Skip to content

Assay vs manual QA sampling

Reviewing a sample of calls by hand compared with checking every conversation automatically against the same scorecard.

About Manual QA sampling

Most teams review a few conversations per agent each month, using a spreadsheet scorecard.

Best for

Very small teams, or deep coaching sessions on a single call.

About Assay

Checks every call, chat and email against your QA scorecard, links each result to the line in the transcript, and sends uncertain ones to a person.

Best for

QA, compliance and team leads who need every conversation checked, with evidence they can show an auditor.

Side by side

QuestionManual QA sampling Assay
Conversations checkedA sample, usually a few per agent each monthEvery call, chat and email
ConsistencyVaries between reviewersThe same rule applied the same way every time
EvidenceReviewer notesThe exact line in the transcript
Reviewer timeSpent listening to routine callsSpent only on the conversations that need a decision
Spotting patternsSlow, from a small sampleAcross every conversation, as they happen

Descriptions of other products are based on how they describe themselves publicly and may change. If something here is out of date, tell us and we’ll correct it.

Questions

Is automated QA as accurate as a human reviewer?
It depends on the rule. Assay reports how often it agrees with your reviewers for each rule, and anything it isn’t sure about goes to a person, so you can see the accuracy rather than take it on trust.
Do we still need QA reviewers?
Yes. Reviewers stop listening to routine calls and spend their time on the conversations that failed a rule or need a judgement call.

Try it on your own calls

Tell us about your team. We’ll reply within two working days to set up a trial with 50 of your transcripts and your current scorecard.

  • No card or contract for the trial
  • Your conversations are never used to train models
  • We go through the results with you