CallMiner Scoring Metric Creation

I am working inside CallMiner Eureka and I need a specialist who truly understands how category-level scores are calculated, weighted, and stored. My goal is to take those existing scores and roll them up into a brand-new metric focused on Agent performance. I already know which Agent-performance categories should feed the metric; what I do not have is the technical blueprint for extracting the raw scores, confirming their accuracy, and combining them into a single, repeatable KPI.

Here is what I would like from you:

• Walk me through CallMiner’s scoring logic for the selected Agent-performance categories, pointing out any nuances that can affect weighting or normalisation.
• Provide the exact steps (API calls, data export, or within-platform methods) to pull category scores at scale.
• Design and document the formula for the new composite metric, including any thresholds or scaling so it remains consistent as new calls are analysed.
• Supply a short validation report using a small data sample to prove the metric behaves as intended.

Acceptance criteria
– Metric uses only the agreed Agent-performance categories and updates automatically when new calls are processed.
– Calculation steps are fully documented so I can reproduce them without additional help.
– Results align with spot-checked manual calculations to within ±1%.

If you have deep, hands-on experience with CallMiner’s category analytics and can turn those insights into a practical agent KPI, I’d love to work with you.