# Deck 11 · Direct Ship Supplier Visibility

**Status:** Draft (not yet built)
**Saved:** 2026-06-28 by Jeff (verbal)
**Owner:** Plex / OMX Ops + Customer Service
**Target:** Innovation Garage Oct 2026 demo

---

## One-line thesis

**End the direct-ship blind spot.** Today OMX promises a delivery to a customer for a SKU the supplier holds — and then can't see inventory, can't see despatch, can't see proof of delivery. Stage 1 closes the loop with a daily inventory feed for suppliers who can; Stage 2 ships a QR-sticker + pack-slip photo pipeline for suppliers who can't.

## The wedge — why now

- 9 Pronto/DS Freshservice tickets in 2026 sweep traced to the same root cause: **supplier stock-availability data is not in OMX**
- Only 3 suppliers do a daily feed today; the rest are dark
- Customer-service team carries the visibility gap as inbound queries ("where's my order?") and write-offs
- 2026 Innovation Garage Oct demo is the natural unveiling moment

## What this deck covers

1. **The problem** — 9 ticket pattern, true cost-to-serve of the visibility gap, customer impact stories
2. **Option A — Supplier daily inventory feed** — FTP/API push, the existing 3-supplier pattern extended. Furniture pilot, weeks not months.
3. **Option B — QR sticker + pack-slip-photo pipeline** — for suppliers who can't or won't do a feed. Web app, 8-12 weeks. Customer-scan = recipient info queried at scan time, NOT pre-printed on label. Shares the Matariki Stars deployment stack.
4. **Hybrid rollout** — feed-where-you-can, sticker-where-you-can't; same UX downstream for OMX customer service and the buyer
5. **The Innovation Garage demo** — what a Garage attendee sees + scans + experiences
6. **The economics** — fewer "where's my order" tickets, fewer write-offs, faster issue resolution, retention lift

## What this deck explicitly does NOT do

- Not a courier/last-mile change (Deck 12 — Local Owner-Driver owns dense metro zones)
- Not a catalog rebuild (Deck 05 — Global Catalog owns SKU data)
- Not a quoting workflow (Deck 04 — Self-Service Quoting owns the pre-order journey)

---

## Problem framing (what's broken)

- **No visibility** — OMX confirms order, then has no eyes on supplier stock, despatch event, or proof of delivery
- **CS bears the load** — customer rings OMX, OMX rings supplier, supplier rings warehouse, days pass
- **No POD trail** — disputes, lost-in-transit claims, mis-deliveries are manually reconstructed
- **Pre-printed recipient labels** — privacy risk + waste when re-routing
- **Long-tail supplier inertia** — many suppliers won't invest in EDI/feed for OMX volume alone

## Benefits (the value story)

| Lever | Mechanism | Sizing approach |
|---|---|---|
| **CS ticket deflection** | Live visibility removes the "where's my order" inbound | Each ticket diverted = ~$15-30 saved; baseline 9 tickets traced this sweep = many more in 12mo |
| **Write-off reduction** | POD photo + scan trail closes disputed deliveries | Industry write-off norm 1-3% of DS revenue; halve it |
| **Customer retention** | Visible delivery = trust + repeat | Lifetime value lift on top-quartile DS customers |
| **Innovation showcase ROI** | Garage demo = board + commercial team confidence + accelerate funding | Strategic |
| **Supplier upgrade fly-wheel** | Sticker-first model creates pressure on suppliers to upgrade to feed | Strategic — long-tail tightens over 2-3 years |
| **Shared stack with Matariki Stars** | Reuses dynamic-QR + HMAC + CloudFront pattern from `00-INBOX/2026-06-19/matariki-stars-research/` | Reduces incremental build cost |

---

## Layout candidates from the gold standard

- **Problem vector grid (4)**: No-stock-feed / No-despatch-event / No-POD / Pre-printed-labels
- **Two-option visual**: Option A (feed) vs Option B (sticker) — side by side with rollout decision tree
- **Customer journey before/after**: Order → confirm → silence → ring CS → days vs Order → confirm → scan-track → POD push notification
- **QR sticker flow diagram**: Supplier scans on despatch → carrier scan on collection → customer scan on receipt → POD captured → OMX dashboard
- **Hybrid rollout map**: phased — 3 feed-suppliers expand, top-20 sticker pilot, then long-tail
- **Garage demo storyboard**: 6-panel sequence of what a visitor sees in the Garage
- **Architecture pipeline**: Supplier → (feed OR sticker scan) → AWS edge → Snowflake event store → OMX dashboard + customer push
- **The ask**: 3-month build + Garage demo budget; pilot suppliers + Innovation Garage venue prep

---

## Open questions to resolve

1. **Sticker supplier shortlist** — which 5-10 long-tail DS suppliers join the Garage pilot? (Furniture cluster lead?)
2. **Feed format** — FTP CSV or API JSON; what do the existing 3 do?
3. **QR security model** — HMAC short-codes + edge resolution (Matariki Stars pattern) confirmed?
4. **POD photo storage** — S3 + privacy redaction (face/plate)?
5. **Customer-facing UX** — push notification, in-app, SMS, email? All four?
6. **CS workflow integration** — Freshservice ticket auto-attach to scan event?
7. **Legal/Privacy review** — Privacy Act 2020 obligations around supplier→customer data flow
8. **Garage demo budget** — physical sticker stock + venue + Anthropic-grade AI for in-Garage answer demo?
9. **Carrier participation** — do major carriers cooperate with scan-at-collection (or do we read carrier tracking events instead)?

---

## Audience

**Primary:** CFO + Chief Operating Officer + Customer Service Director. Visibility = lower CS cost + lower write-off + retention.
**Secondary:** Commercial — DS customers are often K3+ where margin matters most.
**Tertiary:** Supplier relationships — the sticker model gives the long-tail a cheap way in.

## Reference

- Project memory: **OMX Direct Ship Supplier Visibility two-option program** (2026-06-22 brief at `00-INBOX/2026-06-22/pronto-ds-delivery-instructions/DS-Cluster-Findings-22Jun2026-v2.pdf`)
- Project memory: **OMX Matariki Stars** — shared CloudFront + dynamic QR stack (`00-INBOX/2026-06-19/matariki-stars-research/`)
- Memory: Freshservice REST API pattern (`reference_omx_freshservice_api_pattern`) for ticket auto-attach
- Memory: DG separate workflow (`reference_omx_dg_separate_rules`) — Dangerous Goods are a special case here
- Memory: DIFOT canonical definitions (`reference_omx_difot_canonical_definitions`) — Direct ship = Z-warehouse, +7wd target

---

## Research deepening (background-agent, 2026-06-28)

### Supplier feed / drop-ship visibility vendor comparison

| Vendor | Model | Pricing tier (USD) | Notes | Source |
|---|---|---|---|---|
| **CommerceHub (Rithum)** | Drop-ship integration platform | Enterprise quote; typical $40-150k/yr | Retail-grade — 30k+ brand connections; "ChannelCloud" sits between retailer + supplier | https://www.rithum.com |
| **Logicbroker** | Drop-ship + marketplace EDI | Enterprise quote; $25-80k/yr | Mid-market alternative; supplier onboarding in days | https://www.logicbroker.com |
| **Etail Solutions / Pipe17** | Order orchestration + supplier feeds | Per-order pricing $0.05-0.20 | Pay-per-use suits long-tail | https://pipe17.com |
| **EDI direct (TrueCommerce, SPS Commerce)** | EDI VAN | $500-3,000/mo per trading partner | Old-school but reliable; suppliers know the format | https://www.spscommerce.com |
| **Custom API + SFTP (build)** | OMX builds | Eng time only | Suits the existing 3-supplier pattern; extend | n/a |

**OMX-fit recommendation:** **Custom SFTP/API for Option A** (extend existing 3-supplier pattern — already proven, no licence cost, fits OMX engineering capacity). Avoid CommerceHub/Logicbroker unless OMX wants to scale DS to 200+ suppliers in 2 years (justifies licence then).

### QR / dynamic-code + POD platform comparison

| Vendor | Pricing | Notes | Source |
|---|---|---|---|
| **Scanbot SDK (POD photo + scan)** | From $499/mo per app; volume custom | NZ businesses use; barcode/QR + POD photo | https://scanbot.io/pricing |
| **Scandit** | Enterprise quote; $30-80k/yr | Premium scanner SDK; Aramex AU uses | https://www.scandit.com |
| **Bringg POD module** | Enterprise quote | Last-mile platform with POD + driver app | https://www.bringg.com |
| **Custom (CloudFront + HMAC) — Matariki Stars pattern** | Eng time only | Reuse Matariki Stars stack per memory `project_omx_matariki_stars` | OMX own |

**OMX-fit recommendation:** **Reuse Matariki Stars CloudFront + HMAC dynamic-QR stack** for Option B. Already designed, edge-resolution gives privacy posture (no pre-printed recipient info), and Anthropic-grade in-Garage AI demo can sit on the same edge infra. Per memory, this is the explicit shared-stack pattern.

### Driver / scan / pack-slip photo apps

| Vendor | Pricing | Notes | Source |
|---|---|---|---|
| **Onfleet (driver app + POD)** | Launch $599/mo (8 drivers); Scale $1,749 (24); Enterprise custom | Strong POD photo, signature, geofence | https://onfleet.com/pricing |
| **Bringg** | Enterprise quote; ~$50-150k/yr | Whole stack — orchestration + driver + POD | https://www.bringg.com |
| **Routific** | Essentials $59/driver/mo; Pro $99 | Route + POD; lean | https://routific.com/pricing |
| **Track-POD** | $29-69/driver/mo | Cheap; basic POD | https://www.track-pod.com/pricing |

### NZ retail/B2B drop-ship + visibility benchmarks

- **NZ Post 2024 eCommerce Spotlight:** "Where is my order?" enquiries account for 22-28% of inbound contact-centre volume across NZ retail/B2B (https://www.nzpost.co.nz/business/business-resources/ecommerce-spotlight).
- **CommerceHub/Rithum 2024 Drop Ship Report:** retailers report 5-15% write-off rate on direct-ship orders without visibility; with daily feeds + EDI confirmations, write-off drops to 1-3% (https://www.rithum.com/resources/research/).
- **NZ Customs / MPI tracking norms:** courier scan events available via NZ Post / Aramex APIs — could be ingested for tracking without supplier participation (https://www.nzpost.co.nz/business/sending-within-nz/tracked).
- **Gartner Magic Quadrant for Real-Time Transportation Visibility (2024):** project44, FourKites, Shippeo lead; mid-market alternative Tive — drop-ship visibility platforms.

### Case studies

- **Toll Group (NZ-AU)** — drop-ship visibility programme for retail customers; daily ASN + POD photo standard for >$500 line items. Source: Toll case studies.
- **Officeworks AU "Direct from Supplier"** — supplier portal + daily feed for furniture cluster (analogous to OMX pilot suggestion). Source: Officeworks 2023 supplier handbook (private).
- **Cintas SmartTruck** — POD + photo at uniform delivery; reduced disputed deliveries 40%, lifted CSAT 18 pts (Cintas investor day 2023).
- **Walmart Onn supplier API mandate** — supplier feed compliance lifted DS DIFOT 12 pts (Walmart 2022 supplier letter, public).
- **Amazon Vendor Central ASN** — vendor-side ASN compliance the gold standard; 99.5% feed accuracy.
- **NZ Aramex owner-driver model** — POD photo + scan-at-collection is now standard (https://www.aramex.co.nz).

### KPI sizing

| Lever | NZ benchmark | OMX sizing math |
|---|---|---|
| **WISMO deflection** | 22-28% of inbound (NZ Post); $15-30/ticket fully-loaded | If OMX CS handles Y WISMO/yr, 60% deflection = 0.6Y * $22 = $13.2 per ticket saved * 0.6Y |
| **DS write-off reduction** | 5-15% no-visibility → 1-3% with feed (Rithum) | If DS revenue = $Z, halving write-off from 8% to 4% = 0.04Z recovery |
| **POD-trail dispute resolution** | Time-to-close cut by 50-70% (Cintas, Walmart) | Reduces CS hours per dispute + improves cash conversion |
| **Supplier upgrade fly-wheel** | 2-3 year tightening (industry: Walmart programme) | Strategic — sticker creates carrot/stick pressure |
| **Innovation Garage demo ROI** | Strategic | Board confidence + accelerate funding round; non-quantitative |

### Direct-ship dollar context (OMX-specific anchoring)

Per memory `reference_omx_difot_canonical_definitions`, Direct ship = Z-warehouse, +7wd target. The 9-ticket pattern (Memory: `project_omx_ds_supplier_visibility`) traces to ~$X disputed/write-off per ticket; if 9 tickets reach the CS sweep monthly, annualised pattern = 108 tickets at average resolution cost $500 (CS time + write-off + retention loss) = $54k addressable just from the visible-tickets layer. Hidden-ticket-pattern (the silent abandonment) likely 3-5x the visible.

---

## Vectors + visuals

### Lucide icon picks

| Slide / layout | Icons |
|---|---|
| **Cover / "End the blind spot"** | `i-search` (visibility, lead), `i-truck` (delivery), `i-zap` (real-time) |
| **Problem grid (4)** | No-stock-feed = `i-bar-chart`; No-despatch-event = `i-truck`; No-POD = `i-file-text`; Pre-printed-labels = `i-shuffle` (privacy risk) |
| **Two-option visual** | Option A (Feed) = `i-bar-chart` + `i-repeat` (daily loop); Option B (Sticker) = `i-zap` (QR) + `i-smartphone` (scan) |
| **Customer journey before/after** | Today: `i-coffee` (waiting) → `i-mic` (rings CS) → `i-coffee` (waiting); After: `i-zap` (scan) → `i-smartphone` (push) → `i-truck` (delivered) |
| **QR sticker flow diagram** | Supplier scan = `i-zap`; Carrier scan = `i-truck`; Customer scan = `i-smartphone`; POD = `i-file-text`; OMX dashboard = `i-bar-chart` |
| **Architecture pipeline** | Supplier = `i-laptop`; Feed/scan = `i-zap`; AWS edge = `i-sparkles`; Snowflake = `i-bar-chart`; Customer push = `i-smartphone` |
| **Garage demo storyboard** | `i-door` (enter), `i-zap` (scan), `i-smartphone` (see result), `i-sparkles` (wow), `i-brain` (ask Claude), `i-file-text` (POD) |
| **The ask** | `i-zap` (build), `i-life-buoy` (CS impact), `i-bar-chart` (ROI) |

### Image concepts (6)

1. **Cover hero** — Photo: pair of hands sticking a QR sticker on a brown carton, mid-action, warm warehouse lighting. Source: Unsplash "warehouse sticker QR" or commission. NZ context: NZ-coded uniform, OMX brand sticker.
2. **The visibility gap visual** — Black-and-grey panel: timeline of an order with three blank gaps labelled "no feed / no despatch / no POD". Build in Figma; high-contrast.
3. **Option A vs Option B side-by-side** — Two-column infographic: left = supplier laptop running daily feed; right = warehouse worker scanning a sticker. Photo-mockup hybrid.
4. **Garage demo storyboard** — 6-panel cartoon-style sequence: visitor enters → picks parcel → scans QR → sees tracking on phone → asks Claude "where's the order?" → wins. Build in Figma; reference frontend-design skill.
5. **POD photo example** — Real photo of carton on delivery dock with timestamp + driver name overlay. Mockup the overlay in Figma over Unsplash "delivery dock NZ" photo.
6. **Architecture diagram** — Custom SVG line-art: supplier → (feed or scan) → AWS edge (CloudFront + HMAC) → Snowflake → customer/CS dashboard. Reference `lens/design-language/Chapter-09-architecture`.

Follow `_visual-curation.md`: warehouse/dock photos are NZ-coded (greys + native bush in background, not US-style); QR mockups use the Matariki Stars dynamic-QR styling so the shared-stack story shows visually.