API or hosted MCP server connections. MCP tool support in the Agent step allows agents to consume any hosted MCP server for expanded tool access; local MCP servers are not supported. Notable gap: no native Salesforce CRM sync without custom API code. Pricing was restructured in 2026, replacing the previous per-editor tier model (Sandbox/Pro/Business/Enterprise) with two tracks. The Agencies and Partners track offers transparent usage-based billing with a free trial, multi-client workspace management, white-labeling, and client handoff tools. The Businesses track requires contacting sales and includes implementation support (self-serve or fully managed), multi-channel deployment across voice and chat, real-time observability and performance analytics, and team collaboration with roles and permissions. No public dollar amounts are listed for either track. Prior to the restructure, Pro was $60 per editor per month and Business was $150 per editor per month with per-seat add-on costs. Who Voiceflow is not for: developer-first teams who want open-source control and deploy-anywhere flexibility should evaluate Botpress (free to self-host, Cloud from $89/month) for code-first customization without per-editor fees. Microsoft Copilot Studio at $200/month for 25,000 messages fits Microsoft 365 organizations needing IT governance and SharePoint integration. Customer support teams that need a purpose-built helpdesk agent with live chat and ticketing built in should evaluate Intercom Fin at $0.99/resolution or Zendesk AI, both of which include the ticketing layer that Voiceflow requires you to connect externally. Teams needing fully autonomous outbound voice agents at high call volumes should look at Bland AI or Vapi where phone infrastructure is the core product. Current state Q2 2026: Voiceflow launched its first proprietary model (Voiceflow Core) in May 2026, optimized for agentic tasks including tool calling and multi-turn reasoning. The Environments feature enables production pipelines from development to staging to production. Query re-writing improves retrieval accuracy for conversational inputs. Topics auto-categorize conversations for analytics dashboards. The pricing model was restructured from per-editor tiers to usage-based (agencies) and demo-based (businesses) tracks with no public prices. SOC 2 Type II, ISO 27001, GDPR, and HIPAA certifications confirmed. G2 rating holds at 4.6/5 from 110 reviews.

Pricing

custom

View pricing ↗

Segment

b2b

Setup

easy

Verified

Jun 24, 2026

Transparency

Quote Only

Contract

Monthly or Annual

Data training

Not Disclosed

Autonomy

Human Optional

Capabilities

workflow-builderconversation-intelligenceno-codeautonomousmultilingual

Pros & Limitations

Editorial assessment

Pros

  • Industry-leading visual agent builder with production pipelines (dev, staging, production), version control, and prototype testing makes Voiceflow the strongest fit for CX and product teams that want full control over agent behavior without engineering bottlenecks
  • Voiceflow Core proprietary model launched May 2026, purpose-built for agentic tasks (tool calling, multi-turn reasoning, instruction following), with option to use GPT-4, Claude, Gemini, or bring your own model for maximum flexibility
  • Proven enterprise results with published case studies: Trilogy automated 60% of support across 90 products in 12 weeks, eSnipe automated 70% of interactions, StubHub International launched in 90 days with a non-technical team

Limitations

  • Pricing is no longer publicly listed: the 2026 restructure removed dollar amounts from the pricing page, making total cost of ownership difficult to estimate without contacting sales or starting a free trial on the Agencies track
  • No native Salesforce CRM integration listed on the integrations page, which may require custom API code and adds engineering overhead compared to platforms like Botpress Cloud or Microsoft Copilot Studio that offer native CRM connectors
  • Platform positioning has shifted from broad conversational AI (Alexa, Google Assistant) to enterprise CX focus, which means teams building consumer voice skills or non-support use cases may find fewer templates and community resources tailored to their needs

Technical Details

Deployment
cloud-platformapi
Model architectureMulti-LLM support across OpenAI GPT models, Anthropic Claude, plus custom NLU models with built-in conversation orchestration
Avg setup time30 minutes
Autonomous rateVoiceflow agents run autonomously based on agentic playbooks and deterministic workflows powered by Voiceflow Core or third-party LLMs. Agents handle customer questions, route conversations, and execute tool calls without per-action human approval. Designers configure escalation to human agents when AI confidence is low. Published results show 60 to 70% automation rates at enterprise scale.
MCP compatibleYes
Integrations
ZendeskTwilioVonageWhatsAppShopifyHubSpotDiscordLinearNotionManychatSendGridSlackMicrosoft Teams
Security
SOC 2 Type IIISO 27001GDPRHIPAA

Similar agents

Rating

4.2/ 5

Editorial score

How we score this →

Score breakdown

AutCap 4 · IntDepth 5 · PriceTrans 2 · IndEvid 4 · SetupAcc 5 = 4.20

Industries

SaaSB2BEnterpriseStartupsAgencies

Leave a review

Never displayed publicly.

Agent Stacks

See workflow stacks that feature Voiceflow.

Compare

Related Content