Store Layout Optimization at Scale
EdgeRed built a data-driven space optimisation model on Google Cloud Platform for a national retailer — analysing product placement, store layout and inventory patterns across a large store network to deliver measurable improvement in sales performance and a significant reduction in stock loss.
The Challenge
A major Australian retailer was building a next-generation AI solution to dynamically recommend optimal department and category space allocation across their store network. Each store carries unique physical characteristics, and recommendations needed to account for a defined set of business constraints and commercial objectives.
The challenge wasn’t just modelling complexity — it was delivering production-ready data assets, machine learning models and a working front-end UI that could operate reliably at scale, week in week out, without manual intervention.
What We Built
EdgeRed delivered data engineering, ML modelling and front-end development across three parallel workstreams — building the foundations that power the retailer’s AI-driven space optimisation platform.
-
Data foundations
Production-ready datasets built in Google BigQuery — cleansed, structured and purpose-built to power the AI recommendation engine with reliable, fit-for-purpose inputs -
ML modelling with commercial guardrails
Machine learning models overlaid with business constraints and commercial objectives — so recommendations reflect real-world trade-offs, not just statistical optima -
Pipeline integrity
Data ELT pipelines and lineage built in Dataform with unit tests and git-committed code — ensuring every data asset is traceable, testable and maintainable at scale -
End-to-end solution delivery
Front-end UI, back-end APIs and metadata logging built in parallel — capturing user interactions and model outputs to support ongoing diagnostics and improvement
Augmented Services
EdgeRed supplied a dedicated onshore team to extend the client’s internal data capability – without the overhead of a full-time hire. The team brought a blend of seniority and experience – 2 ML / Data Engineers, 1 Data Analyst, 1 Data Scientist and 0.5 Analytics Lead.
What that means in practice:
- Senior capability on demand - right skills, right time, no long-term headcount commitment
- 100% onshore delivery - every person on your project is based in Australia
- Flexible resourcing - scale the team up or down as project needs change
- Deep bench - draw on the collective knowledge of a 50+ person consultancy, not just the people in the room
Technology Used
- Google BigQuery — cloud data warehouse; production-ready datasets built to power the AI recommendation engine
- Google Cloud Platform (GCP) — core cloud infrastructure
- Dataform — ELT pipelines and data lineage with unit tests and git-committed code
- Python — ML modelling and back-end API development
The Outcome
EdgeRed delivered data foundations, modelling capability and a working front-end to power space allocation recommendations across a large national store network. The retailer shifted from manual, inconsistent planning to automated, constraint-aware recommendations at store level — with a scalable AI platform grounded in governed, reliable data. The solution now runs automatically week in week out, generating optimised store layouts that have delivered measurable improvement in sales performance and a significant reduction in stock loss.
