Multi-Location Business Intelligence Rollup

1. The Problem

Businesses operating across multiple locations — retail chains, franchise networks, multi-site hospitality, regional professional services — face a fundamental data problem: each site generates its own performance data in its own system, and the picture only becomes coherent when someone compiles it all together. That someone is typically a head office analyst who spends two days per week in spreadsheets rather than generating insight. The intelligence exists. The visibility does not. And without visibility, the decision to invest in Site A and divest in Site B is made on gut feeling rather than comparative data.

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 `multi-location-bi`. This automation requires API connections to every location's operational system — configure each location as a named data source in the `LOCATION_REGISTRY` section with its system type, API credentials, and the metric fields to extract.
2. 2Define your `NORMALISATION_RULES`: the most common multi-location data problem is inconsistent metric definitions across sites. Configure rules for revenue (including or excluding VAT, tips, adjustments), footfall counting methodology, and any site-specific adjustments required for fair comparison (e.g., a flagship site that always trades differently from a standard location).
3. 3Set your `PERFORMANCE_TIERS`: define how you categorise location performance — for example, Top Tier (top 20% of locations by composite score), Core Performing (middle 60%), and Focus Required (bottom 20%). The agent assigns each location to a tier weekly and tracks tier movement over time.
4. 4Configure your `ANOMALY_DETECTION` sensitivity: the percentage deviation from a location's own rolling average (not the network average — a site that always trades slowly should be compared to its own baseline, not a flagship's) that triggers a flag. Default: 15% above or below the location's own 4-week rolling average.
5. 5Set your distribution workflow: the weekly pack goes to the full leadership team. The monthly strategic deep-dive goes to the board. Configure a separate stream for individual location managers — each manager receives their own site's data with network percentile benchmarks, but not the full multi-location view.

3. The Context Payload (context.json)

Save this file as: .openclaw/agents/multi-location-business-intelligence-rollup/context.json

{
  "automation_id": "29",
  "title": "Multi-Location Business Intelligence Rollup",
  "level": 3,
  "tier": "Enterprise Playbook",
  "setup_time": "1–2 weeks",
  "estimated_api_cost": "~$90–$180/mo",
  "client_price_range": "$2,000–$10,000",
  "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 G.

# ── STEP 1: Set up location registry ──
$ openclaw template multi-location-bi --location-registry --locations {{N}}
# Edit the generated template with each location's credentials.
$ 
# ── STEP 2: Test connection for each location ──
$ openclaw run multi-location-bi --test-all-locations
# Fix any failing locations before proceeding.
$ 
# ── STEP 3: Test data normalisation (dry run) ──
$ openclaw run multi-location-bi --dry-run --week last
# Review: are all metrics comparable? Adjust normalisation_rules if needed.
$ 
# ── STEP 4: Generate first weekly pack (no distribution) ──
$ openclaw run multi-location-bi --week last --no-send
$ 
# ── STEP 5: Activate scheduled reporting ──
$ openclaw schedule multi-location-bi --data-pull --cron "0 4 * * 1"
$ openclaw schedule multi-location-bi --report --cron "0 7 * * 1"
$ openclaw schedule multi-location-bi --deep-dive --cron "0 8 1 * *"
$ 
# ── Add a new location to the registry ──
$ openclaw run multi-location-bi --add-location --interactive
$ 
# ── View current tier standings for all locations ──
$ openclaw run multi-location-bi --tiers --current
$ 
# ── Generate a one-off location performance report ──
$ openclaw run multi-location-bi --location-id <id> --weeks 12