ALHENA RESEARCH LAB
Head-to-head tests of ecommerce AI agents
Each study runs the same shopping and support conversations in real storefront chat widgets and scores resolution, answer quality and speed. Summaries are public; the conversations open with a verified work email.
This is a separately versioned Alhena Research Lab method. It is not the Gorgias leaderboard or the three-storefront quality pilot.
Read the versioned methodology · Browse current research results
We ran the same scripted shopping and support conversations with each AI agent on its own live storefronts, 10 in total, then scored every answer for resolution, quality and speed. A study of shopping and support on ten storefront deployments, with conversation evidence, public merchant policies, fixed quality criteria and separate response timing.
policy-resolution-v1 · Published 2026-09-22
We ran the same scripted shopping and support conversations with each AI agent on its own live storefronts, 10 in total, then scored every answer for resolution, quality and speed. Comparison assembled from published evaluations; no new conversations were run for this comparison.
policy-resolution-v1 · Published 2026-09-23
We ran the same scripted shopping and support conversations with each AI agent on its own live storefronts, 10 in total, then scored every answer for resolution, quality and speed. Comparison assembled from published evaluations; no new conversations were run for this comparison.
policy-resolution-v1 · Published 2026-09-23
We ran the same scripted shopping and support conversations with Sierra's AI agent on 5 live storefronts, then scored every answer for resolution, quality and speed. Live storefront evaluation of policy-compliant resolution, answer quality and response speed.
policy-resolution-v1 · Published 2026-09-23