← Discovery / c_seed_acme

Acme Logistics

Logistics & Supply Chain · budget $50,000 · 8-10 weeks

Qualified
Created 7/29/2026, 6:58:49 AM
Complexity
MEDIUM
Problems
5
identified
Data sources
3
classified
Agent runs
7
on this case
Discovery Analyst

Structured analysis

MEDIUM

Acme Logistics operates in EU logistics and is looking to automate shipment & invoice workflows. The intake suggests a mid-complexity engagement with clear scope and a defined compliance boundary (GDPR + EU residency).

Objectives
  • Reduce manual paperwork by 70%
  • Improve visibility into operational bottlenecks
  • Establish foundation for future automations (driver scheduling, customer notifications)
Problems to solve (5)
PTS-1 · Operations team
Manual data entry between email, HubSpot and Google Sheets
15 hours/week lost to copy-paste
PTS-2 · Sales lead
No real-time visibility into shipment status
Slow response time, leads go cold
PTS-3 · Compliance officer
Audit trail is fragmented across spreadsheets and chat
Quarterly audits take 2 weeks to assemble
PTS-4 · Customer
Slow response on routine "where is my shipment" requests
NPS drops by ~10 points on weekends
PTS-5 · Finance
Reconciliation between Stripe payouts and issued invoices is manual
Books close 5 days late every month
Open questions (3)
HIGH Is there an existing data warehouse or are we creating the first structured store?
HIGH Who is the technical stakeholder on the client side for integration access?
MEDIUM Are there any industry-specific compliance requirements (e.g. AEO customs) beyond GDPR?
Solution Architect

Target architecture

Components
C-1 Intake form (SvelteKit) frontend
C-2 Workflow runtime (n8n) external-saas
C-3 Agent runtime sveltekit-server
C-4 Postgres (Supabase EU) infrastructure
C-5 HubSpot sync external-saas
C-6 Stripe integration external-saas
Workflows (to-be)
WF-1 hybrid
n8n for integrations + custom code for AI glue (n8n-rules.md#rule-5). Bulk integrations (HubSpot, Stripe, courier APIs) favor n8n; LLM parsing benefits from custom code with structured schemas.
Key decisions
Managed Postgres in EU (Supabase EU region)
Fastest setup, pgvector available, EU residency, predictable cost
→ database-rules.md#rule-1, rule-2
n8n cloud as primary orchestrator
Bulk integrations (5+ external systems) + non-technical ops team can own + modify workflows
→ n8n-rules.md#rule-1, rule-2
Agent Layer abstraction with provider registry
Provider-agnostic, swappable per agent, easy to switch to EU-region models
→ ADR-001
Stripe webhooks for reconciliation trigger
Near real-time, lower Stripe API cost, fewer rate-limit issues
→ integration-catalog.md#stripe
Vendor lock-in: MEDIUM

Workflows in n8n format (portable to Make/custom code with effort). AI providers abstracted and swappable. Storage uses standard SQL.

Cost Estimator

Investment

Person-days
38
30–50 days
Monthly
$680
operational
Support
$900
per month

Agent runs

Every agent invocation on this case, in order.

  • discovery-analyst
    mock/mock-1 · 1240→820 tok · 152ms
    Done
  • workflow-analyst
    mock/mock-1 · 1410→1150 tok · 184ms
    Done
  • security-analyst
    mock/mock-1 · 1120→940 tok · 198ms
    Done
  • solution-architect
    mock/mock-1 · 1540→1320 tok · 224ms
    Done
  • cost-estimator
    mock/mock-1 · 980→720 tok · 138ms
    Done
  • proposal-writer
    mock/mock-1 · 1180→1640 tok · 246ms
    Done
  • report-generator
    mock/mock-1 · 860→2120 tok · 312ms
    Done