What is work management analytics, and how do you build it in Metabase?
Work management analytics turns tasks, projects, boards, status changes, knowledge, and collaboration events into shared metrics about how work moves through an organization. In Metabase, build it by syncing source tools into a SQL database, modeling clean work-item, content, and activity layers, and shipping dashboards that help teams see risk early.
TL;DR — Use three adjacent models: work items and status history; content items and versions; collaboration events and threads. Current-state data supports completion, overdue, ownership, and freshness snapshots. Change and event history unlock cycle time, trends, and response-time analysis.
What does work management analytics measure?
- Are teams finishing what they committed to?
- Where is work piling up or aging?
- Which projects are at risk before a deadline is missed?
- Is work distributed sanely across owners and teams?
- How long does work take from start to done?
- Is important knowledge current, owned, and complete?
- Which workspaces are active, stale, or slow to respond?
Which tools feed work management analytics?
| Tool | Best for | Getting data into Metabase |
|---|---|---|
| Notion | Docs, databases, lightweight project tracking | API, connector, or official MCP for exploration |
| Asana | Tasks, projects, portfolios, goals | API, Graph Export, connector, or official MCP |
| ClickUp | Lists, spaces, goals, docs, all-in-one work | API, connector, or community MCP |
| Slack | Conversation, incidents, project-channel activity | Web API, exports, connector, or MCP |
| Trello | Boards, cards, checklists, lightweight kanban | REST API, connector, or community MCP |
| monday.com | Boards, items, timelines, cross-functional programs | GraphQL API, connector, or community MCP |
| Linear | Product and engineering issue tracking | API, connector, or Linear MCP |
| Jira | Agile delivery, sprints, issue changelogs | Atlassian APIs, connector, or Rovo MCP |
| Airtable | Record throughput by workflow status; Overdue records by owner and base | Airtable Web API, pipeline, or Airtable MCP Server for exploration |
| Confluence | Knowledge-base freshness by space; Coverage and ownership gaps | Confluence Cloud REST API, pipeline, or Atlassian Rovo MCP Server for exploration |
| Wrike | Task throughput and completion; Project and portfolio health | Wrike API and Datahub, pipeline, or Wrike MCP Server for exploration |
| Miro | Board and workspace activity; Stale boards by owner | Miro REST API, pipeline, or Miro MCP Server for exploration |
| Microsoft Teams | Channel and team activity; Collaboration response time | Microsoft Graph Teams APIs, pipeline, or Work IQ Teams MCP Server for exploration |
| Figma | Design-file activity and freshness; Design-system component coverage | Figma REST API, pipeline, or Figma MCP Server for exploration |
| Teamwork.com | Client project health; Task and milestone completion | Teamwork.com APIs, pipeline, or Teamwork.com MCP Server for exploration |
| Fibery | Entity throughput by workflow; Workspace and database activity | Fibery API, pipeline, or Fibery MCP Server for exploration |
| Coda | Table workflow completion; Doc and page freshness | Coda API, pipeline, or Coda MCP for exploration |
| SharePoint | Knowledge-base freshness by site; Content ownership and coverage | Microsoft Graph SharePoint APIs, pipeline, or Work IQ SharePoint MCP for exploration |
| Dropbox | File and folder activity; Stale content by owner | Dropbox API, pipeline, or Dropbox MCP Server for exploration |
| Box | Content activity and freshness; Ownership and collaboration coverage | Box Platform API, pipeline, or Box MCP Server for exploration |
| Excel | Workbook inventory and freshness; Table and sheet ownership | Microsoft Graph Excel APIs, pipeline, or Excel MCP Server for exploration |
What shared data models should you build?
Build clean models on top of raw source tables so each dashboard uses the same definitions.
work_items— one row per task, card, issue, page, or itemprojects— project, board, database, list, or portfolio metadataowners— people, teams, and assignment historystatus_history— one row per status transition when availabledaily_work_item_snapshots— daily state for WIP and aging trendscontent_items— pages, files, docs, designs, or workbooks with owners and lifecycle timestampscontent_versions— meaningful revisions and review events for freshness trendscollaboration_events— messages, comments, edits, shares, and other qualifying activitycollaboration_threads— thread creation and first-response timestamps for response-time analysis
Which work management metrics matter most?
- Task completion rate — completed work divided by due or committed work.
- Task throughput — completed work per period.
- Overdue rate — open work past due divided by open work.
- Workload balance — distribution of open work across owners or teams.
- Project health score — risk score from overdue, blocked, stale, and incomplete work.
- Cycle time — started to done, requiring a started timestamp or status history.
- Documentation freshness — active content reviewed inside the required window.
- Knowledge-base coverage — required topics with current, owned documentation.
- Workspace activity rate — meaningful activity across eligible boards, spaces, sites, or channels.
- Collaboration response time— time from a request or thread to the first useful human response.
How do you connect tools to Metabase?
- MCP + CLI — use MCP for a scoped live export, then
mb upload csvfor quick analysis. - Warehouse-backed pipeline — sync with APIs, connectors, or exports for durable dashboards.
- Modeled layer — map each source into the relevant work-item, content, and collaboration models while preserving source-specific extension tables.
Which dashboards should you build first?
- Work overview — completed, open, overdue, unassigned, and blocked work.
- Project health — at-risk projects, deadline risk, stale work, and missing owners.
- Workload distribution — WIP by owner, team, project, and status.
- Flow and bottlenecks — cycle time, lead time, time in status, and aging work.
- Knowledge-base health— freshness, coverage, ownership, stale content, and content activity.
Common mistakes
Trying to compute flow metrics from current-state data.→ Cycle time and time in status need a changelog or snapshots.
Treating task count as productivity.→ Task count ignores scope and complexity. Use it for load and flow, not individual ranking.
Letting each team define done differently.→ Normalize statuses into shared groups before reporting completion or throughput.
Ignoring missing due dates and owners.→ Missing metadata is itself a project-health signal; track it explicitly.
Related
Integrations
FAQ
What is the difference between work management analytics and software delivery analytics?
Work management analytics covers broad operational work across tools like Notion, Asana, ClickUp, Slack, Trello, and monday.com. Software delivery analytics is the engineering-specific subset focused on issues, PRs, deploys, and incidents.
Can I start with CSV uploads?
Yes. CSV uploads are a good first pass for exploration. Move to a database-backed sync when the dashboard needs scheduled refreshes, history, permissions, and trust.
What data do I need first?
For work tools, start with item ID, project, status, owner, created, due, and completed dates. For knowledge tools, start with content ID, container, owner, created, updated, and archived dates. For communication tools, start with event or thread ID, actor, container, created time, and first-response time. Add history next for trends and durations.