What you get back, what the machine handles, and why you can stop worrying.
This dashboard tracks the performance of your WorkOS toolchain across three tiers.
<aside>
12.4 hrs reclaimed · 3 min meeting to action · 97% agent success · €16.80 total cost (€0.23/action)
</aside>
<aside>
Manual setup: Replace this callout with a /linked view of database → KPI Snapshots → create a new Board view named "This week" grouped by nothing, filtered to Period contains this week, showing only: Hours reclaimed, Meeting → Action (min), Agent success rate (%), Cost (€). Set layout to List for a compact single-row summary.
</aside>
If a machine can do it, a person should not have to.
The headline numbers. They answer: what did I get back?
| KPI | What it measures | Data source |
|---|---|---|
| Hours reclaimed / week | Total time saved by automated workflows vs. manual baseline | n8n execution logs x estimated manual duration per task type |
| Meeting to Action latency | Median minutes from meeting end to tasks appearing in Notion | Meeting Notes Date to linked Tasks Created timestamp |
| Auto-triage rate | Percentage of tasks created and triaged without manual input | Tasks where Source = Meeting, Automation, or Slack |
If something interrupts your thinking, it is wrong.
The system health numbers. They answer: is the machine actually doing the work?
| KPI | What it measures | Data source |
|---|---|---|
| Decision capture rate | Percentage of meetings producing a logged decision within 24h | Meeting Notes count vs. Decision Log entries |
| Handoff automation | Percentage of task transitions done by agents vs. manual | Audit Log: auto status changes vs. manual |
| Knowledge freshness | Percentage of KB articles reviewed in last 90 days | Knowledge Auditor agent runs |
| Doc coverage | Percentage of active projects with linked KB articles | Projects to Knowledge Base relation completeness |
Proactive beats reactive.
The trust numbers. They answer: can I stop worrying about this?