Performance Dashboard

1. The Problem

Most businesses run off three to five disconnected data systems: a CRM, an analytics platform, a financial tool, an ad platform, and a project management tool. Compiling a coherent weekly performance view requires logging into each system, pulling specific metrics, entering them into a shared spreadsheet or presentation template, and distributing to stakeholders — a process that consumes 3–5 hours every week, usually on a Friday afternoon when a senior operations or finance person should be doing something more valuable.

2. Integration & Webhook Setup

Follow the exact steps below to configure and deploy this automation inside your OpenClaw workspace.

1. 1Create a new agent named `performance-dashboard`. This automation requires the most API connections of any Level 2 build — budget 5–6 hours for first setup. Connect each of your data sources: CRM (HubSpot/Salesforce/Pipedrive), analytics (Google Analytics 4), advertising (Google Ads and Meta Ads APIs), finance (Xero or QuickBooks), and project management (Asana, Monday.com, or Jira).
2. 2Configure your `METRICS_MAP` section: define exactly which metric from each system appears in which section of the dashboard. Avoid the trap of including every available metric — limit each section to 4–5 KPIs. More data is not more insight.
3. 3Set your `COMPARISON_PERIODS`: the agent compares each metric against last week, the 4-week average, and the current month's target. Configure your target values in the `PERIOD_TARGETS` section — leave blank for any metric without a formal target.
4. 4Define your `ANOMALY_THRESHOLD`: the percentage deviation that triggers a highlighted flag in the report. Default is 15% — adjust to 20–25% for businesses with naturally volatile metrics (e-commerce, event-based revenue) to reduce noise.
5. 5Configure the distribution list and delivery format. The default output is a formatted HTML email with embedded charts. If your leadership team prefers a PDF attachment or a Notion page, alternative output formats are supported via your prompt instructions.

3. The Context Payload (context.json)

Save this file as: .openclaw/agents/performance-dashboard/context.json

{
  "automation_id": "17",
  "title": "Performance Dashboard",
  "level": 2,
  "tier": "Agency Foundation",
  "setup_time": "5–6 hours",
  "estimated_api_cost": "~$20–$45/mo",
  "client_price_range": "$500–$2,000",
  "agents": [
    {
      "role": "orchestrator",
      "model": "claude-3-5-sonnet",
      "temperature": 0.2,
      "max_tokens": 4096
    }
  ],
  "memory": "session",
  "output_format": "structured_json",
  "human_review_gate": true,
  "documentation_standard": "required"
}

4. Execution Commands

Run these commands from your openclaw-workshop/ directory to validate, test, and schedule this automation. Commands are taken directly from The OpenClaw Income Engine, Appendix F.

# ── STEP 1: Test each data source individually ──
$ openclaw run performance-dashboard --test-source ga4
$ openclaw run performance-dashboard --test-source google_ads
$ openclaw run performance-dashboard --test-source meta
$ openclaw run performance-dashboard --test-source hubspot
$ openclaw run performance-dashboard --test-source xero
$ 
# ── STEP 2: Generate last week's report (no sending) ──
$ openclaw run performance-dashboard --week last --dry-run
$ 
# ── STEP 3: Send test report to yourself only ──
$ openclaw run performance-dashboard --recipients "you@domain.com"
$ 
# ── STEP 4: Schedule weekly Monday delivery ──
$ openclaw schedule performance-dashboard --cron "0 7 * * 1"
$ 
# ── Generate a report for a specific date range ──
$ openclaw run performance-dashboard --start 2025-12-01 --end 2025-12-07
$ 
# ── Add a new data source without rebuilding ──
$ openclaw config edit performance-dashboard
# Add new source under data_sources and new metrics to metrics_map