Lorikeet vs Decagon
Lorikeet vs Decagon compared for fintech and regulated support teams: channels, compliance, deployment, implementation, and pricing, plus where Runtime fits behind the ticket.
TL;DR
Both are strong AI support agents. Lorikeet is focused on regulated mid-market companies with published pricing, while Decagon serves large consumer brands with natural-language procedures and a broader certification list. Neither covers the payment ops, finance, and risk work behind the ticket.
| Feature | |||
|---|---|---|---|
| What it is | |||
| Industry focus | |||
| Channels | |||
| How agents are configured | |||
| QA | |||
| Certifications | |||
| Where it runs | |||
| Implementation | |||
| Pricing |
Lorikeet is the better fit for regulated mid-market and growth companies that want a support concierge with published pricing, a free trial, and SMS and WhatsApp coverage. Decagon is stronger for enterprise-scale support teams that want natural-language procedures, Git-backed engineering workflows, and PCI on the certification list. Runtime is the AI agent harness for payment and fintech teams, and it works behind either one on the payment investigation a ticket starts.
Fintech and other regulated companies often shortlist both, because each resolves conversations end to end, takes actions on the customer's account, and puts compliance front and center. They differ in who they are built for, how you configure them, and how you buy them.
Lorikeet vs Decagon at a glance
Lorikeet is an AI customer concierge for complex support. Its Concierge agent works across chat, email, voice, SMS, and WhatsApp, processes refunds and account updates through your APIs, and escalates when judgment is needed. Coach scores every conversation, human or AI. Lorikeet positions around regulated industries: financial services, healthcare, and insurance. Its customers include Airwallex, Linktree, Flex, and Hnry.
Decagon builds AI agents for customer experience at large consumer brands. Agent behavior is written as Agent Operating Procedures (AOPs) in natural language, deployed across voice, chat, and email. Its tooling includes Duet for building and improving agents, Watchtower for always-on QA, Experiments for A/B testing, and Insights. Customers include Chime, Duolingo, American Airlines, and Deutsche Telekom.
Product scope
Lorikeet. A concierge plus a coach. The Concierge processes refunds, updates accounts, and coordinates multi-step workflows, with "pockets of determinism" for regulated steps. Coach reviews every ticket against your quality standards, and Lorikeet adds simulations and guardrails before go-live. It plugs into Zendesk, Intercom, Salesforce, and HubSpot.
Decagon. A broader CX platform for enterprise scale. AOPs let support teams write workflow logic in plain language, while engineers get Git-backed agent logic and versioned staging and production workspaces. Experiments, Watchtower, and Insights give large teams testing and analytics at enterprise scale.
Fintech and compliance depth
Lorikeet. Regulated industries are the core of its pitch, and fintechs are prominent among its customers and investors. It publishes SOC 2 Type II, ISO 27001:2022, HIPAA, and GDPR, signs BAAs, redacts PII automatically, and has zero-data-retention agreements with model vendors. It does not list PCI DSS.
Decagon. Financial services is one of several industries it serves, with fintech customers such as Chime. It lists SOC 2 Type II, HIPAA, GDPR, PCI, and ISO 27001, with automatic PII redaction, RBAC, and SSO.
Deployment and implementation
Lorikeet. Hosted on Google Cloud with US primary hosting and data residency available in Australia and the EU. A forward-deployed support team configures and tests on your real tickets before customers see it. Lorikeet cites time to first live ticket of as little as hours, averaging under 29 days.
Decagon. A hosted platform that can also be deployed headlessly inside your existing infrastructure. It says most enterprises go live in weeks and includes a forward-deployed team with every engagement.
Quality and testing
Lorikeet. Quality is built around Coach, which reviews every ticket, human or AI, against your standards, plus simulations and Auto QA before changes reach customers. Runtime guardrails detect sensitive topics and escalate to your team, and every interaction, model choice, and action is tracked.
Decagon. Quality tooling is split into several products: Testing and QA before launch, Watchtower for always-on monitoring, Experiments for A/B tests on agent behavior, and Insights for reporting. Large support teams get more knobs; smaller teams may not need them all.
Where Runtime enters the picture
Both products are built for one team: support. They answer the customer and take the actions you expose through APIs. When the answer depends on what happened to the money, the ticket leaves their world. Take an ACH return on a customer's deposit. Support gets the ticket, payment ops checks the return code with the processor, finance sees a recon break in the ledger, and risk or compliance may review the account. Today each team investigates separately in different tools.
Runtime is the AI agent harness for payment and fintech teams. On Runtime that ACH return is one investigation with one record: agents read the ledger, the processor, and the bank file, draft the finding with evidence, and send support an answer to give the customer. You can start with support's payment tickets, then add agents for payment ops, finance, and risk on the same platform. Runtime runs alongside either Lorikeet or Decagon: the support agent talks to the customer, and Runtime does the investigation behind the ticket.
| Lorikeet or Decagon | Runtime | |
|---|---|---|
| Job | Resolve the customer conversation | Investigate the payment behind it |
| Teams | Support | Support, payment ops, finance, risk, compliance |
| Where agents run | Vendor platform | Your cloud or self-hosted |
- No single vendor to depend on. Agents run in your AWS, GCP, or Azure account or fully self-hosted, on the sandbox provider you choose with fallbacks, and on Claude, GPT, Gemini, or open-weight models, with routing across Claude Code, Codex, and OpenCode.
- Guardrails you set. Agents start read-only, a person approves before money moves, access follows roles, credentials are masked, card numbers and SSNs are stripped, and every run is recorded end to end.
- Built for teams. Support, payment ops, finance, risk, and compliance share agents, templates, and one memory, so an edge case one team solves helps the next.
- An engineer, not an account executive. A forward-deployed AI engineer who built payment infrastructure at Hulu's payments team at Disney, Finix, Modern Treasury, or BlackRock builds the first agents with your team.
Create a banking ops agent that handles RFIs, hold-harmless requests, and returns of funds from our partner bank, verifies against our ledger, and drafts replies for my approval.
bank-ops-agent
Tools picked from your prompt
Pricing
Lorikeet publishes pricing. Start is $2,100 per month and Scale is $5,100 per month, billed annually, with per-ticket charges for triage, resolution, and QA on top. Signature is custom. A 30-day free trial is available.
Decagon does not publish prices. It offers per-conversation or per-resolution pricing, quoted through sales, with rates that vary by channel and volume.
Which should you choose
- Choose Lorikeet if you are a regulated mid-market or growth company, want published pricing and a free trial, and value SMS and WhatsApp alongside chat, email, and voice.
- Choose Decagon if you run support at enterprise scale, want natural-language AOPs with Git-backed engineering workflows, and need PCI on the certification list.
- Choose Runtime if your tickets lead into ledger, processor, and risk investigations, other teams need agents too, and data must stay in your cloud with approvals before money moves.
Frequently asked questions
Which is better for fintech support, Lorikeet or Decagon?
Both serve fintech. Lorikeet centers its positioning on regulated industries like financial services, healthcare, and insurance. Decagon serves large brands across many industries, including financial services customers like Chime, and lists PCI among its certifications. The better fit depends on your size, channels, and compliance requirements.
How does Lorikeet pricing compare to Decagon?
Lorikeet publishes plans starting at $2,100 per month billed annually, plus per-ticket charges, with a 30-day free trial. Decagon does not publish prices; it offers per-conversation or per-resolution pricing through sales.
Do Lorikeet and Decagon hold PCI DSS?
Decagon lists PCI alongside SOC 2 Type II, HIPAA, GDPR, and ISO 27001. Lorikeet publishes SOC 2 Type II, ISO 27001:2022, HIPAA, and GDPR and does not list PCI DSS.
Can Lorikeet or Decagon investigate a stuck payment?
Both can take actions through APIs you connect and escalate to people. Investigating across the ledger, processor, bank files, and risk systems is usually done by other teams, which is the work Runtime agents handle in your own cloud.
Where does Runtime fit next to Lorikeet and Decagon?
Behind the ticket. Lorikeet and Decagon talk to the customer. Runtime is the AI agent harness for payment and fintech teams: support, payment ops, finance, risk, and compliance build agents that investigate the payment behind the ticket in your own cloud, with approvals before money moves. It runs alongside either one.