ReceptionMatched evaluation cohorts
Compare the same suite, evaluator, capability profile, transport, language, caller policy and observation mode. An SMS receipt test and a transfer-media test belong to different cohorts.
RESULTS / RESEARCH PREVIEW
Evidence before
a place on the board.
A leaderboard starts with comparable trials and inspectable outcomes. We’re building the first live evaluation path. No live system results have been published.
01 / PUBLISHED RESULTS
The first results
will arrive with evidence.
NO PUBLISHED RESULTS YET
There are currently no ranked systems or published live scores. Offline examples are simulations and do not measure Receptionaly or another receptionist.
The first booking-link SMS pilot will be a narrow, unranked research result. It will not establish full receptionist quality or independent certification.
Prepare an adapter or propose a case ↗02 / WHAT MAKES A COMPARISON
Same task.
Same conditions.
Every attempt accounted for
Publish sample counts, failures, unsupported tasks, infrastructure errors, completion and uncertainty. A single successful call is a smoke test, not an aggregate rank.
Evidence and independence
Distinguish self-reported, benchmark-operated and independently audited evidence. Receptionaly is the initial reference system; its developer’s results are labelled internal research with the relationship disclosed.
Review before publication
Entries require publication permission, configuration and evidence digests, review provenance, scope and cost limitations. Raw phone numbers, private recordings and credentials stay private.
Numerical ranking opens after reviewed cohorts, repeated trials, completion gates and uncertainty reporting are implemented. Independent certification additionally requires independent operation and hidden-test custody.
Read the publication rules ↗