How can a non-profit retailer optimise their location planning?

How location planning and analysis of open data allowed Oxfam to optimise its retail strategy and maximise the revenue of its stores

By Oxfam, an international confederation of 18 organisations working in approximately 94 countries worldwide, to find solutions to poverty and injustice around the world.


What location factors determine the performance of a store? How can that performance be maximised?

Oxfam GB, the British branch of the non-profit organisation, which operates a network of 600 second-hand high-street stores to raise revenue for its charitable work, undertook this location planning project in 2014. Their aim was to design a model to optimise the location of their stores in order to maximise their revenue. They were looking for new techniques of location planning that would be cheap, durable, and easy to use in the future, in order to ensure improved efficiency in the long run.


Qualitative and quantitative techniques with open data

An important challenge facing the researchers was the budget constraints; Oxfam, as a nonprofit organization, couldn’t necessarily afford costly formal location planning techniques. Instead, they chose to base their analysis on so-called “open data”, or data that is free to use and access. The starting point was surveys and focus groups in order to identify clients’ concerns regarding store location and to inform the design of the model. The dataset collected was put through the statistical technique of multiple regressions, which involved estimating correlation between each of these independent variables and revenue and store performance, to identify the influence of each of these factors on sales and revenue.

So what?

Financial gains and location planning

The resulting model ended up explaining 55% of variance in shop revenue, so the technique proved relatively reliable. The research afforded Oxfam very concrete financial gains: beyond the £40,000 pounds that was saved by using open data instead of more formal location planning methods, it was estimated that the dataset helped identify £193,000 in prospective improvements to the existing network. Beyond location planning, the model ended up being used to inform many more strategies within Oxfam’s retail enterprises, such as the segmentation of stores and tailoring of prices to different areas, distribution of stocks, prioritisation of investment, implementation of knowledge-sharing mechanisms between managers of over- or underachieving stores, planning of acquisition and competitive strategy. The monetary gains from stock cascades and segmentation alone was evaluated at around £6 million.