canteen-seed-import/seed-data/reference_handbook.md

🌱 Seed Data Import — Reference Handbook

Source data: Machine_List.xlsx (raw Cantaloupe export, 1,848 machines, 62 columns) Seed subset: import_seed.csv (headers stripped, sales columns = "Deleted") Handbook version: 2.0 — merged with Excel analysis + Disney address lookup

---

1. Customer Column → Company / Disney Park

Status: ✅ COMPLETE

The Customer column maps to company in the database. For non-Disney customers, it's the direct business name.

Disney Customers (prefix D-)

Customers starting with D- are Disney properties. 469 machines (25% of fleet) are at Disney locations. The customer name tells us which property:

Customer Prefix	Property Name	Category	Base GPS
D-ART OF ANIMATION	Art of Animation Resort	resort	28.3499, -81.5678
D-All Star Music	All-Star Music Resort	resort	28.3362, -81.5735
D-All Star Sports	All-Star Sports Resort	resort	28.3379, -81.5740
D-BAY LAKE TOWER	Bay Lake Tower at Contemporary	resort	28.4145, -81.5714
D-BLIZZARD BEACH	Blizzard Beach Water Park	park	28.3507, -81.5734
D-Boardwalk	Disney's BoardWalk Inn	resort	28.3680, -81.5577
D-Celebration	Celebration (Admin)	office	28.3248, -81.5364
D-ContemporaryHotel	Disney's Contemporary Resort	resort	28.4145, -81.5714
D-DISNEY VENDING	Disney Support Services	support	28.4100, -81.5800
D-DISNEY WORLD SS	Disney World Support Services	support	28.3720, -81.5610
D-DRC	Disney Regional Center	office	28.4666, -81.4305
D-Disney Springs	Disney Springs	park	28.3705, -81.5175
D-FORT WILDERNESS	Fort Wilderness Resort & Campground	resort	28.4100, -81.5480
D-GRAND FLORIDIAN	Disney's Grand Floridian Resort	resort	28.4105, -81.5890
D-POLYNESIAN RESORT	Disney's Polynesian Village Resort	resort	28.4085, -81.5715
D-POP CENTURY	Disney's Pop Century Resort	resort	28.3460, -81.5710
D-PORT ORLEANS FrQt	Port Orleans French Quarter	resort	28.3985, -81.5480
D-Port Orleans Rvsd	Port Orleans Riverside	resort	28.3950, -81.5450
D-RIVIERA RESORT	Disney's Riviera Resort	resort	28.3680, -81.5493
D-Saratoga Springs	Saratoga Springs Resort & Spa	resort	28.3655, -81.5250
D-Treehouse Villas	Treehouse Villas (Saratoga Springs)	resort	28.3650, -81.5250
D-WIDE WORLD SPORTS	ESPN Wide World of Sports	park	28.3275, -81.5920
D-WILDERNESS LODGE	Disney's Wilderness Lodge	resort	28.4120, -81.5860
D-Winter Summerland	Winter Summerland Mini Golf	park	28.3510, -81.5750
D-YACHT AND BEACH	Disney's Yacht & Beach Club Resorts	resort	28.3665, -81.5560

Match rule: Case-insensitive prefix match. D-Art of Animation Guest → Art of Animation. D-Port Orleans Rvsd GUEST → Port Orleans Riverside.

---

2. Place Column → Location Fields

Status: ✅ COMPLETE (Excel data analyzed)

The Place column has two distinct patterns — the parser must detect which applies.

Pattern A: Simple (~1,379 rows, 75%)

Format: Venue Name-Venue Detail-Venue Detail (dash-separated)

The last meaningful segment after the last dash is the place. Common suffixes to strip:

• Breakroom, Br, Break Room, BREAKROOM, BREAK AREA
• Vending Area, Vending
• Break, Breakroom (with trailing spaces)

Examples:

Sygma - Vending-Sygma - Vending Breakroom         → place = "Sygma - Vending"
Imperial Dade-Imperial Dade- Breakroom             → place = "Imperial Dade"
Ancora Apt.-Ancora Apt. Breakroom                  → place = "Ancora Apt."
Delamarre-Delamarre Breakroom                       → place = "Delamarre"
BMW Service-BMW Breakroom                           → place = "BMW"
Kisselback Ford-Kisselback Ford Breakroom           → place = "Kisselback Ford"
Sam's Club-Sam's Club                               → place = "Sam's Club"
BREAK AREA-BREAK AREA                                → place = "BREAK AREA"
Building 1180-Building 1180                         → place = "Building 1180"

136 unique Pattern A tails in the data. Many have floor/building prefixed:

BLDG10/11 - Hilton Worldwide
2ND FL CARDIO WAITING ROOM
13th St-BREAKROOM
1ST FL BREAK
2nd Floor Front Tower
AHS - Flamingo

Extraction rule:

1. Split on -
2. Take the last non-empty segment
3. Strip known suffixes (Breakroom, Vending, Break, etc.)
4. Extract floor info if present: NTH FL, Floor N, Nth Floor
5. Extract building info: BLDG #, Building N
6. Remaining text is the place name

Pattern B: Complex Disney (~469 rows, 25%)

Format: PersonName - Address-G-Zone FLOOR

Structure breakdown:

[Person Name] - [Address]-[G-][Zone] [FLOOR]
   Todd J    - 901 Tinberline Dr -G- REAR 6TH FL
   Becky N   - 200 Celebration  -     Disney Park & Tickets Ki
   Sarah W   - 4401 Floridian Way-G- BUILDING 9 LOBBY

Extraction steps:

1. Strip person prefix: <FirstName> <Initial> - or <FirstName> <LastInit> -

   • Pattern: ^[A-Z][a-z]+\s+[A-Z]?\s*-\s*
   • Removes: Todd J - , Sarah W - , Justin M - , Brenda G -

2. Strip address: Typically a building number + street name at the start

   • Must identify by pattern: ^\d+\s+\w+\s+\w+\s*(?:Dr|Blvd|Way|Ct|Ln|Ave)
   • Addresses seen: 901 Timberline Dr, 4401 Floridian Way, 2101 Epcot Resorts Blvd, etc.

3. Strip G- prefix if present (it's a zone/location marker)

4. Extract floor info from the remainder using patterns:

   • NTH FL → 1ST FL, 6TH FL, 3RD FL, 4TH FL
   • Nth FLOOR → 1ST FLOOR, 2nd FLOOR, 4th FLOOR
   • FL N → FL 2
   • Nth floor → 7th floor
   • NST FL → 1ST FL

5. Extract building info:

   • BLDG #N, Building N, Bldg N
   • DAAR N, DAAR B-N (Disney building codes)
   • TOWER N, Loop N

6. Extract zone/area (the remaining text after stripping floor/building):

   • LOBBY, BUS STOP, POOL, Cafeteria, Wardrobe
   • ePort/NAYAX/CMS structure details

Disney place suffix examples (after stripping person + address):

Customer	Cleaned Place Suffix	Building	Floor	Zone
Wilderness Lodge	G-REAR 6TH FL	REAR	6TH FL	—
Wilderness Lodge	G-South End 7th floor	South End	7th floor	—
Wilderness Lodge	G-CENTER 3RD FL	CENTER	3RD FL	—
Wilderness Lodge	G-FW VILLAS 1ST FL	FW VILLAS	1ST FL	—
Pop Century	G-BLDG 1 50A 1ST FL	BLDG 1 50A	1ST FL	—
Pop Century	G-BLDG 10 70B 2ND FL	BLDG 10 70B	2ND FL	—
Art of Animation	G-DAAR 1-1ST FL	DAAR 1	1ST FL	—
Art of Animation	G-DAAR 10-2ND FL	DAAR 10	2ND FL	—
Art of Animation	G-COZY CONE POOL	—	—	COZY CONE POOL
Art of Animation	G-NEMO POOL	—	—	NEMO POOL
All Star Sports	G-Sports #1 - 2ND FL	SPORTS #1	2ND FL	—
All Star Sports	G-SPORTS GRAND SLAM POOL	—	—	GRAND SLAM POOL
All Star Sports	G-SPORT SURF'S UP LAUNDRY	—	—	SURF'S UP LAUNDRY
Contemporary Hotel	G-CONT HOTEL TOWER 10	TOWER 10	—	—
Contemporary Hotel	G-CONT HOTEL SA FRONT	—	—	SA FRONT
Boardwalk	Hotel 2nd Floor Left	—	2nd Floor	Left
Boardwalk	Villas 3 FL by room	—	3 FL	by room
Grand Floridian	G-BUILDING 5 LOBBY	BUILDING 5	—	LOBBY
Grand Floridian	G-GF 2nd Floor	—	2nd Floor	GF
Riviera	G-2ND FLOOR NORTH	—	2ND FLOOR	NORTH
Saratoga Springs	Carousel Bldg 7501	Carousel 7501	—	—
Saratoga Springs	Congress Bldg 1101-1436	Congress 1101-1436	—	—
Fort Wilderness	G-FW LOOP 1400	FW LOOP 1400	—	—
Fort Wilderness	G-FW Meadows POOL 1400	—	—	Meadows POOL
Polynesian	G-POLY BLDG 10 MOOREA 1ST	MOOREA	1ST	BLDG 10
Polynesian	G-POLY BLDG 4 TUVALU-2ND	TUVALU	2ND	BLDG 4
French Quarter	G-PO FRENCH QTR B2	BLDG B2	—	—
French Quarter	G-Bus Stop A	—	—	Bus Stop A
Riverside	G-PO RIVERSIDE 80S	80S	—	—
Riverside	G-PO RIVERSIDE BLD	—	—	—
Riverside	G-WEST DEPOT BUS STOP	—	—	WEST DEPOT BUS
Port Orleans	Bus Stop A	—	—	Bus Stop A
Yacht & Beach	G-Beach Club Laundry	—	—	Beach Club Laundry
Yacht & Beach	G-Yacht Laundry	—	—	Yacht Laundry

Zone Prefixes (Disney)

Many Disney place suffixes have a G- prefix (Guest) or C- prefix (Cast). These indicate audience:

• G- = Guest area (hotel lobbies, pools, bus stops)
• C- = Cast/Backstage area (wardrobe, breakrooms, maintenance)

Strip G- and C- from the place field but store them as a zone_type metadata field.

Floor Extraction — Unified Rules

The data has floors expressed in these formats:

Raw	Normalized	Digestible
6TH FL	6th Floor	6
7th floor	7th Floor	7
FL 2	Floor 2	2
1ST FL	1st Floor	1
2ND FLOOR	2nd Floor	2
3 RD FLOOR	3rd Floor	3
4th FLOOR	4th Floor	4
14th FLOOR	14th Floor	14
5T (Contemporary tower)	Floor 5	5
2N, 2 fl	Floor 2	2

Note: Contemporary Resort uses tower suffix notation: 5T, 6T, 7T, 8T, 9T, 10, 11, 12, 14 — the T suffix stands for Tower.

---

3. GPS Derivation for Multi-Floor Machines

Status: ✅ COMPLETE

Key insight: It varies by building. Some buildings have machines stacked in the same spot across floors, others have multiple machines on one floor in different locations, or machines in different spots on different floors. GPS derivation must be conservative — never assume two machines share GPS without verification.

Considerations from the data:

• Disney resorts: 469 machines across 25 properties, many multi-floor (4-14 floors)
• Pop Century: 42 machines across 4 floors in 10 buildings (BLDG 1-10)
• Art of Animation: 46 machines across 4 floors in 10 buildings (DAAR 1-10)
• Wilderness Lodge: 20 machines across 3 zones (CENTER, REAR, DOCK) × multiple floors
• Contemporary Resort: Bay Lake Tower (4 floors), Contemporary Tower (floors 5-14), multiple zones
• All Star Sports: Building 9 has a machine on each floor — same spot
• All Star Sports: Has separate pool/laundry machines (different location entirely)
• Saratoga Springs: Multiple buildings spread across the property (Carousel, Congress, Grandstand, etc.)
• Port Orleans Riverside: Machines at bus stops and in buildings — same address, different spots

Base GPS Strategy

1. Property GPS — Use the Disney property GPS from the mapping table as the starting point for address geocoding
2. Existing GPS is authoritative — If a machine already has GPS in the database, preserve it (field-collected is authoritative)
3. Floor assumption — If machine A at 901 Tinberline Dr - REAR 6TH FL has GPS, that GPS is usable for machine B at 901 Tinberline Dr - REAR 5TH FL (same zone, adjacent floor)
4. Zone distinction — Do NOT share GPS across zones. REAR → REAR only. CENTER → CENTER only. Pool → Pool only.
5. Building distinction — Do NOT share GPS across different buildings on the same property. BLDG 1 → BLDG 1 only, not BLDG 2.
6. Bus stops — Each bus stop at Fort Wilderness / Port Orleans needs its own GPS
7. Non-Disney machines — Use geocoded address coordinates from the address column

GPS Proximity Confidence Levels

Situation	Confidence	Action
Same building + same zone + adjacent floor (±1 floor)	✅ High	Auto-apply GPS
Same building + same zone + far floor	⚠️ Medium	Auto-apply with flag for review
Same building + different zone	❌ Low	Do NOT auto-apply
Same address + different building	❌ Low	Do NOT auto-apply
Non-Disney, exact same address	✅ High	Auto-apply with address geocode

---

4. Serial Number Validation & OCR Correction

Status: ✅ COMPLETE (data analyzed from Excel + import_seed.csv)

Standard Serial Formats Found

Format	Example	Likely Make	Notes
All digits, 9-12 chars	168020939, 471011958	Crane / AMS	Standard numeric format
Starts with 1- prefix	1-16018993, 1-1711-0396	AMS	Dash-separated
Starts with 1 (10+ digits)	112034419, 145192104106	Crane / USI	All digits, no dashes
Starts with 222 (12 digits)	222001280013	Newer Crane	Telemetry-equipped
###-###### format	472-052119, 449-011276	Crane	Model-prefixed serial
124473010109 (12 digits)	124473010109	USI Mercato	USI standard
15S49... (suspect OCR)	15S491811573	Crane	S→5 correction needed

Likely OCR Errors Found

Machine ID	Raw Serial	Suspected Correct	Pattern
60017	15S491811573	155491811573	S → 5
60025	15S491811581	155491811581	S → 5
60028	I5s331807595	155331807595	I → 1, s → 5
60032	I5SS331807593	1555331807593	I5SS → 1555
60035	15S491811599	(check — likely same S→5)	S → 5

OCR Character Confusion Table

Mistyped	Should Be	Context
S	5	S looks like 5 in some fonts
s	5	Lowercase s scanned as 5
I	1	Capital I is indistinguishable from 1
O	0	Letter O vs digit zero
B	8	B scanned as 8 (less common)
G	6	G scanned as 6 (less common)

Validation Algorithm

For each serial:

1. Strip non-alphanumeric chars (dashes, spaces, dots)
2. Apply OCR correction: S→5, s→5, I→1, O→0, B→8, G→6
3. Check length: valid serials are 9-14 characters after cleaning
4. Flag as placeholder if serial is < 5 characters after cleaning
5. Flag as invalid if all chars are the same (e.g., 0000000000)

Placeholder / Invalid Serials

Machine ID	Serial	Action
60031	520	Clear to blank
68194	123	Clear to blank
99660	00	Clear to blank

Rule: Any serial < 5 characters after cleaning → set to blank and flag for manual entry.

---

5. Type/Class/Make/Model Consolidation

Status: ✅ COMPLETE (Excel data analyzed)

The Priority Rule

The Type column is the best source for Class/Make/Model. Parse it first, then fill gaps from the standalone Class, Make, Model columns.

Type Column Format

CLASS (MAKE MODEL)

Extraction:

"Snack (AMS Sensit 3)"    → Class=Snack, Make=AMS, Model=Sensit 3
"GF Food (Crane 472)"     → Class=Food,  Make=Crane, Model=472
"Bev (Royal GIII)"        → Class=Bev,   Make=Royal, Model=GIII
"NATIONAL - 168 SERIES"   → Class=Snack, Make=Crane, Model=168
"VENDO - 721"             → Class=Bev,   Make=Vendo, Model=721
"Unknown"                 → All three = Unknown

Full Type Pattern Distribution

Type Pattern	Count	Class → DB	Make → DB	Model → DB
Bev (Royal GIII)	299	Bev	Royal	GIII
Snack (Crane Merchant Media)	287	Snack	Crane	Merchant Media
Bev (Vendo 621/721/821)	235	Bev	Vendo	621/721/821
GF Food (Crane 472)	151	Food	Crane	472
Snack (Crane 15x/16x)	76	Snack	Crane	15x/16x
Unknown	67	Unknown	Unknown	Unknown
GF Bev (DN 200E)	49	Bev	DN	200E
GF Bev (DN 5800)	35	Bev	DN	5800
Snack (Crane 186)	34	Snack	Crane	186
GF Bev (DN BevMax 4)	34	Bev	DN	BevMax 4
Bev (DN 501E/600E/276E)	34	Bev	DN	501E/600E/276E
Snack (Crane 472)	26	Snack	Crane	472
Bev (DN 276E)	25	Bev	DN	276E
GF Bev (DN 3800 BevMax 4)	20	Bev	DN	3800 BevMax 4
GF Bev (DN Baby BevMax)	19	Bev	DN	Baby BevMax
Bev (DN 501E)	18	Bev	DN	501E
GF Bev (DN BevMax)	18	Bev	DN	BevMax
Snack/Bev (Crane 472)	16	Snack/Bev	Crane	472
Snack (Crane 187)	16	Snack	Crane	187
NATIONAL - 168 SERIES	16	Snack	Crane	168
VENDO - 721	14	Bev	Vendo	721
Snack (AMS 3561)	12	Snack	AMS	3561
Snack (USI Mercato)	10	Snack	USI	Mercato
Snack (VE)	10	Snack	VE	Unknown
NATIONAL - 186	10	Snack	Crane	186
NATIONAL - 187	10	Snack	Crane	187

Total unique Type patterns in data: 30+

GF Prefix Rule

Strip GF prefix from the Class field. The GF prefix means "Glass Front" — it's a physical descriptor, not a class.

Raw Class	DB Class
GF Food	Food
GF Bev	Bev
GF Snack	Snack (if present)

The GF prefix IS useful — it tells us the machine has a glass front. Store it as a boolean field is_glass_front = true.

Make Normalization

Raw Value	Normalized To
Crane, Crane Co, Crane Nat, NATIONAL, Crane National	Crane
DN, Dixie Narco	DN
Royal, Royal Vendors	Royal
Vendo, Vendo Co	Vendo
USI, U Select It	USI
AMS, Automatic Merchandising	AMS
VE	VE
AP, Automatic Products	AP (if present)
Faz, Fas	Fas (if present)

Class Normalization

Raw Value	DB Class
Snack, Snack/Food	Snack
Bev	Bev
GF Food	Food (strip GF)
GF Bev	Bev (strip GF)
Food	Food
Snack/Bev	Snack/Bev (combo machine)

Special/Non-standard Type Parses

Raw Type	Rule
NATIONAL - 168 SERIES	Make=Crane (National IS Crane), Model=168, Class=Snack
NATIONAL - 186	Make=Crane, Model=186, Class=Snack
NATIONAL - 187	Make=Crane, Model=187, Class=Snack
VENDO - 721	Make=Vendo, Model=721, Class=Bev
Snack/Bev (Crane 472)	Special combo class. Splits to Class=Snack/Bev
Snack (VE)	Make=VE, let Model = Unknown or check standalone Model column
Snack (no parens)	Class=Snack. Check standalone Make and Model columns
Bev (no parens)	Class=Bev. Check standalone columns
Unknown	Leave all three as Unknown — flag for manual identification

---

6. Remote Pricing Status — Meanings

Status: ✅ COMPLETE (Excel data analyzed)

Distribution (from 1,848 machines)

Status	Count	Meaning
On	614 (33%)	Remote pricing is active and working. CMS/ePort can push price changes.
Action Required	718 (39%)	Issue preventing remote pricing: No Dex Passcode configured, Bad Coil Count mismatch, or configuration error. Needs technician attention.
Incompatible	516 (28%)	Machine does not support remote pricing. Older model requiring EEEPROM upgrade, or pre-telemetry machine.

Practical Meanings

"On" — Working normally. Prices can be updated remotely. No action needed.

"Action Required" — Common causes:

• No Dex Passcode configured for the control board
• Coil count mismatch (machine has physically different coils than what's configured)
• Connection issue between telemetry module and vending controller
• Action: Technician needs to check the DEX cable, passcode, or coil configuration

"Incompatible" — Hardware limitation:

• Machine pre-dates remote pricing support
• Requires EEEPROM upgrade to support remote pricing
• Some older Crane 472, 167, 168 models
• May still accept DEX data for sales — just can't receive price changes
• Action: If revenue justifies it, consider EEEPROM upgrade kit. Otherwise, accept as-is.

Update Policy

When re-importing: auto-overwrite pricing status from the export. This is dynamic data.

---

7. Telemetry ID → Card Reader Decoder

Status: ✅ COMPLETE (Excel data analyzed)

Telemetry Provider Patterns

The Device column (Telemetry ID) contains the telemetry module identifier. The suffix in parentheses reveals the provider:

Telemetry ID Pattern	Provider	Reader Type	Count
VJ... (ePort)	ePort (USI/Crane)	Cantaloupe ePort telemetry	~400
K3CT... (ePort)	ePort (Crane)	Cantaloupe ePort telemetry	~300
VK... (ePort)	ePort	Cantaloupe ePort telemetry	~200
... (NAYAX)	Nayax	Nayax card reader (Myna, Topper, VPOS Touch)	275
... (CMS)	Crane Merchandising Systems	CMS telemetry/reader	628
All-numeric (no suffix)	Unknown	Unknown/other	21

Provider Detail

• ePort — Cantaloupe (formerly USI/ePort) telemetry module. These connect to the machine's control board via DEX/MDB and report sales, inventory, and health data. Supports remote pricing when compatible. Predicts card reader: Cantaloupe ePort built-in or external.
• NAYAX — Nayax cellular telemetry device. Connects via MDB and supports cashless payments. Models: Myna, Topper, VPOS Touch. Nayax devices typically have their own cellular connection and work independently of the machine's control board.
• CMS — Crane Merchandising Systems integrated telemetry. Crane's proprietary system built into Crane machines (Merchant Media, 472, 167/168, etc.). May be integrated or external.

Card Reader Mapping (Rule-Based)

Provider	Reader Brand	Notes
ePort	Cantaloupe	Cantaloupe ePort module includes card reader
NAYAX	Nayax	Nayax reader (Myna, Topper, VPOS Touch)
CMS	CMS	Crane integrated reader

Distribution (1,848 machines)

Provider	Count
ePort	924 (50%)
CMS	628 (34%)
Nayax	275 (15%)
Other/unknown	21 (1%)

Update Policy

Telemetry provider is relatively stable — auto-overwrite if changed, but changes are rare between exports.

---

8. Alert Meanings

Status: ✅ COMPLETE (4 alert types categorized)

Alert Types Found

The raw export (Excel) has a Coil Alerts and Product Alerts column in addition to the general Alerts column.

Alert Text	Severity	Meaning
(empty / none)	—	No alerts
Scheduled service on Monday, May 25, 2026	Info	Future service scheduled
Scheduled for service today	Info/Warning	Same-day service scheduled
Out of touch for 19 hours	Critical	Telemetry offline > 19 hours
Out of touch... (combined with service notice)	Critical	Multiple issues

Coil Alerts (from Excel)

The raw export has a Coil Alerts column (count of coil issues) and Product Alerts column (out-of-stock or jammed products). These are numeric counts, not text strings.

Alert Categorization

Alert	Severity	Show in app list?	Meaning
Scheduled service (future date)	Info	✅ Yes, show badge	A tech has been scheduled to visit. No action needed — informational.
Scheduled service today	Info	✅ Yes, show badge	A tech has been scheduled for today. Nothing to action — heads up.
Out of touch (telemetry lost >19h)	Critical	✅ Yes, show badge	Telemetry offline for 19+ hours. Machine can't report sales or receive price changes. Needs investigation.
Out of touch + service needed	Critical	✅ Yes, show both icons	Combined: critical telemetry issue + info service notice. Both badges shown. Overall state = Critical.

---

9. Sales Priority Scoring

Status: ✅ COMPLETE (1,826 of 1,848 machines have yearly sales data)

Sales Data Available in Excel

Column	Description	Range
Yearly Sales	Primary — Total annual sales per machine	$10.50 - $206,230
Monthly Sales	Last 30 days sales	varies
Weekly Sales	Last 7 days sales	$1.50 - $18,314.50
Daily Average Sales	Per-day average	$0.22 - ~$565
Today Sales	Current day sales	varies
Yesterday Sales	Previous day sales	varies
Sales (Restock)	Sales since last restock	varies
Days Since Restock	Days since last service	0 - 30+

Rule: Priority is computed from Yearly Sales only. Use the standalone columns in the Excel export.

Yearly Sales Distribution (1,826 machines with data)

Percentile	Yearly Sales ($)
10th	$818.50
25th	$1,948.50
50th (median)	$4,425.50
75th	$8,357.30
90th	$15,848.50
95th	$23,747.75
99th	$64,118.50
Max	$206,230.00

Priority Tiers

Priority	Yearly Sales Range	% of Fleet
Low	≤ $2,000	~25% (456 machines)
Mid	$2,001 - $15,000	~65% (1,187 machines)
High	$15,001+	~10% (183 machines)

Update Policy

Priority is a computed field — recalculate on every import. Always auto-overwrite.

---

10. Asset List & Detail Screen Spec

Status: ✅ COMPLETE

Asset List Card

Each asset card in the main app list should show:

• Machine ID
• Class badge (Snack / Bev / GF Food)
• Make · Model
• Location (place, address)
• Yearly sales ($)
• Priority badge (Low / Mid / High)
• Alert indicator (A badge with the text or count)
• Remote pricing status (On / Action Required / Incompatible)

Not shown on card:

• Disney park badge — in detail view only
• GPS badge — in detail view only
• Card reader type — in detail view only
• Last contact / dex time — in detail view only

Asset Detail Screen

The detail view sections in order:

1. Identity — Machine ID, Make · Model, Class badge, Serial number, Photo (if available)
2. Location — Place, Building, Floor, Zone, Address, City, Disney park badge, GPS coordinates
3. Sales & Priority — Yearly sales, Priority badge, Monthly/Weekly/Daily sales
4. Technical — Remote pricing status, Telemetry provider, Card reader type, Glass front indicator
5. Alerts & Status — Alert text, Pricing status badge, Last contact time, Days since restock
6. History — Install date, Last DEX report, Deployed/Pulled dates, Route info

Editable sections (from admin panel):

• Location fields (place, building, floor, GPS)
• Photo upload
• Serial number (manual correction)

Read-only (from import):

• Make, Model, Class
• Sales data & priority
• Telemetry / card reader info
• Alerts

11. Update Policy — Per-Field

Status: ✅ COMPLETE (all per-field policies defined)

For each field, specify how updates from new seed data should be handled:

DB Column	Source	Policy	Notes
machine_id	Machine ID	preserve	Primary key, never changes
serial_number	Serial Number	auto (+ flag)	Accept with OCR correction. Flag placeholder/invalid
name	Generated (Make + Place)	auto	Build from Make + Place fields
make	Type→Make (parsed)	auto	✅
model	Type→Model (parsed)	auto	✅
class	Type→Class (parsed)	auto	Strip GF prefix
is_glass_front	Type→Class	auto	true if raw Class starts with GF
company	Customer	auto	Direct map
place	Place (parsed)	auto	Stripped of breakroom/vending suffixes
building_name	Place (parsed)	auto	From Pattern A/B extraction
floor	Place (parsed)	auto	Normalized floor
zone	Place (parsed)	auto	Lobby, pool, bus stop, etc.
zone_type	Place (parsed)	auto	guest vs cast (G- vs C-)
address	Address	auto	Direct map
location_area	City	auto	Direct map
latitude	Derived	preserve	Field-collected is authoritative
longitude	Derived	preserve	Field-collected is authoritative
photo_path	—	preserve	Never overwrite
dex_report_date	Last Dex Report Time	auto	Fresh data replaces old
install_date	Added Date	preserve	Set once on first import
deployed	—	preserve	Set by field operations
pulled_date	—	preserve	Set by field operations
disney_park	Customer (parsed)	auto	From Disney mapping table
remote_pricing_status	Status (Remote Pricing)	auto	Dynamic data
alerts	Alerts	auto	Fresh alerts replace old
telemetry_provider	Telemetry ID (parsed)	auto	Parsed from suffix
card_reader_brand	Telemetry ID (parsed)	auto	From provider mapping
priority	Computed from sales	auto	Recalculated on every import
yearly_sales	Yearly Sales	auto	✅
monthly_sales	Monthly Sales	auto	✅
weekly_sales	Weekly Sales	auto	✅
days_since_restock	Days Since Restock	auto	✅
prepick_group	Prepick Group	auto	✅
has_cashless	Has Cashless	auto	Direct map

Policy values: auto = always overwrite from new data, preserve = never touch existing value

---

Appendix A: Column Mapping (Raw Excel → DB)

Excel Column	DB Column	Extract Type
Device (Telemetry ID)	telemetry_provider, card_reader_brand	Parse suffix (ePort) / (NAYAX) / (CMS)
Location	—	Redundant (Customer + Address)
Asset ID	machine_id	Direct (primary key)
Place	place, building_name, floor, zone, zone_type	Complex parse (Pattern A or B)
Type	class, make, model, is_glass_front	Parse CLASS (MAKE MODEL)
City	location_area	Direct
Address	address	Direct
Last Contact Time	—	Freshness indicator (not stored)
Last Dex Report Time	dex_report_date	Direct
Last Restock	—	Not stored
Sales (Restock)	—	Not stored
Daily Average Sales	—	Not stored
Today Sales	—	Not stored
Yesterday Sales	—	Not stored
Weekly Sales	—	Not stored
Monthly Sales	—	Not stored
Yearly Sales	yearly_sales	Used for priority scoring
Days Since Restock	days_since_restock	Used for priority scoring
Prepick Group	prepick_group	Direct
Customer	company (non-Disney) or disney_park + company=Disney	Direct + Disney parsing
Management Company	—	Not stored
Route	—	Not stored (routing info)
State	—	Not stored
Postal Code	—	Not stored
Serial Number	serial_number	Validate + OCR correct
Class	class (fallback)	Type column takes priority
Make	make (fallback)	Type column takes priority
Model	model (fallback)	Type column takes priority
Has Cashless	has_cashless	Direct (Yes/No → bool)
Added Date	install_date	Direct (set once)
Status	remote_pricing_status	On / Action Required / Incompatible
Alerts	alerts	Parsed to structured alerts
Deployed	deployed	Yes/No
Pulled Date	pulled_date	Direct

---

Appendix B: Non-Disney Customers (sample identifying patterns)

Many non-Disney customers follow recognizable patterns:

• The Sygma Network — food distribution/manufacturing
• Imperial Dade — janitorial supply distributor
• Wood Springs Suites — hotel
• Ancora Apartments — apartment complex
• Southern Technical College — educational
• AdventHealth — hospital/medical
• BMW Service — automotive
• Kisselback Ford — automotive
• Hilton Worldwide — hotel chain (multiple buildings: BLDG2/3 through BLDG19/12)
• Sam's Club — retail
• Finfrock — manufacturer
• Walt Disney World Cast Credit Union — credit union

---

Version 2.0 — Complete reference handbook for the seed data import pipeline. All sections reviewed and finalized.