Jev AI use cases
Classification & scoring
Sort and score business text, such as sales leads and call transcripts, against criteria you define.
Classification & scoring
Sort and score business text, such as sales leads and call transcripts, against criteria you define.
Lead Qualification
Compare a sales lead with your ideal customer profile to judge fit, buying intent and the next follow-up.
- Lead scoring
- ICP fit
- Buying intent
Sales Call Scoring
Score a call transcript against your own scorecard, item by item, and see which topics were never covered.
- Call scoring
- Scorecards
- Sales coaching
About classification & scoring
Much business text needs the same judgment made many times: is this lead a fit, did this call cover what it should. Classification puts text into categories you define; scoring rates it against levels you describe.
Jev returns both as typed answers with probabilities and confidence, so results can be sorted, averaged and thresholded in a CRM or dashboard. Categories and score levels are written in plain language, so you change the criteria by editing text.
Which page fits your task
- Lead QualificationYou have inbound or enriched leads and want fit, buying intent and a recommended follow-up from your ideal customer profile.
- Sales Call ScoringYou have call transcripts and a scorecard, and want item-by-item scores with the topics the rep never covered.
Common mistakes
- Combining several criteria into one score, which hides what to fix.
- Writing score levels as bare numbers instead of describing each level.
- Automating decisions before comparing scores with past outcomes or manager reviews.
Classification & scoring FAQ
Can I use my own categories and score levels?
Yes. Categories are Choice options and levels are Score criteria, both written as text. Edit them in the playground, save the judge and reuse it in Batch or the API.
Can I get one overall score?
Ask each criterion separately, then combine them with weights in your own code. That keeps each part visible and lets you change the weights without asking again.
Further reading · reviewed 2026-09-23
- Combining independent scores in codeTypeSafe documentation · docs.typesafe.ai/patterns/composite-scoring
- Writing and interpreting Score rubricsTypeSafe documentation · docs.typesafe.ai/primitives/score
One decision model. Your own rules.
Use a Noul for a yes/no probability, a Choice for a fixed set of outcomes, or a Score for an ordered rubric. Ask focused questions about the same input and combine the results in code.