Agentic Index
careCycle vs Delfa (2026)
Both vendors last verified
careCycle and Delfa both run voice and messaging agents in regulated healthcare markets, for different buyers, and neither has public rates. careCycle sells AI employees and a softphone to Medicare and ACA insurance agencies, whose voice agents prequalify leads, warm transfer to licensed agents, answer benefit questions from maintained plan data and run retention campaigns, with a public API, charged per active policy plus usage. That verdict is the Agentic Index coverage score, graded from each vendor's own published materials.
Delfa sells agents and a patient relationship system for clinical trial recruitment, whose agents prescreen, take medical histories, schedule and remind patients by voice, chat, SMS or email, and update the site's CTMS and CRM. careCycle adds custom roles that scope records by agent license number and audit logs of who did what across records, campaigns and calls. Delfa tests its agents before release on synthetic patient profiles and cases written by people. Choose careCycle for an insurance agency's sales and retention calls. Choose Delfa for a research site recruiting trial patients.
On the Agentic Index healthcare agent ranking, neither careCycle nor Delfa clears the bar, which asks for all five chart loop capabilities documented in full. careCycle does not document human oversight and guardrails in full; Delfa documents one of the five in full. 4 of the 69 vendors in the lane clear it. See the healthcare agent ranking
This comparison is published by Agentic Index, an independent agentic AI vendor research platform. careCycle and Delfa are each graded against the same 14 capability Agentic Index taxonomy, from the vendor's own public materials under the Agentic Index verification standard, alongside 1,084 researched vendors. No vendor pays for placement and no vendor has reviewed this page. How this evidence is graded
Choose careCycle if
- You run a Medicare or ACA agency and want leads prequalified and warm transferred.
- Benefit questions should be answered from maintained plan data.
- Your developers need a public API.
Choose Delfa if
- You run a clinical research site or site network recruiting trial patients.
- Prescreening and medical histories should flow into your CTMS and CRM.
- Agents should be tested on synthetic patients and hard cases before release.
| Feature | c careCycle |
D Delfa |
|---|---|---|
| Action & orchestration | ||
|
Integrations & Tool Calling Ability to connect agents to real systems through native integrations, OAuth-authenticated actions, custom tools, APIs, webhooks, or MCP-compatible tools. |
||
|
careCycleIntegrations & Tool Calling The AI transfers calls, books appointments and logs inquiries per scenario, and careCycle connects to MedicarePRO, GoGuruX, EnrollHere, Salesforce and HealthSherpa, syncs customers and policies, and pushes events out by webhook. SourcecareCycle, docs.carecycle.ai Assistant behaviors and carecycle.ai homepageread 2026-09-28 |
||
|
DelfaIntegrations & Tool Calling Agents update the site's CTMS and CRM the way a member of staff would, with integrations named for Realtime Recruitment, Advarra, Zoho, CRIO and HubSpot. A unified inbox gathers participant conversations from every messaging channel in one place. SourceDelfa, delfa.ai/product and delfa.airead 2026-10-06 |
||
|
Workflow Orchestration Ability to sequence, branch, retry, route, and combine deterministic workflow nodes with autonomous agent steps. |
||
|
careCycleWorkflow Orchestration Pulse workflows chain entry, call, SMS, wait, loop and exit nodes with routing on how each call ended, automations chain campaigns into a member journey, and AI employees work beside licensed agents with warm transfers between them. SourcecareCycle, docs.carecycle.ai Pulse nodes and configurationread 2026-09-28 |
||
|
DelfaWorkflow Orchestration Agents prescreen, take medical histories, schedule and remind patients, handling the repetitive calls, workflows and data entry, within Agent Operating Procedures that set guardrails for every workflow and reduce the risk of unsupported outputs. Delfa's team builds each agent. Delfa names no workflow builder, branching or multi agent execution that the site configures. SourceDelfa, delfa.ai/product and delfa.airead 2026-10-06 |
||
|
Triggers & Channel Coverage How agents wake up and where they work: schedules, webhooks, message events, CRM events, inbox events, chat, email, voice, and collaboration tools. |
||
|
careCycleTriggers & Channel Coverage Campaign triggers and automations enroll customers into a Pulse workflow on rules, callback pulses call members back on their own, and a receptionist number answers inbound calls and texts, so work starts without an agent placing it. SourcecareCycle, docs.carecycle.ai Pulse nodes and documentation indexread 2026-09-28 |
||
|
DelfaTriggers & Channel Coverage Agents reach patients by voice, chat, SMS or email in several languages, capturing leads with online chat, prescreening with a call and reminding over SMS. Delfa does not say what starts outreach, such as a new lead, a referral or a schedule. SourceDelfa, delfa.ai and /customers/migraineread 2026-10-06 |
||
| Knowledge & context | ||
|
Knowledge Grounding & RAG Ability to ground agent behavior in company data through document ingestion, retrieval, external knowledge APIs, semantic search, or RAG layers. |
||
|
careCycleKnowledge Grounding & RAG Cora queries a maintained Medicare Advantage plan benefit dataset, joined to the agency's imported book of business by plan code, to verify copays, deductibles, drug coverage and providers before answering. That dataset is careCycle's own, and there is no stated way to add the agency's own documents. SourcecareCycle, docs.carecycle.ai Plan dataread 2026-09-28 |
||
|
DelfaKnowledge Grounding & RAG Agents prescreen patients for each trial and take full medical histories, and Delfa's doctors and engineers build and test them. Delfa does not say how protocol and eligibility knowledge reaches the agent, whether through a maintained store or setup for each trial. Sites can also query their own participant database to find matches for a study, but Delfa does not say the agents draw on it. SourceDelfa, delfa.ai and /customers/migraineread 2026-10-06 |
||
|
Memory & State Persistence Ability to persist context across a run, conversation, workflow, user, team, or longer-term memory layer. |
||
|
careCycleMemory & State Persistence The AI keeps no separate memory with its own scope and lifetime. Member history lives in careCycle's CRM records (customers, policies, calls), which are the agency's business record. SourcecareCycle, docs.carecycle.ai documentationread 2026-09-28 |
||
|
DelfaMemory & State Persistence Patient records live in Delfa's patient relationship management system, its unified inbox and the site's CTMS, which hold the business record. Delfa names no memory that the agent keeps for itself. SourceDelfa, delfa.ai/product and delfa.airead 2026-10-06 |
||
| Control & trust | ||
|
Human Oversight & Guardrails Approval steps, consent checkpoints, escalation rules, structured guardrails, policy constraints, and pause/resume controls. |
||
|
careCycleHuman Oversight & Guardrails The agency decides scenario by scenario whether the AI transfers, books or logs an inquiry, during business hours and after, and warm transfers hand high intent callers to licensed agents. There is no human verification or approval before the AI acts or sends a message. SourcecareCycle, docs.carecycle.ai Assistant behaviorsread 2026-09-28 |
||
|
DelfaHuman Oversight & Guardrails Agent Operating Procedures set guardrails for every workflow, clinical teams review interactions, and the recruitment team stays in control of the patient experience. Participants hear at the start that the call is automated and always have a route to a person, and a clinical question, a possible adverse event or a sign of distress hands the conversation to site staff. Delfa names no step where a person approves an agent action before it commits. SourceDelfa, delfa.ai/product and /news/how-to-validate-ai-in-clinical-researchread 2026-10-06 |
||
|
Security, Identity & Governance RBAC, SSO, auditability, encryption, least-privilege tool access, compliance posture, and data handling policy. |
||
|
careCycleSecurity, Identity & Governance Custom roles set what each teammate can do and data filters scope which records they see by NPN, sessions cap at 12 hours and audit logs record who, what, when and from where. careCycle reports SOC 2 Type 2 certification and HIPAA compliance. SourcecareCycle, docs.carecycle.ai Roles and permissions and SOC 2 and HIPAA pagesread 2026-09-28 |
||
|
DelfaSecurity, Identity & Governance Badges show AICPA certification, GDPR compliance and HIPAA compliance, but Delfa names no SOC 2 report or its type. It names no roles, SSO or other controls over who at the site can direct the agents. SourceDelfa, delfa.ai/product and delfa.airead 2026-10-06 |
||
|
Observability & Auditability Traces, logs, execution histories, metrics, audit events, and debugging detail for production agent behavior. |
||
|
careCycleObservability & Auditability Audit logs track who did what, when and from where across customer records, campaigns, calls and admin actions, calls can be replayed with transcripts and dispositions, appointments carry an audit history through the API, and webhook deliveries are logged. SourcecareCycle, docs.carecycle.ai SOC 2 and HIPAA page and API reference indexread 2026-09-28 |
||
|
DelfaObservability & Auditability Every conversation is logged, and Delfa centralizes every call, note and update in a view the whole team shares in real time. After release Delfa monitors calls and outcomes, keeps interactions under QA and analyzes agent executions against defined quality and safety checks. Delfa does not say whether the site can open a record of the actions the agent took on each call. SourceDelfa, delfa.ai/product and /news/how-to-validate-ai-in-clinical-researchread 2026-10-06 |
||
|
Deployment & Data Residency Deployment modes and options, including SaaS, dedicated cloud, VPC, on-prem, hybrid, local runtime, and self-hosting. |
||
|
careCycleDeployment & Data Residency There is no region choice or deployment in the agency's own environment. careCycle says its AI runs on premises with nothing sent to OpenAI, Anthropic or Google, which describes where careCycle runs its own models. SourcecareCycle, carecycle.ai homepage and docs SOC 2 and HIPAA pageread 2026-09-28 |
||
|
DelfaDeployment & Data Residency Delfa runs as a hosted service. No hosting region is published, and Delfa names no region choice, private deployment or customer environment option. SourceDelfa, delfa.ai/product and delfa.airead 2026-10-06 |
||
| Solution readiness | ||
|
Prebuilt Agents, Templates & Packs Ready-made workflows, packaged employees, templates, blueprints, industry solutions, and role-specific agents that reduce time-to-value. |
||
|
careCyclePrebuilt Agents, Templates & Packs Named campaigns do separate jobs within Sales, Retention and Revenue suites, among them inbound and outbound prequalification, a Welcome Call, an Approval Check-In and ACA specific retention campaigns, and cross sell and ancillary revenue campaigns, with campaign templates available through the API. SourcecareCycle, carecycle.ai homepage and docs.carecycle.ai guides indexread 2026-09-28 |
||
|
DelfaPrebuilt Agents, Templates & Packs Agents are custom built for each site and trial by Delfa's doctors and engineers, not adopted from a catalog. Prescreening, histories and scheduling are jobs of one agent. At Elevate Clinical Research the agent was live within a week, prescreened 106 participants for a complex migraine study in its first week on calls that regularly ran 20 minutes, supported Spanish speaking patients bilingually, and can now be set up for a new study within 24 hours. Customer stories also cover weight loss, weight management and type II diabetes, and cardiovascular studies. SourceDelfa, delfa.ai, /customers and /customers/migraineread 2026-10-06 |
||
| Platform extensibility | ||
|
Model Flexibility & Routing Ability to work across multiple foundation models, route tasks to different models, or let buyers bring their own providers and keys. |
||
|
careCycleModel Flexibility & Routing The customer does not choose a model or provider. careCycle runs its own models on premises and says nothing goes to OpenAI, Anthropic or Google. SourcecareCycle, carecycle.ai homepageread 2026-09-28 |
||
|
DelfaModel Flexibility & Routing No model or provider is published, and Delfa names no customer choice of model. SourceDelfa, delfa.ai/product, delfa.ai and /news/how-to-validate-ai-in-clinical-researchread 2026-10-06 |
||
|
APIs, SDKs & MCP Extensibility Composability layer: stable APIs, SDKs, MCP tool consumption/serving, custom tools, and integration into internal systems. |
||
|
careCycleAPIs, SDKs & MCP Extensibility The careCycle Public API takes an x-api-key header with keys scoped by capability and rate limited per key, and its reference covers customers and policies, campaigns and their behaviors, enrollment, appointments, sources, suppressions and ping post, with webhooks for real time events. SourcecareCycle, docs.carecycle.ai API authentication and API reference indexread 2026-09-28 |
||
|
DelfaAPIs, SDKs & MCP Extensibility No API, SDK or developer documentation for Delfa's platform is published. Delfa connects to a site's own systems itself, and Elevate Clinical Research says it was integrated into their systems within a week. SourceDelfa, delfa.ai/product and /customers/migraine; delfa.ai/sitemap.xmlread 2026-10-06 |
||
|
Testing, Debugging & Optimization Testing, debugging, scoring, retries, fallbacks, quality gates, and optimization loops for improving agent workflows before and after deployment. |
||
|
careCycleTesting, Debugging & Optimization Before launch, staff talk to their AI employee in the browser or take a real test call that exercises the full flow, and a preflight check stops on missing variables. The review is by ear, with no scored test cases or pass or fail criteria. SourcecareCycle, docs.carecycle.ai Test your campaignread 2026-09-28 |
||
|
DelfaTesting, Debugging & Optimization Agents are tested before release on synthetic patient profiles and cases written by people, routine and difficult ones such as a participant confused about a medication, a possible adverse event, distress or a noisy line. The tests check that the agent stays within the approved conversation, avoids prohibited actions and hands over to site staff when required. After release Delfa runs continuous evaluations on agent outputs, with a named point of contact for any issue. Delfa's own published customer results include more than 150 prescreens a week at ALSA Research, 36,600 patient interactions and $42,000 in labor savings at a weight management network, and 20 hours a week saved per recruiter at a migraine site. SourceDelfa, delfa.ai/news/how-to-validate-ai-in-clinical-research and /customersread 2026-10-06 |
||
| Specialist automation | ||
|
Browser & Computer Use Browser, desktop, or remote/local computer control for workflows that cannot be handled through stable APIs alone. |
||
|
careCycleBrowser & Computer Use There is no browser, desktop or computer control. AI employees work by phone and SMS and act through the platform and CRM integrations. SourcecareCycle, docs.carecycle.ai guidesread 2026-09-28 |
||
|
DelfaBrowser & Computer Use Agents work by voice, chat, SMS and email and update systems through integrations, working inside the site's CRIO and HubSpot rather than through a separate portal. Delfa names no browser, desktop or computer control. SourceDelfa, delfa.ai/product and /customers/migraineread 2026-10-06 |
||
Pricing snapshot
Sourced from the Index pricing dataset · open each vendor's profile for full detail.
| Pricing | c careCycle |
D Delfa |
|---|---|---|
|
Entry price Lowest public entry point |
Not published. A flat rate per active policy plus usage, quoted in a pricing call. | Contact sales |
|
Pricing confidence How public the numbers are |
Contact only | Contact only |
|
Billing Primary billing axis |
Per active policy on the book, plus usage | Not published |
|
Variable cost Workload / overage exposure |
Medium variable cost | Medium variable cost |
|
Free tier / trial Try before you buy |
No free tierTrial
|
No free tier
|
|
Buying motion Self-serve vs sales call |
Self-serve | Sales call |
|
Pricing verified When this pricing was last checked |
Product updates
Newest first, each linked to its source.
careCycle
-
· Agent capability
In careCycle 2.0 every call opens with a member fact sheet listing policies with their writing agent, appointments with reference codes, open requests and whether anyone acted on them, and summaries of the last eight calls.
Source -
· Agent capability
careCycle expanded its assistant's plan data to every US county across Medicare Advantage and PDP plan types.
Source -
· Agent capability
careCycle introduced Assembly, a new campaign workspace for configuring AI employees.
Source
careCycle charges a flat rate per active policy plus usage, quoted on a pricing call, and Delfa sells through a demo.
Other matchups in voice agents
Not the pairing you were after? These compare a different set of voice agents on the same 14 capabilities.