Hana Prochazkova
Designer · Marketing
HP
ROI score
62Watch
Efficiency is flat or down vs last month.
Recommendation
Review what changed vs last month (new model, new task type).
AI spend
€272
13% vs prior period
Yield rate
73%
Team average 70%
Cost per accepted unit
€7.59
Team average €7.09
Leverage vs last month
0.88
49 runs · 36 accepted · 4.2M tokens
Weekly AI spend
Human loops
Accepted77%
Rejected 23% · team 83% accepted
Guidance levelMedium
Runs without result5%
Waste€106
ROI score breakdown
Yield 30%
73
Cost efficiency 25%
44
Trend 20%
41
Model fit 10%
84
Capability use 15%
81
Each component is scored 0–100; 50 means on par with the team median or flat vs last month. Good ≥ 70, Watch 40–69, Bad < 40.
Model mix: spend share vs token share
Premium reasoning model39% spend · 14% tokens
Standard model46% spend · 37% tokens
Fast small model15% spend · 49% tokens
Top skills used
- Prompt templates
- Drafting
- Spreadsheet automation
Agents and MCP usage
- Ticket triage agent156 runs
- Calendar MCP92 runs