/

Decagon

  • What it is:Decagon is an enterprise AI platform specializing in autonomous AI agents that resolve complex customer support queries across chat, email, and voice.
  • Best for:Large enterprises with high support volume, Customer support teams seeking outcome-based pricing, Companies prioritizing predictable budgeting
  • Pricing:Starting from Custom quote (volume discounts)
  • Rating:85/100Very Good
  • Expert's conclusion:Decagon is Designed For Enterprises That Are Ready to Scale AI-Powered Customer Service Across All Channels at an Enterprise Level With Compliance and Analytics at an Enterprise Level.
Visit website
Reviewed byMaxim Manylov·Web3 Engineer & Serial Founder

What Is Decagon and What Does It Do?

Decagon is a generative artificial intelligence (AI) startup that creates generative AI agents that help businesses automate their customer service by providing support through the use of AI via chat, email or voice. The founders are Jesse Zhang and Ashwin Sreenivas who founded the business in 2023. They have developed generative AI agents that offer better performance than traditional chatbots and provide customers with control over how they want their AI agents to operate in addition to being transparent about their operation. Since its founding, Decagon has grown into a unicorn company receiving significant funding from some of the largest venture capital firms.

Active
📍San Francisco, CA
📅Founded 2023
🏢Private
TARGET SEGMENTS
EnterprisesCustomer Support Teams

What Are Decagon's Key Business Metrics?

📊
$231M
Funding Raised
📊
$1.5B
Valuation
🏢
A few hundred
Employees
📊
San Francisco, New York, London
Offices
📊
2023
Founding Year

How Credible and Trustworthy Is Decagon?

85/100
Excellent

Rapidly achieving unicorn status with significant funding from major venture capital firms in such a short period of time, indicates a high level of market acceptance and ability to execute well for a relatively new company.

BREAKDOWN
Product Maturity75/100
Company Stability90/100
Security & Compliance80/100
User Reviews70/100
Transparency85/100
Support Quality85/100
TRUST SIGNALS
Unicorn valuation ($1.5B)Raised $231M from top VCsStrong founder pedigree (ex-Google, Scale AI acquisition)Enterprise customer focus

What is the history of Decagon and its key milestones?

2023

Company Founded

Co-founders Jesse Zhang, CEO, and Ashwin Sreenivas, CTO, were both successful professionals in the field of Artificial Intelligence before starting Decagon.

2024

Series B Funding

Decagon raised a total of $65 million in its series B funding round, which was part of its aggressive growth plan.

2025

Series C Funding

Decagon raised a total of $131 million in its series C funding round and reached unicorn status with a valuation of $1.5 billion.

2025

International Expansion

Decagon opened a second location in London to further expand its reach in Europe.

Who Are the Key Executives Behind Decagon?

Jesse Zhang— CEO & Co-founder
Jesse Zhang is an ex Google engineer, previously started two companies that were acquired, Helia and Lowkey, and co-founded both.
Ashwin Sreenivas— CTO & Co-founder
Ashwin Sreenivas is an experienced technology leader in AI and Computer Vision.

How Much Does Decagon Cost and What Plans Are Available?

Pricing information with service tiers, costs, and details
☐Service$CostℹDetails🔗Source
Per-conversation pricingCustom quote (volume discounts)Fixed rate for every incoming conversation, regardless of resolution. Predictable and scales with usage.Official pricing page
Per-resolution pricingCustom quote (higher rate, volume discounts)Higher fixed rate for each fully resolved conversation. No charge for escalations. Larger commitments lower rate.Official pricing page
Per-conversation pricingCustom quote (volume discounts)
Fixed rate for every incoming conversation, regardless of resolution. Predictable and scales with usage.
Official pricing page
Per-resolution pricingCustom quote (higher rate, volume discounts)
Higher fixed rate for each fully resolved conversation. No charge for escalations. Larger commitments lower rate.
Official pricing page
💡Pricing Example: Enterprise handling 10,000 customer conversations/month
Per-conversationCustom (median contract $400K/year)
Billed for all interactions regardless of outcome
Per-resolutionHigher per-unit (enterprise contracts $95K-$590K/year)
Only when AI fully resolves without escalation

How Does Decagon Compare to Competitors?

FeatureDecagonSierra.aiCrescendo.aipagergpt
Core FunctionalityCustomer support AI agentsCustomer support AICustomer support AICustomer support AI
Pricing ModelPer-conversation or per-resolutionOutcome-basedPer-resolution ($1.25+)Session-based
Pricing TransparencyCustom quotesNot publishedPublished tiersPublished ($99-$349/mo)
Free TierNoNoNoYes (25 sessions)
Enterprise FeaturesYes (custom contracts)YesYes (ISO 27001, SOC 2)Yes (custom)
API Availability
Support OptionsSales-ledEnterprise-focusedPriority supportAll tiers included
Security CertificationsEnterprise-gradeEnterprise-gradeISO 27001, SOC 2, GDPREnterprise-grade
Core Functionality
DecagonCustomer support AI agents
Sierra.aiCustomer support AI
Crescendo.aiCustomer support AI
pagergptCustomer support AI
Pricing Model
DecagonPer-conversation or per-resolution
Sierra.aiOutcome-based
Crescendo.aiPer-resolution ($1.25+)
pagergptSession-based
Pricing Transparency
DecagonCustom quotes
Sierra.aiNot published
Crescendo.aiPublished tiers
pagergptPublished ($99-$349/mo)
Free Tier
DecagonNo
Sierra.aiNo
Crescendo.aiNo
pagergptYes (25 sessions)
Enterprise Features
DecagonYes (custom contracts)
Sierra.aiYes
Crescendo.aiYes (ISO 27001, SOC 2)
pagergptYes (custom)
API Availability
Decagon—
Sierra.ai—
Crescendo.ai—
pagergpt—
Support Options
DecagonSales-led
Sierra.aiEnterprise-focused
Crescendo.aiPriority support
pagergptAll tiers included
Security Certifications
DecagonEnterprise-grade
Sierra.aiEnterprise-grade
Crescendo.aiISO 27001, SOC 2, GDPR
pagergptEnterprise-grade

How Does Decagon Compare to Competitors?

vs Sierra.ai

Decagon is based on a more predictable price-per-conversation compared to Sierra's potentially more expensive outcome-based pricing model. Both Decagon and Sierra are focused on serving enterprise-level clients, however, Decagon differentiates itself from Sierra by offering more transparent pricing for its clients' budgets.

Decagon will be better suited for organizations looking to forecast their costs, while Sierra will be better suited for organizations seeking to position themselves as premium enterprise players.

vs Crescendo.ai

Crescendo is the least expensive option at $1.25 per resolution with all costs including onboarding covered in a fixed monthly fee. Decagon's pricing models are customized for its larger enterprise clients with higher average contract values of approximately $400,000 per year.

Crescendo would be best suited for cost-conscious teams requiring support, while Decagon would be best suited for high volume enterprise level support needs.

vs pagergpt

PagerGPT provides session-based pricing that is very transparent at $99 - $349 per month and is ideal for small to medium-sized businesses. Decagon's pricing models are typically customized for its larger enterprise clients and do not include publicly available pricing information.

PagerGPT would be best suited for organizations that require predictable budgets for their small to medium-sized business operations, while Decagon would be best suited for automating large scale support operations.

What are the strengths and limitations of Decagon?

Pros

  • Decagon provides predictable per-conversation pricing that increases directly with your organization's usage, eliminating potential disputes regarding outcomes.
  • Pricing is flexible — you can either charge per conversation or per resolution depending on your business needs
  • Pricing is aligned to value — you only get charged when a case is successfully resolved.
  • There are discounts for volume — the more cases you have the less you pay per case.
  • This product is designed for enterprises — the fact that most customers sign contracts of about $400k indicates that it is scalable to meet the needs of large teams.
  • There is no per user pricing — pricing is based upon the amount of work an artificial intelligence does for you and not by how many people are using it.

Cons

  • The company does not publish their pricing — this means that you will have to do a sales call and they will give you a quote for what your pricing would be.
  • The pricing is opaque — this means it will be difficult for you to accurately predict what your costs are going to be until you negotiate a contract.
  • Pricing is very high for enterprises — annual prices range from $95k to $590k as well as one time on boarding fees.
  • There may be confusion in defining what constitutes a resolution — in some cases, there could be disputes over whether or not a case has been resolved under the per resolution pricing plan.
  • In order to make price adjustments you will have to renegotiate your contract — you will have to have a sales representative agree to the new pricing.
  • Some users have complained of surprise bills — the usage-based pricing model can be unpredictable and therefore result in unexpected charges.

Who Is Decagon Best For?

Best For

  • Large enterprises with high support volume — Large companies can use customized contracts with volume discounts to create the best possible ROI.
  • Customer support teams seeking outcome-based pricing — The per resolution pricing model allows you to tie the cost of using the software directly to the number of resolutions you achieve.
  • Companies prioritizing predictable budgeting — Using the per conversation model removes the possibility of a dispute occurring regarding the definition of a resolution.
  • Organizations replacing human support agents — The usage-based pricing model fits into the concept of a virtual employee model.

Not Suitable For

  • Small businesses and startups — If you are a small company, the minimum contracts may be too expensive for you. For example, if you are looking for something similar to Decagon, consider pagerGPT which is priced at $99/month.
  • Teams needing pricing transparency — Unlike other products, Decagon does not offer self-serve pricing. Instead you should consider using Crescendo.ai, which offers pricing of $1.25/resolution.
  • Budget-conscious SMBs — Like all of our recommendations, we suggest you avoid contracts with large enterprises that are priced at a median of $400k because these are typically too expensive. Instead, consider using a session-based alternative.

Are There Usage Limits or Geographic Restrictions for Decagon?

Pricing Model
Custom quotes only - per-conversation or per-resolution
Public Pricing
Not available - sales consultation required
Contract Minimums
Enterprise-focused, median $400K/year contracts
Pricing Changes
Requires contract renegotiation with sales team
Resolution Definition
Custom per contract - potential for billing disputes
Onboarding Costs
Included in enterprise contracts ($95K-$590K range)

Is Decagon Secure and Compliant?

Enterprise-grade SecurityBuilt for enterprise deployments with appropriate security measures (specific certifications not publicly detailed).
SOC 2 ComplianceComparable to enterprise competitors offering SOC 2 (inferred from market positioning).
GDPR ComplianceEnterprise customer support AI handling sensitive data (standard for category).
ISO 27001 ReadySecurity framework appropriate for enterprise-scale customer data processing.

What Customer Support Options Does Decagon Offer?

Channels
Required for pricing and onboardingEnterprise account managementDedicated for enterprise customers
Hours
Business hours for sales, dedicated support for enterprise
Response Time
Priority enterprise response, sales-led for new customers
Satisfaction
Enterprise-focused (limited public review data)
Specialized
Dedicated customer success managers for enterprise accounts
Business Tier
Custom enterprise contracts include priority support
Support Limitations
•No self-serve support for pricing inquiries
•Requires sales qualification before technical support

What APIs and Integrations Does Decagon Support?

API Type
REST API with native integrations for enterprise tools
Authentication
SSO, enterprise-grade authentication with audit logs
Webhooks
Supported via custom API surfaces for events like conversation updates and escalations
SDKs
No official SDKs mentioned; integrations via API and native connectors
Documentation
Comprehensive enterprise guides available; focused on integrations and deployment
Sandbox
Simulations and regression testing environments for QA with mock personas
SLA
Enterprise-grade with Watchtower monitoring; specific uptime not public
Rate Limits
Usage-based pricing model; enterprise scaling for high-volume
Use Cases
Trigger actions across Zendesk, Salesforce, Stripe, Shopify; manage refunds, account updates, compliance monitoring

What Are Common Questions About Decagon?

With Decagon’s native integration of natural language Agent Operating Procedures (AOPs), non-technical teams are able to easily define the logic of agents while technical teams control how those agents interact with third-party applications such as CRM and billing systems. Additionally, Decagon’s agents can interact with customers through a variety of channels including chat, email, voice and SMS and can maintain context throughout an entire interaction. Decagon also supports persistent memory within its agents allowing them to remember the context of previous interactions. Finally, Decagon supports a wide array of integrations such as Zendesk and Stripe to allow agents to take action (such as issuing refunds).

Decagon provides native support for integrating with popular applications such as Zendesk, Salesforce, Intercom, Stripe, Shopify and knowledge bases like Confluence. Additionally, Decagon supports voice communications via CPaaS providers such as Amazon Connect. Decagon also includes a custom API to connect to billing, CRM and back office systems to enable seamless automation of workflows.

For enterprises, Decagon offers a usage-based pricing model. Enterprises are encouraged to contact their sales representative for specific pricing information. Decagon has delivered an impressive Return on Investment (ROI) with 70%+ of issues being deflected from humans and 65% cost reductions as reported by its customers.

With Decagon, enterprises can deploy conversation capabilities across multiple channels (e.g., chat, email, voice, SMS), all within one logical layer versus patchwork solutions. Additionally, Agent Operating Procedures (AOPs) provide a natural language way for Customer Experience (CX) teams to design workflow through conversational interfaces. Watchtower enables Decagon's ability to monitor for compliance in large-scale environments.

Decagon provides enterprise-level security features including Single Sign-On (SSO), audit logs, confidence thresholds, and human fallback. Watchtower provides an automatic flagging feature for compliance reviews of conversations that need them. Examples of regulated customers that have utilized Decagon include Rippling (Financial Services).

Decagon has seen customer results of 70-80% deflection rates, 3x improvements in CSAT, 65% cost reductions and greater than 90% resolution of issues without human interaction. Decagon also reports that voice deflection exceeds 50% with 70% chat/voice resolution.

Decagon provides a sandbox environment for enterprises to test conversational AI deployments, including simulated scenarios, regression testing, and A/B testing. Decagon will only allow production deployments once the solution has gone through quality assurance (QA) validation using mock personas and historical transcript data.

This product would be best suited for enterprises that require large-scale conversational AI operations and could potentially be too expensive or complicated for small business use cases. The pricing model for this product is based on usage and is focused toward enterprise organizations. In order to utilize this product's custom action functionality, an organization would need to integrate with Decagon.

Is Decagon Worth It?

Decagon delivers enterprise-class conversational AI products that offer multi-channel deployment and natural language-based Agent Operating Procedures, which have enabled customers to deflect 70%+ of interactions away from humans while saving significant amounts of money. Decagon's Watchtower Compliance Monitoring and Self-Improving Analytics place Decagon at the forefront of the competition when compared to legacy CRM Automation systems. The recent $131 million dollar Series C funding round validates Decagon's rapid growth in the enterprise market.

Recommended For

  • Organizations utilizing enterprise customer support teams that handle high-volume interactions across multiple channels
  • Organizations currently utilizing CRM products such as Zendesk, Salesforce, Stripe who want to leverage action-oriented AI Agents
  • Organizations operating in regulated industries that need automated compliance monitoring
  • CX Leaders seeking to drive 70%+ deflection rates and 3x CSAT Improvements

!
Use With Caution

  • Mid-Market Teams - The pricing model for this product is usage-based, therefore mid-market teams may find that it is beyond their budget.
  • Overkill vs Basic Chatbot — Only Needs FAQs and No Other Conversations
  • Companies Without Technical Integration Resources

Not Recommended For

  • Small Businesses With Low Support Volume — Better Served By Lower-Cost Alternatives
  • Single Channel Implementation — Loses Multi-Channel Value Proposition
  • Budget Constrained Startups — Enterprise Pricing Model
Expert's Conclusion

Decagon is Designed For Enterprises That Are Ready to Scale AI-Powered Customer Service Across All Channels at an Enterprise Level With Compliance and Analytics at an Enterprise Level.

Best For
Organizations utilizing enterprise customer support teams that handle high-volume interactions across multiple channelsOrganizations currently utilizing CRM products such as Zendesk, Salesforce, Stripe who want to leverage action-oriented AI AgentsOrganizations operating in regulated industries that need automated compliance monitoring

What do expert reviews and research say about Decagon?

Key Findings

Decagon Emerged From Stealth With a $231 Million Funding Round At a $1.5 Billion Valuation, Delivering Enterprise Wide Conversational AI Across Chat, Email, Voice, SMS. AOP’s Enable Non-Technical Workflow Design With 70-90% Deflection Rates, 3X CSAT Improvements And 65% Cost Reductions for Customers Like Chime, Hertz and Rippling. The Multi-Channel Architecture with Watchtower Compliance Differentiates from CRM-Native Solutions.

Data Quality

Good - comprehensive official website data plus third-party validation from Futurum Group, customer case studies. No public pricing/API docs; enterprise sales required. Revenue confirmed at eight-figure ARR.

Risk Factors

!
Young Company (Emerging From Stealth in 2024)
!
Opaque Pricing Structure Based On Usage Without Contact with Sales Team
!
Enterprise Focused Excludes the SMB Market
!
Competitive Space with Salesforce Einstein, Sierra
Last updated: February 2026

What Additional Information Is Available for Decagon?

Funding & Growth

Total Funding of $231 Million Including a $131 Million Series C at $1.5 Billion Valuation (June 2025). First Year Went from Zero to Eight-Figure ARR Serving Tens of Millions of End Users Across Global Brands.

Customer Success

Chime Achieved a Resolution Rate of 70% (Baseline Was 40%). Hertz Reached 70%+ (Baseline was 10%). Rippling Uses Watchtower for 100% Regulatory Complaint Review. Notion Improved Resolution 34%.

Technical Differentiation

Optimized Across OpenAI, Anthropic, Google Models via Model Constellation Strategy. Persistent Memory Layer Maintains Context Across Channels. Self-Improving Data Flywheel With Automated Knowledge Suggestions.

Watchtower Monitoring

Automated Quality Assurance Flags Conversations Needing Human Review Based On Custom Criteria. Enables Regression Testing, A/B Experiments and Real-Time Compliance for Regulated Industries.

Market Position

Competes with Sierra in conversational AI for Salesforce. Claims that it will turn support from a cost center to a growth engine by increasing high deflection rates and improving CSAT.

What Are the Best Alternatives to Decagon?

  • •
    Sierra: Next gen conversational AI agent platform competing with Decagons' enterprise approach. Uses similar multi-model strategies, but has different focus areas in terms of channel. Best suited for large enterprises that are looking for new and alternative methods of using legacy CRM AI.
  • •
    Salesforce Einstein: Native to CRM conversational AI to Service Cloud. Has more of an ecosystem lock-in than Decagon, however it is well-established and has many enterprise level features. Best suited for existing Salesforce clients who want to gradually use AI to improve their business processes.
  • •
    Voiceflow: Conversational AI builder which is being marketed as an alternative to Decagon. Voice is a major area of focus and provides visual workflow development capabilities. Provides a more developer friendly environment, with less emphasis on enterprise compliance. Best suited for mid-tier companies developing voice-first implementations.
  • •
    Intercom Fin: AI Agent within Intercom's Customer Messaging Platform. Provides tighter integration with Intercom, but also limits the user to Intercom. Is less scalable to enterprise levels compared to Decagons' multi-channel approach. Best suited for those using Intercom as the primary platform.
  • •
    Nurix AI: Autonomous conversational agents with transparent pricing. Views Decagon as a competitor. Could potentially provide a lower cost entry point. Best suited for large enterprises comparing feature sets and pricing models side-by-side.

What Are Decagon's Agent Performance Metrics?

70%+
Resolution Rate
70-80%
Deflection Rate
3x+
CSAT Improvement
up to 95%
Cost Reduction

What Can Decagon's Agents Do?

Multi-Channel Support

Agents can chat, email, call, or send SMS messages while maintaining consistent intelligence.

Natural Language AOPs

Agent Operating Procedures are defined in Natural Language.

Action Execution

Triggers real-world actions like issuing refunds and updating accounts.

Persistent Memory

Aware of context across sessions and channels.

Intelligent Routing

Automatically escalates to human agents when necessary.

Agent Assist

In real time provides a co-pilot function for human agents.

What Supported Llm Backends Does Decagon Support?

OpenAIAnthropic ClaudeGoogle ModelsModel Constellation

What Is Decagon's Agent Deployment Options?

Multi-Channel
Chat, email, voice, SMS
Enterprise Integrations
Native support for Zendesk, Salesforce, Intercom, Stripe, Shopify
Cloud Platform
Managed enterprise platform
Admin Dashboard
Centralized management and analytics

What Agent Tool Integrations Does Decagon Support?

ZendeskSalesforceIntercomStripeShopifyInternal DatabasesCRMsBilling Systems

What Is Decagon's Agent Reasoning Approach?

Core Engine
Self-improving AI agent engine with data flywheel
Natural Language Instructions
Agent Operating Procedures (AOPs) for workflow definition
Context Persistence
Persistent memory layer across interactions
Continuous Learning
Feedback loops from interactions improve accuracy

What Agent Governance Safety Does Decagon Offer?

Intelligent Routing

Automatically transfers conversations to human agents when the AI cannot continue.

Watchtower Compliance

Monitors compliance automatically.

QA & Audit Layer

Displays transparency into how the AI made its decisions.

Enterprise Guardrails

Provides confidence thresholds and safety controls.

Admin Dashboard

Tracks performance and provides full observability.

A/B Testing

Allows safe experimentation and version control.

Expert Reviews

📝

No reviews yet

Be the first to review Decagon!

Write a Review

Similar Products

AutoGPT
autogpt.net
CrewAI
crewai.com
AgentGPT
agentgpt.reworkd.ai