Back to Report Blocks

CostAnalysis

Cost
Report block

Summarizes score-result cost and call-count distribution across scorecards, scores, or windows.

What It Answers

  • Which scores are expensive?
  • What is average cost per item?
  • Where should we optimize model or prompt cost?

Use When

  • Cost review.
  • Model migration planning.
  • Investigating unusually expensive scoring runs.

Avoid When

  • You need accuracy or alignment quality.
  • You need per-feedback curation.

Run From The CLI

Direct report commands are the simplest path for one-off usage. Saved report configurations use `plexus report config create` and `plexus report run`.

plexus report run --config "Cost Review" days=7
plexus report config create --name "CostAnalysis Example" --file cost-analysis.md
plexus report run --config "CostAnalysis Example"

Minimal Configuration

class: CostAnalysis
scorecard: "Customer Service QA"
days: 7
group_by: score

Live Rendered Example

Rendered with the same dashboard component used by generated reports.

Cost per score result
$0.01420
LLM calls per score result
3.2
Cost per item
$0.04260
LLM calls per item
3.2
Total cost:$34.08000Total LLM calls:7,680
Score results:2,400Items analyzed:800
ScoreAverage costTotal costScore resultsAvg LLM calls
Medication Review: Dosage$0.01500$12.00000800-
Agent Misrepresentation$0.01100$6.60000600-
Cost distribution per score
AgentMisrepresentationMedication Review:Dosage$0.01$0.02$0.03$0.04

How To Interpret It

  • Look for high average cost per item before optimizing total cost.
  • A score with high total cost may simply have high volume.