01 — Business Snapshot
Your business,
as we documented it.
Everything in this assessment was drawn directly from what you shared during the Blueprint session. This section reflects your operation — its strengths, its friction points, and the areas we've identified as meaningful leverage opportunities.
Business Profile
SunWest is a B2B e-commerce company specializing in construction safety equipment, with over 20 years of operating history. You sell your own branded products alongside third-party manufacturers — including 3M — operating simultaneously as a wholesaler and a custom order fulfiller. Your client base includes major national organizations such as Amtrak, and you operate out of California's Silicon Valley area.
Annual Revenue
~$1M
Established 20+ year operation
Profit Margin
~15%
Constrained by operational overhead
Team
1 + 5
Owner + five VAs, local & overseas
Orders That Are Custom
~80%
Handled manually through email
Active Supplier Relationships
~50
Of 100+ available to the business
Annual Tooling Spend
$24K
Zapier, Make.com, automation specialist
Core Friction Points
Six operational pain points emerged during the Blueprint session. These are not isolated problems — they compound. Each one limits capacity, and together they impose a ceiling on what the business can reach.
01
Inventory data doesn't reflect reality — fulfillment failures are the downstream consequence.
Discontinued products and supplier price changes — including those driven by tariff shifts — are not reflected in your Shopify records in real time. The gap between what the system shows and what suppliers can actually deliver surfaces as a customer-facing problem before it surfaces internally. By the time an issue is caught, an order has already failed.
02
Half your supplier network is effectively inaccessible — and that's a margin problem on every order.
You have established relationships with 100+ suppliers, but only approximately 50 are actively monitored and compared. The remaining 50+ are not being checked because the manual bandwidth required to do so doesn't exist. Every sourcing decision made without a full market comparison is a pricing decision made with incomplete information. The cost is real — it's just not tracked.
03
Manual order processing is the structural bottleneck — and it scales linearly with headcount.
Your stated goal is a 75% reduction in order processing time. That reduction is not achievable through incremental improvement of the current workflow. The current approach applies human judgment to every step regardless of whether the step requires it. Order capacity is a direct function of VA count, which means growth requires proportional hiring — indefinitely.
04
80–90% of your revenue flows through email with no central tracking and no visibility.
Custom orders — the dominant share of your business — are managed end-to-end through email threads. Specs to manufacturers, prototype reviews, customer approvals, production confirmations — all scattered across inboxes with no unified status view. As volume increases, so does the probability that something gets missed. At the current scale, it's manageable. At the next scale, it isn't.
05
Daily portal checks are a fixed daily cost with no leverage ceiling.
Your VAs manually log into hundreds of customer procurement portals each day to check for new orders. This is a task with no variability, no judgment component, and an output that scales directly with client roster size. As you add clients, the daily check time compounds — with no mechanism to contain it short of additional headcount.
06
$24,000 per year is being spent on tooling that hasn't changed the underlying dynamic.
The automation specialist and the Zapier/Make stack represent a legitimate investment toward the right goal. The constraint is architectural, not effort-based. The tools have a ceiling on what they can do, and that ceiling is below what your operation requires. More configuration of the same stack produces a more refined version of the same limitation.
Areas Identified for Consideration
Beyond the friction points, the Blueprint session surfaced three structural opportunities that sit above the operational layer — areas where the right infrastructure doesn't just fix problems, it creates new capability that didn't previously exist.
Supplier Network Leverage
The full 100+ supplier relationship set is already in place. The opportunity isn't to acquire new suppliers — it's to operationalize the ones you have. Real-time comparison across all 100+ on every order changes your margin profile on the sourcing side without any change to your pricing strategy.
Order Capacity as a Growth Lever
If order processing time drops by 75% and VA bandwidth is freed from repeatable tasks, your current team can handle significantly more order volume. Capacity becomes infrastructure-constrained rather than headcount-constrained — which means growth no longer requires proportional hiring.
Custom Order Visibility as a Quality Signal
Centralizing the custom order lifecycle — specs through production — creates traceability and accountability that currently doesn't exist. For enterprise clients like Amtrak, that visibility is a service differentiator. For your team, it's a risk reduction on the revenue that matters most.
02 — Current Operations
How your operation
runs today.
Seven workflows were mapped during the Blueprint Assessment. Below you'll find the end-to-end order flow — as it operates now, and as it would operate with the proposed infrastructure in place. Toggle between the two states. The data sources behind each step are identified explicitly.
End-to-End Order Workflow
Manual portal checks + RFQ email alerts
VAs manually log into hundreds of customer procurement portals daily looking for new orders. The open.gov subscription sends automated email alerts when a matching RFQ is identified, but all other channels require active manual monitoring. No orders are captured between check cycles.
Data Sources
Customer procurement portals (manual login, daily)
open.gov subscription alerts (email notification)
● Manual — significant daily VA overhead
⚠ Orders missed between check cycles
⚠ Cost scales with client roster size
VA manually reads, reviews, and categorizes each order
Orders arrive through two channels: direct email for small and custom orders, and procurement portal notifications for larger orders ($10K+). A VA manually reads each incoming order, determines its type, and categorizes it before any downstream action occurs. There is no automated classification layer.
Data Sources
Email inbox (unstructured, manual review)
Customer procurement portals (notification-based)
● Manual — required for every order
⚠ VA bottleneck on every intake
VA logs into Shopify — data may not reflect supplier reality
For each order, a VA manually logs into Shopify and checks product availability. Because Shopify is not synchronized against supplier portals in real time, the records may be stale — reflecting stock levels, pricing, or product availability that no longer matches what suppliers can actually deliver. Fulfillment failures trace directly to this step.
Data Sources
Shopify inventory (manually updated, potentially stale)
Physical warehouse stock
● Manual login required per order
● Inventory data drifts from reality
⚠ Fulfillment failures originate here
VA contacts suppliers one at a time — ~50 of 100+ are reached
When a product is not in stock, a VA manually contacts suppliers individually to find the best available price and lead time. Given the bandwidth required, only approximately 50 of the 100+ available supplier relationships are realistically checked. Every sourcing decision is made against an incomplete view of the market. The cost of the unchecked suppliers is real — it simply isn't tracked.
Data Sources
Supplier portals (manual access, ~50 of 100+ checked)
Direct supplier communication (ad hoc outreach)
● Manual — hours of outreach per order
● 50+ suppliers never compared
⚠ Best pricing routinely missed
VA builds each quote manually and sends via email or portal
Once sourcing is resolved, a VA manually constructs the quotation and delivers it to the customer via email or their procurement portal. Invoice follows. Quote turnaround time is measured in hours — in some cases, days. In competitive procurement environments, response speed is itself a selection criterion. The current process cannot compete on that dimension.
Data Sources
Supplier pricing (as manually researched)
Customer contact details
Email / procurement portal (manual send)
● Built manually per order
⚠ Hours-to-days turnaround
⚠ Slow response loses bids
Specs → manufacturer → prototype → customer approval — all via email
Custom orders are handled through a sequence of emails: specs to manufacturer, prototype returned and forwarded to customer, customer approval collected, production commenced. There is no centralized tracking. Status lives across inboxes. The risk of a missed communication, a delayed approval, or a lost thread is directly proportional to volume — and the dominant share of your revenue operates with no system of record.
Data Sources
Email inbox (unstructured)
Manufacturer communications (ad hoc)
Customer approval threads (scattered across inboxes)
● No central system of record
● Entirely email-dependent
⚠ Lost threads = lost revenue
⚠ 80–90% of revenue at risk
Manual coordination from warehouse to shipment — no automated follow-up
Orders are prepped from warehouse stock or sourced externally. Custom branding or print work is completed if applicable. The order ships. There is no automated customer notification, no delivery status communication, and no post-delivery follow-up sequence to create conditions for repeat business.
Data Sources
Warehouse inventory (physical)
Shipping providers (manual coordination)
Customer contact (manual email only)
● Manual end-to-end coordination
⚠ No automated customer updates
⚠ No post-delivery follow-up
What This Costs — Quantified
Supplier Margin Lost
~50%
of your supplier network is never compared on any given order. The pricing delta is untracked but present on every transaction.
Annual Tool Spend, No Ownership
$24,000
Zapier, Make.com, and the automation specialist — recurring annually. No infrastructure built, no equity accrued at year end.
Order Capacity Ceiling
VA-bound
Every additional order requires proportional VA time. Growth means hiring. There is no leverage in the current model.
AI agents monitor all channels continuously — no daily checks required
AI agents monitor all customer procurement portals continuously. RFQ notifications, email-based orders, and portal alerts are captured, classified, and routed automatically the moment they appear. The daily manual check cycle is eliminated entirely. No orders are missed between check windows — because there are no check windows.
Data Sources
All customer procurement portals (AI agents, 24/7)
Email inbox (automated capture)
open.gov RFQ alerts (integrated)
✓ Automated — 24/7 continuous
✓ Zero orders missed
AI Browsing + Agent Intelligence
Every inbound order read, classified, and queued automatically
All inbound orders — regardless of channel — are processed through a document intelligence layer that reads the order, extracts the relevant details, classifies the type, and places it into the appropriate queue in the Intelligence Portal. No VA review is required for standard intake. The queue is ready for matching and action.
Data Sources
Email
Customer portals
Shopify website orders
RFQ services — all unified into one queue
✓ Instant automated classification
✓ No VA intake review required
Document Intelligence Engine
Live database queried in real time — Shopify always reflects actual availability
The live supplier database — updated continuously from all 100+ supplier portals — is queried the moment an order enters the system. Shopify is kept in sync via API, making it an accurate reflection of the live database rather than a separate record that drifts. Stale inventory data is structurally eliminated.
Data Sources
Live Supplier Database (100+ portals, continuously updated) → Shopify API sync (real-time push)
✓ Real-time — always current
✓ No stale data, no drift
Live DB + Shopify API Sync
All 100+ suppliers compared simultaneously — best option surfaced for your confirmation
The Smart Order Matching Engine compares every available supplier against the incoming order — across all 100+ relationships, simultaneously — evaluating price, availability, lead time, and margin. The highest-margin viable option is surfaced to you for one-click confirmation. No supplier is skipped. No comparison is made on incomplete information.
Data Sources
Live Supplier Database (full 100+, real-time pricing & availability)
Parsed order specs from intake module
✓ All 100+ suppliers compared
✓ Best margin surfaced automatically
Inference + Matching Engine
Quote auto-generated and delivered within minutes of your confirmation
Once you confirm the supplier selection, the system generates the quotation automatically — formatted, priced, and delivered to the customer via email or their portal with a payment link and delivery estimate included. Turnaround goes from hours to minutes. No VA drafting required. No delay between decision and delivery.
Data Sources
Confirmed supplier pricing
Order specs from intake
Customer contact and portal delivery details
✓ Auto-generated on confirmation
✓ Minutes, not hours
AI Assistant Intelligence
Entire custom order lifecycle managed centrally — specs through production
Custom order communications are managed entirely within the Intelligence Portal. Specs are transmitted directly to the manufacturer. Prototype receipt is logged. Customer approval is routed and tracked. Production status is visible in real time. Nothing lives in an inbox. Nothing can be missed or lost between threads. The 80–90% of revenue that flows through custom orders is under full visibility and control.
Data Sources
Custom Order module
Manufacturer comms (portal-based)
Customer approval records
Production status tracking
✓ Fully centralized
✓ Zero email dependency
Custom Order Management Module
Automated confirmations, delivery updates, and follow-up sequences
Fulfillment triggers automated customer communication: order confirmation on dispatch, delivery status updates in transit, and a post-delivery follow-up sequence designed to generate repeat orders. The customer relationship extends beyond shipment without any incremental manual effort.
Data Sources
Fulfillment system
Shipping provider data
Customer contact records
Post-delivery sequencing engine
✓ Full automated comms layer
✓ Post-delivery follow-up included
Customer Communication Intelligence
03 — Stack Assessment & Alternatives
Current stack,
honest assessment.
What's in place today, what each tool is actually capable of at an architectural level, and how the available alternatives compare.
Current Stack
Workflow Automation Zapier & Make.com Works for linear triggers — not for reasoning ▼
Event-driven integration platforms. Zapier connects 6,000+ apps via pre-built connectors; Make offers a visual scenario builder with more complex conditional branching. Both execute deterministic, pre-defined workflows when a trigger fires.
What it does well
Reliable point-to-point data movement between supported apps. Well-suited for notification routing, CRM updates, form submissions, and linear multi-step automations where the logic path is fixed and the data is structured. Mature error logging and retry handling. Low engineering overhead to maintain.
Structured data routing
App-to-app triggers
Linear workflow execution
Architectural boundary
Neither platform has a reasoning layer. All logic is defined ahead of time as conditional branches — they cannot dynamically evaluate supplier options, parse unstructured email content into structured order data, or make multi-variable decisions. Web scraping of supplier portals at scale is outside their design intent; both rely on official API connectors or webhooks. Stateful memory across sessions does not exist natively. Complex branching workflows hit execution limits and error-handling complexity quickly.
No LLM reasoning
No autonomous browsing
No stateful memory
No unstructured data parsing
Ongoing Shopify Inventory Restructuring Right direction — wrong anchor point ▼
Active effort to normalize and clean the Shopify product and inventory database as a foundation for downstream automation. Structurally sound as a prerequisite step — the limitation is Shopify's data model, not the cleanup effort.
What it does well
Shopify's Admin API provides read/write access to products, inventory levels, variants, and locations. A well-normalized product catalog makes programmatic inventory updates possible and reduces the surface area for sync errors. This groundwork is not wasted — it's a prerequisite for Shopify API integration regardless of what sits upstream.
Clean data reduces sync errors
Shopify API integration prereq
Architectural boundary
Shopify's inventory system was designed as a retail storefront — its data model assumes inventory is managed from within, not continuously updated by an external source of truth. It has no native mechanism to ingest real-time supplier pricing feeds, flag discontinued products from portal data, or manage a relational supplier database. Using Shopify as the canonical data store for a 100+ supplier procurement operation requires fighting its architecture at every integration point.
No external real-time data ingestion
Not a supplier database
Storefront data model, not procurement
Active open.gov & RFQ Subscription Services Solves discovery — nothing downstream ▼
Subscription services that monitor federal and state procurement portals and emit email alerts when a matching RFQ is published. Approximately $1,000/year. The discovery layer works as intended.
What it does well
Automated monitoring of government procurement portals with keyword and category matching is genuinely useful and would otherwise require manual daily checks. The alert delivery is reliable. For the specific problem of RFQ discovery, this is an appropriate point solution.
Automated RFQ discovery
Appropriate cost for scope
Architectural boundary
Scope is limited to alert delivery. The system outputs an email notification — everything after that (intake, sourcing, quoting, fulfillment) remains manual. It does not integrate with your order queue, does not trigger downstream workflow, and does not replace monitoring of non-government procurement portals. One input channel addressed; the processing bottleneck unchanged.
No downstream integration
Email output only
Human Automation Specialist via WeShop.co — $2,000/month Execution is sound — tooling ceiling is the constraint ▼
A contractor engaged since January to build automations using Zapier and Make.com, sourced through the WeShop.co freelance platform. The practitioner's implementation is technically correct for the platforms in use.
What it does well
Skilled configuration of Zapier and Make within their design constraints. On-demand access to technical talent through WeShop is a reasonable model for incremental automation work — low commitment overhead, flexible scope. Appropriate for extending what the current stack can do.
Sound implementation
Flexible engagement model
Architectural boundary
The constraint is not the practitioner — it is the ceiling of Zapier and Make as described above. More hours of configuration within these platforms produces a more refined version of the same architecture. The specialist cannot build what the platforms don't support: autonomous browsing, stateful supplier databases, or LLM-mediated matching logic. The investment is valid; the return is bounded by the tools.
Ceiling = platform capabilities
Cannot extend beyond Zapier/Make limits
Alternatives Landscape
At a glance — capability comparison
| Approach | Autonomous browsing | Unstructured data parsing | Multi-supplier matching | Stateful memory / DB | Custom order mgmt | Ownership |
|---|---|---|---|---|---|---|
| Zapier / Make | ✕ | ✕ | ✕ | ✕ | ✕ | Subscription |
| B2B ERP (Cin7 etc.) | ✕ | Partial | Partial | ✓ | Limited | Subscription |
| ChatGPT / Claude / Gemini | Plugin only | ✓ | Manual prompt | ✕ Stateless | ✕ | Subscription |
| Claude Cowork | ✓ | ✓ | With configuration | Limited persistence | General-purpose | Subscription |
| OpenClaw / Agent platforms | ✓ | ✓ | With custom build | Depends on config | Buildable | Partial |
| In-house developer | ✓ Buildable | ✓ Buildable | ✓ Buildable | ✓ Buildable | ✓ Buildable | Full |
| Youtiva | ✓ | ✓ | ✓ | ✓ | ✓ | Full — owned |
B2B ERP Cin7 / QuickBooks Commerce / Brightpearl Handles structured inventory — not heterogeneous portal extraction ▼
Purpose-built B2B inventory and order management platforms with supplier relationship modules, purchase order workflows, and multi-channel sales support. Legitimate products with real enterprise adoption.
What it does well
Structured supplier catalogs with EDI or API-based data feeds, PO generation and tracking, multi-warehouse inventory management, and B2B order workflows. If your suppliers provided standardized data feeds (EDI 846/832 or REST APIs), these platforms would solve the inventory sync problem competently. Reporting and audit trails are strong. Implementation cost is lower than custom builds.
Structured supplier data ingestion
PO and fulfillment workflows
Multi-channel inventory
Why it doesn't fit here
Your 100+ supplier portals do not offer EDI or standardized API access — they are heterogeneous web interfaces requiring autonomous browsing to extract data. ERPs assume structured supplier data flows; they have no mechanism for web-based portal extraction. Custom order management with prototype loops, manufacturer communication, and approval tracking is outside their standard feature set. AI-mediated order matching is not a native capability in any of these platforms.
Assumes EDI/API supplier access
No autonomous web extraction
No custom order lifecycle mgmt
LLM Platform ChatGPT / Claude / Gemini (direct use) Strong reasoning — no persistence, no orchestration ▼
General-purpose large language models accessed via chat interface or API. Genuinely capable of reading and reasoning over unstructured content — emails, documents, order specs. Widely used for task assistance and workflow augmentation.
What it does well
Parsing unstructured email content into structured fields, drafting quotations from loosely specified inputs, summarizing supplier communications, classifying order types from freeform text. The inference quality on procurement-adjacent tasks is high. Used via API, these models can be embedded as reasoning components inside larger systems.
Unstructured text parsing
Document reasoning
Strong API integration potential
Architectural boundary
These are stateless inference endpoints — they have no persistent memory between sessions, no ability to query or maintain a live supplier database, no autonomous browsing capability without explicit tool integrations, and no built-in workflow orchestration. Using them to build what your operation requires means constructing the full surrounding system: database layer, agent orchestration, browser automation, API integrations, and custom UI. That system is the build — the LLM is one component inside it.
Stateless — no session memory
Not a system — an inference endpoint
No native orchestration
AI Agent Platform Claude Cowork (Anthropic) Capable desktop agent — general-purpose, not domain-specific ▼
Anthropic's desktop automation product. Claude Cowork allows Claude to operate a computer — opening applications, navigating browsers, reading screens, and executing multi-step tasks via natural language instruction. Targets non-developer operators who want to automate file and task management without writing code.
What it does well
Computer-use capability — Claude can navigate a browser, log into a supplier portal, read the page, and extract data without requiring an API. This is architecturally relevant to your portal extraction problem. For one-off or low-frequency tasks, this works. The underlying model's reasoning quality on procurement documents and unstructured email is strong. No code required for basic automation.
Autonomous browser navigation
Screen reading and data extraction
No-code task automation
Architectural boundary
Cowork is a general-purpose desktop agent — it is not configurable into a persistent, domain-specific system. It does not maintain a live supplier database between sessions, does not run scheduled background jobs against 100+ portals, and does not orchestrate multi-step procurement workflows autonomously without human initiation. Each task requires a human to invoke it. There is no persistent state layer, no custom order management module, and no Shopify API sync capability built in. It is a powerful tool for assisted automation — not an autonomous operational system.
No persistent background jobs
Human-initiated only
No stateful supplier database
General-purpose — not domain-configured
AI Agent Platform OpenClaw / Agent Orchestration Frameworks Closest technical category — still requires a full custom build ▼
OpenClaw and similar agent orchestration platforms (LangChain, CrewAI, AutoGen, n8n with AI nodes) provide frameworks for building multi-agent systems — coordinating LLM calls, tool use, browser automation, and memory across complex workflows. These are developer-facing frameworks, not end-user products.
What it does well
Agent orchestration frameworks give developers the building blocks to construct exactly the kind of system your operation requires — multi-agent pipelines, tool-using agents with browser access, vector memory stores, and structured output parsers. The architectural primitives are there: an agent can be configured to log into a supplier portal, extract pricing data, compare across sources, and write results to a database. This is the correct technical category for your use case.
Agent orchestration primitives
Browser automation support
LLM + tool use pipelines
Configurable memory layers
What this actually means in practice
These are frameworks — they provide the scaffolding, not the system. Building what your operation requires on top of OpenClaw or LangChain still demands: a software engineer to architect and implement the agent pipelines, a database design for the supplier data layer, custom handling for each supplier portal's DOM structure and authentication, a Shopify API integration, a custom order management UI, and ongoing maintenance as supplier portals update. The framework reduces boilerplate; it does not replace the engineering effort. This path converges with the in-house developer option — it just starts with better scaffolding.
Framework only — full build still required
Requires engineering team
No out-of-box procurement logic
Human / Internal In-House Developer or Engineering Team Full capability — full cost and timeline ▼
Hiring a mid-to-senior software engineer or small team to build the required system from scratch or using agent frameworks. The only path other than Youtiva that can produce fully owned, fully custom infrastructure.
What it does well
Full ownership of the codebase, full flexibility to modify scope, and no external vendor dependency for ongoing changes. An experienced engineer with the right stack (Python, Playwright or Puppeteer for browser automation, PostgreSQL, Shopify API, LangChain or similar) can build every component described in this proposal. This is the correct path if you have an appetite for building and maintaining an engineering function.
Full ownership
Maximum flexibility
No vendor dependency
The cost and timeline reality
A mid-senior engineer in this stack costs $130–160K/year in salary alone, plus benefits and tooling. The Phase 1 equivalent — supplier data extraction pipeline, live database, Shopify sync, analytics layer — is a 4–6 month build for one engineer working on it as a primary project, assuming no ramp-up time on your business domain. You would also be acquiring an ongoing maintenance obligation for every supplier portal that updates its authentication or DOM structure, which they do routinely. The total two-year cost of this path exceeds the Youtiva project investment before accounting for the opportunity cost of the build period.
$130–160K+/year salary cost
4–6 month build before Phase 1 equivalent
Ongoing maintenance obligation
04 — Proposed Solution
The SunWest
Intelligence Portal.
A private AI operating system built around your business, your data, and your workflows. Not a subscription. Not a configured SaaS product. Infrastructure that you own — purpose-built to run SunWest's procurement and fulfillment operation.
System Architecture
The portal has two operational sides — a Supplier Side and a Customer Side — connected through an intelligence layer that handles the matching, reasoning, and decision support between them. All three components are part of one privately hosted system.
Supplier Side — Live Database
100+ supplier portals connected via AI browser agents
Real-time pricing per product across all relationships
Live stock levels & availability — updated per portal cadence
Discontinued product flags detected automatically
New product additions logged as they appear
Tariff and market changes tracked and reflected in pricing
→ Shopify synced via API. The storefront becomes an accurate output of this database.
Intelligence Layer
Smart Order Matching
Reads every order. Compares all 100+ suppliers simultaneously. Evaluates price, margin, availability, lead time. Surfaces best option for your one-click confirmation.
Automated Quotation Engine
Generates and delivers professional quotes on confirmation. Payment link, delivery estimate, order reference — sent automatically.
Communication Layer
Confirmations, delivery updates, payment follow-ups, post-delivery sequences — triggered by order lifecycle events.
Predictive Analytics
Demand forecasting, price shift prediction, supplier performance scoring, margin opportunity detection, and risk alerting.
Customer Side — Order Intelligence
Email orders — captured, read, classified automatically
Customer procurement portals — monitored 24/7 by AI agents
RFQ notifications from open.gov — integrated and routed
Shopify store orders — unified into the same queue
Custom order lifecycle — specs, manufacturer comms, prototype review, approval, production tracking — all centralized
→ All channels. One queue. Nothing routed manually.
What Gets Replaced
Eliminated
Manual VA portal monitoring — daily check cycle
Manual supplier price checks per order
Manual quotation and invoice assembly
Manual Shopify inventory updates
Email-based custom order tracking
Zapier / Make.com subscriptions ($24K/year)
What You Own Instead
Private SunWest Intelligence Portal — hosted, branded
Live database of all 100+ suppliers, updated continuously
AI order matching and margin optimization engine
Auto-generated quotations and invoices
Centralized custom order management system
Predictive analytics and supplier performance scoring
05 — Execution Roadmap
Two phases.
One sequenced build.
Your entire operation depends on supplier data. The customer-facing order intelligence system built on top of an incomplete supplier foundation would route orders into an unreliable base. Phase 1 is therefore completed in full and validated before Phase 2 begins.
Full Timeline Overview
Phase 1 — Supplier Intelligence Foundation
Phase 1
| Stage | Scope & Deliverable | Timeline |
|---|---|---|
Stage 1 Portal Development |
The SunWest Intelligence Portal is built as the private, branded environment that everything else operates within. Database architecture is established to receive and store supplier data. User access and admin controls are configured for your team. Private cloud hosting environment is stood up and secured with zero external data exposure. |
Days 1–45 |
Stage 2 Supplier Data Extraction & Live DB |
All 100+ supplier portals connected using your provided credentials — scaled from a single-supplier pilot through full deployment. Real-time data extracted from each portal: pricing, availability, stock levels, new products, discontinued items. Live supplier database built and populated. Refresh cycles established per portal capability. Discontinued products and pricing changes flagged automatically. |
Days 46–90 |
Stage 3 Shopify API Sync |
API connection built between the Intelligence Portal and Shopify. Live supplier database pushes real-time stock levels and pricing to Shopify continuously. Shopify remains the customer-facing storefront — the portal becomes the source of truth it reflects. Sync accuracy tested and validated across all product lines. |
Days 91–105 |
Stage 4 Supplier Intelligence & Analytics |
Intelligence layer added on top of clean, real-time data. Includes: opportunity analysis across supplier margins, demand forecasting from historical order patterns, price prediction from market trends and tariff data, automatic margin-optimized supplier surfacing per order, supplier performance scoring over time, inventory opportunity detection, and risk alerts for volatility or instability. |
Days 106–120 |
Phase 1 Completion Gate: Phase 2 does not begin until Phase 1 is fully delivered and validated. At Day 120, you have a live supplier database, a functioning Shopify sync, and a working analytics layer — in your hands, operational, and demonstrably performing. The Phase 2 decision is made at that point, informed by working infrastructure rather than a projection.
Phase 2 — Customer Order Intelligence
Phase 2
| Stage | Scope & Deliverable | Timeline |
|---|---|---|
Stage 1 Order Intake & Centralization |
All inbound order channels connected to the portal: email inbox for automatic capture and classification, customer procurement portals for automated monitoring, RFQ notification services, and Shopify website orders. All orders flow into a single unified queue — one place, all channels, nothing routed manually. |
Days 121–165 |
Stage 2 Smart Order Matching Engine |
Matching engine built and connected to the live supplier database from Phase 1. For each incoming order: system reads and parses specs, compares against all available suppliers simultaneously, surfaces viable options with price, availability, and delivery time side by side, highlights highest-margin option. You review and confirm with one click. |
Days 166–210 |
Stage 3 Automated Quotation & Invoicing |
On your confirmation, the system generates a professional quotation automatically. Quotation and invoice are sent to the customer via email or their portal — including payment link, delivery estimate, and order reference. No manual drafting. Turnaround measured in minutes. |
Days 211–220 |
Stage 4 Custom Order Management |
Full custom order loop brought into the portal. Specs transmitted to manufacturer directly from the system. Manufacturer prototypes received and logged. Prototypes routed to customer for approval with one action. Production status tracked from approval through fulfillment. 80–90% of your revenue moves out of email dependency. |
Days 221–230 |
Stage 5 Customer Communication Intelligence |
Full automated communication layer deployed: order confirmation emails sent on approval, delivery status updates in transit, payment follow-up sequences for outstanding invoices, and post-delivery follow-ups designed to generate repeat order opportunities. The customer relationship is maintained automatically throughout and after the order lifecycle. |
Days 231–240 |
Total Project
Total Duration
240 days
~11 months across two milestone-gated phases
Total Investment
$115,000
$50K Phase 1 · $65K Phase 2
What You Own
Permanently
No recurring platform fees. No subscription dependency.
06 — Return on Investment
Your numbers,
projected forward.
All figures are derived from data you provided during the Blueprint Assessment. The projections apply conservative operational improvement assumptions at each phase. Adjust any variable below — the calculator uses your actual baseline.
Before & After — Full Project
Your Current State
Annual Revenue
~$1,000,000
Profit Margin
~15%
Annual Profit
~$150,000
Suppliers Monitored
~50 of 100+
Order Processing
Hours per order
Fulfillment Failures
Regular occurrences
Annual Tool Cost
$24,000
→
→
→
→
→
→
→
After Full Deployment
Annual Revenue
~$2,000,000+
Profit Margin
~21–25%
Annual Profit
~$420,000–500,000
Suppliers Monitored
100+ in real time
Order Processing
Minutes per order
Fulfillment Failures
Near zero
Annual Tool Cost
Eliminated
Adjust the Variables
Pre-loaded with your actual figures. Adjust any variable to see how the output changes.
Current Annual Revenue$1,000,000
Current Profit Margin15%
Revenue Multiplier After Deployment2.0×
Margin Improvement (percentage points)+8pp
$420K
Est. Annual Profit
3.7×
ROI on Investment
2.7 yrs
Payback Period
Projections are estimates based on figures you provided during the Blueprint Assessment. Actual results depend on implementation quality, market conditions, and team adoption. ROI reflects profit gain over current baseline relative to total project investment.
The Investment in Context
Current Tool Spend / Year
$24,000
Recurring annually. Zapier, Make.com, and the automation specialist. No infrastructure accrued at year end.
Portal Investment — One Time
$115,000
Milestone-gated. No recurring platform fee. No subscription dependency. The infrastructure is yours permanently.
5-Year Comparison
$5K less
$120K in tool spend over five years yields no ownership. $115K invested once yields permanent, owned infrastructure.
Over five years, you will spend $120,000 on the current tool stack and own nothing at the end of it. The portal costs $5,000 less over that same window — and is owned outright from day one of deployment.
Current path — $120,000 over 5 years, zero ownership
Intelligence Portal — $115,000 once, owned permanently
The Cost of Staying Put
The $115,000 has a visible price tag. What the current path costs doesn't arrive as an invoice — it arrives as margin that never materializes, bids that go to faster competitors, and customers who don't return.
Foregone Orders
Every order above VA bandwidth that doesn't get processed in time. At 15% margin, a missed $10K order is $1,500 in gross profit — not deferred, gone.
Untapped Supplier Margin
~50 supplier relationships that are never compared on any order. If 10 of them carry 3–5% better margin on common products, the annual delta is real and entirely untracked.
CC Refund Processing Fees
Stale inventory triggers failed fulfillments. Refunds carry non-recoverable processor fees of 0.5–1% per transaction. Chargebacks add $15–100 per dispute. A direct tax on data drift.
Lost Bids to Faster Quotes
In enterprise procurement, a quote that arrives in 20 minutes beats one in two days regardless of price. Hours-to-days turnaround is a structural competitive disadvantage.
Negative Reviews
Fulfillment failures from stale inventory data reach customers before they're caught internally. Enterprise vendors are evaluated on reliability. One failure is recoverable. A pattern isn't.
Repeat Orders Left Uncaptured
The workflow ends at shipment. No automated follow-up, no reorder prompt. Repeat orders from existing enterprise accounts cost a fraction of new acquisition — and are left to chance.
Every one of these compounds with volume. The current architecture's failure modes don't smooth out at scale — they amplify.
07 — Deal Terms & Next Steps
How this works.
What you're agreeing to.
The full project is $115,000 across two phases. The decision today is Phase 1 — $50,000, 120 days, backed by a full money-back guarantee. Everything below is the structure of the engagement.
The 120-Day Money-Back Guarantee
120-Day Money-Back Guarantee
Phase 1 is delivered in 120 days — or you get your money back. Full stop.
The Phase 1 engagement carries a 120-day delivery commitment backed by a full refund guarantee. At Day 120, you will have a live supplier database connected to all 100+ of your supplier portals, a functioning Shopify API sync, and an operational predictive analytics layer — all validated and in your hands. If Phase 1 is not delivered as described within 120 days, you receive a full refund of the Phase 1 investment. No partial credit, no scope renegotiation, no extended timeline offered as a substitute. The guarantee exists because the milestone is achievable and we are prepared to stand behind it financially. You are not being asked to take a position on trust alone.
Engagement Structure
Phase 1 — The Foundation
$50,000
120 days · full money-back guarantee
✓ Private Intelligence Portal built
✓ 100+ suppliers connected, live
✓ Shopify API sync validated
✓ Predictive analytics operational
→
Phase 2 — Order Intelligence
$65,000
120 days · begins after Phase 1 gate
→ All order channels unified
→ Smart Order Matching Engine
→ Auto quotation & invoicing
→ Custom order management
→ Communication intelligence
Terms & Commercial Details
Ownership
You own the infrastructure built under this engagement outright. The code, the database, the portal, the integrations — all transferred to SunWest. No ongoing licensing fee. No subscription required to continue using what was built. This is not a SaaS product with a vendor dependency. It is custom software, and it is yours.
Hosting & Data Privacy
The Intelligence Portal is deployed on a private cloud environment. Zero data is shared externally by design. Supplier pricing data, customer order records, and business financials remain within your infrastructure. Data sensitivity is classified as Medium-High and handled accordingly throughout the build and deployment process.
Payment Structure
Phase 1 ($50,000) is invoiced at engagement commencement. Phase 2 ($65,000) is invoiced only after Phase 1 is delivered and validated at the 120-day gate. You are not billed for Phase 2 before Phase 1 is in your hands and operational. Payment terms and milestone schedule are formalized in the engagement agreement.
Credentials & Access
Phase 1 requires access to your 100+ supplier portal credentials to enable the extraction and database build. These are held securely and used solely for the purposes described in this assessment. All access protocols are documented and agreed upon at kickoff. You retain full control of all credentials throughout the engagement.
How We Proceed
01
Agreement executed on Phase 1. Formal proposal issued and signed. Phase 1 investment of $50,000 confirmed. Engagement agreement documents scope, timeline, deliverables, and Phase 2 gate criteria.
02
Day 1 kickoff. Portal development begins. Your team access is established. Supplier credential handoff is completed. Kick-off documentation confirms the build sequence and communication cadence.
03
Day 120 delivery and validation. Live supplier database, Shopify API sync, and predictive analytics are delivered and validated in your hands. Phase 1 deliverables reviewed against the agreed criteria. Gate signed off.
04
Phase 2 decision. With working Phase 1 infrastructure as the reference point, the Phase 2 engagement is initiated. Timeline, staging, and scope confirmed. Phase 2 invoice issued at commencement.
SunWest Intelligence Portal · Blueprint Assessment · March 2026 · Confidential