admin-baked/skills/dispensary/daily-dispensary-ops-review

Daily Dispensary Ops Review

Purpose

Give a dispensary operator a fast, opinionated read of yesterday's performance — what moved, what didn't, what needs action today — so they can run the store, not dig through dashboards.

When to Use

• Morning briefing playbook fires (daily, ~7 AM local time)
• Operator manually asks for yesterday's or today's store performance
• A metric alert fires and needs context before routing to an action memo
• End-of-day check before closing

When NOT to Use

• Weekly or multi-period trend analysis → use executive-brief with a wider date range
• Competitor performance → route to Ezal for competitive context
• Campaign performance deep-dive → route to low-performing-promo-diagnosis
• Loyalty segment health → route to loyalty-reengagement-opportunity-review
• Technical issues (POS sync errors, missing data) → route to Linus before interpreting numbers

Required Inputs

• org_id — required
• date — the review date; default: yesterday
• store — store name (for multi-location orgs)
• baseline_window — comparison period; default: prior 7-day average

Reasoning Approach

Work through five performance dimensions in order. Each one can independently surface an action.

1. Revenue & traffic

• Total revenue vs baseline (7-day avg, same weekday prior week)
• Transaction count vs baseline
• Average basket size vs baseline
• Basket drives can reveal product mix shifts even when traffic is flat

2. Menu performance

• Top 5 SKUs by revenue vs prior period — anything surprising?
• Any category with significant velocity change (vapes up, flower down)?
• Out-of-stock flags that likely cost sales

3. Promotional impact

• Was any campaign active? Track: send → open → transaction (attributed revenue)
• Promo redemption rate vs expected
• Unintended margin drag from broad discounts

4. Loyalty signals

• Points earned vs redeemed ratio — high earn + low redeem = program not landing
• New enrollments today vs daily average
• Any loyalty tier drops (champion → engaged) in last 24 hours

5. Anomalies

• Any metric > 15% above or below its baseline → flag as anomaly, assess urgency
• Cross-dimension: traffic up but revenue flat → basket size problem or product mix issue
• Apply URGENCY tier: Immediate (act today) / This Week / Monitor

Confidence rule: If any input data is incomplete or POS sync was delayed, state it and reduce confidence on affected dimensions. Never fill gaps with assumptions.

Output Contract

## Daily Ops Review — [Store] — [Date]

REVENUE: $X,XXX ([+/-]X% vs baseline)
TRANSACTIONS: N ([+/-]X%)
AVG BASKET: $XX.XX ([+/-]X%)
DATA QUALITY: [Complete / POS lag noted on: ...]

### Top Actions Today
| # | Action | Owner | Urgency |
|---|--------|-------|---------|
| 1 | ...    | ...   | ...     |

### Performance Snapshot
**Revenue & Traffic:** [2 sentences — what moved and why]
**Menu:** [1–2 sentences — top performers, any gaps]
**Promos:** [1 sentence — active campaign impact or "no active campaign"]
**Loyalty:** [1 sentence — enrollment and redemption signal]

### Anomalies
[Bulleted list of flagged deviations with urgency tier. "None detected" if clean.]

### Tomorrow's Watch List
[1–2 items to monitor or confirm at next review]

Total length: 200–350 words. Scannable in under 60 seconds.

Edge Cases

• Sunday / short-hours day: Compare to prior Sundays, not weekday average — flag if Sunday baseline not established
• Holiday: Note the holiday; adjust baseline comparison; high traffic days should compare to same holiday prior year
• POS sync failure: Do not produce a memo on incomplete data — alert operator, state expected sync time
• New store (< 30 days open): No reliable baseline yet; report absolute numbers with "baseline not yet established"
• Multi-location org: Produce one memo per location; do not blend metrics across stores

Escalation Rules

• Revenue < 50% of baseline with no known cause: Escalate to Immediate + route anomaly-to-action-memo
• Out-of-stock on top-3 revenue SKU: Flag for procurement same day
• POS data missing > 2 hours: Escalate to Linus before producing memo
• Loyalty redemption rate > 40% in one day: Flag for margin review — may indicate gaming or error

Compliance Notes

• Do not include customer PII in the memo — aggregate counts and revenue figures only
• Promotional attribution figures are estimates based on send-to-transaction window; label as attributed, not proven
• Revenue projections or forecasts are estimates — label clearly