本文へ移動
cccskills
無料GitHub で公開

restaurant-menu

Menu engineering analysis — item placement, description quality, pricing psychology, upsell opportunities, photo presence, Kasavana matrix (Stars/Plowhorses/Puzzles/Dogs)

インストール方法を見る

含まれるファイル(1)

  • SKILL.md10.1 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Menu Engineering Analysis

You analyze a restaurant menu using the Kasavana & Smith menu engineering framework, scoring each item on popularity and contribution margin, then producing concrete recommendations to lift average check size and overall menu profitability.

DISCLAIMER: AI-generated menu analysis based on publicly available menu data and industry benchmarks. The restaurant owner should verify item-level cost and sales data.


When to use

  • /restaurant menu <name-or-url> — full menu engineering audit
  • "menu analysis for [name]"
  • "what should I change about my menu"

The Kasavana & Smith Matrix

Every menu item plots on a 2x2 grid:

High MarginLow Margin
High PopularitySTARS — protect, featurePLOWHORSES — re-engineer cost or raise price
Low PopularityPUZZLES — re-merchandise or relocate on menuDOGS — remove

Definitions:

  • Margin = menu price - food cost
  • Popularity = % of category orders > category mean

Execution Pipeline

Step 1: Acquire Menu

Sources in priority order:

  1. Restaurant website (most accurate)
  2. Google Business Profile menu
  3. Yelp menu
  4. Third-party (Uber Eats, DoorDash) — note prices may include markup
  5. Photos of physical menu
WebSearch("[name] [city] menu prices")
WebSearch("[name] [city] menu pdf")

Step 2: Catalog Every Item

For each item capture:

  • Category (appetizer / entrée / side / dessert / drink)
  • Name
  • Description (full text)
  • Price
  • Photo present Y/N
  • Position in category (1st, 2nd, 3rd, etc.)
  • Any descriptors (chef's special, popular, gluten-free, signature)
  • Estimated food cost % (use industry standards if actual unknown):
    • Pasta dishes: 18-25%
    • Steaks/seafood: 30-40%
    • Burgers/sandwiches: 25-32%
    • Pizza: 20-30%
    • Appetizers: 20-30%
    • Desserts: 15-25%
    • Cocktails: 18-22%
    • Wine: 25-35%

Step 3: Apply Kasavana Matrix

For each item, estimate:

  • Contribution margin = price - (price × est. food cost %)
  • Popularity = best guess based on menu position, descriptors, photo presence, review mentions

Plot each item as Star / Plowhorse / Puzzle / Dog.

Step 4: Score 5 Sub-Dimensions

DimensionWhat to CheckScore (0-20)
Menu Structure & LayoutCategories logical? Length appropriate? Star placement in golden triangle?...
Description QualitySensory language, origin stories, ingredient highlights, length 15-30 words?...
Pricing PsychologyCharm pricing ($14.95 vs $15)? Decoy items? Anchoring (high-price first)? Currency symbols dropped?...
Photo PresencePhotos for high-margin items? Quality? Consistency?...
Upsell / Combo EngineeringAdd-ons, combos, dessert nudges, drink pairings, sides upgrades?...

Step 5: Identify Quick Wins

The 5 highest-leverage menu changes:

  1. Reposition Stars to the golden triangle (top-right, top-left, first item in category)
  2. Rewrite descriptions for Plowhorses to boost margin perception
  3. Re-merchandise Puzzles with photos and "chef's pick" badges
  4. Remove or replace Dogs (free up real estate)
  5. Add 2-3 strategic combos / add-ons for 8-12% ticket lift

Pricing Psychology Reference

TacticWhen to useExample
Charm pricing (.95, .99)Casual / value-driven concepts$14.95 vs $15.00
Round pricing (whole dollars)Premium / fine dining"32" vs "$32.00"
Drop currency symbolsRemoves "spending" psychology"Burger 16"
Decoy effectAnchor a premium item to make others look reasonable$48 ribeye makes $32 strip steak feel cheap
Bracket pricing3 price tiers per category (low/mid/high) — mid sells most$12 / $18 / $26
Avoid price columnsRight-aligned price columns invite price comparisonInline price after description
Avoid dollar signsEspecially in fine dining$14 → 14

Description Quality Rubric

A great menu description (15-30 words):

✅ Uses sensory language (crispy, slow-braised, hand-rolled, charred) ✅ Names 2-3 key ingredients ✅ Mentions origin or technique (Sicilian, wood-fired, 36-month aged) ✅ Hints at preparation (chef's signature, family recipe) ✅ Differentiates from competitors

❌ Generic ("delicious chicken with vegetables") ❌ Just a list of components ❌ Over 40 words (decision fatigue) ❌ Under 8 words for items > $20 (feels cheap)


Golden Triangle / Eye Tracking

Eye-tracking studies show diners read menus in this pattern:

1. Top-right corner  ← place 2 highest-margin items here
2. Top-left corner   ← place signature / story-driven items here
3. Center            ← middle ground
4. Bottom-left
5. Bottom-right      ← place items you want LESS attention on

Within categories: first item gets 20% more attention. Last item gets 15% more attention. Middle items get the least.


Output Template

Save to RESTAURANT-MENU-[Name].md:

# Menu Engineering Analysis: [NAME]

> **Generated:** [DATE] | **Menu Score:** [X]/100 | **Total Items Analyzed:** [N]

**DISCLAIMER: AI-generated menu analysis. Verify item-level cost data with the owner.**

## Menu Health Snapshot

| Metric | Value | Industry Benchmark |
|--------|-------|---------------------|
| Total items | X | 20-30 (sweet spot) |
| Avg item price | $X.XX | ... |
| Items with photos | X% | 30-50% (digital), 0-20% (print) |
| Avg description length | X words | 15-30 words |
| Estimated avg food cost | X% | 28-32% |
| Pricing strategy | [Charm / Round / Mixed] | ... |

## Kasavana Matrix Distribution

| Quadrant | Item Count | % of Menu |
|----------|------------|-----------|
| Stars (high pop, high margin) | X | X% |
| Plowhorses (high pop, low margin) | X | X% |
| Puzzles (low pop, high margin) | X | X% |
| Dogs (low pop, low margin) | X | X% |

## Item-by-Item Analysis

### Entrées

| Item | Price | Est. Margin | Pop. | Classification | Recommendation |
|------|-------|-------------|------|----------------|----------------|
| Spaghetti Carbonara | $19 | $14.25 | High | Star | Move to top-right of menu; add chef story |
| Lasagna | $17 | $11.50 | High | Plowhorse | Raise to $18.50 (price elasticity weak here) |
| Branzino | $32 | $22.40 | Low | Puzzle | Add photo, badge "Chef's Pick" |
| Caesar Salad Entrée | $14 | $9.80 | Low | Dog | Remove or convert to half-portion side |

### Appetizers, Sides, Desserts
[same table structure]

## 5 Quick-Win Recommendations

### 1. Reposition Stars to the Golden Triangle
- Move Spaghetti Carbonara to top-right position
- Move Margherita Pizza (signature) to top-left
- Expected impact: +5-8% category sales lift

### 2. Rewrite Plowhorse Descriptions
- Current Lasagna description: "Traditional lasagna with meat sauce." (8 words)
- Suggested: "Hand-layered with Bolognese braised 6 hours, fresh ricotta, and Grana Padano — Nonna Maria's recipe since 1962." (20 words)
- Expected impact: support price increase from $17 → $18.50 without complaints

### 3. Re-Merchandise Puzzles
- Branzino — add photo + "Chef's Pick" badge
- Veal Saltimbocca — pair with house Chianti as a "Tuscany Tasting"
- Expected impact: 30-50% lift in low-pop high-margin item sales

### 4. Remove or Convert Dogs
- Caesar Salad Entrée — remove (low margin, cannibalizes appetizer Caesar)
- Three Cheese Ravioli — replace with seasonal special slot
- Expected impact: freed menu real estate + lower waste

### 5. Add 3 Strategic Add-ons
- "Make it a Bolognese" — add ground beef to any pasta for $4 (90%+ margin)
- "Truffle upgrade" — shaved truffle on any pasta for $7
- "Tiramisu duo" — small tasting + espresso for $8
- Expected ticket lift: +$3.50/cover = +$X,XXX/month at current covers

## Pricing Psychology Audit

| Element | Current State | Recommendation |
|---------|---------------|----------------|
| Charm vs round pricing | Mixed | Standardize on round (concept is mid-premium) |
| Currency symbols | Used | Drop "$" — less price-focus |
| Price alignment | Right-aligned column | Move to inline at end of description |
| Decoy item | None | Add $42 Ribeye to anchor entrée prices down |

## Photo Audit

| Item | Has Photo? | Should It? |
|------|------------|------------|
| Carbonara (Star) | Yes | Yes — keep |
| Branzino (Puzzle) | No | YES — add immediately |
| Lasagna (Plowhorse) | Yes | Yes — but quality is poor, reshoot |
| Caesar (Dog) | No | No — being removed |

## Menu Score Breakdown

| Dimension | Score | Notes |
|-----------|-------|-------|
| Structure & Layout | X/20 | ... |
| Description Quality | X/20 | ... |
| Pricing Psychology | X/20 | ... |
| Photo Presence | X/20 | ... |
| Upsell / Combo Engineering | X/20 | ... |
| **Total** | **X/100** | |

## Revenue Impact Projection

If the 5 quick wins are implemented:
- Avg check size lift: +$2.80 (8.5%)
- At [N] daily covers × 30 days: **+$X,XXX/month revenue lift**
- No change in food cost % expected

## What to Implement First

1. **This week:** Move Stars to golden triangle (zero cost, just menu redesign)
2. **This week:** Add 3 strategic add-ons (zero food cost change)
3. **Within 2 weeks:** Rewrite Plowhorse descriptions, raise prices
4. **Within 30 days:** Reshoot photos for Puzzles + signature items
5. **Within 60 days:** Full menu reprint with all changes

**DISCLAIMER: AI-generated analysis. Always verify item-level food cost and sales mix data with the owner before raising prices or removing items.**

Quality Standards

  • Every item gets a classification (no skipping)
  • Every recommendation includes a revenue impact estimate
  • Pricing recommendations consider local competition and concept tier
  • Photo recommendations factor in cost ($500-$2,000 for a 20-item shoot)

DISCLAIMER: For educational/research purposes only. AI-generated.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Launches 5 parallel AI agents to produce a comprehensive restaurant audit with composite Restaurant Health Score (0-100), grade, and prioritized action plan

日本語の概要は準備中です。原文の説明を表示しています。

zubair-trabzada/ai-restaurant-claude332026年5月21日 更新

Facebook/Instagram ad copy variations — 10+ ad copy variants across happy hour, lunch, weekend brunch, catering, birthday/anniversary angles with A/B testing framework

日本語の概要は準備中です。原文の説明を表示しています。

zubair-trabzada/ai-restaurant-claude332026年5月21日 更新

Top 5 local competitor analysis — menu, pricing, review, social, and SEO comparison with positioning gap identification

日本語の概要は準備中です。原文の説明を表示しています。

zubair-trabzada/ai-restaurant-claude332026年5月21日 更新

Email and SMS sequences — loyalty program emails, birthday emails, win-back campaigns, special event invites, abandoned cart for online orders

日本語の概要は準備中です。原文の説明を表示しています。

zubair-trabzada/ai-restaurant-claude332026年5月21日 更新

Local SEO audit — "best [cuisine] near me" targeting, Google Business Profile optimization, NAP consistency, citation building, restaurant schema markup

日本語の概要は準備中です。原文の説明を表示しています。

zubair-trabzada/ai-restaurant-claude332026年5月21日 更新

Online presence audit — Google Business Profile completeness, Yelp listing quality, website assessment, online ordering setup, third-party platform presence

日本語の概要は準備中です。原文の説明を表示しています。

zubair-trabzada/ai-restaurant-claude332026年5月21日 更新

zubair-trabzada のスキルをすべて見る

このスキルの問題を報告する