Sainsbury’s: Property Analytics

Property Analytics

The University has a long history of collaboration with the Property Analytics function at Sainsbury’s.  Dr Andy Newing explains how this programme of research has developed over the past ten years:

“The Location Planning and Property Analytics teams support new store development and assessment of the store network from a spatial perspective, answering questions such as:

  • What is the trading potential for this site?
  • What is the right store format for this location?
  • Where are customers most likely to shift to online purchasing?
  • How are our competitors performing in this location?

Research collaborations spanning the past decade have supported several studentships for PhD study, enabling researchers to work closely with Sainsbury’s colleagues to undertake cutting-edge research to answer some of these questions.

These collaborations draw on the University’s expertise in retail modelling and have facilitated the development of bespoke spatial modelling tools to support store location planning. These collaborations include considerable data-sharing and opportunities for researchers to work alongside members of the Sainsbury’s team to understand the business context.

The formalised working practices and data-sharing infrastructure provided by LIDA has enabled a new phase of collaboration focusing on more computationally intensive model building. These collaborations capture evolving consumer behaviours and reflect the ever-changing retail supply side and the shift to e-commerce within this sector.

There remains considerable potential for ongoing collaboration within this strand of work with mutual benefits in terms of operational improvements for the Property Analytics theme and genuine academic impact.”

 

Tim Rains, Senior Spatial Data Scientist at Sainsbury’s commented: “Our collaborations with LIDA over the past decade have enabled us to grow our understanding in several important strands of the geography of retail including fluctuating consumer demand and e-retailing. But perhaps more importantly, the partnership has greatly helped our own team’s development, who get an insight into novel techniques and research, and that of the PhD researchers, who benefit from working with real-world data and problems.”

 

Project Examples

 

Understanding e-commerce channel use in the grocery market

The grocery sector is at the forefront of applied spatial modelling capturing traditionally predictable and habitual consumer behaviours and interactions such as the consumer traveling to a store.

The nature of these interactions between demand and supply are changing following the introduction of home delivery/click-and-collect facilities. Increased grocery sector E-commerce operations have resulted in significant supply side investment, however location planning modelling tools have not kept pace with the changing nature of consumer behaviour and retail operations.

This project takes the approach of updating traditional spatial interaction models to capture these behaviours and relationships and accurately predict multi-channel consumer behaviours within a spatial modelling framework.

“Working with the team at Sainsbury’s has allowed me to really focus on the applied element of my research. Invaluable insights and knowledge into how e-commerce services are operated, monitored and evaluated by a national brand have greatly supported my understanding of the subject area. Along with access to commercial data to underpin my research the partnership fostered discussion into the issues they are looking to tackle. This has fundamentally shaped the focus of the project in line with industry needs.”

Ryan Urquhart, PhD Student, ESRC Data Analytics and Society Centre for Doctoral Training

 

Improving Understanding of e-commerce channel use in the grocery market – updating traditional spatial interaction models to capture changing behaviours and relationships and to accurately predict multi-channel consumer behaviours within a spatial modelling framework.

An investigation into the geography of corporate e-commerce sales in the UK grocery market
Elena Kirby-Hawkins, Mark Birkin, Graham Clarke

 

Increasing the accuracy of the revenue forecasts by accounting for Temporal Demand Variations in Retail Location Models.

Accounting for Temporal Demand Variations in Retail Location Models
Tom Waddington, Graham Clarke, Martin C. Clarke, Nick Hood, Andy Newing

Open all hours: spatiotemporal fluctuations in U.K. grocery store sales and catchment area demand
Thomas B. P. Waddington, Graham P. Clarke, Martin Clarke & Andy Newing

 

Identifying seasonal variations in store‐level visitor grocery demand – using spatial analysis to identify revenue originating from outside the store catchment, and explore the spatial and temporal nature of the visitor demand recorded in‐store.

Applied spatial modelling for retail planning in tourist resorts
Andy Newing, Graham Clarke, Martin Clarke

Exploring Small Area Demand for Grocery Retailers in Tourist Areas
Andy Newing, Graham Clarke, Martin Clarke.

 

Evaluating the potential of agent-based modelling to capture consumer grocery retail store choice behaviours – enabling the complexity of consumer behaviours to be captured and simulated within a novel modelling framework.

Evaluating the potential of agent-based modelling to capture consumer grocery retail store choice behaviours
Charlotte Sturley, Andy Newing, Alison Heppenstall