Shopify + ERP: scaling fashion growth with cost control and margin clarity

Executive summary
Download the short version of this case study, with the key metrics, client outcomes, and roadmap. 

⬇ Download executive summary PDF

At a glance

• Industry: Mid-sized fashion retailer (France)
• Channels: D2C (Shopify Plus), legacy B2C (WooCommerce), Wholesale (Zalando, La Redoute)
• Employees: ~60 (incl. 5-person IT team)
• Annual turnover: ~€35M

The client had burned budget in the past with a “big consultancy” that promised full ERP digital transformation, but delivered over-engineered systems they couldn’t maintain. They came to us cautious: “Prove you can fix my problems without locking me into costs I can’t control.”

How the client found us

The client discovered GetDevDone through a referral from a partner agency that had worked with us on complex MarTech and eCommerce integrations. They were looking for a partner they could trust to deliver results reliably, after experiencing fragmented delivery and hidden costs with previous vendors.

Initial Challenges (Before engagement)

1. Overselling & stock conflicts
   • Shopify + WooCommerce stock not syncing correctly.
   • Backorders ~6–8% of total orders → refunds, unhappy customers.
2. ERP disconnect
   • Odoo ERP used for finance + warehouse.
   • No direct integration → manual CSV imports.
   • Warehouse often packed products that were already “sold out” online.
3. Cost fear
   • Prior consultancy had pitched “middleware + SaaS stack” costing €8–10K/month in licenses.
   • Client lost trust in “expensive software subscriptions.”
4. Margin blindness
   • Leadership couldn’t see margin differences between wholesale (Zalando) and D2C (Shopify).
   • Promotions sometimes ran at a loss without visibility.

Our Approach: Transparency First

We acknowledged these client fears directly and structured delivery to minimize risk:

• Pilot-first delivery → 6-week, fixed-price scope. No hidden costs.
• Tool independence → prefer open-source or client-owned cloud (avoid SaaS lock-in).
• Step-by-step roadmap → prove ROI at each stage, only then scale.
• Knowledge transfer → goal was NOT dependency on us, but empowerment of client’s IT.

Phased Solution Delivery

Phase 1: Pilot (6 weeks, fixed cost)

• Built custom Shopify ↔ Odoo connector in Python/Node.js.
• Covered top 200 SKUs only (high turnover).
• Added basic reconciliation dashboard in Metabase (open-source BI).
• Hosted on client’s own AWS → no SaaS fees.

Value created (pilot):

• Overselling dropped 70% in top SKUs.
• Warehouse no longer double-packed.
• Leadership approved scaling based on hard ROI.

Phase 2: Scale-up (3 months)

• Extended integration to all SKUs + WooCommerce + Zalando + La Redoute.
• Added Python forecasting model (Prophet) running on AWS EC2.
  • Forecast horizon: 8 weeks.
  • Accuracy: ~82% vs historical baseline.
• Built margin visibility dashboards in Metabase:
  • Split by channel (Shopify vs Zalando vs wholesale).
  • Promotion profitability tracking.

Value created 

• Overselling across all SKUs reduced from ~7% → 1.5%.
• Saved ~€4K/month by avoiding SaaS subscriptions.
• Management now makes pricing decisions based on actual margin data.

Phase 3: Governance & Compliance 

• Added GDPR audit trail (PostgreSQL + audit logs).
• Role-based access for dashboards (finance vs operations vs exec).
• PCI DSS v4.0 alignment for payment data handling and encryption.
• WCAG 2.1 accessibility review on customer-facing storefronts.
• Training sessions for client’s IT:
  • 80% of operations now self-managed.
  • GetDevDone provides “escalation support” only.

Value created 

• Client passed GDPR audit in 2024 without external consultants.
• IT team independent for daily operations, vendor dependency reduced.

Tech Stack (Transparent Choices)

Area	Choice	Why not SaaS?	Cost impact
Commerce	Shopify Plus, WooCommerce	N/A	Existing licenses
ERP	Odoo ERP	Already in use	N/A
Integration	Custom Python/Node.js connectors	Avoid Linnworks (~€1.5K/mo)	One-time build, client-owned
Forecasting	Prophet (Python, AWS EC2)	Avoid AWS Forecast (higher cost, black-box)	€150/mo infra
Dashboards	Metabase (open-source)	Avoid Snowflake + Tableau (~€3K/mo)	Free (self-hosted)
Compliance	PostgreSQL audit logging, PCI DSS v4.0 evidence packs, WCAG 2.1 accessibility checks	Simple, transparent	Minimal dev time

Total SaaS savings: ~€4,000/month vs big-agency approach.

Key lessons for our next Clients

• Agency rates ≠ agency bloat: Yes, our engineering rates are not “freelancer cheap.” But we deliver ROI-focused pilots first, so you see value before scaling spend.
• We don’t oversell: Case proves we can start with 200 SKUs → expand to full inventory, only after ROI is visible.
• You won’t be trapped: Everything built runs on your AWS, your ERP, your dashboards. No hidden SaaS subscriptions.
• Shared ownership: 80% of operations now handled by client’s IT — dependency risk eliminated.

Business impact

• Overselling: ↓ from 7% → 1.5% (quarterly average)
• SaaS savings: €4K/month avoided
• Forecast accuracy: ~82% vs baseline
• Exec margin visibility: 100% across channels
• IT independence: 80% ops handled internally

Hidden aspects we’re open about

• Forecast model is good, not perfect — ~82% accuracy is solid, but demand shocks (e.g., influencer campaign) still require human judgment.
• Metabase dashboards lack polish compared to Tableau, but the client accepted “80% functionality at 0% license cost.”
• Knowledge transfer takes effort — client’s IT had to invest ~60 hours in training.

Client testimonial (unedited)

We were burned by a big consultancy before, so we demanded a pilot-first, no hidden costs. GetDevDone delivered exactly that. We saved thousands in SaaS fees and got real margin visibility. Today, my IT team runs 80% of the system. That independence is priceless. — COO, French Fashion Retailer

Future growth roadmap

• Expansion into digital product passports (EU requirement for apparel in 2026).
• Preparing to connect marketplaces beyond Europe (Farfetch, ASOS).