Lorikeet · Lorikeet: AI Customer Concierge for Fintech, Health and Insurance Support Across Chat, Email, SMS and Voice, With Dreamweaver Proactive Outreach
Lorikeet is an AI customer concierge for regulated businesses such as fintechs, digital health companies and insurers that resolves support issues end to end across chat, email, SMS, WhatsApp and voice, acting in your back-office systems through APIs and MCP servers with payments and identity checks run as fixed code. Its Dreamweaver add-on, launched November 4, 2026, works toward goals like activation and retention by deciding per customer whether to send a message, create a task for your team or do nothing.
Best for
Support and CX teams at fintechs, digital health companies and insurers handling a few thousand or more tickets a month who need an AI agent that takes real account actions under strict controls, holds SOC 2, ISO 27001 and HIPAA attestations, and can also run goal-driven proactive outreach
Not ideal for
Teams that need PCI DSS scope or card payments taken on calls, small businesses that cannot commit to an annual plan of $2,100 a month or more, organizations that require on-premises hosting, or companies wanting a simple FAQ chatbot
Who it's for
CX, support operations and lifecycle teams at fintech, lending, digital health, telehealth and insurance companies, from growth-stage startups to large scale-ups
Lorikeet's pitch is that regulated support needs an agent that can act, not just answer, and its design follows from that: the steps that move money or verify identity are fixed code, guardrails are enforced in code as well as by AI checks, and every conversation is scored. Published pricing and a self-serve trial set it apart from most enterprise rivals, though the $2,100-a-month annual floor is easier to justify at higher ticket volumes. Dreamweaver is the more novel part, moving the same agent from reacting to tickets to working toward retention and activation goals, but it is days old and its headline result comes from a single pilot, so run your own holdout test. Check two things early: whether any flow needs card capture, since Lorikeet is outside PCI scope, and whether you need data outside the US on a Start or Scale plan.
Who should use it
Support and lifecycle leaders at fintechs, lenders, digital health and insurance companies who want one AI agent that resolves account-changing requests under strict controls across chat, email, SMS and voice, and who want to test proactive, goal-driven outreach on the same platform.
Who should skip it
Teams whose support flows require taking card payments, small businesses that cannot commit to an annual plan, organizations that need on-premises hosting or non-US data residency on a mid-tier plan, and anyone who only needs an FAQ chatbot.
Start
$2,100/month (paid annually) plus per-resolution rates
Billed annual
Scale
$5,100/month (paid annually) plus per-resolution rates
Billed annual
Signature
Custom (annual billing)
Billed annual
Free tier limits: No free plan. A 30-day free trial is available with a work email; a card is taken at checkout and nothing is charged until the trial ends. For the Dreamweaver launch, the first 100 sign-ups receive $1,000 of free credits.
Note: Lorikeet charges only for successfully resolved tickets: escalations to a person are not charged, and resolutions you mark as bad are not billed. Voice rates assume a 3-minute average call, with full voice pricing set in an outcomes consultation. Lorikeet describes the bill as the plan price plus metered resolutions, while its worked examples set resolution totals against the plan price, so confirm with Lorikeet whether usage draws down the plan fee. Dreamweaver has no separate line on the pricing page; the published resolution rates cover inbound and outbound chat, email, SMS and voice. All prices in USD.
Modalities
Text, Voice
Fintech account and payment support
The concierge checks live account data and runs payment and identity steps as fixed code, so a refund or repayment change follows the same path every time and escalates with context when policy says so.
Activation outreach for new customers
Dreamweaver watches usage after sign-up and contacts customers who stall by SMS, email or phone at times they tend to reply, within contact-hour and consent limits, with a holdout group showing the effect.
Phone support on existing telephony
Voice runs on local US, UK or Australian numbers through Twilio, Amazon Connect, Five9 and similar systems, with the same workflows and guardrails as chat.
Staged rollout on replayed tickets
Simulations on historical tickets show which topics the agent can handle before go-live, and Coach scores every live conversation afterward.
Lorikeet vs. Decagon
Decagon also resolves support across chat, voice, email and SMS, with behavior written as natural-language Agent Operating Procedures, and is sold through sales without public pricing. Lorikeet publishes plan and per-resolution prices, has a 30-day self-serve trial, targets regulated fintech, health and insurance teams with payments and identity checks run as fixed code, and adds Dreamweaver for proactive outreach.
Lorikeet vs. Sierra
Sierra is a sales-led customer-facing agent platform for large consumer brands with outcome-based pricing and inbound and outbound voice, and it lists PCI DSS Level 1 among its compliance programs, which matters if calls must take card payments. Lorikeet publishes its rates, offers a self-serve trial and is aimed at scale-ups in regulated industries, but does not hold PCI DSS, so card capture stays in your own payment system.
Lorikeet vs. Fini
Fini is also built for fintech and other regulated support, runs on top of existing help desks with deterministic rule workflows, and publishes per-resolution pricing, but is sales-led with no self-serve sign-up and documents no SMS or WhatsApp channel. Lorikeet covers SMS, WhatsApp and outbound voice, has a self-serve trial, and adds goal-driven proactive outreach through Dreamweaver.
Lorikeet vs. Fin
Fin (now part of Salesforce) runs inside Intercom or on top of another help desk, answers from help center content and can take actions through APIs, Data Connectors and MCP, at $0.99 per outcome with no plan fee on standalone deployments and a 14-day trial. Lorikeet charges an annual plan fee from $2,100 a month plus per-resolution rates, is built around regulated fintech, health and insurance workflows with payment and identity steps run as fixed code, and adds Dreamweaver for goal-driven proactive outreach.
What does Lorikeet do?
Lorikeet is an AI customer support platform for complex and regulated businesses such as fintechs, digital health companies and insurers. Its concierge resolves customer issues end to end across chat, email, SMS, WhatsApp and voice: it pulls live context from your systems, follows your policies, takes actions through your APIs and MCP servers, runs sensitive steps such as payments and identity checks as fixed code, and escalates to your team with full context when it should.
What is Lorikeet Dreamweaver?
Dreamweaver is a proactive AI account manager built into Lorikeet's concierge, launched on November 4, 2026. You set a goal such as activation, retention, conversion or lifetime value. Dreamweaver reviews each customer's product usage and past conversations on a schedule and whenever something changes, then sends a message on the customer's preferred channel (voice, SMS or email), creates a task for your team, or does nothing. It checks contact hours, limits and consent before every message, learns from notes your team leaves, and measures impact with a holdout group. Lorikeet reports that in an early pilot with PayJoy, 30% more new customers used their credit line within 14 days than in a control group.
How much does Lorikeet cost?
Lorikeet's Start plan is $2,100 a month and Scale is $5,100 a month, both paid annually, and Signature is custom with a $75K platform fee. On top of the plan, Lorikeet meters each successful resolution: $0.99 (Start) or $0.90 (Scale) for chat, email or SMS, and $1.50 or $1.20 for voice, based on a 3-minute average call. Triage tagging and automated QA cost $0.30 or $0.27 per action. Escalations to a person and resolutions you mark as bad are not charged. There is a 30-day free trial; a card is taken at checkout and nothing is charged until the trial ends.
Is Lorikeet SOC 2 and HIPAA compliant?
Lorikeet states it is SOC 2 Type II audited, ISO 27001:2022 certified, HIPAA compliant with Business Associate Agreements, and GDPR aligned, with reports available under NDA from its Trust Center. It has zero-data-retention agreements with its model vendors and does not train models on customer data. It does not hold PCI DSS, so card capture should stay in your own payment system.
Which help desks and phone systems does Lorikeet work with?
Lorikeet integrates natively with Zendesk, Intercom, Salesforce, HubSpot, Front, Help Scout, Kustomer, Slack and Genesys Cloud, connects other webhook-based help desks through Flexible Ticketing, and offers its own website chat for teams without a ticketing system. For voice it works with Amazon Connect, Twilio, Vonage, Five9, Aircall and Dialpad. Knowledge can come from Confluence, Notion, Google Docs and Guru.
Lorikeet vs Decagon: what is the difference?
Both build AI agents that resolve support across chat, email, SMS and voice. Decagon is sales-led with no public pricing and defines behavior in natural-language Agent Operating Procedures. Lorikeet publishes plan and per-resolution prices, offers a 30-day self-serve trial, focuses on regulated fintech, health and insurance support with payments and identity checks run as deterministic code, and adds Dreamweaver for goal-driven proactive outreach.
Decagon
Large consumer brands automating high-volume, multi-channel customer support with on-brand, measurable AI agents
CustomSierra
Enterprises that want branded, action-taking AI agents as the front door to their customer experience, with deep control over tone and guardrails
Custom
Fini Technologies
Support and operations teams at fintechs, banks, insurers and healthcare companies on Zendesk, Intercom, Salesforce, HubSpot or Gorgias that want an autonomous agent taking real account actions under written rules, with audit trails, compliance attestations and a plan for traffic from customers' personal AI agents
PaidSalesforce (Fin, formerly Intercom)
Support teams on Intercom, Zendesk, Salesforce, HubSpot, or Freshdesk who want to automate resolution of repetitive questions at a transparent $0.99-per-outcome price
PaidAda
Mid-market and enterprise teams that want a vendor-independent AI support platform working across all channels
CustomLorikeet is built for support that involves money, health or policy decisions rather than FAQ lookups. A Lorikeet concierge plugs into the help desk you already run (native integrations cover Zendesk, Intercom, Salesforce, HubSpot, Front, Help Scout, Kustomer, Slack and Genesys Cloud, other webhook-based help desks connect through Flexible Ticketing, and a Lorikeet chat widget works without any ticketing system), learns from your help docs, SOPs and brand guidelines, and reaches your business systems through APIs, webhooks and MCP servers. Behavior is written as natural-language instructions and skills, while sensitive steps such as payments and identity checks run as deterministic code the agent can call but never alter. Before launch, teams replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps, then go live topic by topic; Lorikeet quotes time to first live ticket as little as hours and under 29 days on average, with forward-deployed support on every plan (included on Start and Scale, a paid package on Signature). In production, agent guardrails check every reply before it is sent, message checks screen incoming messages for prompt-injection attempts, a grounding guardrail keeps answers to approved sources, and Coach scores every AI-handled conversation. Escalations land in Lorikeet's own agent workspace (Resolution Loop) or your help desk with full context. Voice answers and places calls on local numbers in the US, UK and Australia, using your existing telephony such as Amazon Connect, Twilio, Five9 or Dialpad. Dreamweaver, launched November 4, 2026 and built into the concierge, adds proactive work: you set a goal such as activation, retention, conversion or lifetime value, and it reviews each customer's product usage and history on a schedule and whenever something changes, then sends a message on the customer's preferred channel (voice, SMS or email), creates a task for your team, or does nothing, checking contact hours, limits and consent before every message and measuring impact against a holdout group. Lorikeet states SOC 2 Type II, ISO 27001:2022, HIPAA with BAAs, and GDPR alignment, with zero-data-retention agreements with its model vendors. Pricing is published: a Start plan at $2,100 a month or Scale at $5,100 a month (both paid annually) plus a per-resolution rate, with a 30-day self-serve free trial. The tradeoffs: Lorikeet does not hold PCI DSS, so card capture has to stay in your own payment flow; Start and Scale keep data in the US; the annual plan fee is a real floor for small teams; and most product documentation sits behind a customer sign-in.
Lorikeet launched Dreamweaver, built into its Concierge agents. A business sets goals such as activation, retention or lifetime value, and Dreamweaver checks each customer on a schedule and when something changes, then sends a message by voice, SMS or email, creates a task for the team, or does nothing. Every message is checked against contact hours, limits and consent, and a holdout group measures impact. It is available through self-serve sign-up, and the first 100 sign-ups receive $1,000 of free credits.
Lorikeet introduced Workforce Manager, which builds support rosters from ticket and call volume, checks each shift against contracted hours, rest periods, channel skills and leave, and flags coverage gaps and hiring needs ahead of time. It is aimed at support teams of 20 or more agents running live chat or phones.
Lorikeet launched B2A (business to agent), which lets a business deploy a concierge endpoint for customers' personal AI agents with a code snippet. Agents reach the same knowledge, workflows and actions as human customers, under the same guardrails, identity verification and rate limits.
Are you the founder? Claim this listing →