Executive Reporting

# Weekly Support Operations Report
## Week of [date range]

### Highlights
- [2-3 key wins or concerns from the week]

### Volume and SLA
- Total tickets: [number] ([+/-]% vs last week)
- First response time (avg): [time] (target: [time])
- SLA compliance: [%] (target: [%])
- Forecast next week: [number] tickets

### Quality
- Average QA score: [number]/100 ([+/-] vs last week)
- Conversations scored: [number] ([%] coverage)
- Top coaching theme: [theme from QA trends]

### Customer Health
- Average sentiment: [score] ([+/-] vs last week)
- Escalation alerts triggered: [number]
- At-risk customers flagged: [number]

### AI Performance
- Reply draft acceptance rate: [%]
- Triage routing accuracy: [%]
- KB gaps identified: [number] (closed this week: [number])

### Action Items
- [ ] [Specific action with owner and deadline]
- [ ] [Specific action with owner and deadline]

Required agents = Forecasted daily volume / Tickets per agent per day
Buffer          = Required agents * 1.15 (15% buffer for PTO, meetings, etc.)
Gap             = Buffer - Current headcount

### AI Investment ROI - [Month]

**Costs**
- Total LLM spend across all services: $[amount]
- Infrastructure: $[amount]
- Total AI cost: $[amount]

**Savings**
- Agent hours saved (AHT reduction): [hours] = $[amount]
- QA analyst hours saved (automated scoring): [hours] = $[amount]
- Reporting hours saved (Pulse automation): [hours] = $[amount]
- Total savings: $[amount]

**Net ROI: $[savings - costs] ([X]x return)**

**Quality Impact (not dollarized)**
- CSAT: [trend]
- Escalation rate: [trend]
- QA coverage: [before]% -> [after]%

"""Weekly report generator. Run via cron or scheduled task."""

import httpx
from datetime import datetime, timedelta

SERVICES = {
    "pulse": "http://localhost:8001",
    "qa": "http://localhost:8002",
    "sentiment": "http://localhost:8003",
    "reply": "http://localhost:8004",
    "triage": "http://localhost:8005",
    "kb": "http://localhost:8006",
}

def fetch_json(base_url: str, path: str, params: dict | None = None) -> dict:
    """Fetch JSON from a service endpoint."""
    resp = httpx.get(f"{base_url}{path}", params=params or {})
    resp.raise_for_status()
    return resp.json()

def generate_weekly_report() -> str:
    end = datetime.utcnow()
    start = end - timedelta(days=7)
    date_params = {
        "start": start.isoformat(),
        "end": end.isoformat(),
    }

    # Gather data from each service
    metrics = fetch_json(SERVICES["pulse"], "/metrics", date_params)
    sla = fetch_json(SERVICES["pulse"], "/sla", date_params)
    forecast = fetch_json(SERVICES["pulse"], "/forecast")
    qa_scores = fetch_json(SERVICES["qa"], "/scorecards", date_params)
    sentiment = fetch_json(SERVICES["sentiment"], "/trends", date_params)
    reply_feedback = fetch_json(SERVICES["reply"], "/feedback", date_params)
    kb_gaps = fetch_json(SERVICES["kb"], "/gaps")

    # Build the report
    report = f"""# Weekly Support Operations Report
## Week of {start.strftime('%b %d')} - {end.strftime('%b %d, %Y')}

### Volume and SLA
- Total tickets: {metrics.get('total_tickets', 'N/A')}
- First response time (avg): {metrics.get('avg_first_response_time', 'N/A')}
- SLA compliance: {sla.get('compliance_rate', 'N/A')}%
- Forecast next week: {forecast.get('predicted_volume', 'N/A')} tickets

### Quality
- Average QA score: {qa_scores.get('average_score', 'N/A')}/100
- Conversations scored: {qa_scores.get('total_scored', 'N/A')}

### Customer Health
- Average sentiment: {sentiment.get('average_score', 'N/A')}
- Escalation alerts: {sentiment.get('escalation_count', 'N/A')}

### AI Performance
- Reply draft acceptance rate: {reply_feedback.get('acceptance_rate', 'N/A')}%
- KB gaps identified: {kb_gaps.get('total_gaps', 'N/A')}
"""
    return report

if __name__ == "__main__":
    report = generate_weekly_report()
    filename = f"report_{datetime.utcnow().strftime('%Y-%m-%d')}.md"
    with open(filename, "w") as f:
        f.write(report)
    print(f"Report written to {filename}")

# Every Monday at 8am UTC
0 8 * * 1 python3 /path/to/generate_report.py