Voice AI Benchmark: Retell Ranks #1 for Medicare Workflow Accuracy

Cekura, an independent voice-agent evaluation company, tested a byte-identical Medicare third-party marketing organization agent across six platforms in 414 calls, measuring compliance-sensitive workflow performance in a regulated insurance setting. Retell ranked first with 95.7% workflow accuracy, passing 22 of 23 evaluators, and retained the same score under strict end-to-end grading that also counted runtime failures; ElevenLabs followed at 91.3%, while the remaining platforms scored from 65.2% to 82.6% on workflow accuracy. The benchmark evaluated required call openings and permissions, customer needs assessment, lead qualification, and consumer-protection safeguards, with each evaluation requiring three successful runs to count. It highlights that platform reliability can affect whether an agent delivers disclosures, obtains consent, avoids unauthorized personalized advice, correctly uses tools, and transfers callers to licensed representatives. Retell attributes its result to workflow execution, tool calling, knowledge-base responses, warm transfers, and post-call quality monitoring, while acknowledging that the findings reflect one independently run workflow and that results can vary with agent design and maintenance.

No tracked trend matches for this post yet.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.