Decagon
AI agent platform purpose-built for customer support, with deep integration into support toolchains.
Decagon builds enterprise AI agents for customer support, positioning them as an AI concierge that resolves issues end to end rather than a chatbot that only suggests articles. Its agents handle conversations across chat, email, voice, and SMS, drawing on a company's knowledge bases, help articles, product documentation, past conversations, and internal policies to answer accurately, and escalating to a human with full context when a request falls outside what they should handle.
The centerpiece of the platform is what Decagon calls Agent Operating Procedures, or AOPs. Instead of building rigid coded decision trees, customer-experience teams write support workflows in plain language, when to issue a refund, when to escalate, how to handle a particular complaint, and Decagon compiles those instructions into executable agent logic.
The effect is a division of labor: CX operators own the business logic and can change agent behavior without waiting on an engineering sprint, while engineers manage the underlying integrations and guardrails. Crucially, AOPs trigger real backend operations, so an agent can retrieve and update customer data, process a refund, cancel a subscription, or query a third-party system, not just talk.
Decagon unifies its channels under one intelligence layer with cross-channel memory, so an agent recognizes a returning customer and recalls earlier issues, and a conversation that starts in chat can continue by voice without the customer repeating themselves. Its voice agents handle real-time calls with interruption handling and brand-matched voices, and can also run outbound calls for things like renewals and follow-ups. Increasingly the company routes traffic through language models it has trained in-house on customer-support interactions, alongside leading foundation models.
For the teams operating these agents, Decagon adds tooling for reliability and oversight: monitoring and quality assurance, guardrails, a trace view that shows how a given procedure executed for a specific conversation, and tools that help debug and draft new procedures. It is an enterprise-focused, managed offering, deployed with the help of Decagon's own product managers and engineers rather than through self-serve signup, and priced around resolved conversations rather than per agent seat, aimed at large support operations in sectors like fintech, retail, travel, and consumer technology.
Vendor details
Canonical URL
https://decagon.ai
Category
Customer support agent
Company status
independent
Use cases & customers
Target customers
Deployment options
In practice
Your CX team wants to change how the agent handles refunds without filing an engineering ticket. Decagon's Agent Operating Procedures let them write the workflow in plain language, and the platform compiles it into executable agent logic.
A customer starts in chat, then calls back later. Decagon unifies chat, email, and voice with cross-channel memory, so the voice agent already knows the prior issue and the customer doesn't repeat themselves.
You need an agent that resolves issues, not just answers them. Decagon's agents take real actions like processing refunds, updating accounts, and querying your systems, and escalate to a human with full context when needed.
Agentic Index coverage score
10.0 / 14 capabilities · 71%
| Integrations & Tool Calling | Full |
|---|---|
|
Integrations page documents connectors across distinct classes: CRM, helpdesk and ticketing (Salesforce, Intercom, Zendesk), knowledge bases (Confluence, Contentful, Kustomer), CPaaS and telephony (Amazon Connect, RingCentral, SIP trunking), plus custom endpoints and MCP connectivity for arbitrary systems, with agents both retrieving data and taking action. Sourcedecagon.ai/product/integrationsread 2026-09-05 |
|
| Workflow Orchestration | Full |
|
Agent Operating Procedures compile natural-language inbound and outbound workflows into executable agent logic that pulls live data and triggers actions across connected systems, with Git-based version tracking and staged rollout across agent versions. Sourcedecagon.ai/product/aopread 2026-09-05 |
|
| Knowledge Grounding & RAG | Full |
|
Decagon maintains the agent's own knowledge base, populated by sync from customer systems including Confluence, Contentful and Kustomer, and keeps it current with Suggestions, which runs conversation-driven gap analysis, ranks coverage impact and issues monthly content updates; traces name the knowledge articles the agent referenced. A vendor-built structure that persists, not context assembled per conversation. Sourcedecagon.ai/product/suggestionsread 2026-09-05 |
|
| Human Oversight & Guardrails | Partial |
|
Guardrails are configurable for brand voice, escalation and hallucination, a supervisor model revises ungrounded responses before they send, bad-actor detection escalates to a human, and Watchtower flags conversations for triage against customer criteria. Oversight is automated gating and QA after the fact; no customer facing approval queue holds an agent action pending sign off. Sourcedecagon.ai/product/aopread 2026-09-05 |
|
| Security, Identity & Governance | Full |
|
Security page documents role-based access control, SSO with Okta and Microsoft Entra, tamper-protected audit logs, short-lived scoped JWTs for agent access, AES-256 at rest and TLS 1.2+ in transit, DLP-based PII redaction and zero-day retention with model providers, alongside SOC 2, ISO 27001, PCI DSS, HIPAA, CCPA and Data Privacy Framework marks and a live trust portal. SourceDecagon security page and trust.decagon.ai; decagon.ai/securityread 2026-09-05 |
|
| Observability & Auditability | Full |
|
Traces explain how and why the agent decided as it did at any point in a conversation, covering model calls, workflow triggers, the knowledge articles referenced and runtime and latency figures, and Watchtower reviews every conversation against customer-defined criteria and flags outliers for triage. Reconstruction of reasoning rather than reporting on outcomes. Sourcedecagon.ai/product/overviewread 2026-09-05 |
|
| Memory & State Persistence | Full |
|
Memory is a named component of the agent engine, holding customer context across sessions and channels alongside the customer's systems of record with stated data portability and control, so a conversation started in chat continues by voice without repetition. Retention lifetime and per customer deletion are not documented publicly. Sourcedecagon.ai/product/overviewread 2026-09-05 |
|
| Deployment & Data Residency | Partial |
|
Security page names multi-region infrastructure with autoscaling and auto-failover as a resiliency property. No customer selectable region, VPC, single tenant or on premises option is published, so residency rests on the vendor's architecture rather than a customer choice, and the product is delivered only as SaaS hosted by Decagon. Sourcedecagon.ai/securityread 2026-09-05 |
|
| Prebuilt Agents, Templates & Packs | Not documented |
|
Agents are built by the customer writing Agent Operating Procedures in natural language, or by Duet generating them from the customer's own conversation transcripts and applying Decagon's deployment learnings behind the scenes; the seven industry pages are positioning and case studies rather than adoptable solutions. No template gallery, named prebuilt agent set, blueprint library or industry pack is published. Sourcedecagon.ai/product/aopread 2026-09-05 |
|
| Triggers & Channel Coverage | Full |
|
Chat, voice and email run on one intelligence layer, with inbound calls routed over SIP and CPaaS platforms, email intake through Zendesk and Intercom, live chat escalation through Zendesk Sunshine and Salesforce, and outbound and proactive agents documented as first-class. Sourcedecagon.ai/product/overviewread 2026-09-05 |
|
| Model Flexibility & Routing | Partial |
|
The agent engine runs on a network of frontier and specialized models, the security page names OpenAI and Anthropic as providers held to zero-day retention, and model redundancy is listed as a platform property. Routing is handled inside Decagon, with no customer or admin selection of the model that powers the product. Sourcedecagon.ai/product/overviewread 2026-09-05 |
|
| APIs, SDKs & MCP Extensibility | Partial |
|
Decagon documents a REST API for its own platform at api.decagon.ai, authenticated with a customer-issued key passed as a bearer token and rate limited, and an API key can be generated from the Developer page of the Decagon dashboard. The documentation at docs.decagon.ai sits behind a customer login and nothing on the public site links it, while the public Integrations page describes only inbound connectivity to the customer's own systems. SourceDecagon developer documentation at docs.decagon.airead 2026-09-05 |
|
| Testing, Debugging & Optimization | Full |
|
Testing & QA and Experiments are shipped customer-facing products: agents are validated before launch with end-to-end simulated conversations and unit testing, then live traffic is routed across agent versions to A/B test changes and measure impact on CSAT, deflection rate and other metrics at scale. The artifact under test is the customer's own agent and the result is readable and comparable. Sourcedecagon.ai/product/overviewread 2026-09-05 |
|
| Browser & Computer Use | Not documented |
|
The agent acts entirely through programmatic surfaces: API and MCP connectors, CRM and ticketing integrations, CPaaS platforms and SIP telephony. No browser control, virtual desktop or operation of software lacking a programmatic interface appears anywhere on the product surface. Sourcedecagon.ai/product/integrationsread 2026-09-05 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Recent platform changes
Decagon released Voice 3, a voice agent rebuilt so a fast conversational model listens and speaks while a stronger model handles reasoning, tools and guardrails in parallel. The agent can narrate progress and take follow up questions during long tasks, and it speaks through Chord, Decagon's own speech model for customer calls, switching automatically across more than 70 languages.
Bears on: Agent capability
View sourceDecagon introduced Browser Actions, a feature that allows its AI agents to access and complete tasks inside any web-based system. This capability enables the agent to navigate and interact with web interfaces directly.
Bears on: Browser/computer use
View sourceDecagon has announced its availability on the AWS Marketplace, enabling enterprises to deploy its conversational AI agents directly through their AWS environments. This move extends existing AWS data residency, compliance, and access controls to Decagon deployments from day one. Additionally, it allows teams using Amazon Connect to plug Decagon directly into their current contact center infrastructure.
Bears on: Deployment / data residency
View sourcePricing
Contact sales
hybrid
What is public
Decagon (decagon.ai - enterprise AI customer-support platform: autonomous agents resolve tickets end-to-end across chat/email/voice/SMS with cross-channel memory; Agent Operating Procedures (AOPs) = natural-language workflows; Voice 2.0 sub-second latency (+ outbound voice, Spring 2026); Watchtower QA; AI Actions for Stripe/Shopify/Salesforce) does NOT publish a price card. The pricing MODEL is public; figures are third-party only.
Billing mechanics
Usage-based (agents priced as 'workers,' not seats), layered on an annual platform fee. Two usage models: per-conversation (default/most popular - pay for every interaction whether resolved or not) OR per-resolution (higher rate, pay only when the AI fully resolves without human escalation). All custom-quoted; annual contracts; white-glove onboarding included. No native helpdesk (sits on Zendesk/Salesforce/Intercom/Kustomer).
Cost watchouts
~$50,000 annual platform fee baseline BEFORE any usage (corroborated by eesel/Quiq/Featurebase/Intercom/Fin); per-conversation model charges for unresolved/escalated interactions too; 'resolution' definition is ambiguous (Decagon's own glossary admits gray areas can cause billing disputes); no native helpdesk - you still pay for Zendesk/Salesforce/etc.; professional-services/integration fees for custom ERPs.
Variable cost rationale
Cost scales with conversation/resolution volume on top of the fixed annual platform fee; per-conversation exposes you to paying for failed/escalated interactions.
Additional watchouts
Enterprise-only economics (~$50K platform-fee redline + six-figure median ACV); per-conversation billing pays for failures; resolution definition can spark disputes; no public pricing - sales-led, slow procurement.
Overage / add-ons
Usage billed per conversation or per resolution on top of the annual platform fee; volume discounts on larger commitments; per-conversation means paying even for interactions that escalate to a human.
Sales call required
Yes, required for paid access
Free / trial
n/p
Lowest paid plan
n/p
Commercial notes
Founded 2023 (Jesse Zhang/Ashwin Sreenivas); ~$481M raised; $4.5B valuation (Jan 2026 Series D, $250M, Coatue/Index - tripled from $1.5B in ~6mo; employee tender at same valuation Mar 2026); ~$35M ARR; 100+ enterprise customers (Duolingo, Chime, Hertz, Affirm, Dropbox, Notion, Rippling, Oura, Classpass); competes with Sierra, Ada, Intercom Fin, Salesforce Agentforce, Zendesk AI
Key ambiguities
No official figures. Third-party (NOT official): ~$50K annual platform fee; per-conversation ~$0.99 and per-resolution ~$0.50 (a negotiated enterprise rate); Vendr median annual contract ~$386,120 (range ~$95,000-$590,000+); ~$50K redline (below = not a fit).
Cancellation / refund
No free trial, no self-serve signup; annual (sometimes multi-year) contracts; white-glove implementation included in contract price.
Support SLA / resale
White-glove onboarding (dedicated Agent PMs + Forward-Deployed Engineers); integrates with Zendesk/Salesforce/Intercom/Kustomer, Amazon Connect/RingCentral/SIP (voice), Shopify/Stripe; Decagon University for existing customers
Missing data
No official figures (the pricing model is public: a ~$50K annual platform fee + usage via per-conversation ~$0.99 or per-resolution; all dollar figures are third-party). Vendr median annual contract ~$386,120 (range ~$95,000-$590,000+); ~$50K redline below which you're not a fit. 'Resolution' definition can be ambiguous (billing-dispute risk).
Related vendors
- 5.Y — Singapore early-stage platform whose GLUCOSE product deploys…
- Ada — Agentic customer experience platform whose AI agents autonomously…
- Aisera — Enterprise agentic AI platform that orchestrates specialized agents…
- ASAPP — Generative AI for enterprise contact centers that plugs into…
- Assembled — AI customer support orchestration platform that unifies autonomous…
- BIK — Agentic AI CRM for eCommerce that consolidates marketing and…
Alternatives to Decagon
The closest documented capability profiles to Decagon among customer support agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- ASAPP11.5 / 14Adds documented Prebuilt Agents, Templates & Packs
- Sierra11.5 / 14Adds documented Prebuilt Agents, Templates & PacksDecagon vs Sierra →
- Ada11.0 / 14Adds documented Prebuilt Agents, Templates & PacksDecagon vs Ada →
- Lorikeet11.0 / 14Adds documented Prebuilt Agents, Templates & PacksDecagon vs Lorikeet →
- Nurix AI9.0 / 14Adds documented Prebuilt Agents, Templates & Packs
- Cognigy11.5 / 14Adds documented Prebuilt Agents, Templates & Packs
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded